ABSTRACT
Benzene is a ubiquitous environmental pollutant. Recent population-based studies suggest that benzene exposure is associated with an increased risk for cardiovascular disease. However, it is unclear whether benzene exposure is sufficient to induce cardiovascular toxicity. We examined the effects of benzene inhalation (50 ppm, 6 h/day, 5 days/week, 6 weeks) or HEPA-filtered air exposure on the biomarkers of cardiovascular toxicity in male C57BL/6J mice. Benzene inhalation significantly increased the biomarkers of endothelial activation and injury including endothelial microparticles, activated endothelial microparticles, endothelial progenitor cell microparticles, lung endothelial microparticles, and activated lung and endothelial microparticles while having no effect on circulating levels of endothelial adhesion molecules, endothelial selectins, and biomarkers of angiogenesis. To understand how benzene may induce endothelial injury, we exposed human aortic endothelial cells to benzene metabolites. Of metabolites tested, trans,trans-mucondialdehyde (10 μM, 18h) was most toxic. It induced caspases-3, −7 and −9 (intrinsic pathway) activation, and enhanced microparticle formation by 2.4-fold. Levels of plateletleukocyte aggregates, platelet macroparticles, and proportion of CD4+ and CD8+ T-cells were also significantly elevated in the blood of the benzene-exposed mice. We also found that benzene exposure increased the transcription of genes associated with endothelial cell and platelet activation in the liver; and induced inflammatory genes and suppressed cytochrome P450s in the lungs and the liver. Together, these data suggest that benzene exposure induces endothelial injury, enhances platelet activation and inflammatory processes; and circulatory levels of endothelial cell and platelet-derived microparticles and platelet-leukocyte aggregates are excellent biomarkers of cardiovascular toxicity of benzene.
Highlights
Inhaled benzene exposure increases the levels of blood endothelial microparticles.
In vitro, benzene metabolite trans, trans-mucondialdehyde induces endothelial cell apoptosis and microparticles formation.
Inhaled benzene exposure decreases the levels of hematopoietic progenitor cells in the bone marrow.
Inhaled benzene exposure augments the circulating levels of platelet-leukocyte adducts.
INTRODUCTION
Environmental pollution accounts for 9 million pre-mature deaths worldwide, and two-third of these deaths are attributed to air pollution (1). Benzene, a volatile organic compound (VOC), is abundant both in outdoor and indoor air. Ranked number sixth on the Agency for Toxic Substances and Disease Registry (ATSDR) priority list, benzene is one of the top twenty chemicals generated by industrial sources in the United States. It is used to produce industrial chemicals, rubbers, dyes, lubricants, detergents, etc. (2). The United States Occupational Safety and Health Administration has set the occupational benzene exposure limit of 1 ppm (3), however, benzene exposure in excess of 100 ppm is still prevalent in the developing countries (4). High levels of benzene (>50 ppm) are also generated by tobacco products such as water pipes, cigars, pipe tobacco, and cigarettes (5, 6). Petroleum products and automobile exhaust also contain copious amount of benzene, especially near the emission source (7–9). Indoor sources of benzene include vapor or gases released by benzene containing products such as paints, furniture wax, and detergents (2). The atmospheric benzene exposure is likely to be higher in people living near gasoline refineries, petrochemical industries, gasoline fueling stations, and Superfund and other hazardous waste sites.
Excessive rates of type 2 diabetes and stroke have been found in an evaluation of 720,000 individuals living within a half-mile of 258 Superfund sites that were associated with excessive VOC (such as benzene and trichloroethylene) exposure (10). We observed that environmental benzene exposure is associated with increased CVD risk scores and augmented levels of sub-clinical markers of cardiovascular disease (11–14). Others have shown that benzene exposure increases the risk for arterial hypertension (15, 16), rhythm abnormalities (15), and heart failure(17). An assessment of excessive amount of VOC exposure and cardiovascular disease (CVD) mortality shows that in a single-pollutant model, benzene, propylene, and xylene are all significantly associated with CVD mortality (18). In a cohort study of intra-urban variation in VOCs and mortality, similar associations were found between CVD mortality and exposure to benzene, hexane, and total hydrocarbon (19). However, it is unclear whether benzene exposure is sufficient to cause cardiovascular disease or injury. Therefore, using a well-controlled mouse model, we systematically examined the effect of inhaled benzene exposure on biomarkers of cardiovascular toxicity.
