Benzene metabolites increase vascular permeability by activating heat shock proteins and Rho GTPases
Data files
Nov 12, 2025 version files 95.01 GB
-
C1_1.fq.gz
1.64 GB
-
C1_2.fq.gz
1.70 GB
-
C2_1.fq.gz
1.56 GB
-
C2_2.fq.gz
1.61 GB
-
C3_1.fq.gz
1.55 GB
-
C3_2.fq.gz
1.61 GB
-
C4_1.fq.gz
1.56 GB
-
C4_2.fq.gz
1.62 GB
-
C5_1.fq.gz
1.63 GB
-
C5_2.fq.gz
1.71 GB
-
C6_1.fq.gz
1.52 GB
-
C6_2.fq.gz
1.56 GB
-
C7_1.fq.gz
1.46 GB
-
C7_2.fq.gz
1.54 GB
-
C8_1.fq.gz
1.75 GB
-
C8_2.fq.gz
1.82 GB
-
C9_1.fq.gz
1.53 GB
-
C9_2.fq.gz
1.59 GB
-
DEG.MA_HAEC_24h.csv
1.34 MB
-
DEG.MA_HAEC_2h.csv
1.30 MB
-
DEG.MA_HAEC_6h.csv
1.37 MB
-
DEG.MA_mCMVEC_6h.csv
1.98 MB
-
enrichGO_HAEC.csv
23.69 KB
-
Reactome_mCMVEC.csv
2.18 KB
-
README.md
8.19 KB
-
T1_1.fq.gz
1.60 GB
-
T1_2.fq.gz
1.66 GB
-
T2_1.fq.gz
1.47 GB
-
T2_2.fq.gz
1.51 GB
-
T3_1.fq.gz
1.47 GB
-
T3_2.fq.gz
1.51 GB
-
T4_1.fq.gz
1.51 GB
-
T4_2.fq.gz
1.53 GB
-
T5_1.fq.gz
1.46 GB
-
T5_2.fq.gz
1.49 GB
-
T6_1.fq.gz
1.47 GB
-
T6_2.fq.gz
1.53 GB
-
T7_1.fq.gz
1.46 GB
-
T7_2.fq.gz
1.48 GB
-
T8_1.fq.gz
1.54 GB
-
T8_2.fq.gz
1.60 GB
-
T9_1.fq.gz
1.66 GB
-
T9_2.fq.gz
1.73 GB
-
Z1_1.fq.gz
2.47 GB
-
Z1_2.fq.gz
2.53 GB
-
Z2_1.fq.gz
2.40 GB
-
Z2_2.fq.gz
2.46 GB
-
Z3_1.fq.gz
2.01 GB
-
Z3_2.fq.gz
2.06 GB
-
Z4_1.fq.gz
2.02 GB
-
Z4_2.fq.gz
2.08 GB
-
Z5_1.fq.gz
2.09 GB
-
Z5_2.fq.gz
2.15 GB
-
Z6_1.fq.gz
3.01 GB
-
Z6_2.fq.gz
3.09 GB
-
Z7_1.fq.gz
2.34 GB
-
Z7_2.fq.gz
2.39 GB
-
Z8_1.fq.gz
2.60 GB
-
Z8_2.fq.gz
2.67 GB
Abstract
Benzene is a hazardous air pollutant and environmental contaminant emitted into the atmosphere from motor vehicle exhaust, fuel evaporation at gas stations, and during man-made and natural disasters such as wildfires and military burn pit incinerations. Chronic exposure to benzene, even at lower levels, is associated with elevated risk for cardiovascular diseases, including atherosclerosis and cardiac dysfunction. However, the mechanisms of benzene-induced cardiovascular complications remain unknown. Our data suggest that intradermal injections of benzene metabolite trans,trans-muconaldehyde (MA) increased vascular permeability by 54% in C57BL6 mice, while intravenous administration of MA propagated endothelial injury measured by elevated circulating endothelial-specific microparticles. The exposure of primary cardiac microvascular endothelial cells to MA increased vascular permeability, which was detected by transendothelial monolayer electrical resistance and by fluorescently labeled dextran’s diffusion. To investigate the molecular mechanisms of MA-induced endothelial permeability, we exposed mouse cardiac microvascular endothelial cells (CMVEC) to 10 μM MA for 6 hours and human aortic endothelial cells (HAEC) for 2, 6, and 18 hours. Then, we analysed differentially regulated genes (DEG) using bulk RNA sequencing. The transcriptomic analysis indicated that MA increased the expression of stress-response and chaperone genes and perturbed GTPase regulation in endothelial cells. The functional importance of HSP and GTPase signaling pathway in MA-induced endothelial permeability was confirmed in vitro and in vivo. In conclusion, benzene metabolites increased vascular permeability through activation of HSP and GTPase signaling pathways.
