key: cord-0992893-5r0u2mmo
authors: Fadason, Tayaza; Gokuladhas, Sreemol; Golovina, Evgeniia; Ho, Daniel; Farrow, Sophie; Nyaga, Denis; Pan, Hong; Karnani, Neerja; Wong, Conroy; Cooper, Antony; Schierding, William; O’Sullivan, Justin M.
title: A transcription regulatory network within the ACE2 locus may promote a pro-viral environment for SARS-CoV-2 by modulating expression of host factors
date: 2020-04-24
journal: bioRxiv
DOI: 10.1101/2020.04.14.042002
sha: d3fdec7303d8bd1cb93fa825d8df9fbdee7f04ea
doc_id: 992893
cord_uid: 5r0u2mmo

Introduction A novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was recently identified as the pathogen responsible for the COVID-19 outbreak. SARS-CoV-2 triggers severe pneumonia, which leads to acute respiratory distress syndrome and death in severe cases. As reported, SARS-CoV-2 is 80% genetically identical to the 2003 SARS-CoV virus. Angiotensin-converting enzyme 2 (ACE2) has been identified as the main receptor for entry of both SARS-CoV and SARS-CoV-2 into human cells. ACE2 is normally expressed in cardiovascular and lung type II alveolar epithelial cells, where it positively modulates the RAS system that regulates blood flow, pressure, and fluid homeostasis. Thus, virus-induced reduction of ACE2 gene expression is considered to make a significant contribution to severe acute respiratory failure. Chromatin remodeling plays a significant role in the regulation of ACE2 gene expression and the activity of regulatory elements within the genome. Methods Here, we integrated data on physical chromatin interactions within the genome organization (captured by Hi-C) with tissue-specific gene expression data to identify spatial expression quantitative trait loci (eQTLs) and thus regulatory elements located within the ACE2 gene. Results We identified regulatory elements within ACE2 that control the expression of PIR, CA5B, and VPS13C in the lung. The gene products of these genes are involved in inflammatory responses, de novo pyrimidine and polyamine synthesis, and the endoplasmic reticulum, respectively. Conclusion Our study, although limited by the fact that the identification of the regulatory interactions is putative until proven by targeted experiments, supports the hypothesis that viral silencing of ACE2 alters the activity of gene regulatory regions and promotes an intra-cellular environment suitable for viral replication.

triggers severe pneumonia, which leads to acute respiratory distress syndrome and death in 23 severe cases. As reported, SARS-CoV-2 is 80% genetically identical to the 2003 SARS-CoV 24

virus. Angiotensin-converting enzyme 2 (ACE2) has been identified as the main receptor for 25 entry of both SARS-CoV and SARS-CoV-2 into human cells. ACE2 is normally expressed in 26 cardiovascular and lung type II alveolar epithelial cells, where it positively modulates the RAS 27 system that regulates blood flow, pressure, and fluid homeostasis. Thus, virus-induced 28 reduction of ACE2 gene expression is considered to make a significant contribution to severe 29 acute respiratory failure. Chromatin remodeling plays a significant role in the regulation of 30 ACE2 gene expression and the activity of regulatory elements within the genome. 31

Methods: Here, we integrated data on physical chromatin interactions within the genome 32 organization (captured by Hi-C) with tissue-specific gene expression data to identify spatial 33 expression quantitative trait loci (eQTLs) and thus regulatory elements located within the 34 ACE2 gene. 35

Results: We identified regulatory elements within ACE2 that control the expression of PIR, 36 CA5B, and VPS13C in the lung. The gene products of these genes are involved in inflammatory 37 responses, de novo pyrimidine and polyamine synthesis, and the endoplasmic reticulum, 38

respectively. 39

Conclusion: Our study, although limited by the fact that the identification of the regulatory 40

interactions is putative until proven by targeted experiments, supports the hypothesis that viral 41 silencing of ACE2 alters the activity of gene regulatory regions and promotes an intra-cellular