MATERIALS AND METHODS
Murine Benzene Exposure
Seven-week-old male C57BL6/J mice were obtained from Jackson laboratories, Bar Harbor, ME. Mice were treated according to American Physiological Society Guiding Principles in the Care and Use of Animals, and all protocols were approved by University of Louisville Institutional Animal Care and Use Committee. Mice (n=24/group) were housed under pathogen-free conditions in the University of Louisville vivarium under controlled temperature and 12h light/12h dark cycle. Mice were maintained on a standard chow diet (Rodent Diet 5010, LabDiet, St. Louis, MO) containing 4.5% fat by weight). Starting at eight weeks of age mice were exposed to 50 ppm benzene (6 h/day, 5days/week) for 6 weeks as described before(20, 21). Mice exposed to HEPA-filtered air only served as a control. To examine the effect of benzene exposure on the susceptibility to inflammation, a sub-set of benzene and air-exposed mice (n=24/group) were treated with 0.5 mg/kg lipopolysaccharides (LPS, Sigma Cat# 2630, Lot# 028M4022V; i.p). At the end of the exposure protocol, mice were euthanized with sodium pentobarbital (150 mg/kg body weight; i.p.) and blood and tissues were harvested.
RNA-seq analysis
One μg of DNase-I-treated total RNA, isolated from the liver and lung tissues, was used for the cDNA library construction for poly-A RNA-seq at Novogene, Sacramento, CA using NEBNext Ultra II RNA Library Prep Kit for Illumina (New England BioLabs, #E7775) according to manufacturer’s protocol. After a series of terminal repair, poly-adenylation, and sequencing adaptor ligation, the double-stranded cDNA library was completed following size selection and PCR enrichment. The resulting 250-350 bp insert libraries were quantified using a Qubit 2.0 fluorometer (Thermo Fisher Scientific) and quantitative PCR. The size distribution was analyzed using an Agilent 2100 Bioanalyzer. Qualified libraries were sequenced on an Illumina Novaseq 6000 system using a paired-end 150 run (2×150 bases). A minimum of 20 million raw reads were generated from each library. The RNA-seq data generated in this study were deposited in the Gene Expression Omnibus (GSE).
For data analysis, FASTQ files were trimmed using fastp(22) (version 0.20.0) and the following parameters: --cut_by_quality5 --cut_by_quality3 --detect_adapter_for_pe --overrepresentation_analysis --correction --trim_front1 7 --trim_front2 7. The alignment to the mouse genome (GRCm38.90) was performed using STAR(23) (version 020201) and the following parameters: --runMode alignReads --runThreadN 8 --outSAMstrandField intronMotif -- outSAMmode Full --outSAMattributes All --outSAMattrRGline --outSAMtype BAM SortedByCoordinate --limitBAMsortRAM 45000000000 --quantMode GeneCounts -- outReadsUnmapped Fastx -outSAMunmapped within. Differential expression analysis was performed using edgeR(24) (version 3.10). The RLE (relative log expression) method was used to normalize the data. Gene ontology analysis was performed with the DAVID Bioinformatics Resources 6.8 (25). A Venn diagram was drawn using Bioinformatics & Evolutionary Genomics website (26). Heatmaps were created with Multiple Experiment Viewer (MEV) (27).
Bone marrow derived stem cells
Bone marrow cells were isolated from the femur and tibia and separated by Ficoll gradient. The cells were washed twice with PBS containing 1% BSA (PBS/BSA) and incubated with Fc Block (CD32/CD16) anti-mouse antibody for 10 minutes at 4°C to prevent non-specific binding. Samples were incubated (30 min, 4°C) with antibody cocktails containing Lineage-Pacific Blue, ckit-APC-Cy7, Sca-FITC, and CD34-Alexa Fluor 700 antibodies, and analyzed on an LSR II flow cytometer for 90 seconds on high speed. Cell populations were gated using the FlowJo software and normalized to the total number of cells.
