key: cord-0800900-f17pshwm
authors: Ordonez, Alvaro A.; Bullen, C. Korin; Villabona-Rueda, Andres F.; Thompson, Elizabeth A.; Turner, Mitchell L.; Davis, Stephanie L.; Komm, Oliver; Powell, Jonathan D.; D’Alessio, Franco R.; Yolken, Robert H.; Jain, Sanjay K.; Jones-Brando, Lorraine
title: Sulforaphane exhibits in vitro and in vivo antiviral activity against pandemic SARS-CoV-2 and seasonal HCoV-OC43 coronaviruses
date: 2021-03-25
journal: bioRxiv
DOI: 10.1101/2021.03.25.437060
sha: c81393b62f4b7a0e0fc1dad608044caf4823e298
doc_id: 800900
cord_uid: f17pshwm

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), the cause of coronavirus disease 2019 (COVID-19), has incited a global health crisis. Currently, there are no orally available medications for prophylaxis for those exposed to SARS-CoV-2 and limited therapeutic options for those who develop COVID-19. We evaluated the antiviral activity of sulforaphane (SFN), a naturally occurring, orally available, well-tolerated, nutritional supplement present in high concentrations in cruciferous vegetables with limited side effects. SFN inhibited in vitro replication of four strains of SARS-CoV-2 as well as that of the seasonal coronavirus HCoV-OC43. Further, SFN and remdesivir interacted synergistically to inhibit coronavirus infection in vitro. Prophylactic administration of SFN to K18-hACE2 mice prior to intranasal SARS-CoV-2 infection significantly decreased the viral load in the lungs and upper respiratory tract and reduced lung injury and pulmonary pathology compared to untreated infected mice. SFN treatment diminished immune cell activation in the lungs, including significantly lower recruitment of myeloid cells and a reduction in T cell activation and cytokine production. Our results suggest that SFN is a promising treatment for prevention of coronavirus infection or treatment of early disease.

The multi-functional phytochemical sulforaphane (SFN) is the isothiocyanate derived from enzymatic hydrolysis of its precursor glucoraphanin, a glucosinolate found in high concentrations in broccoli (Brassica oleracea italica). SFN is a potent naturally occurring activator of the transcription factor nuclear factor erythroid 2-related factor 2 (NRF2), with welldocumented antioxidant and anti-inflammatory effects 1,2 . Treatment with SFN increases phagocytic activity of alveolar macrophages 3 and reduces lung injury in animal models of acute respiratory distress syndrome (ARDS) 4 . SFN also lowers the levels of IL-6 and viral load in human subjects infected with live attenuated influenza virus 5, 6 . Numerous clinical trials utilizing SFN demonstrate favorable pharmacokinetics after oral dosing and document excellent tolerability and safety 1, 7, 8 .

Our aim was to interrogate SFN for efficacy against human coronaviruses. We report here that SFN inhibits in vitro seasonal coronavirus HCoV-OC43 and Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infections of mammalian host cells and appears to have a synergistic interaction with remdesivir. In addition, SFN reduces viral load and pulmonary pathology in a mouse model of SARS-CoV-2 infection.

To evaluate the potential virus-inhibitory activity of SFN, cells were exposed in vitro to the test drug for 1 -3 hours before inoculation with coronaviruses. In this near-simultaneous druginfection scenario, SFN effectively inhibited both HCoV-OC43 and SARS-CoV-2-Wuhan-Hu-1 virus-associated cell death in non-human primate Vero C1008 cells in a dose-dependent manner revealing comparable median inhibitory concentrations (IC50 = 10 µM, 95% CI 4.7 -20.4, and 12 µM, 95% CI 4.7 -30, respectively), and virus selectivity (TI50 = 7 and 7, respectively) ( Figures 1A, 2A) . When the same assay was performed using human diploid fibroblasts, MRC-5, SFN treatment of HCoV-OC43 infection produced similar results (IC50 = 18 µM, 95% CI 9.7 -33.5, TI50 = 5) ( Figure 1B) . SFN cytotoxicity was also dose-dependent, the median cytotoxic dose (TD50) remained within the range of 73 -89 µM ( Figures 1A-B, 2A) .

infection did not result in measurable cell death. Instead, we quantified viral RNA from SARS-CoV-2 infected human intestinal Caco-2 cells treated with SFN. A dose-dependent reduction was observed with an IC50 of 2.4 µM ( Figure 2C ).