Dataset DOI: 10.5061/dryad.12jm63z8q
Description of the data and file structure
We analyzed gene expression profiles in mouse cardiac microvascular endothelial cells (mCMVEC) and human aortic endothelial cells (HAEC) exposed to 10 μM of trans,trans-muconaldehyde (MA) for the indicated time using bulk RNA sequencing. Specific analysis details are available in the associated manuscript.
Description of the data and file structure
Raw fastq files
Name Cell type Treatment Time
Z1_1.fq.gz mCMVEC Control 6h
Z1_2.fq.gz mCMVEC Control 6h
Z2_1.fq.gz mCMVEC Control 6h
Z2_2.fq.gz mCMVEC Control 6h
Z3_1.fq.gz mCMVEC Control 6h
Z3_2.fq.gz mCMVEC Control 6h
Z4_1.fq.gz mCMVEC Control 6h
Z4_2.fq.gz mCMVEC Control 6h
Z5_1.fq.gz mCMVEC MA 6h
Z5_2.fq.gz mCMVEC MA 6h
Z6_1.fq.gz mCMVEC MA 6h
Z6_2.fq.gz mCMVEC MA 6h
Z7_1.fq.gz mCMVEC MA 6h
Z7_2.fq.gz mCMVEC MA 6h
Z8_1.fq.gz mCMVEC MA 6h
Z8_2.fq.gz mCMVEC MA 6h
C1_1.fq.gz HAEC Control 2h
C1_2.fq.gz HAEC Control 2h
C2_1.fq.gz HAEC Control 2h
C2_2.fq.gz HAEC Control 2h
C3_1.fq.gz HAEC Control 2h
C3_2.fq.gz HAEC Control 2h
C4_1.fq.gz HAEC Control 6h
C4_2.fq.gz HAEC Control 6h
C5_1.fq.gz HAEC Control 6h
C5_2.fq.gz HAEC Control 6h
C6_1.fq.gz HAEC Control 6h
C6_2.fq.gz HAEC Control 6h
C7_1.fq.gz HAEC Control 24h
C7_2.fq.gz HAEC Control 24h
C8_1.fq.gz HAEC Control 24h
C8_2.fq.gz HAEC Control 24h
C9_1.fq.gz HAEC Control 24h
C9_2.fq.gz HAEC Control 24h
T1_1.fq.gz HAEC MA 2h
T1_2.fq.gz HAEC MA 2h
T2_1.fq.gz HAEC MA 2h
T2_2.fq.gz HAEC MA 2h
T3_1.fq.gz HAEC MA 2h
T3_2.fq.gz HAEC MA 2h
T4_1.fq.gz HAEC MA 6h
T4_2.fq.gz HAEC MA 6h
T5_1.fq.gz HAEC MA 6h
T5_2.fq.gz HAEC MA 6h
T6_1.fq.gz HAEC MA 6h
T6_2.fq.gz HAEC MA 6h
T7_1.fq.gz HAEC MA 24h
T7_2.fq.gz HAEC MA 24h
T8_1.fq.gz HAEC MA 24h
T8_2.fq.gz HAEC MA 24h
T9_1.fq.gz HAEC MA 24h
T9_2.fq.gz HAEC MA 24h
Counts data were analyzed using the DESeq2 R-package:
- File "DEG.MA_mCMVEC_6h.csv" contains differential expression data generated from DESeq2 comparing normalized counts for mCMVEC following 6 hours of exposure to 10 uM MA vs vehicle (0.1% DMSO).
- File "DEG.MA_HAEC_2h.csv" contains differential expression data generated from DESeq2 comparing normalized counts for HAEC following 2 hours of exposure to 10 uM MA vs vehicle (0.1% DMSO).