Microparticles
Microparticles in the peripheral blood were measured as described before (28, 29) with slight modifications. Briefly, plasma was centrifuged for 2 min (11,000xg at 4°C) to remove residual cells and debris, and the supernatant was aspirated and centrifuged for 45 min (17,000xg at 4°C). The resulting microparticle pellet was resuspended in Annexin V Buffer pre-filtered through 0.22μm syringe filter and incubated with the anti-mouse FcBlock (CD32/CD16) for 10 minutes. Endothelial microparticles were stained with the antibody cocktail containing Annexin V-Pacific Blue, Flk-APC, Sca-PECy7, CD62E-PE and CD143-FITC for 30 min. Platelet microparticles were stained in a separate tube with Annexin V-Pacific Blue and platelet CD41-FITC antibody. Identical samples with no antibodies were utilized as controls for the gating. Counting beads, added to individual samples were used for data normalization. Samples were analyzed on BD LSR II flow cytometer for 5 min at low speed. Microparticle numbers were quantified in gated populations <1μm in size and positive for Annexin V staining using the FlowJo software. Microparticle subpopulations were further identified based on expression of various surface markers.
To examine the effect of benzene metabolites on endothelial cell apoptosis and microparticles formation in vitro, human aortic endothelial cells (HAEC) were incubated with hydroquinone - HQ, Catechol - Cat, and MA (10 μM each) for 18h, and the apoptosis was examined by western blotting using anti-cleaved-caspase-3, -cleaved caspase-7, -cleaved caspase-8, and cleaved caspase-9 antibody (Cell Signaling Technology, Danvers, MA). The microparticles (<1μm, Annexin V+) released in the cell culture medium were analyzed by flow cytometry.
Synthesis of trans,trans-Mucondialdehyde
trans,trans-Mucondialdehyde (MA) was prepared from muconic acid (Sigma-Aldrich, St. Louis, MO) by a recently developed one-pot acid-to-aldehyde reduction protocol (30)
Markers of endothelial function and inflammation
Levels of soluble adhesion molecules, markers of angiogenesis, and cytokines and chemokine in the plasma were measured by multiplex arrays at Eve Technologies (Calgary, Alberta, Canada).
Immune cells
Circulating immune cells were analyzed by flow cytometry as described before (28, 31). Briefly, lysed whole blood was centrifuged and washed twice with PBS containing 1% BSA (PBS/BSA). The cell pellets were re-suspended in the same buffer and incubated with CD32/CD16 for 10 min at 4°C to prevent unspecific binding. The cells were then incubated with an antibody cocktail consisting of FITC-anti-Nk1.1, PE-anti-Ly6C, PerCPe710-anti-CD8, PECy7-anti-CD62, APC-anti-CD19, Alexa 700-antiGr-1, APCe780-anti-CD3, eVolve605-CD11b, and e650-anti-CD4. After 30 min on ice, the cells were washed, re-suspended in PBS/BSA, and analyzed on an LSR II flow cytometer for 90 sec on high speed. Cell numbers were analyzed using the FlowJo software and normalized to the total leukocyte numbers.
Platelet-leukocyte aggregates
Platelet-leukocyte aggregates were identified by flow cytometry and quantified as events double positive for CD41 (platelets) and CD 45 (leukocytes) as described before(28).
Statistics
Data are expressed as mean ± standard error of mean (SEM). Statistical significance was accepted at P<0.05 level. Student’s two-tailed t test with unequal variance was used to compare the data sets.
