key: cord-0828542-et0wy22z
authors: Singh, Raman Deep; Barry, Michael A.; Croatt, Anthony J.; Ackerman, Allan W.; Grande, Joseph P.; Diaz, Rosa M.; Vile, Richard G.; Agarwal, Anupam; Nath, Karl A.
title: The spike protein of SARS-CoV-2 virus induces heme oxygenase-1: Pathophysiologic implications
date: 2021-12-14
journal: Biochim Biophys Acta Mol Basis Dis
DOI: 10.1016/j.bbadis.2021.166322
sha: 76e9d626be13dde0d30914e26df43e23a9876941
doc_id: 828542
cord_uid: et0wy22z

BACKGROUND: Acute kidney injury (AKI) is both a consequence and determinant of outcomes in COVID-19. The kidney is one of the major organs infected by the causative virus, SARS-CoV-2. Viral entry into cells requires the viral spike protein, and both the virus and its spike protein appear in the urine of COVID-19 patients with AKI. We examined the effects of transfecting the viral spike protein of SARS-CoV-2 in kidney cell lines. METHODS: HEK293, HEK293-ACE2(+) (stably overexpressing ACE2), and Vero E6 cells having endogenous ACE2 were transfected with SARS-CoV-2 spike or control plasmid. Assessment of gene and protein expression, and syncytia formation was performed, and the effects of quercetin on syncytia formation examined. FINDINGS: Spike transfection in HEK293-ACE2(+) cells caused syncytia formation, cellular sloughing, and focal denudation of the cell monolayer; transfection in Vero E6 cells also caused syncytia formation. Spike expression upregulated potentially nephrotoxic genes (TNF-α, MCP-1, and ICAM1). Spike upregulated the cytoprotective gene HO-1 and relevant signaling pathways (p-Akt, p-STAT3, and p-p38). Quercetin, an HO-1 inducer, reduced syncytia formation and spike protein expression. INTERPRETATION: The major conclusions of the study are: 1) Spike protein expression in kidney cells provides a relevant model for the study of maladaptive and adaptive responses germane to AKI in COVID-19; 2) such spike protein expression upregulates HO-1; and 3) quercetin, an HO-1 inducer, may provide a clinically relevant/feasible protective strategy in AKI occurring in the setting of COVID-19. FUNDING: R01-DK11916 (KAN), R01-AI100911 (JPG), P30-DK079337; R01-DK059600 (AA).

The human host infected with the virus SARS-CoV-2 displays a heterogeneity of responses that range from an asymptomatic state to fulminant organ dysfunction and ensuing mortality (1) (2) (3) . Such organ dysfunction reflects viral infection and as well as regional and systemic inflammatory responses, all of which may involve the lungs, kidney, and other major organs and tissues (1) (2) (3) .

SARS-CoV-2 invades cells through a process that is dependent upon the engagement of the spike S glycoprotein of the virus with the ACE2 receptor abundantly present on host cells, including those in the kidney (1) (2) (3) (4) . Following such engagement, the following steps sequentially occur: the spike protein is proteolytically cleaved and primed by TMPRSS2 (and other proteases) also present on host cells; ACE2 receptor-mediated, viral transmission via the plasma membrane into the intracellular compartment occurs; viral replication ensues; and offspring virions are released to neighboring cells (1) (2) (3) (4) (5) .

In addition to enabling viral cell entry through the ACE2 receptor, and as increasingly recognized, the spike protein may exert diverse effects, including interaction with innate immune receptors (6) , activation of the inflammasome (7) and the alternative complement system (8) , endothelial injury and increased endothelial permeability (9) , endothelial dysfunction attended by oxidative stress and mitochondrial injury (10) , procoagulant processes including the upregulation of PAI-1 (11) , and exhaustion of NK cells (12) .

As one of the main targets in COVID-19, the kidney, when acutely injured, significantly contributes to COVID-19-associated morbidity and mortality (1, 2) . Acute kidney injury (AKI) uncovered a strategy that may be considered as an approach in protecting against AKI in COVID-19.

Cell culture and chemicals. HEK293-IIIA native (HEK293) cells and HEK293-IIIA-ACE2 (HEK293-ACE2 + ) stably overexpressing ACE2 cells were generated and provided by Dr.

Michael A. Barry's laboratory. Vero E6 cells were provided by Dr. Richard Vile's laboratory.

These cell lines were maintained in DMEM containing 10% FBS with the selection antibiotic G418 (0.5 mg/ml) added to maintain HEK293-ACE2 + cells. Pharmaceutical grade quercetin (PHR1488, Sigma Aldrich) and Zn (II) protoporphyrin (Zn625-9, Frontier Specialty Chemicals) were used in this study. All other chemicals were from Sigma Aldrich.