- File "DEG.MA_HAEC_6h.csv" contains differential expression data generated from DESeq2 comparing normalized counts for HAEC following 6 hours of exposure to 10 uM MA vs vehicle (0.1% DMSO).
- File "DEG.MA_HAEC_24h.csv" contains differential expression data generated from DESeq2 comparing normalized counts for HAEC following 24 hours of exposure to 10 uM MA vs vehicle (0.1% DMSO).
In each of the files, column "baseMean" represents the mean of normalized counts for all samples, log2FoldChange represents the Log base 2 of the fold change, lfcSE represents the standard error of the log base 2 of the fold change, column “stat” represents the difference in deviance between the reduced model and the full model, which is compared to a chi-squared distribution to generate a p-value, pvalue represents the Wald test p-value: condition treated vs untreated, and padj represents the BH adjusted p-values.
The files above were used to identify significant enrichment of upstream regulators and signaling pathways among the differentially expressed genes in the datasets.
The file "Reactome_mCMVEC.csv" contains the results of the Reactome analysis of DEGs in the mCMVEC_6h (MA vs Control) dataset.
Description of column names:
ID: The unique identifier for the Reactomr term
Description: The descriptive name of the Reactome term
GeneRatio: The ratio of genes in this list that are annotated with this term, formatted as (M/N) (where (M) is the number of your genes in the set and (N) is the total number of your genes in the set).
BgRatio: The ratio of the number of genes in the background universe that are annotated with this term, formatted as (M/N) (where (M) is the number of background genes in the set and (N) is the total number of background genes).
pvalue: over-representation assessed using hypogeometric distribution, which is equivalent to a one-sided Fisher's exact test.
p.adjust: The adjusted p-value, which corrects for multiple testing to control the false discovery rate (FDR).
qvalue: An estimate of the FDR calculated from the p-values.
geneID: The gene IDs that overlap with the functional term.
Count: Total number of genes from the input gene list that match the functional term.
The file "enrichGO_HAEC.csv" contains the results of the analysis using the enrichGO function in the clusterProfiler R package. The analysis was performed on all three time points for HAEC exposed to MA vs Control.
Description of column names:
ID: The unique identifier for the GO term
Description: The descriptive name of the GO term
GeneRatio: The ratio of genes in this list that are annotated with this term, formatted as (M/N) (where (M) is the number of your genes in the set and (N) is the total number of your genes in the set).
BgRatio: The ratio of the number of genes in the background universe that are annotated with this term, formatted as (M/N) (where (M) is the number of background genes in the set and (N) is the total number of background genes).
pvalue: over-representation assessed using hypogeometric distribution, which is equivalent to a one-sided Fisher's exact test.
p.adjust: The adjusted p-value, which corrects for multiple testing to control the false discovery rate (FDR).
qvalue: An estimate of the FDR calculated from the p-values.
geneID: The gene IDs that overlap with the functional term.
Count: Total number of genes from the input gene list that match the functional term.
Cell culture: Mouse cardiac microvascular endothelial cells (mCMVEC) and human aortic endothelial cells (HAEC) were purchased from CellBiologics, Chicago, IL. Cells were cultured in a complete endothelial growth medium without phenol red (CellBiologics) and used up to 8 passages for the experiments.
RNA Isolation and RNAseq analysis: HAEC and mCMVEC were incubated with MA (10 µM in 0.1% DMSO) or vehicle (0.1% DMSO) for 2, 6, and 24 hours (mCMVEC were incubated only for 6 hours), and total RNA was isolated using an RNeasy Mini Kit (Qiagen). RNA quality was measured by Agilent 2100 bioanalyzer (Thermo Fisher Scientific, MA, USA), and samples with high RNA integrity were used for subsequent RNAseq analysis. RNA samples were processed by Novogene using mRNA sequencing services (Novogene, Beijing, China). The resultant raw reads of the FASTQ files were aligned to the human genome (hg38) or mouse genome (mm39) using the HISAT2 R package. The mRNA differentially regulated genes (DEG) and pathway enrichment analysis were performed using DESeq2 and ReactomePA enrichPathway R packages.