RESULTS
Inhaled benzene exposure and endothelial microparticles formation
The endothelium is a critical regulator of vascular homeostasis, vascular tone, angiogenesis, and thrombosis. Our recent studies demonstrate that exposure to benzene depletes circulating endothelial progenitor cells (EPCs, also known as circulating angiogenic cells) in human and mice (11, 14). Mobilization of EPCs from the bone marrow and their homing to the injury sites can be affected by exogenic factors such as aging, disease, and an unhealthy lifestyle (32–36), and therefore changes in EPC levels are reflective of endothelial health. A decrease in the levels of blood EPCs reflects endothelial injury and impaired repair. To assess the benzene exposure-induced endothelial toxicity, we measured circulating endothelial microparticles. Endothelial microparticles, 0.1-1.0 μm vesicles shed from activated or injured cells, are surrogate markers of endothelial activation and injury and comprise 5-15% of microparticles in the blood. Circulating endothelial microparticles are positively associated with coronary artery disease and stroke (37, 38). We observed that inhaled benzene exposure significantly increases the levels of circulating endothelial microparticles (<1μm; Annexin V+/CD41-/Flk+), activated endothelial microparticles (<1μm; Annexin V+/CD41-/CD62E+ [E-selectin]), EPC-derived microparticles (<1μm; Annexin V+/CD41-/Flk+/Sca+), lung endothelial microparticles (<1μm; Annexin V+/CD143+), and activated lung endothelial microparticles (<1μm; Annexin V+/CD143+/CD62E+) by 1.8-3.8-fold (Fig. 1). To assess the effect of inhaled benzene exposure on endothelial activation we measured the circulating levels of soluble adhesion molecules. Our data showed that benzene exposure modestly decreased the levels of soluble intra cellular adhesion molecule-1 (sICAM-1; Supplemental Table 1), whereas other soluble adhesion molecules - platelet endothelial cell adhesion molecule-1 (sPECAM-1), endothelial selectin (sE-selectin), and platelet selectin (sP-selectin) were comparable with air-exposed controls. Together these data suggest that inhaled benzene exposure does not induce endothelial activation and most of the endothelial microparticles are derived from benzene-induced endothelial injury. Since benzene inhalation also depletes circulating angiogenic cells mice (11, 14), we also quantified circulating angiogenesis markers. However, our data show that levels of angiogenesis markers in benzene-exposed mice are comparable to the air-exposed controls (Supplemental Table 1).
Because toxicity of benzene is mediated by its metabolism to reactive metabolites, next we directly examined the effect of benzene metabolites hydroquinone, catechol, and MA on HAEC apoptosis. As shown in Fig. 2, hydroquinone and catechol only modestly increased the activation of the pan apoptosis marker caspase-7 in HAEC, whereas MA profoundly increased caspase-7 cleavage. MA also robustly increased caspase-3 activation, suggesting that it is the most toxic benzene metabolite for endothelial cells. To examine the mechanisms by which MA exerts its toxicity, we measured the activation of caspase-8 and caspase-9. As shown in Fig. 2, MA had no effect on caspase-8, but robustly activated caspase-9. Together, these data suggest that MA doesn’t affect the extrinsic pathway of apoptosis, but selectively activates the intrinsic pathway.
Inhaled benzene exposure and hematopoietic progenitor cells
Benzene is a well-known hematopoietic toxin (39). Metabolites of benzene such as hydroquinone, catechol, and MA can diffuse from their sites of generation and exert the toxicity at distal sites. We observed that under our experimental conditions, benzene exposure did not affect the levels of common myeloid progenitor (CMPC) and multipotent progenitor cells (MPC) in the bone marrow but significantly decreased the levels of hematopoietic progenitor cells (HPC; Fig. 3). Because the vascular niche of HPC is critical for hematopoiesis and endothelial cell materialization, benzene-inhalation-induced depletion of HPC could affect EPC formation in the bone marrow and compromise endothelial repair.
Inhaled benzene exposure and platelet activation
The surface of quiescent endothelial cells (luminal surface) which face the blood is normally anti-adhesive. However, injury to the endothelial cells promotes platelet adhesion for repair. We observed that inhaled benzene exposure augments platelet-leukocyte aggregate formation by 3-fold (Fig. 4). This was accompanied by 1.6-fold increase in circulating levels of platelet microparticles in benzene-exposed mice. Together these data suggest that benzene exposure enhances platelet activation and platelet-derived microparticles could serve as a biomarker of pro-thrombotic response of inhaled benzene.
