Interleukin-2 alters distribution of CD144 (VE-cadherin) in endothelial cells

Abstract

Background: High-dose IL-2 (HDIL2) is approved for the treatment of metastatic melanoma and renal cell carcinoma, but its use is limited in part by toxicity related to the development of vascular leak syndrome (VLS). Therefore, an understanding of the mechanisms that underlie the initiation and progression of HDIL2-induced increases in endothelial cell (EC) permeability leading to VLS are of clinical importance.

Methods: We established a novel ex vivo approach utilizing primary human pulmonary microvascular ECs to evaluate EC barrier dysfunction in response to IL-2.

Results: Complementary in vitro studies using exogenous IL-2 and ex vivo studies using serum from patients treated with IL-2 demonstrate that HDIL2 induces VLS through CD144 (vascular endothelial (VE)-cadherin) redistribution.

Conclusions: These findings provide new insight into how IL-2 induces VLS and identifies VE-cadherin as a potential target for preventing IL-2-related VLS.

Figure 1

IL-2 induces an increase in EC permeability. A) IL-2 receptor α, β, and γ subunit levels as determined using RT-PCR. Lane 1: reference cDNA, lane 2: cDNA isolated from peripheral blood mononuclear cells, lanes 3–5: cDNA isolated from ECs from three different donors. Data are representative of five experiments with similar results. B) Expression of IL-2 receptor α, β, and γ subunits on the surface of ECs as examined using flow cytometry. Specific staining was compared with background staining using an isotype control antibody. Data represent three experiments with similar results. C) As a positive control, the effect of TNF-α (20 ng/ml) or vehicle [PBS] on EC permeability was measured by determining the flux of dextran across ECs 24 hours after treatment. D) ECs were treated with 100 IU/ml IL-2 or without [Vehicle: PBS], and the flux of dextran across ECs was measured 24 hours after treatment. Data in (C-D) are presented as mean + S.E.M. (n = 4 in each group), are representative of three experiments with similar results, and *p < 0.05 examined using ANOVA.

Figure 2

IL-2 enhances VE-cadherin endocytosis. A) ECs were treated with Vehicle (PBS without IL-2), 100 IU/ml IL-2, or 100 IU/ml IL-2 in the presence of a functional blocking anti-IL-2 antibody for 24 hr. VE-cadherin and F-actin were examined using confocal microscopy. The effect of TNF-α was examined as a control. B) Data from (A) were quantified using imagej64 software (NIH). VE-cadherin and F-actin intracellular redistribution (i.e., intracellular intensity in arbitrary units) is shown. Gray line indicates image background intensity. C) ECs were treated with Vehicle (PBS without IL-2) or 100 IU/ml IL-2 for 24 hrs. Cell surface VE-cadherin was labeled at 4°C to inhibit endocytosis. Cells were then incubated at 37°C for 1 hr to initiate endocytosis. VE-cadherin was examined using confocal microscopy. Data shown represent two independent experiments.

Figure 3

Serum IL-2 results in altered CD144 (VE-cadherin) distribution in the EC cytoskeleton. A) Primary human pulmonary microvascular ECs were treated with paired pre- or post-IL-2 serum, and the rate of dextran flux across an EC monolayer was measured after 24 hr of incubation. Individual data points and mean values ± S.E.M. are shown in the graph. * p was determined using paired t-test. B) Pre- and post-IL-2 serum was cultured with a functional blocking anti-IL-2 antibody or an isotype control IgG for 30 minutes and then used to treat ECs for 24 hr. The effect on the distribution of VE-cadherin and F-actin was examined using confocal microscopy. An image in which Patient 5 serum was used is shown and is representative of 4 patients tested. C) Data from (B) were quantified using imagej64 software (NIH). VE-cadherin and F-actin intracellular redistribution (i.e., intracellular intensity in arbitrary units) is shown. Gray line indicates image background intensity. Data represent two independent experiments.