Abstract

Thiocyanate (SCN−) is, unexpectedly, the principal physiologic substrate for eosinophil peroxidase (EPO) and a major (i.e., accounting for 50% of H2O2 consumed) substrate for myeloperoxidase (MPO). The product of these reactions is HOSCN, a weak, exclusively sulfhydryl-reactive oxidant that we have previously shown to be a uniquely potent (up to 100-fold) oxidant transcriptional inducer of human umbilical vein endothelial cell (HUVEC) tissue factor (TF), ICAM-1, E-selectin, and VCAM-1 expression via a mechanism dependent upon NF-κB p65/p50 activation. Histone deacetylase inhibitors (HDACi) have recently been found capable of gene-specific transcriptional regulation by acetylating lysine residues on transcription factors, including p65. Because of previous reports of beneficial effects of the HDACi trichostatin (TSA) in vivo in murine models of SLE and asthma inflammation, we hypothesized that TSA might exert antiinflammatory effects by downregulating endothelial expression of proinflammatory mediators induced by physiologic agonists such as HOSCN. We analyzed the effects of HOSCN and TSA on the HUVEC transcriptome by incubating HUVEC monolayers (n=3, single donor-derived preparations) 4 hours in M199 medium containing 10% FCS supplemented with buffer control, 150 μM HOSCN, 100 nM TSA, or HOSCN + TSA prior to total RNA extraction and analysis using Affymetrix® U133 2.0 Plus Microarray Chips. HOSCN significantly (i.e., > 2-fold) upregulated 0.5% of HUVEC genes, but most strikingly stimulated (10–100 fold) the adhesion molecules VCAM-1, ICAM-1, and E-selectin, chemokines IL-8, MCP-1, CXCL-1 and CXCL2 and COX2, all NF-kB – regulated genes. TSA suppressed 0.03% of genes in HOSCN-treated HUVEC, including, notably, VCAM-1 (but not ICAM-1), IL-8 and CCL2, and COX2. Pursuing the discordant effects of TSA upon HOSCN-stimulated VCAM-1 and ICAM-1 mRNA levels, we confirmed by qRT-PCR that HOSCN increases VCAM-1 mRNA 10–16x at 3–4 h and TSA inhibits this by 85%. In contrast, ICAM-1 mRNA increases 7x in response to HOSCN but is stimulated another 4-fold by TSA. We assessed HUVEC expression of VCAM-1 and ICAM-1 protein by western blot after a 4h exposure to 150 μM HOSCN and found them upregulated 10- and 5-fold, respectively. TSA (ED50 30 nM) suppressed HOSCN-mediated VCAM-1 expression by >90% but increased expression of ICAM-1 2–3x. EMSA and anti-p50 and anti-p65 supershift confirmed HOSCN activation of p65/p50 binding to VCAM-1 and ICAM-1 sequence-derived NF-kB motif oligo probes but, paradoxically, TSA inhibited p65/p50 binding to both VCAM-1 and ICAM-1 probes. In direct contrast, chromatin immunoprecipitation using anti-p65 showed that TSA decreased HOSCN-induced p65 binding to the endogenous HUVEC genomic DNA VCAM-1 NF-kB binding site but did not diminish its binding to the ICAM-1 site. In a static adhesion assay human eosinophils bound to HUVEC exposed 4h to 150 μM HOSCN increased to 4x baseline and 1 μM TSA completely blocked this increase whereas a blocking anti VCAM-1 antibody diminished it by 50%. We conclude that the HDACi TSA is a potent and relatively specific inhibitor of several NF-kB-dependent pro-inflammatory genes, but not ICAM-1, in HUVEC activated by the physiologic oxidant agonist HOSCN. We hypothesize that differential TSA regulation of VCAM-1 and ICAM-1 may be attributable to variations in the nucleotide sequence of their NF-kB-binding motifs or, alternatively, differential recruitment of NFkB transcriptional cofactors such as p300/CBP. We propose that HDACi have significant therapeutic potential as anti-inflammatory agents, particularly in those disease states most dependent upon VCAM-1.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author