Inhaled benzene exposure and inflammatory markers
In humans, polymorphism of cytokines and endothelial activation markers increases the susceptibility to benzene-induced hematopoietic toxicity (40). We have recently shown that in addition to depleting circulating EPCs, benzene inhalation also suppresses the levels of leukocytes, lymphocytes, monocytes, and neutrophils in the peripheral blood in mice (11). However, our flow cytometric analysis of T-lymphocytes shows that inhaled benzene exposure modestly increases the circulating levels of CD3+, CD4+ and CD8+ T-cells (Fig 5). Blood CD19+ B cells, NK1.1+ natural killer cells, Gr1+ granulocytes and Ly6C+ monocytes (Fig. 5) in benzene-exposed mice were comparable to the corresponding air-exposed controls. Quantitation of plasma cytokines showed that IL-6 levels were significantly lower in benzene-exposed mice. All the other circulating cytokines in the benzene-exposed mice were comparable with the air-exposed controls (Supplemental Table 1). Stimulation with low dose LPS, 18h before euthanasia, significantly increased the levels of cytokines and chemokines such as GM-CSF, IL-6, IL-10, KC, MCP-1 etc. in the peripheral blood. However, benzene exposure did not affect the LPS-induced cytokine formation.
Inhaled benzene exposure and pulmonary and hepatic metabolism
While benzene is primarily metabolized in the liver, lungs are the first target of inhaled benzene. We, therefore, examined gene transcription in the liver and the lungs of benzene exposed mice. Although six weeks of benzene exposure did not affect the expression of CYP2E1, which plays a pivotal role in benzene metabolism, benzene exposure significantly down-regulated 185 genes in the lungs and 29 genes in the liver, whereas transcription of 301 genes was increased in the lungs and 43 genes in the liver (>1.5-fold and P<0.05, Fig 6). The Heat map of genes associated with cardiometabolic toxicity showed the suppression of cytochrome P450s and up-regulation of inflammatory genes in the lungs and the liver, increased transcription of glycolysis associated genes in the lungs, and induction of the transcription of genes associated with oxidative stress, endothelial activation, and platelet activation in the liver (Fig. 6). Gene ontology analysis showed strong association with the lipid metabolic process, cardiac contractility genes, and keratinization in the lungs, and activation of NF-kB, inflammatory response, leukocyte cell-cell adhesion and apoptosis in the liver (Fig. 6). These observations are consistent with benzene-induced endothelial apoptosis and our recent studies demonstrating that the benzene-induced insulin resistance is mediated by NF-kB activation, inflammatory signaling, and oxidative stress in the liver (12).
DISCUSSION
The major findings of this study are that benzene exposure induces endothelial injury and augments platelet activation, as assessed by a panel of blood endothelial cell and platelet microparticles and platelet-leukocyte adduct formation. This was accompanied by the differential regulation of genes associated with xenobiotic metabolism, endothelial function, platelet activation, and inflammatory signaling associated genes in the liver and the lung, suppression of hematopoietic progenitor cells in the bone marrow, and increase in the levels of T-cells in the peripheral blood.
Although little is known about the direct effect of VOCs such as benzene on vascular injury and thrombosis, the endothelium has been shown to be particularly vulnerable to the effects of tobacco smoke which contains high levels of benzene and other VOCs. In smokers, endothelial dysfunction is the most primitive sign of injury and precedes morphological changes in the vessel wall (41). A dysfunctional endothelium affects vascular homeostasis, blood pressure regulation, thrombosis, atherogenesis, plaque stability, and cardiac functions (42, 43). To examine the effect of benzene exposure on endothelial changes, we measured the levels of microparticles that are released from activated or apoptotic endothelial cells (44) and are a sensitive index of vascular injury (45, 46). Increased levels of circulating endothelial microparticles correlate with endothelial dysfunction in patients with coronary artery disease (36), end stage renal failure (47), obesity (48), and type-2 diabetes (37). Augmented activated endothelial microparticles in the blood are associated with cardiovascular events (49), and enhanced plasma lung endothelial microparticle levels in healthy smokers precede changes in pulmonary function (50). Our data demonstrating that benzene exposure increases the circulating levels of endothelial microparticles, activated endothelial microparticles, EPC microparticles, lung endothelial microparticles and activated lung endothelial microparticles, suggest that these microparticles are sensitive and robust surrogate markers of benzene-induced endothelial injury.