Within months of the first reports [1], the COVID-19 outbreak has become a pandemic infecting and killing thousands of people worldwide [2] . COVID-19 is an infectious disease 48 associated with acute respiratory distress syndrome (ARDS) that is caused by SARS-CoV-2, a 49

Betacoronavirus that is 80% identical to the SARS-CoV virus [3] . Betacoronaviruses, 50

including SARS-CoV, Murine Hepatic Virus (MHV), and SARS-CoV-2, utilize the ACE2 51 protein for cell entry [4, 5] . The Spike protein on SARS-CoV-2 has a 10 to 20 fold higher 52 affinity for the ACE2 protein than its SARS-CoV homologue [3, 6] . 53

The ACE2 protein is highly expressed in cardiovascular and lung type II alveolar epithelial 54 cells [3, 7, 8] , where ACE2 is a primary modulator of the renin-angiotensin (RAS) system that 55 regulates blood flow, pressure and fluid homeostasis [9] . The ACE2 protein and the products 56 of the reactions it catalyzes have also been implicated in immune responses and anti-57 inflammatory pathways [10] [11] [12] . 58 SARS-CoV infection reduces ACE2 gene expression and this is thought to contribute to severe 59 acute respiratory failure [4] by triggering an imbalance in the RAS system that causes a loss of 60 fluid homeostasis, induces inflammatory responses [10, 13, 14] , and results in severe acute 61 injury in heart and lung [3,15,16]. As mentioned above, both SARS-CoV and SARS-CoV-2 62 utilize the ACE2 protein for cell entry. Poor prognoses in elderly SARS-CoV-2 patients (≥65 63 years old) are frequently associated with a pre-existing reduction in ACE2 expression and 64 imbalance in ACE2-related host derived pathways [17, 18] . ACE2 is an X-linked gene whose 65 expression is regulated by chromatin structure. Brg1, a chromatin remodeler, and the FoxM1 66 transcription factor recognize the ACE2 promoter and reduce expression through a mechanism 67 respectively. VPS13C, encoding a factor required for late stage endosome maturation, is also 95 controlled by a putative enhancer located in intron 11 of BMX, adjacent to ACE2. We propose 96 that ACE2 repression by SARS-CoV-2 trips a chromatin-based switch that coordinates the 97 activity of these regulatory elements and thus the genes they control. Collectively, these 98 changes inadvertently lead to the development of a pro-viral replication environment. 99 100 101 102

We selected all common single nucleotide polymorphisms (SNPs) from dbSNP (build153) 105 with a minor allele frequency (MAF) > 1% that were located within chrX:15,519,996-106 15,643,106, which included the ACE2 gene and its flanking region (hereafter ACE2 locus). 107 SNP positions are as reported for the human genome build hg38 release 75 (GRCh38). 108

We used the CoDeS3D algorithm [23] to identify putative spatial regulatory interactions for all 110 SNPs at the ACE2 locus (S1 Fig). CoDeS3D integrates data on physical chromatin interactions 111 within the genome organization (captured by Hi-C) with tissue-specific gene expression data 112 to identify spatial expression quantitative trait loci (eQTLs). To get lung-specific spatial 113 connections, we identified SNP-gene pairs across lung-specific Hi-C libraries using published 114 data for IMR90, A549, and NCI-H460 cell lines and lung tissue (GEO accession numbers 115 GSE35156, GSE43070, GSE63525, GSE105600, GSE105725, GSE92819, GSE87112, S1 116 for their potential to act as spatial eQTLs. None of the common SNPs we tested affected ACE2 149 expression levels in lung tissue (S2 Table) . 150

