Trauma promotes heparan sulfate modifications and cleavage that disrupt homeostatic gene expression in microvascular endothelial cells

Abstract

Introduction: Heparan sulfate (HS) in the vascular endothelial glycocalyx (eGC) is a critical regulator of blood vessel homeostasis. Trauma results in HS shedding from the eGC, but the impact of trauma on HS structural modifications that could influence mechanisms of vascular injury and repair has not been evaluated. Moreover, the effect of eGC HS shedding on endothelial cell (EC) homeostasis has not been fully elucidated. The objectives of this work were to characterize the impact of trauma on HS sulfation and determine the effect of eGC HS shedding on the transcriptional landscape of vascular ECs. Methods: Plasma was collected from 25 controls and 49 adults admitted to a level 1 trauma center at arrival and 24 h after hospitalization. Total levels of HS and angiopoietin-2, a marker of pathologic EC activation, were measured at each time point. Enzymatic activity of heparanase, the enzyme responsible for HS shedding, was determined in plasma from hospital arrival. Liquid chromatography-tandem mass spectrometry was used to characterize HS di-/tetrasaccharides in plasma. In vitro work was performed using flow conditioned primary human lung microvascular ECs treated with vehicle or heparinase III to simulate human heparanase activity. Bulk RNA sequencing was performed to determine differentially expressed gene-enriched pathways following heparinase III treatment. Results: We found that heparanase activity was increased in trauma plasma relative to controls, and HS levels at arrival were elevated in a manner proportional to injury severity. Di-/tetrasaccharide analysis revealed lower levels of 3-O-sulfated tetramers with a concomitant increase in ΔIIIS and ΔIIS disaccharides following trauma. Admission levels of total HS and specific HS sulfation motifs correlated with 24-h angiopoietin-2 levels, suggesting an association between HS shedding and persistent, pathological EC activation. In vitro pathway analysis demonstrated downregulation of genes that support cell junction integrity, EC polarity, and EC senescence while upregulating genes that promote cell differentiation and proliferation following HS shedding. Discussion: Taken together, our findings suggest that HS cleavage associated with eGC injury may disrupt homeostatic EC signaling and influence biosynthetic mechanisms governing eGC repair. These results require validation in larger, multicenter trauma populations coupled with in vivo EC-targeted transcriptomic and proteomic analyses.

Keywords: angiopoietin-2; endotheliopathy; glycocalyx; heparanase; sulfatase; sulfotransferase; transcriptome; vascular endothelium.

FIGURE 1

Heparan sulfate levels, heparanase activity, and relative abundance of heparan sulfate di-/tetrasaccharides in plasma from adult trauma subjects relative to healthy controls. (A) Circulating levels of heparan sulfate, as measured by ELISA, in trauma subjects at the time of hospital arrival (n = 49) and 24 h after hospitalization (n = 48) relative to healthy controls (n = 25). Comparisons were performed using the Kruskal-Wallis 1-way ANOVA followed by Dunn’s multiple comparisons test. *p < 0.05. (B) Heparanase activity, as measured using homogeneous time resolved fluorescence, in plasma from 10 randomly selected trauma subjects at the time of hospital arrival relative to 10 healthy controls. Comparison was made using the Mann-Whitney U test. **p < 0.01. (C) Relative abundance of heparan sulfate disaccharides or tetrasaccharides (resistant to heparin lyase I, II digestion), as measured using liquid chromatography-tandem mass spectrometry, in plasma from trauma subjects at hospital arrival (n = 20) and 24 h after hospitalization (n = 19) relative to healthy controls (n = 10). Comparisons were performed using the Kruskal-Wallis 1-way ANOVA followed by Dunn’s multiple comparisons test. *p < 0.05; **p < 0.01; ****p < 0.0001.

FIGURE 2

Spearman’s rank correlations of plasma levels of heparan sulfate at arrival with plasma levels of angiopoietin-2 measured 24 h following hospital admission in adults who suffered a traumatic injury. Total levels heparan sulfate (A) and relative abundance of the ∆IIIS heparan sulfate disaccharide (B) in plasma at hospital arrival correlated with angiopoietin-2 levels 24 h after hospitalization. Conversely, the representation of the heparan sulfate tetrasaccharides tetra-2 (C) and tetra-4 (D) inversely correlated with 24-h levels of angiopoietin-2 in circulation following trauma.