Plasmids and Transfections. Plasmids used in this study were provided by Dr. Michael A.

Barry. Briefly, a codon optimized cDNA encoding the original wild-type spike protein from severe acute respiratory syndrome coronavirus 2 (pcDNA1-SARS-CoV-2 spike; 7788bp) isolate 2019-nCoV_HKU-SZ-002a_2020, accession number MN938384.1 and empty vector (pcDNA1; 4033bp) plasmids were synthesized by Genewiz. HEK293 cells, HEK293-ACE2 + , and Vero E6 cells were transfected using Lipofectamine LTX and Plus reagent kit as per the manufacturer's instructions. Briefly, cells were plated and grown to ~60-75% confluence before transfection.

DNA complexes were prepared in Opti-MEM reduced serum media (Cat # 31982-070, ThermoFisher Scientific) and added to cells pre-incubated with Opti-MEM reduced serum media. Full media was added at 4-6 hours post transfection and cells were collected at the indicated times (8-48 hours) . For studies examining the effect of quercetin, full media containing either quercetin (10 µM) or vehicle was added at 4-6 hours post transfection. In a subsequent experiment, Zn (II) protoporphyrin (10 µM) was added 30 minutes prior to quercetin treatment.

J o u r n a l P r e -p r o o f Journal Pre-proof Western analysis. Assessment of protein expression by Western blot analysis was performed as we have described previously (19, 20) . Primary antibodies employed in overnight 4°C The Student's t-test was used for parametric data and the Mann-Whitney U-test was employed for nonparametric data.

For these studies, we utilized HEK293, HEK293-ACE2 + cells, and Vero E6 cells that express endogenous ACE2 protein and binds to SARS-CoV-2 spike protein (5, 21) (see methods). These cells were transfected with SARS-CoV-2 spike and empty vector plasmid using Lipofectamine LTX and Plus reagent kit. Spike protein and ACE-2 protein expression was assessed by Western blot analysis. Two major bands were observed with the 180 kDa band corresponding to full length SARS-CoV-2 spike protein and the 90 kDa band representing the cleaved SARS-CoV-2 spike protein (5) . Transfection of all the cell lines with SARS-CoV-2 spike plasmid resulted in expression of spike protein (Fig 1A, B , and C). We confirmed robust expression of ACE2 in HEK293-ACE2 + cells whereas no such expression occurred in HEK293 cells. In addition, Vero E6 cells also showed endogenous ACE-2 expression ( Fig 1C) .

Interestingly, following spike protein expression in both HEK293-ACE2 + and Vero E6 cells, expression of ACE2 noticeably decreased (Fig 1B and C) . 

Spike protein expression in these cells led to upregulation of certain genes that are known to contribute to the pathogenesis of AKI. TNF-α, MCP-1, and ICAM1 mRNA expression was assessed by quantitative RT-PCR in HEK293-ACE2 + cells upon spike and EV transfections at 8 and 24 hours post transfection. TNF-α mRNA levels were significantly increased both at 8 and 24 hours post transfection (Fig 3A) , whereas MCP-1 mRNA expression was significantly higher at 8 hours, and ICAM1 mRNA expression at 24 hours after transfection with spike protein (Fig 3 B and C, respectively). In additional studies, we assessed the nephrotoxic gene expression in HEK293 cells ( Supplementary Fig 2A) and HEK293-ACE2 + cells ( Supplementary   Fig 2B) with 24-hour spike or EV transfection. We again observed significantly higher TNF-α and ICAM1 mRNA expression upon spike transfection in HEK293-ACE2 + cells, while no significant differences were observed in HEK293 cells.

We considered the possibility that cells expressing ACE2 may shed extracellular vesicles containing spike, and that spike protein from such vesicles upon binding to cellular ACE2, becomes internalized and may thereby lead to nephrotoxic gene expression. We, therefore, collected media from HEK293 and HEK293-ACE2 + cells 24 hours post transfection with spike or EV plasmid and analyzed for the presence of spike protein. Notably, we observed the cleaved spike proteins (90kDa) in media from HEK293-ACE2 + cells indicating that spike J o u r n a l P r e -p r o o f Journal Pre-proof protein was processed and cleaved before secretion into the media. However, spike transfection in HEK293 cells did not result in secretion of spike protein (Supplementary Fig 3) .

