Inflammation determines the pro-adhesive properties of high extracellular d-glucose in human endothelial cells in vitro and rat microvessels in vivo

Abstract

Background: Hyperglycemia is acknowledged as an independent risk factor for developing diabetes-associated atherosclerosis. At present, most therapeutic approaches are targeted at a tight glycemic control in diabetic patients, although this fails to prevent macrovascular complications of the disease. Indeed, it remains highly controversial whether or not the mere elevation of extracellular D-glucose can directly promote vascular inflammation, which favors early pro-atherosclerotic events.

Methods and findings: In the present work, increasing extracellular D-glucose from 5.5 to 22 mmol/L was neither sufficient to induce intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) expression, analyzed by flow cytometry, nor to promote leukocyte adhesion to human umbilical vein endothelial cells (HUVEC) in vitro, measured by flow chamber assays. Interestingly, the elevation of D-glucose levels potentiated ICAM-1 and VCAM-1 expression and leukocyte adhesion induced by a pro-inflammatory stimulus, such as interleukin (IL)-1beta (5 ng/mL). In HUVEC, high D-glucose augmented the activation of extracellular signal-regulated kinase 1/2 (ERK 1/2) and nuclear transcription factor-kappaB (NF-kappaB) elicited by IL-1beta, measured by Western blot and electromobility shift assay (EMSA), respectively, but had no effect by itself. Both ERK 1/2 and NF-kappaB were necessary for VCAM-1 expression, but not for ICAM-1 expression. In vivo, leukocyte trafficking was evaluated in the rat mesenteric microcirculation by intravital microscopy. In accordance with the in vitro data, the acute intraperitoneal injection of D-glucose increased leukocyte rolling flux, adhesion and migration, but only when IL-1beta was co-administered.

Conclusions: These results indicate that the elevation of extracellular D-glucose levels is not sufficient to promote vascular inflammation, and they highlight the pivotal role of a pro-inflammatory environment in diabetes, as a critical factor conditioning the early pro-atherosclerotic actions of hyperglycemia.

Figure 1. Effect of IL-1β and D-glucose on the expression of adhesion molecules in HUVEC.

Cells incubated in a medium containing 5.5 mmol/L D-glucose were challenged for 18 h with IL-1β (0.1 to 10 ng/mL) and the levels of (A) ICAM-1 and (B) VCAM-1 were determined at the cell surface by flow cytometry. In another set of experiments, cells were incubated for 18 h in medium containing 5.5 or 22 mmol/L D-glucose in the presence or absence of IL-1β (5 ng/mL), after which (C) ICAM-1 and (D) VCAM-1 levels were quantified. L-glucose was used as an osmotic control. *P<0.05 versus 5.5 mmol/L D-glucose without IL-1β; †P<0.05 versus 5.5 mmol/L D-glucose with IL-1β. (E) ICAM-1 (left panel) and VCAM-1 (right panel) were visualized by indirect immunofluorescence in HUVEC cultured for 18 h in culture medium containing 5.5 or 22 mmol/L D-glucose in the presence or absence of IL-1β (5 ng/mL). Nuclei were counterstained in blue with 4′-6-diamidino-2-phenylindole (DAPI) (x400).

Figure 2. ERK 1/2 activation and its impact on ICAM-1 and VCAM-1 levels in HUVEC.

(A) Cells were incubated in medium containing 5.5 mmol/L or 22 mmol/L D-glucose with or without IL-1β (5 ng/mL) for 5–60 min, after which ERK 1/2 activation was determined by Western blotting. Representative gels are shown on the top. (B) Involvement of ERK 1/2 in CAMs expression. HUVEC were cultured for 18 h in the above-described conditions. PD 98059 (30 µmol/L) was used as an inhibitor of ERK 1/2 activation. *P<0.05 versus 5.5 mmol/L D-glucose; †P<0.05 versus 5.5 mmol/L D-glucose in the presence of IL-1β; ‡P<0.05 versus matched treatment without PD 98059.

Figure 3. NF-κB activation in HUVEC and its impact on ICAM-1 and VCAM-1 levels.

(A) HUVEC were incubated in medium containing 5.5 mmol/L or 22 mmol/L D-glucose in the presence or absence of IL-1β (5 ng/mL) during 1, 4, 6 and 18 h, after which NF-κB binding activity was quantified by EMSA. Representative EMSAs are shown on the left. (B) Translocation of NF-κB from cytoplasm to nucleus was visualized by indirect immunofluorescence in HUVEC cultured for 1 h as mentioned above (x1000). (C) Involvement of NF-κB in CAMs expression. HUVEC were cultured for 18 h in the above-mentioned conditions. PDTC (100 µmol/L) was used as NF-κB inhibitor. *P<0.05 versus 5.5 mmol/L D-glucose; †P<0.05 versus 5.5 mmol/L D-glucose with IL-1β; ‡ P<0.05 versus matched treatment without PDTC.

Figure 4. Adhesion of HL60 leukocytes to HUVEC under flow conditions in vitro.

HUVEC monolayers were exposed to either 5.5 or 22 mmol/L extracellular D-glucose in the presence or absence of IL-1β (5 ng/mL) for 18 h prior to leukocyte perfusion. *P<0.05 versus 5.5 mmol/L D-glucose without IL-1β; †P<0.05 versus IL-1β in the presence of 5.5 mmol/L D-glucose. Representative micrographs showing HL60 adhesion to HUVEC monolayers are shown on the top (x200).

Figure 5. In vivo leukocyte trafficking in rat mesenteric post-capillary venules.

Animals were i.p. injected with 10 mL of PBS alone (C) or supplemented with D-glucose (DG; 40 mg/kg), either in the absence or presence of IL-1β (200 ng/kg). L-glucose (LG; 40 mg/kg) was used as an osmotic control. After 18 h, (A) leukocyte rolling flux, (B) leukocyte rolling velocity, (C) leukocyte adhesion and (D) leukocyte migration were determined by intravital microscopy. *P<0.05 versus matched group without IL-1β; †P<0.05 versus IL-1β in PBS. Panel E shows representative photomicrographs showing ICAM-1 and VCAM-1 immunolocalization and leukocyte adhesion and transmigration in post-capillary venules. The brown reaction product indicates positive staining. Adhered and transmigrated leukocytes are marked with arrows in the right column. All panels are lightly counterstained with hematoxylin (x400).