We further evaluated SFN for activity against a second reference strain of SARS-CoV-2 as well as two clinical strains that carry the spike D614G (614G+) substitution that is found in the majority of variants of concern currently in circulation ( Figure 2D ) 9 . SFN inhibited USA-WA1/2020, (IC50 = 31 µM, 95% CI 14.7 -66.4) and the two 614G+ clinical strains, USA/MDHP-20/2020 (MD) and USA/DCHP-7/2020 (DC) (IC50 = 28 µM, 95% CI 14.9 -52.9, and 29 µM, 95% CI 8.2 -102.3, respectively), with comparable efficacy to that reported above for reference strain Wuhan-Hu-1.

We next investigated whether SFN could affect an established virus infection. As shown in Figures 1C and 2B , SFN effectively inhibited both an HCoV-OC43 and a SARS-CoV-2-Wuhan-Hu-1 infection that had been allowed to replicate for 24 hours before addition of drug. The IC50

for both viruses was in the lower micromolar range, 18 µM (95% CI 4 -84.1) and 13 µM (95% CI 8.6 -20), respectively. Interestingly, these results show that the anti-virus specific activity, i.e., the therapeutic index (TI), of SFN is similar whether the drug is added just before or 24 hours after virus inoculation ( Figures 1A, 1C , 2A-B) suggesting an effect on both extracellular entry and intracellular post-entry viral processes. We also determined whether a single application of SFN could protect from the cytopathic effects (CPE) of subsequent viral infection lasting 4 days. As shown in Figure 1D , SFN pretreatment of Vero C1008 host cells resulted in measurable inhibition of HCoV-OC43 CPE with an IC50 = 21 µM (95% CI 9.3 -49.3) and TI50 =

Finally, we examined the potential synergistic effects of SFN combined with the anti-viral drug remdesivir, an inhibitor of viral RNA-dependent RNA polymerase recently reported to shorten the time to recovery in adults who were hospitalized with COVID-19 10 . As shown in Figure 2E , remdesivir effectively inhibits in vitro replication of SARS-CoV-2 (IC50 = 4 µM) as well as HCoV-OC43, albeit at a higher concentration (IC50 = 22 µM) ( Figure 1E ). In two-drug combination assays, SFN and remdesivir interacted synergistically at several combination ratios to inhibit replication of both HCoV-OC43 ( Figure 1F ) and SARS-CoV-2-Wuhan-Hu-1 ( Figure 2F ) at concentrations below the corresponding IC50 for each drug.

To evaluate the ability of SFN treatment to reduce viral titers and inflammation in vivo, K18-hACE2 transgenic male mice were inoculated intranasally with 8.4x10 5 tissue culture infectious dose 50 (TCID50) of SARS-CoV-2/USA/WI1/2020 11 . SFN was administered daily via oral gavage to a subgroup of infected animals starting one day prior to viral inoculation ( Figure 3A) .

A marked weight loss was observed in the infected animals starting at four days post inoculation. By day 6 post inoculation, SFN-treated mice lost significantly less weight compared to controls ( Figure 3B , P<0.0001). As a measure of lung injury, the protein concentration in the bronchoalveolar lavage (BAL) was significantly lower in the SFN-treated infected mice compared to untreated infected controls ( Figure 3C , P<0.0001) suggesting a measure of protective effect of drug pretreatment. The viral burden measured in the alveolar fluid was also significantly lower in treated animals compared to untreated controls, with a 1.15 log reduction in viral titers ( Figure 3D, P=0.04) . Similarly, a 1.5 log reduction in viral lung titers was observed in SFN-treated mice compared to untreated controls, when normalized to Pol2Ra ( Figure 3E , P=0.004). Data on pulmonary viral burden without normalization are presented in figure S1.

Analysis of hematoxylin and eosin-stained lung sections from these animals showed an inflammatory process similar to what has been previously described for this model after SARS-CoV-2 infection 11-13 ( Figure 3F ). SFN-treated mice had a lower degree of pulmonary pathology with less alveolar and peribronchiolar inflammation, compared to infected untreated mice ( Figure S2 ). Histopathology analysis showed a significant reduction a lung inflammation in SFNtreated mice (histopathology score of 1/16) over untreated controls (histopathology score of Given the known immunomodulatory effects of SFN, we employed high-dimensional flow cytometry to evaluate the changes in the immune response of SARS-CoV-2-infected mice treated with SFN and untreated controls, as compared to uninfected mice ( Figure S3 , Table S1 ). 