HO-1 mRNA expression was significantly increased in HEK293-ACE2 + cells after spike transfection (as compared with EV transfection) at both 8 and 24 hours post transfection ( Fig   4A) . Notably, spike transfection induced significantly higher HO-1 protein expression in both HEK293-ACE2 + cells and Vero E6 cells as assessed by Western blot analysis (Fig 4 B) .

We then examined the activation of signaling molecules upstream of HO-1 which can promote HO-1 gene transcription. Following spike transfection, and as compared with EV transfection, expression levels of p-Akt, p-STAT3 and p-p38 were all increased as assessed by Western blot analysis at 24 hours post transfection (Fig 5A and B) .

Since HO-1 expression is induced at both the mRNA and protein levels upon S protein expression in HEK293-ACE2 + cells, we questioned whether inducing HO-1 in these cells would influence syncytia formation. To this end we employed quercetin, a recognized inducer of HO-1 that is widely used in preclinical studies in vitro and in vivo (22) . Cells were transfected with spike and empty vector plasmids and then exposed to quercetin or vehicle 4 hours post transfection. Such exposure to quercetin significantly decreased syncytia formation (see arrows indicating syncytium size) (Fig 6A) . To determine if we had attained efficient transfection within Quantitation of the size and area covered by syncytia revealed that both were significantly and markedly decreased by quercetin (Fig 6B and C) . Notably, we also observed significantly decreased spike protein expression (Fig 7A and B) , and as expected, HO-1 expression was prominently increased in quercetin-exposed cells (Fig 7A and C) .

We attempted to determine whether HO-1 mediated the anti-syncytial effects of quercetin by examining the effects of quercetin in the presence of the cpmpetitive inhibitor of HO activity, zinc protoporphyrin. However, zinc protoporphyrin exerted cytotoxic effects in these cells, even in cells transfected with the empty vector. Such toxic and confounding effects of zinc protoporphyrin also precluded determining whether HO-1, as induced in these cells when they express the spike protein (and in the absence of exposure to quercetin), represented an adaptive, cytoprotective response.

The present study demonstrates that expression of the spike protein in HEK293-ACE2 + cells leads to cell fusion, progressive syncytia formation, and the lifting off and sloughing of sheets of fused cells from the monolayer, the latter giving rise to focal areas of denudation. (27, 28) , and ICAM1 (29, 30) to AKI have all been confirmed by approaches employing either genetically deficient murine strains or by inhibitory antibodies, peptides, or relevant chemical inhibitors. We focused on these three genes because of their established role in nephrotoxicity in rodent models of AKI. ICAM1 may be of particular interest in view of its capacity to promote adhesion of cells, especially those in the kidney with AKI. We speculate that the upregulation of these genes in HEK293-ACE2 + cells expressing the spike protein reflects an injurious effect of the spike protein in these cells relevant to AKI. Studies that examine the pathophysiologic effects of expression of these specific genes are of interest but are beyond the scope of the present work. We also speculate that the failure of HEK293 cells to evince such gene expression (TNF-α and ICAM1) and syncytia formation when these cells are transfected with spike plasmid suggest a linkage between these two cellular responses.

In addition to potentially nephrotoxic genes, we also examined expression of a nephroprotective gene, HO-1. There is abundant literature that supports induction of HO-1 in models of ischemic and nephrotoxic AKI and that such induction confers protection against diverse forms of AKI (31, 32) . In the present studies we observed that HO-1 mRNA and HO-1 protein as well as upstream signaling species which elicit HO-1 expression -p-Akt, p-STAT3, and p-p38, depending upon the experimental context (31, 32) -were all induced. Thus, the HO- induction, as provided by the current study, is yet to be reported in any study to date. Moreover, such induction of HO-1 also robustly occurred in another cell type, Vero E6 cells (which also endogenously express ACE2 protein), when these cells are transfected with the spike plasmid.

There are multiple reasons for this interest in HO-1 in COVID-19 for at least four considerations (33) (34) (35) (36) (37) . First, it has been postulated that increased amounts of free heme exist in plasma or tissues in patients with COVID-19, a hypothesis that clearly merits examination.