Increased circulating endothelial microparticles have also been observed in humans following episodic fine particulate matter exposure (29). Nonetheless, unlike murine exposure to benzene, fine particulate matter exposure in humans did not increase blood activated endothelial microparticles, suggesting that activated endothelial cells are more sensitive to benzene exposure than fine particulate matter.
Endothelial microparticles contain Von Willebrand factor and factor VIII which promote platelet activation (51, 52). Therefore, the observed increase in platelet-leukocyte adduct formation in benzene-exposed mice could be secondary to benzene-induced endothelial injury and microparticle formation. Moreover, hypercholesterolemia following benzene exposure (11) could also augment platelet activation. Induction of thromboxane A synthase 1 (Tbxas1) in the liver of benzene-exposed mice also corroborates hyper platelet activation, whereas hepatic induction of guanylate cyclase soluble subunit β-1 (Gucyb1), the receptor of nitric oxide, could reflect an adaptive response to benzene exposure-induced platelet activation.
MA-induced endothelial cell apoptosis and endothelial microparticle formation suggest that the observed toxicity of benzene is likely to be mediated by its reactive metabolites. Hepatic induction of the orphan nuclear receptor Nr4a1 (Nurr77), a molecular regulator of apoptosis and inflammation (53–59), in benzene-exposed mice further support that benzene exposure affects apoptotic and inflammatory processes. Although the contribution of benzene-induced Nr4a1 transcription in endothelial toxicity is unknown, Nr4a1 has been suggested to prevent TNFα and IL-1β-induced endothelial activation (60), and endothelial deficiency of Nr4a1suppresses oxLDL-induced apoptosis (57). Unlike Nr4a1, benzene exposure suppressed the hepatic expression of Snai2, a transcription factor involved in the endothelial to mesenchymal transition and implicated in pathological angiogenesis and atherosclerosis (61, 62). Further studies are required to examine the contribution of Snai2 in benzene-induced endothelial toxicity.
Benzene-induced increase in the expression of Thromboxane A synthase 1 (Tbxas1) in the liver corroborate benzene-induced platelet activation. Because thromboxanes play a critical role in modulating vasoconstriction and platelet aggregation, increased formation of thromboxanes can disrupt vascular homeostasis and promote thrombotic vascular events. Increased hepatic transcription of guanylate cyclase 1 soluble subunit beta 1(Gucy1b1) could and guanylate cyclase 1 soluble subunit beta 1(Gucy1b1), the receptor for nitric oxide, could be an adaptive response to mitigate benzene-induced pro-thrombotic responses. However, additional studies are required to examine which cells in the liver induce Tbxas1 and Gucy1b1 and how do these proteins affect benzene-induced vascular homeostasis and thrombosis. Likewise, additional studies are also required to examine the contribution of benzene inhalation-induced transcription of an array of inflammatory genes in the lungs and the liver on endothelial toxicity and platelet activation.
Together, these studies suggest that inhaled benzene exposure induces endothelial injury and affects platelet activation and inflammatory processes. Because benzene is a pervasive and abundant air pollutant, decreasing its exposure can significantly reduce air pollution-induced cardiovascular disease.
Funding
This study was supported in parts by NIH grants P42 ES023716, R01 HL149351, R01 HL137229, R01 HL146134, R01 HL156362, R01 HL138992, R01 ES029846, R21 ES033323, U54 HL120163, and the Jewish Heritage Foundation grant OGMN190574L.
Footnotes
Conflict of interest statement: The authors have declared that no conflict of interest exists