Smoke and Spike: Benzo[a]pyrene Enhances SARS‐CoV‐2 Infection by Boosting NR4A2‐Induced ACE2 and TMPRSS2 Expression

Abstract Cigarette smoke aggravates severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) infection. However, the underlying mechanisms remain unclear. Here, they show that benzo[a]pyrene in cigarette smoke extract facilitates SARS‐CoV‐2 infection via upregulating angiotensin‐converting enzyme 2 (ACE2) and transmembrane protease serine 2 (TMPRSS2). Benzo[a]pyrene trans‐activates the promoters of ACE2 and TMPRSS2 by upregulating nuclear receptor subfamily 4 A number 2 (NR4A2) and promoting its binding of NR4A2 to their promoters, which is independent of functional genetic polymorphisms in ACE2 and TMPRSS2. Benzo[a]pyrene increases the susceptibility of lung epithelial cells to SARS‐CoV‐2 pseudoviruses and facilitates the infection of authentic Omicron BA.5 in primary human alveolar type II cells, lung organoids, and lung and testis of hamsters. Increased expression of Nr4a2, Ace2, and Tmprss2, as well as decreased methylation of CpG islands at the Nr4a2 promoter are observed in aged mice compared to their younger counterparts. NR4A2 knockdown or interferon‐λ2/λ3 stimulation downregulates the expression of NR4A2, ACE2, and TMPRSS2, thereby inhibiting the infection. In conclusion, benzo[a]pyrene enhances SARS‐CoV‐2 infection by boosting NR4A2‐induced ACE2 and TMPRSS2 expression. This study elucidates the mechanisms underlying the detrimental effects of cigarette smoking on SARS‐CoV‐2 infection and provides prophylactic options for coronavirus disease 2019, particularly for the elderly population.


TMPRSS2
(A) The effect of cigarette smoke exposure on the mRNA level of ACE2 was evaluated with the use of public databases regarding GSE994, TCGA LUAD, GSE8987, and GSE10718.(B) The effect of cigarette smoke exposure on the mRNA level of TMPRSS2 was evaluated with public databases regarding GSE994, TCGA LUAD, GSE8987, and GSE10718.Mann-Whitney test,

Figure S2 .
Figure S2.Effects of single nucleotide polymorphisms (SNPs) on the activities of the ACE2 and TMPRSS2 transcriptional regulatory sequences in cells stimulated with CSE (A) The activity of the ACE2 enhancer with A (served as reference) or G at rs112312217.(B) The activity of the ACE2 promoter with A (served as reference) or G at rs113208650.(C) The activity of the TMPRSS2 enhancer with A (served as reference) or G at rs9981570.(D) The activity of the TMPRSS2 promoter with A (served as reference) and G at rs12481984 and rs812074.Control, cells stimulated with DMSO (0.1%, 12 hours).CSE, cells stimulated with 20% CSE for 12 hours.The luciferase activity of promoter/enhancer sequence with reference genotype, which was detected in cells treated with DMSO, served as a reference to calculated relative luciferase activity.Student's t-test, * P<0.05, ** P<0.005, *** P<0.0005.Error bars represent standard deviation.

Figure S3 .
Figure S3.BaP upregulated the expression of SARS-CoV-2 receptors in human primary human alveolar type Ⅱ (AT Ⅱ) cells (A) Representative image of Western blot tests detecting the protein levels of ACE2 and TMPRSS2 in primary AT Ⅱ cells.Cells were stimulated with BaP at 15μM for 0, 12, 24, and 48 hours before being subjected to Western blot tests.(B) Representative results of RT-qPCR evaluating SARS-CoV-2 receptors in primary AT Ⅱ cells.Cells were stimulated with BaP at 15μM or 0.1% DMSO for 12 hours.All experiments were repeated three times with three replicates and one of the representative results is shown here.RT-qPCR data is normalized by reference gene GAPDH.The intensity of Western blot band is measured by Image J software and normalized to the band intensity of GAPDH.Relative mRNA level and protein level are presented as a ratio of BaP group to DMSO group.Student's t-test, *P<0.05,**P<0.005,***P<0.0005.Error bars represent standard deviation.

Figure S4 .
Figure S4.Effects of cigarette smoke extract (CSE) on the transcriptional levels of NR4A2, CREAM, and SNAI2 in Calu3 and H1650 cell lines Results of RT-qPCR tests evaluating the transcriptional levels of NR4A2, CREAM, and SNAI2 in cells stimulated with 20% CSE for 12 hours.All experiments were repeated three times with three replicates and one of the representative results is shown here.RT-qPCR data is normalized by reference gene GAPDH.Relative mRNA level is presented as a ratio of the CSE group to the control group.Student's t-test, * P<0.05, ** P<0.005, *** P<0.0005.Error bars represent standard deviation.

Figure S5 .
Figure S5.Effects of SNPs on the activities of the ACE2 and TMPRSS2 transcriptional regulatory sequences in cells with NR4A2 knockdown (A) NR4A2 knockdown significantly decreased the activity of the ACE2 promoter sequences with different genotypes at rs113208650.NR4A2 knockdown significantly decreased the activities of the TMPRSS2 promoter sequences with different genotypes at rs12481984 and rs8128074.(B) NR4A2 knockdown attenuated the positive effect of BaP on the activity of the ACE2 promoter sequences with different genotypes at rs113208650.(C) NR4A2 knockdown attenuated the positive effect of BaP on the activity of the TMPRSS2 promoter sequences with different genotypes at rs12481984 and rs8128074.siScramble, cells transfected with siRNA targeting scramble sequence for 48 hours.siNR4A2, cells transfected with siRNA targeting NR4A2 for 48 hours.Luciferase activity of cells treated with DMSO and transfected with siScramble was applied as a reference to calculate relative luciferase activity.Student's t-test, * P<0.05, ** P<0.005, *** P<0.0005.Error bars represent standard deviation.