There is at least one study demonstrating that heme levels are increased in COVID-19 patients with reduced oxygen saturation (38) . Increased amounts of free heme may originate from destabilized hemoglobin released from lysed RBCs (as occurs in hemolysis) or destabilized myoglobin released from necrotic muscle (as occurs in rhabdomyolysis), both types of cell lysis having been described in COVID-19. Additionally, cytochrome P450 proteins, which reside in microsomes of all cells, are inherently unstable heme proteins, and when cells are injured from whatever cause -be it hypoxia, inflammation, or toxin-induced -these heme proteins are prone to release their heme prosthetic group (37) . It is also suggested that heme levels may increase as a consequence of the postulated high affinity binding of SARS-CoV-2 to porphyrins, the latter being essential moieties in the synthesis of heme proteins (34) . Increased levels of heme, postulated to occur from these mechanisms, is relevant to tissue injury in COVID-19 because heme is a potent prooxidant, proapoptotic, and proinflammatory species. In light of these considerations, it is theorized that induction of HO-1 is beneficial in COVID-19 because, by virtue of its catabolism of heme, HO-1 would remove this potentially cytotoxic, proinflammatory species. Second, the products of HO-1, namely, bile pigments and carbon monoxide, are recognized cytoprotectants by virtue of their antioxidant, anti-inflammatory, anti-thrombotic, and anti-apoptotic action; in essence, HO-1 would replace a toxic species (heme) by cytoprotective ones (bile pigments and carbon monoxide). Third, the SARS-CoV-2 open reading frame 3 can bind to the HO-1 protein thereby vitiating its potential protective effects (33) . Fourth, for a J o u r n a l P r e -p r o o f Journal Pre-proof number of viruses other than SARS-CoV-2, HO-1 exerts anti-viral effects, an effect that, putatively, may extend to SARS-CoV-2 (33) (34) (35) (36) (37) .

In light of these considerations and our finding that HO-1 is induced in kidney cells expressing the spike protein, we sought to determine whether induction of HO-1 is functionally significant. We thus questioned whether a known and widely employed inducer of HO-1, quercetin (22) , would exert functional effects in these cells after transfection with the spike protein. We observed that such treatment with quercetin markedly reduced syncytia formation that otherwise occurred. Quercetin also reduced spike protein expression in these cells. We suggest that the reduction in spike protein expression with quercetin occurs because of the following considerations. Adhesion of the plasma membranes of spike protein-expressing cells to other cells that express spike protein, either in lesser amounts or not at all, may promote transcellular movement of the spike vector to these cells with lesser or no spike protein expression. These cells now with increased amounts of the spike vector will then augment their expression of spike protein. When syncytia formation is inhibited, as it occurs with quercetin, spike protein expression will accordingly and concomitantly decrease.

We confirmed, as expected, that quercetin markedly increased HO-1 expression.

However, we were unable to resolve whether HO-1 indeed mediated the anti-syncytial effects of quercetin because the inhibitor of HO activity, zinc protoporphyrin, exerted cytotoxic effects on Quercetin has numerous cellular effects besides the induction of HO-1. As for other naturally occurring compounds, especially those with anti-oxidant and anti-inflammatory effects, the suggestion has been made that quercetin may provide a therapeutic approach in COVID-19 (39) . Several studies have indicated that quercetin may be protective in models of AKI in settings not related to COVID-19 (40) . Based on such protective effects in AKI in these settings, it has been suggested that quercetin would be protective against AKI occurring specifically in COVID-19 (41) . Quite remarkably, the possibility that quercetin may be protective in AKI occurring in COVID-19 is supported by theoretic analyses based on network pharmacology and molecular docking studies (42) . Our present findings provide, to the best of our knowledge, the first experimental data in support of these speculations and analyses.

In summary, we introduce an in vitro model for the study of AKI in COVID-19 based on expressing the spike protein in kidney cells. This "reductionist" approach is supported by the burgeoning biology that attests to cellular effects of the spike protein per se relevant to the pathobiology of COVID-19; by the fact that SARS-CoV-2 virus infects the kidney in  and by the fact that the spike protein is present in the urine in patients with AKI. In the course of the present studies, we uncover the induction of the HO-1 system in spike protein-expressing ACE2 + kidney cells, and that quercetin, which induces HO-1 and other molecular species and pathways, can reduce syncytia formation caused by spike protein expression. 

Extrapulmonary manifestations of COVID-19

COVID-19-associated acute kidney injury: consensus report of the 25th Acute Disease Quality Initiative (ADQI) Workgroup

Severe Acute Respiratory Syndrome Coronavirus 2, COVID-19, and the Renin-Angiotensin System: Pressing Needs and Best Research Practices

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Characterization of spike glycoprotein of SARS-CoV-2 on virus entry and its immune cross-reactivity with SARS-CoV

SARS-CoV-2 Spike Protein Interacts with Multiple Innate Immune Receptors

SARS-CoV-2 Entry Receptor ACE2 Is Expressed on Very Small

Hematopoietic and Endothelial Cells and in Response to Virus Spike Protein Activates the Nlrp3 Inflammasome