The ongoing SARS-CoV-2 pandemic has created the immediate need for effective therapeutics that can be rapidly translated to clinical use. Despite the introduction of vaccines, effective antiviral agents are still necessary, particularly considering the potential effects of viral variants 16 . While great efforts have been made to develop drugs that target the virus, this approach can be affected by the emergence of viral variants that change the affinity of the drug to the viral protein 17 . An alternative approach is to also target host mechanisms required by the virus to infect cells and replicate 18 . Host-directed therapy is advantageous as it allows preexisting drugs to be repurposed, may provide broad-spectrum inhibition against multiple viruses, and is generally thought to be more refractory to viral escape mutations 19, 20 .

Following a preliminary examination of a small group of clinically approved drugs, experimental novel compounds, and nutritional supplements, SFN was identified as a promising candidate to target the host cellular response, given that it is orally bioavailable, commercially available, and has limited side-effects 8, 21 . We observed that SFN has dual antiviral and anti-inflammatory properties against coronaviruses. We determined that SFN has potent antiviral activity against The pathogenesis of many viral infections is associated with increased production of reactive oxygen species (ROS) which leads to cell death 23, 24 . Conversely, SFN increases antioxidant, anti-inflammatory, and antiviral defenses primarily via activation of the cap'n'collar transcription factor NRF2 25 . Under normal conditions, NRF2 remains in an inactive state by association with its inhibitor protein Kelch-like ECH-associated protein 1 (KEAP1) 26 . In response to oxidative stress, KEAP1 is inactivated and NRF2 is released to induce NRF2-responsive genes that subsequently protect against stress-induced cell death 27 . SFN has been extensively studied in humans for its anti-cancer properties, has been shown to activate the NRF2 pathway in upper airways 28 , and improves the phagocytic ability of alveolar macrophages 3 . The dual antiviral and anti-inflammatory properties of SFN have also been previously described for other viral infections. In vitro antiviral activity has been reported against influenza virus 29 and SFN treatment significantly limited lung viral replication and virus-induced inflammation in respiratory syncytial virus-infected mice 30 .

Targeting the NRF2 pathway has been considered a promising approach to develop therapeutics for COVID-19 for multiple reasons 31 . NRF2 deficiency is known to upregulate the angiotensin-converting enzyme 2 (ACE2), the primary mechanism of cell entry for SARS-CoV-2.

The NRF2 activator oltipraz reduces ACE2 levels, suggesting that NRF2 activation might reduce the availability of ACE2 for SARS-CoV-2 entry into the cell 32 . Increased NRF2 activity also reportedly inhibits IL-6 and IL-1β gene expression 33 , two cytokines known to play key roles in promoting the hyperactive immune response in severely ill COVID-19 patients 34 . Conversely, NRF2 activity is dysregulated in disease states that have been associated with increased severity of COVID-19, e.g. diabetes 35 . Further, NRF2 activity declines in older patients who are more susceptible to severe COVID-19 36 . Recent reports suggest that NRF2-dependent genes are suppressed in SARS-CoV-2 infected cells and in lung biopsies from COVID-19 patients 31 .

Similarly, treatment of cells with NRF2 agonists 4-octyl-itaconate and dimethyl fumarate inhibited replication of SARS-CoV-2 in vitro 31 . SFN also inhibits inflammation through NRF-2 independent pathways, such as reducing the pro-inflammatory nuclear factor kappa B (NF-κB) 37 . NF-κB activation has been described as a key component of the inflammatory response to multiple viral infections, including COVID-19 38 . There are also other pathways affected by SFN (e.g. STING) that could play a role in its antiviral response to coronaviruses. These additional pathways are the subject of continuing investigation 39 .

As a potent NRF2 activator, SFN can modulate the host's immune response while also providing direct antiviral effects. In contrast to therapeutics that inhibit a single cytokine (e.g. IL- 

We documented that SFN can inhibit the in vitro and in vivo replication of SARS-CoV-2 at pharmacologically and therapeutically achievable doses and can modulate the inflammatory response thereby decreasing the consequences of infection in the animal model when administered prior to infection. Given that SFN is orally bioavailable, commercially available, and has limited side-effects, our results suggest it could be a promising and easily scalable approach for the prevention and treatment of COVID-19 as well as other coronavirus infections.

Drugs L-SFN, 10mg/mL in ethanol (56mM), was obtained from Cayman Chemical (Ann Arbor, MI). D,L-SFN was obtained from Millipore Sigma (St. Louis, MO); stock solution of 5mM was prepared in DMSO. Remdesivir was obtained from MedChemExpress or Cayman Chemical and stock solutions, 5 or 20mM, respectively, were prepared in DMSO. Drug stock solutions were stored at -25°C. 

