Effects of Sulforaphane on SARS‑CoV‑2 infection and NF‑κB dependent expression of genes involved in the COVID‑19 'cytokine storm'

Abstract

Since its spread at the beginning of 2020, the coronavirus disease 2019 (COVID‑19) pandemic represents one of the major health problems. Despite the approval, testing, and worldwide distribution of anti‑severe acute respiratory syndrome coronavirus 2 (SARS‑CoV‑2) vaccines, the development of specific antiviral agents targeting the SARS‑CoV‑2 life cycle with high efficiency, and/or interfering with the associated 'cytokine storm', is highly required. A recent study, conducted by the authors' group indicated that sulforaphane (SFN) inhibits the expression of IL‑6 and IL‑8 genes induced by the treatment of IB3‑1 bronchial cells with a recombinant spike protein of SARS‑CoV‑2. In the present study, the ability of SFN to inhibit SARS‑CoV‑2 replication and the expression of pro‑inflammatory genes encoding proteins of the COVID‑19 'cytokine storm' was evaluated. SARS‑CoV‑2 replication was assessed in bronchial epithelial Calu‑3 cells. Moreover, SARS‑CoV‑2 replication and expression of pro‑inflammatory genes was evaluated by reverse transcription quantitative droplet digital PCR. The effects on the expression levels of NF‑κB were assessed by western blotting. Molecular dynamics simulations of NF‑kB/SFN interactions were conducted with Gromacs 2021.1 software under the Martini 2 CG force field. Computational studies indicated that i) SFN was stably bound with the NF‑κB monomer; ii) a ternary NF‑kB/SFN/DNA complex was formed; iii) the SFN interacted with both the protein and the nucleic acid molecules modifying the binding mode of the latter, and impairing the full interaction between the NF‑κB protein and the DNA molecule. This finally stabilized the inactive complex. Molecular studies demonstrated that SFN i) inhibits the SARS‑CoV‑2 replication in infected Calu‑3 cells, decreasing the production of the N‑protein coding RNA sequences, ii) decreased NF‑κB content in SARS‑CoV‑2 infected cells and inhibited the expression of NF‑kB‑dependent IL‑1β and IL‑8 gene expression. The data obtained in the present study demonstrated inhibitory effects of SFN on the SARS‑CoV‑2 life cycle and on the expression levels of the pro‑inflammatory genes, sustaining the possible use of SFN in the management of patients with COVID‑19.

Keywords: COVID‑19; NF‑kB; SARS‑CoV‑2; SFN; nutraceuticals; pro‑inflammatory genes.

Figure 1

Quantification of SARS-CoV-2 genomes in infected Calu-3 cells. SARS-CoV-2 has been quantified by reverse transcription-digital-droplet PCR. (A) Representative 2D plots obtained by the quantification of human RNAse P and N1-SARS-CoV-2 sequence are reported. (B) Detected copies/µl in each reaction well are reported, for technical issue starting cDNA employed for N1 and N2-SARS-CoV-2 sequences have been diluted 1:10,000 before performing the amplification, while as regard RNAse P 1 µl of starting cDNA has been amplified. (C) Example of 1D plot. SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Figure 2

Effects of Calu-3 exposure to SARS-CoV-2. (A and B) Expression of mRNAs coding for (A) NF-κB p50 and (B) NF-κB p65 following 24- and 48-h infection of Calu-3 cells with SARS-CoV-2, as indicated. Results represent the fold increase with respect to uninfected control Calu-3 cells in three independent cultures. (C) Representative example of the western blot data obtained. The membrane was first incubated with an antibody recognizing NF-kB p105/p50, then stripped, washed and incubated with a NF-kB p65 antibody, then stripped, washed and incubated with a control antibody recognizing β-actin. The antibodies used are enlisted in Table II. The original uncut autoradiograms are shown in Figs. S1 and S2. (D) The protein/β-actin ratios obtained after densitometry analysis of the data shown in panel C; ChemiDoc instrument and Image Lab software (both from Bio-Rad Laboratories, Inc.) were used for densitometric analysis of the obtained bands (n=3). The Ponceau staining of the membrane is shown in Fig. S3, confirming the loading quality of the experiment. ^*P<0.05 and ^**P<0.01. SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Figure 3

Increased expression of NF-kB-regulated genes in SARS-CoV-2 infected cells. (A-C) Reverse transcription-quantitative PCR analysis was performed using primers/probes detecting (A) IL-1β, (B) IL-8 and (C) IL-6 mRNAs. (D) Quantification of IL-6 content in Calu-3 SARS-CoV-2 infected cells by ELISA test. IL-6 content expressed in pg/ml was determined 48 h after the infection with SARS-CoV-2. ^*P<0.05 and ^**P<0.01. SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Figure 4

Summary of docking results. Unoccupied pockets (yellow, pink, green and blue transparent surfaces) and predicted binding site for SFN (atom colored spheres) in NF-κB (blue cartoon) as predicted by SeeSAR. (A) Detail of the interaction between NF-κB residues and SFN in the predicted binding site. (B) H-bonds are highlighted as dashed red lines. SFN, sulforaphane.