Figure S6 .
Figure S6.ChIp-qPCR followed Sanger sequencingThe representative image of Sanger sequencing chromatogram (the result of Calu3 cells).The product of qPCR conducted with DNA binding to NR4A2 was subjected to Sanger sequencing.The sequence of amplified ChIP-qPCR products was confirmed to be the ACE2 and TMPRSS2 promoter region.

Figure
Figure S7.mRNA level of Nr4a2 in lung tissues from mice The lung tissues from 3-month old mice and 20-month old mice (n = 3 per group) were subjected to RT-qPCR.Relative mRNA level was calculated as a ratio of the value of 20-month old mice to 3-month old mice.Student's t-test, * P<0.05, ** P<0.005, *** P<0.0005.Error bars represent standard deviation.

Figure S8 .
Figure S8.The pyrosequencing data of mouse lung tissues at the Nr4a2 promoter region Genomic DNA extracted from lung tissues of 3-month-old mice and 20-month-old mice were subjected to the pyrosequencing assay.The pyrosequencing signal of each nucleotide site was shown on the X axis.The sequence of the Nr4a2 promoter was shown on the Y axis.The light blue region represented the position of CpG islands.The methylation rate was marked on the top of the CpG island region.

Figure S9 .
Figure S9.Effects of inflammatory factors on the expression of ACE2 and TMPRSS2 (A) The mRNA levels of ACE2 and TMPRSS2 in Calu3 cells treated with inflammatory factors (50ng/ml, 12 hours), respectively.(B) The protein levels of ACE2 and TMPRSS2 in Calu3 cells treated with inflammatory factors (50ng/ml, 12 hours), respectively.(C) The mRNA levels of ACE2 and TMPRSS2 in H1650 cells treated with inflammatory factors (50ng/ml, 12 hours), respectively.(D) The protein levels of ACE2 and TMPRSS2 in H1650 cells treated with inflammatory factors (50ng/ml, 12 hours), respectively.All experiments were repeated three times with three replicates and one of the representative results is shown here.RT-qPCR data is normalized by reference gene GAPDH and relative mRNA level is presented as a ratio of inflammatory factors treated group to control group.The intensity of Western blot band is measured by Image J software and normalized to band intensity of GAPDH.Relative band intensity is presented as a ratio of inflammatory factors treated group to control group.Student's t-test, * P<0.05, ** P<0.005, *** P<0.0005.Error bars represent standard deviation.

Figure S10 .
Figure S10.Infection assay of VSV-based pseudoviruses of SARS-CoV-2 Viral infection was evaluated by detecting luciferase activity.(A) Effect of BaP on infection of VSV-based pseudovirus carrying G protein (control pseudoviruses of pseudoviruses for SARS-CoV-2).Calu3 cells were treated with 15μM BaP or 0.1% DMSO for 12 hours.Then, cells were infected with control pseudoviruses for 24 hours before being subjected to luciferase assay.(B) Effect of anti-Spike antibody on infection of VSV-based pseudovirus of SARS-CoV-2.Calu3 cells were categorized into four groups and treated with 0.1% DMSO, 15μM BaP, 0.1% DMSO combined with anti-Spike antibody, and 15μM BaP combined with anti-Spike antibody, respectively.Twelve hours later, cells were infected with VSV-based pseudoviruses of Omicron BA.5 for 24 hours and then subjected to luciferase assay.Student's t-test, * P<0.05, ** P<0.005, *** P<0.0005.Error bars represent standard deviation.

Figure S11 .
Figure S11.Effective of BaP on the expression of Nr4a2, Ace2, and Tmprss2 in lung and testis tissues of hamsters (A) Representative image of Western blot tests.(B) Representative image of IHC (scale bar, 50μm).Hamsters in BaP group (n = 3) were injected with BaP (125mg/kg) into the left pleural cavity and scrotum.The hamsters in DMSO group were injected with the equal value of DMSO in the same way.Three weeks later, lung and testis tissues were collected and subjected to Western blot and IHC analysis.The intensity of Western blot band is measured by Image J software and normalized to band intensity of GAPDH.Relative band intensity is presented as a ratio of BaP group to DMSO group.

Figure S12 .
Figure S12.BaP facilitated the infection of SARS-CoV-2 pseudovirus in lung tissues of mice Ex vivo live images and quantitative analysis of image data for lung tissues of healthy C57BL/6 mice (n = 3 per group).C57BL/6 mice were injected with BaP (125mg/kg) or DMSO into left pleural cavity.Three weeks later, mice were injected with lentivirus-based Omicron pseudovirus into left pleural cavity (100 μL, 2×10 7 U/mL).GFP fluorescence was detected by two days later.The flux of GFP fluorescence in the region of interest (ROI) was measured to evaluate the infection level of pseudovirus.Student's t-test, *P<0.05,**P<0.005,***P<0.0005.Error bars represent standard deviation.
Transcription factors were predicted by using RegulomeDB ChIP-seq database.The peak value of ChIP-seq and location were listed here.
*Transcription factors were predicted by using JASPAR database.The relative score and putative binding sequences were listed here.The robust results were identified as those with relative score >0.95.†