The wider ACE2 locus (chrX: 15,519,996-15,643,106; GRCh38/hg38) sits within a 151 topologically associating domain (TAD) that is conserved across some tissues, e.g. IMR90 (Fig  152   1A ). Therefore, it was not surprising that we identified control elements within this ACE2 locus 153 ( Fig 1A) . The distribution of targets for the putative control elements we identified is consistent 154 with previous studies that show that while the majority of significant eQTLs fall within 100 kb 155 of the transcription start site of a gene, only 60% of all eQTLs are upstream of the gene they 156 regulate [32] . Notable amongst the elements we identified are long distance trans-regulatory 157 interactions involving: 1) rs1399200:VPS13C (chr15:61,852,389-62,060,473; encodes 158 vacuolar protein sorting-associated protein 13C); and 2) rs6632680:PHKA2 (chrX:18,892,300-159 18,984,598; encodes phosphorylase kinase regulatory subunit alpha 2) (S2 Table) . 160

We identified eighty genetic variants within the ACE2 locus as cis-acting spatial eQTLs that 161 physically modulate the expression of genes PIR (encodes Pirin), CA5BP1 (a pseudogene of 162 CA5B), and CA5B (encodes mitochondrial carbonic anhydrase) in lung tissues (S2 Table) . CA5B genes (Fig 1B) . 172

The common variants that we tested show an unusual ancestry associated pattern of minor 173 allele frequencies ( Fig 1B) . Specifically, the East Asian population (1K Genomes project) 174 displays little variation across the bulk of the variants we analyzed. This observation is 175 supported by measures of genetic diversity (FST) between the Indian, Chinese and Malay 176 populations within the Growing Up in Singapore Towards healthy Outcomes (GUSTO) cohort 177 (S3 Table) . However, this pattern breaks down at several positions across the ACE2 gene 178 (including rs4646142, rs2285666, and rs2106809, which show significant selection towards 179 the reference allele) in all of the tested populations, indicating potential selective pressure at 180 these loci (Fig 1B) . Notably, two of these variants alter potential transcription factor binding 181 sites (i.e. rs2285666 alters HNF1, and Ncx motifs, rs2106809 alters a CEBPB motif; S4 Table) . 182

All three variants (rs4646142, rs2285666, and rs2106809) have previously been associated 183 with allele, sex and ethnicity specific impacts on hypertension, blood pressure, hypertrophic 184 cardiomyopathy, type 2 diabetes, myocardial infarction (reviewed in [33] We identified transcription regulatory elements for CA5B and PIR that are active in lung tissue 190 and are located within the ACE2 gene. We also identified a transcription regulatory element 191 (located in the BMX gene, adjacent to ACE2) for the PIR and VPS13C genes. It is sterically 192 impossible for a single DNA sequence to simultaneously be transcribed and regulate another 193 gene through a physical connection. Therefore, we propose that SARS-CoV-2-induced 194 chromatin-dependent repression of ACE2 expression in lung enables the regulatory sites, 195

repressing PIR and activating CA5B, to exhibit increased functionality in infected cells (Fig 2) . 196

We hypothesize that this regulatory change extends to coordinate changes in the expression of 197 VPS13C and PHKA2 in ways that promote viral proliferation. This host regulatory network has 198 not evolved to benefit the virus but rather, these regulatory changes inadvertently produce an 199 environment advantageous for the virus. . Therefore, it is notable that repressor regulatory sequences for PIR sit within 221 the ACE2 gene (Fig 2A) . We postulate that the chromatin modifications that silence ACE2 222 expression upon early stage infection activate the PIR repressor ( Fig 2B) . This reduces 223 responsiveness of NF-κB, and thereby delays the expected and needed anti-viral response. the GUSTO study group for comments on the manuscript. This work has been released as a 296

pre-print. 297

The authors declare that the research was conducted in the absence of any commercial or 299 financial relationships that could be construed as a potential conflict of interest. 300 capable of downregulating the expression of PIR, which is involved in the NF-κB pathway. 533