Figure 5

Results from CG-MD simulations. (A) Time-dependent distance between com-DNA and com-chain A (blue) and between com-DNA and com-chain B (red) of the original NF-kB/DNA complex. (B) Time-dependent distance between com-DNA and com-chain A (blue) and between com-DNA and com-chain B (red) of the chain-A displaced NF-kB/DNA complex in the absence of any ligand. (C) Time-dependent distance between com-DNA and com-chain A (blue) and between com-DNA and com-chain B (red) of the chain-A displaced NF-kB/DNA complex in the presence of SFN bound to chain A. (D) Time-dependent distance between com-DNA and com-chain A (blue) and between com-DNA and com-chain B (red) of the chain-A displaced NF-kB/DNA complex in the presence of a decoy ligand (n-nonane) bound to chain A. (E) Time-dependent SFN RMSD (nm) bound to chain A alone (blue) and bound to the entire NF-kB/DNA complex (red). (F) Detail of the interaction between NF-kB/DNA and SFN at the end of the 100 ns of CG-MD simulation. H-bonds are highlighted as dashed red lines. Note that the H-bond with 54Arg was retained along the entire simulation. A video of the CG-MD simulation is also available (Video S1). CG-MD, coarse-grained molecular dynamics; SFN, sulforaphane.

Figure 6

Quantification of SARS-CoV-2 genomes in Calu-3 SARS-Cov-2-infected and SFN-treated cells. (A-C) The amount of SARS-CoV-2-released genomes was determined by absolute reverse transcription-quantitative PCR and compared in (A) Calu-3 infected cells analyzed 24 and 48 hpi, or in Calu-3 infected cells treated with SFN or with the SFN vehicle DMSO and analyzed (B) 24 and (C) 48 hpi (n=3) ^*P<0.05, significant; ^**P<0.01, highly significant. (D and E) Representative reverse transcription-digital-droplet dPCR 2D plots quantifying intracellular SARS-CoV-2 genomic N1 and N2 sequences (as indicated); samples were isolated at (D) 24 and (E) 48 hpi time points. (F and G) Two-way ANOVA, followed by post hoc Bonferroni test (see also Fig. S7) performed on samples from SARS-CoV-2-infected Calu-3 cells in the absence or in the presence of SFN, as indicated. Cells were harvested at 24 and 48 hpi. The results represent the mean ± S.D. (n=6). The analysis of the data shown in panels F and G is further detailed in Fig. S7. For SARS-CoV-2 genome quantification, the N1 (F) and N2 (G) sequences have been considered. The P-values reported in panels F and G (two-way ANOVA, followed by post hoc Bonferroni test) were obtained comparing untreated infected cells (SARS-CoV-2) vs. SFN-treated infected cells (SARS-CoV-2 + SFN). Results represent the mean ± S.D. SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; hpi, h post-infection; SFN, sulforaphane.

Figure 7

Quantification of NF-κB gene expression in Calu-3 SARS-Cov-2 infected, SFN-treated cells. (A and B) For quantification of NF-κB mRNAs, reverse transcription-quantitative PCR was employed to quantify the expression of transcripts coding for (A) NF-κBp50 subunit and (B) NF-κBp65 subunit. (C and D) NF-κB proteins were quantified by western blotting. (C) Representative Western Blot membranes hybridized with antibodies against NF-κB (p105/50, p65) and b-actin (as endogenous control). The original uncropped gels used for panel C are shown in Figs. S4 and S5. Representative Ponceau S staining of the membranes to verify extracts loading in each sample is presented in Fig. S6. (D) Quantitative analysis of the protein/β-actin ratios obtained after densitometric analysis of the data shown in panel C; ChemiDoc instrument and Image Lab software (both from Bio-Rad Laboratories, Inc.) were used densitometry analysis of the obtained bands (n=3). ^*P<0.05 and ^**P<0.01. SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; SFN, sulforaphane.

Figure 8

Expression of pro-inflammatory genes in SARS-CoV-2-infected, SFN-treated Calu-3 cells. A short list of pro-inflammatory mRNAs encoding cytokines belonging to the 'COVID-19 cytokine storm' were quantified by reverse transcription-quantitative PCR in SFN-treated Calu-3 cells: (A) IL-1β mRNA, (B) IL-6 mRNA and (C) IL-8 mRNA. (D-F) Binding sites of key transcription factors interacting with the (D) IL-1β, (E) IL-6 and (F) IL-8 gene promoters. ^*P<0.05 and ^**P<0.01. SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; SFN, sulforaphane.