FIGURE 3

Top differentially expressed genes in flow conditioned primary human lung microvascular endothelial cells (HLMVEC) following exposure to heparinase III (HepIII) relative to vehicle control. Messenger RNA was collected from confluent monolayers of HLMVEC that were conditioned with 15 dyn/cm² for 48 h followed by exposure to heparinase III 500 mU/mL or vehicle for 6 h while remaining under shear stress (n = 4 biological replicates per condition; two replicates were pooled to generate two samples per condition for RNAseq). (A) Heatmap representing top 40 differentially expressed genes in HLMVEC between heparinase III and vehicle. Each treatment group contains n = 2 RNA samples that were combined from HLMVEC within two ibidi channel slides, thus representing a total of n = 4 per condition. Colors represent gene expression z-score with red corresponding to upregulated and blue to downregulated. (B) Volcano plot depicting differential gene expression between HLMVEC exposed to heparinase III (positive log2 fold change) and vehicle (negative log2 fold change). Red genes meet figure thresholds of p ≤ 1 × 10⁻³ and log2 fold change ≥|1| for the purposes of visualization. (C) Expression of the flow-responsive genes Krüppel-like factor 2 and 4 (KLF2,4), endothelial nitric oxide synthase (NOS3) and solute carrier family nine isoform A3 regulatory factor 2 (SLC9A3R2) is reduced following heparinase III treatment. (D) Expression of angiopoietin-2 (ANGPT2), endothelial cell-specific molecule-1 (ESM1, also known as endocan), and thrombospondin (THBS1), markers of endothelial cell activation, is increased following heparinase III treatment.

FIGURE 4

Targeted representation of gene set enrichment analysis (GSEA) in flow conditioned (15 dyn/cm² for 48 h) primary human lung microvascular endothelial cells (HLMVEC) after 6-h exposure to vehicle or heparinase III (HepIII, 500 mU/mL) while remaining under shear stress (n = 4 biological replicates per condition; two replicates were pooled to generate two samples per condition for RNAseq). GSEA was performed using (A) GO: Biological Process and (B) KEGG datasets. Figure displays up to twenty pathways from GSEA that are most relevant to endothelial cell organization and function with lowest False discovery rate (FDR)-adjusted p values (FDR q value). Pathways were organized according to their contribution to cellular maintenance and bioenergetics; cell organization and adhesion; angiogenesis and wound healing; or response to biophysical cues. Pathways on presented on the left were enriched in HLMVEC after exposure to vehicle whereas pathways on the right were enriched in HLMVEC after exposure to heparinase III. Circle size corresponds with number of genes present in experimental samples that overlap with respective dataset pathways, and circle shading represents the −log10 (FDR q value) with darker shades representing lower q values.

FIGURE 5

Differentially expressed genes that govern synthesis of heparan sulfate proteoglycans and glycosaminoglycans in flow conditioned (15 dyn/cm² for 48 h) primary human lung microvascular endothelial cells treated for 6 h with vehicle or heparinase III (HepIII, 500 mU/mL) while remaining under shear stress (n = 4 biological replicates per condition; two replicates were pooled to generate two samples per condition for RNAseq). (A) Of the heparan sulfate proteoglycans found in the vascular endothelial apical glycocalyx, expression of syndecan 3 (SDC3) and SDC4 were downregulated by heparinase III treatment. (B) Of the enzymes regulating hyaluronan expression in the endothelial glycocalyx, hyaluronan synthase isoform 2 (HAS2) was upregulated while hyaluronidases 1 and 2 (HYAL1,2) were downregulated by heparinase III treatment. (C) Of the enzymes that synthesize chondroitin sulfate expressed in the endothelial glycocalyx (commonly observed in SDC1 and SDC3) and that modify its sulfation, chondroitin sulfate synthase isoform 3 (CHSY3) and chondroitin sulfate N-acetylgalactosaminylsulfotransferase isoform 1 (CSGALNACT1) were upregulated while carbohydrate sulfotransferase isoform 15 (CHST15, catalyzing 6-O-sulfation of 4-O-sulfated N-acetylgalactosamine in chondroitin sulfate disaccharides) was downregulated following heparinase III treatment. (D) Of the enzymes that synthesize and modify heparan sulfate expressed in the endothelial glycocalyx, expression of N-deacetylase/N-sulfotransferase isoform 1 (NDST1) and glucuronic acid C5-epimerase (GLCE) (which also contributes to glucuronic acid epimerization to iduronic acid in chondroitin sulfate) were downregulated while heparan sulfate 3-O-sulfotransferase isoform 1 (HS3ST1) and heparan sulfate 6-O-sulfotransferase isoform 3 (HS6ST3) were upregulated following heparinase III treatment. We also found that heparinase III treatment suppressed heparanase (HPSE) expression. False discovery rate (FDR)-adjusted p values (FDR q values) are presented.