Enhancer elements within ACE2 are poised to upregulate CA5B expression, which encodes an 534 enzyme important for pyrimidine synthesis. In addition to this, an enhancer region within the 535 BMX gene (still within the same TAD) contributes to VPS13C regulation. (B) We hypothesize 536

that upon viral infection, SARS-COV-2 represses ACE2 expression, which increases the 537 activity of the PIR repressor and CA5B enhancer. This results in a reduction in the production 538 of PIR -the redox switch necessary for NF-κB activation, while also increasing pyrimidine 539 synthesis, which is necessary for viral replication. 540 541 Supplementary Tables 542 S1 specific Hi-C libraries were interrogated to identify genes in fragments that spatially interact 554 (in cis-and trans-) with SNP-containing fragments. The identified spatial SNP-gene pairs were 555 further used to query GTEx lung tissue (dbGaP Accession phs000424.v8.p2, 556

UBERON:0008952). The Benjamini-Hochberg FDR control algorithm was applied to adjust 557 the p values of the resulting eQTL associations to identify only significant (FDR < 0.05) lung-558 specific SNP-gene spatial relationships in the ACE locus. 559 560 S2 Figure. The eQTL data used in this study was obtained from lung samples taken from 561 middle-aged individuals. To assess the correlation of genetic variation with the changes in 562 gene expression, the GTEx project (https://gtexportal.org/home/) collected and analysed lung 563 samples from donors who were densely genotyped. The age-distribution graph illustrates that 564 approximately 70% of the lung samples that were obtained were from donors aged between 50 565 and 60. 566

Individual Variation of the SARS-326

CoV2 Receptor ACE2 Gene Expression and Regulation. 2020. 327 2. WHO. Coronavirus disease 2019

The origin, 330 transmission and clinical therapies on coronavirus disease 2019 (COVID-19) outbreak 331 -an update on the status

A crucial role of angiotensin 333 converting enzyme 2 (ACE2) in SARS coronavirus-induced lung injury

336 SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is

Cryo-EM 339 structure of the 2019-nCoV spike in the prefusion conformation

Single-cell RNA expression profiling 342 of ACE2, the putative receptor of Wuhan 2019-nCov

CPS1 maintains pyrimidine pools 449 and DNA synthesis in KRAS/LKB1-mutant lung cancer cells

Pyrimidine 452 pathways in health and disease

455 Inhibition of Polyamine Biosynthesis Is a Broad-Spectrum Strategy against RNA 456

Solaimanzadeh I. Acetazolamide, Nifedipine and Phosphodiesterase Inhibitors: 458 Rationale for Their Utilization as Adjunctive Countermeasures in the

COVID-19)

Severe Acute Respiratory Syndrome 461 (SARS) Coronavirus-Induced Lung Epithelial Cytokines Exacerbate SARS 462 Pathogenesis by Modulating Intrinsic Functions of Monocyte-Derived Macrophages and 463 Dendritic Cells

The NF-κB-465 dependent and -independent transcriptome and chromatin landscapes of human 466 coronavirus 229E-infected cells

Pirin is an iron-dependent 469 redox regulator of NF-B

NF-κB signaling in inflammation

478 VPS13A and VPS13C are lipid transport proteins differentially localized at ER contact 479 sites

SARS-coronavirus replication is supported by a reticulovesicular 482 network of modified endoplasmic reticulum

Spatiotemporal Coupling of the Hepatitis C Virus Replication Cycle by Creating a Lipid

Replication Depends on Endosomal Cholesterol Homeostasis

Coronavirus infection, ER stress, apoptosis and innate immunity

Genetic profile of Egyptian hepatocellular-carcinoma associated with hepatitis C virus 495

Genotype 4 by 15 K cDNA microarray: Preliminary study

Members of the nuclear receptor 498 superfamily regulate transcription from the hepatitis B virus nucleocapsid promoter

Corticosteroid Resistance in Airway Disease

Selective 503 Binding of Steroid Hormone Receptors to Octamer Transcription Factors Determines 504

Transcriptional Synergism at the Mouse Mammary Tumor Virus Promoter

Direct triggering of the type I interferon system by virus infection: 507 activation of a transcription factor complex containing IRF-3 and CBP/p300

Targeted 510 exploration and analysis of large cross-platform human transcriptomic compendia