Direct activation of the alternative complement pathway by SARS-CoV-2 spike proteins is blocked by factor D inhibition

The SARS-CoV-2 spike protein alters barrier function in 2D static and 3D microfluidic in-vitro models of the human blood-brain barrier

SARS-CoV-2 Spike Protein Impairs Endothelial Function via Downregulation of ACE 2

ZMPSTE24 Regulates SARS-CoV-2 Spike Protein-enhanced Expression of Endothelial Plasminogen Activator Inhibitor-1

SARS-CoV-2 Spike 1 Protein Controls Natural Killer Cell Activation via the HLA-E/NKG2A Pathway

SARS-CoV-2 renal tropism associates with acute kidney injury

Does SARS-CoV-2 Infect the Kidney?

Multiorgan and Renal Tropism of SARS-CoV-2

SARS-CoV-2 in the kidney: bystander or culprit?

Viral Load in Urine Sediment Correlates with Acute Kidney Injury and Poor COVID-19

Evidence for SARS-CoV-2 Spike Protein in the Urine of COVID-19 patients

Antithrombotic effects of heme-degrading and heme-binding proteins

Heme oxygenase-2 protects against ischemic acute kidney injury: influence of age and sex

SARS-CoV-2 envelope protein causes acute respiratory distress syndrome (ARDS)-like pathological damages and constitutes an antiviral target

Naturally Derived Heme-Oxygenase 1 Inducers and Their Therapeutic Application to Immune-Mediated Diseases

Association between Postmortem Kidney Biopsy Findings and Acute Kidney Injury from Patients with SARS-CoV-2 (COVID-19)

Syncytia formation by SARS-CoV-2-infected cells

The anti-HIV drug nelfinavir mesylate (Viracept) is a potent inhibitor of cell fusion caused by the SARSCoV-2 spike (S) glycoprotein warranting further evaluation as an antiviral against COVID-19 infections

The renal injury and inflammation caused by ischemia-reperfusion are reduced by genetic inhibition of TNF-αR1: a comparison with infliximab treatment

Gene therapy expressing amino-terminal truncated monocyte chemoattractant protein-1 prevents renal ischemia-reperfusion injury

CCR2 signaling contributes to ischemia-reperfusion injury in kidney

Antibody to intercellular adhesion molecule 1 protects the kidney against ischemic injury

Intercellular adhesion molecule-1-deficient mice are protected against ischemic renal injury

Adaptive responses to tissue injury: role of heme oxygenase-1

Heme oxygenase-1 and acute kidney injury

The HMOX1 Pathway as a Promising Target for the Treatment and Prevention of SARS-CoV-2 of 2019 (COVID-19)

Genetic Polymorphisms Complicate COVID-19 Therapy: Pivotal Role of HO-1 in Cytokine Storm

Heme oxygenase-1 (HO-1) cytoprotective pathway: A potential treatment strategy against coronavirus disease 2019 (COVID-19)-induced cytokine storm syndrome

Heme oxygenase-1 modulation: A potential therapeutic target for COVID-19 and associated complications

Targeting the Heme-Heme Oxygenase System to Prevent Severe Complications Following COVID-19 Infections

Desaturation and heme elevation during COVID-19 infection: A potential prognostic factor of heme oxygenase-1

Anti-inflammatory potential of Quercetin in COVID-19 treatment

Effect of bioflavonoids quercetin and curcumin on ischemic renal injury: a new class of renoprotective agents

Mechanistic Aspects and Therapeutic Potential of Quercetin against COVID-19-Associated Acute Kidney Injury

Quercetin as a potential treatment for COVID-19-induced acute kidney injury: Based on network pharmacology and molecular docking study

Conceptualisation, data curation, formal analysis, investigation, methodology, project administration, resources, software, validation, visualisation, writing -original draft, and writing -review & editing

Conceptualisation, Methodology, Resources, Validation, writingreview & editing

Croatt: Conceptualisation, data curation, methodology, resources, software, supervision, validation, visualisation, writing -original draft, and writing -review & editing. Allan W. Ackerman: Conceptualisation, data curation, formal analysis, methodology, software, validation, visualisation

Grande: Conceptualisation, Formal analysis, Validation, and writingreview & editing

Resources, writing -review & editing

Vile: Conceptualisation, Validation, and writing -review & editing

Anupam Agarwal: Conceptualisation, Validation, and writing -review & editing

Conceptualisation, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, validation, visualisation, writing -original draft, and writing -review & editing

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:J o u r n a l P r e -p r o o f