We used a colorimetric assay that interrogates both antiviral and anti-host cell activities to evaluate compounds 56 

Animal studies were carried out based on the recommendations in the Guide for the Care and and uninfected controls also received daily oral gavage with 2% ethanol in water. After induction of anesthesia with ketamine hydrochloride and xylazine, the animals received 8.4x10 5 TCID50 of SARS-CoV-2/USA/WI1/2020 intranasally. Uninfected animals received intranasally the same volume of vehicle (2% ethanol and water). Weights were monitored daily, the animals were sacrificed 6 days post-infection by isoflurane overdose, and the tissues were harvested. Tissues were perfused with PBS after serum collection via cardiac puncture and before tissue harvest.

Broncheoalveolar lavage (BAL) was obtained by cannulating the trachea with a 20-gauge catheter. The right lung was lavaged twice (each aliquot 1 ml; calcium-free PBS); total returns averaged 1-1.5 ml/mouse. BAL was centrifuged at 600 g for 8 minutes at 4°C. The cell-free supernatants were stored at -80°C for total protein quantification using the BCA protein assay (Sigma).

Lungs were minced and incubated at 37°C in an enzyme cocktail of RPMI containing 2.4 mg/ml collagenase I and 20 μg/ml DNase (Invitrogen), then mashed through a 70-μm nylon cell strainer (BD Falcon). All flow cytometry antibodies used for phenotypic and metabolic analysis can be found in Table S1 . FCS files were analyzed using Flowjo v10.6.2 software (BD). Manual gating strategies for all the panels can be found in Figure S3 . High-dimensional unbiased analysis of cell phenotypes was performed using Flowjo plugins DownSample v3 and UMAP.

After euthanasia, tissues were fixed on 10% neutral-buffered formalin. Tissues were embedded 

Broccoli or Sulforaphane: Is It the Source or Dose That Matters?

Sulforophane glucosinolate. Monograph

Targeting Nrf2 signaling improves bacterial clearance by alveolar macrophages in patients with COPD and in a mouse model

Protective mechanism of sulforaphane in Nrf2 and anti-lung injury in ARDS rabbits

Effect of Broccoli Sprouts on Nasal Response to Live Attenuated Influenza Virus in Smokers: A Randomized, Double-Blind Study

Sulforaphane: Its "Coming of Age" as a Clinically Relevant Nutraceutical in the Prevention and Treatment of Chronic Disease

Can Activation of NRF2 Be a Strategy against COVID-19?

Sulforaphane reduces hepatic glucose production and improves glucose control in patients with type 2 diabetes

SARS-CoV-2 spike D614G change enhances replication and transmission

Remdesivir for the Treatment of Covid-19 -Final Report

COVID-19 treatments and pathogenesis including anosmia in K18-hACE2 mice

SARS-CoV-2 infection of human ACE2-transgenic mice causes severe lung inflammation and impaired function

Lethality of SARS-CoV-2 infection in K18 human angiotensinconverting enzyme 2 transgenic mice

COVID-19 and myeloid cells: complex interplay correlates with lung severity

Metabolic programs define dysfunctional immune responses in severe COVID-19 patients

SARS-CoV-2 variants and ending the COVID-19 pandemic

SARS-CoV-2 Viral Variants-Tackling a Moving Target

SARS-CoV-2 dependence on host pathways

Combating emerging viral threats

Genetic Screens Identify Host Factors for SARS-CoV-2 and Common Cold Coronaviruses

Sulforaphane treatment of autism spectrum disorder (ASD)

Plitidepsin has potent preclinical efficacy against SARS-CoV-2 by targeting the host protein eEF1A

Therapeutic Modulation of Virus-Induced Oxidative Stress via the Nrf2-Dependent Antioxidative Pathway

Oxidative stress during viral infection: A review

Direct evidence that sulfhydryl groups of Keap1 are the sensors regulating induction of phase 2 enzymes that protect against carcinogens and oxidants

The Nrf2 regulatory network provides an interface between redox and intermediary metabolism

Nrf2 as a master regulator of tissue damage control and disease tolerance to infection

Oral sulforaphane increases Phase II antioxidant enzymes in the human upper airway

Nrf2 expression modifies influenza A entry and replication in nasal epithelial cells

Antiviral Activity of Nrf2 in a Murine Model of Respiratory Syncytial Virus Disease

SARS-CoV2-mediated suppression of NRF2-signaling reveals potent antiviral and anti-inflammatory activity of 4-octyl-itaconate and dimethyl fumarate

Nrf2 deficiency upregulates intrarenal angiotensin-converting enzyme-2 and angiotensin 1-7 receptor expression and attenuates hypertension and nephropathy in diabetic mice

Nrf2 suppresses macrophage inflammatory response by blocking proinflammatory cytokine transcription

Cytokine release syndrome in severe COVID-19

Dysregulation of Nrf2/Keap1 redox pathway in diabetes affects multipotency of stromal cells

Redox regulation by NRF2 in aging and disease

Sulforaphane inhibits endothelial lipase expression through NF-κB in endothelial cells

NF-κB Pathway as a Potential Target for Treatment of Critical Stage COVID-19 Patients

Nrf2 negatively regulates STING indicating a link between antiviral sensing and metabolic reprogramming

Tocilizumab in Patients Hospitalized with Covid-19 Pneumonia

COVID-19: consider cytokine storm syndromes and immunosuppression

Pathological findings of COVID-19 associated with acute respiratory distress syndrome

Characteristics of peripheral lymphocyte subset alteration in COVID-19 pneumonia

Sulforaphane Inhibits Prostate Carcinogenesis and Pulmonary Metastasis in TRAMP Mice in Association with Increased Cytotoxicity of Natural Killer Cells

Elevated lnterleukin-1 Release by Human Alveolar Macrophages during the Adult Respiratory Distress Syndrome

Interleukin-8 and development of adult respiratory distress syndrome in at-risk patient groups

Alveolar macrophage transcriptional programs are associated with outcomes in acute respiratory distress syndrome

Synergism of TNF-α and IFN-γ Triggers Inflammatory Cell Death, Tissue Damage, and Mortality in SARS-CoV-2 Infection and Cytokine Shock Syndromes

Safety, tolerance, and metabolism of broccoli sprout glucosinolates and isothiocyanates: a clinical phase I study

Current landscape of NRF2 biomarkers in clinical trials

In vivo pharmacokinetics and regulation of gene expression profiles by isothiocyanate sulforaphane in the rat

Absorption and chemopreventive targets of sulforaphane in humans following consumption of broccoli sprouts or a myrosinase-treated broccoli sprout extract

Quantitative determination of dithiocarbamates in human plasma, serum, erythrocytes and urine: pharmacokinetics of broccoli sprout isothiocyanates in humans

Rapid and sustainable detoxication of airborne pollutants by broccoli sprout beverage: results of a randomized clinical trial in China

Repeat COVID-19 Molecular Testing: Correlation of SARS-CoV-2 Culture with Molecular Assays and Cycle Thresholds

Total Synthesis of the Natural Product (+)-trans-Dihydronarciclasine via an Asymmetric Organocatalytic [3+3]-Cycloaddition and discovery of its potent anti-Zika Virus

Guideline to reference gene selection for quantitative real-time PCR

Antiviral effects of SFN against SARS-CoV-2. Median effect plot and dose-effect curves calculated for (A) Vero C1008 cells infected with SARS-CoV-2/Wuhan-Hu-1 after 1-3h incubation with increasing concentrations of SFN; (B) Vero C1008 cells infected with SARS

Antiviral data displayed in red; antihost cell activity (cytotoxicity) displayed in blue. (C) The antiviral activity in human Caco-2 cells was determined by measuring viral RNA by qPCR. The cells were incubated with SFN for 1h before viral inoculation. (D) Effects of SFN evaluated in Vero C1008 cells exposed to drug for 1h followed by viral inoculation. A reference strain (USA-WI1/2020) and two 614G+ clinical strains of SARS-CoV-2 were evaluated for cytopathic effects using a bioluminescence readout. (E) Effects of SFN and remdesivir evaluated in Vero C1008 cells exposed to drug for 1h followed by viral inoculation. (F) Normalized isobologram showing combination index (CI) for combinations of various doses of SFN and remdesivir

Additive effect (CI=1)

replicates within each experiment, except experiment shown in E which was performed once. infected treated (n=9). (F) Hematoxylin and eosin (H&E) staining and immunostaining for SARS-CoV-2 spike protein, of histological sections of the lungs of a representative uninfected control, an infected untreated, and an infected treated mouse. Regions of the lung anatomy where alveolar and peribronchiolar inflammation was assessed are highlighted in boxes. Images show low (left panels; scale bar, 1mm) and high-power magnification (right panels; scale bar, 50µm) of the same tissue section. (G) Histopathological severity scoring was evaluated according to the pathological changes outlined in the methods section. Data from one independent experiment

Quantification of the SARS-CoV-2 spike protein immunostaining showed a 4.41x lower % area in the lungs of SFN-treated mice compared to infected untreated controls (P=0.01). Data from one independent experiment, uninfected (n=4), infected untreated (n=8), infected treated (n=5)