Size and targeting to PECAM vs ICAM control endothelial delivery, internalization and protective effect of multimolecular SOD conjugates

Abstract

Controlled endothelial delivery of SOD may alleviate abnormal local surplus of superoxide involved in ischemia-reperfusion, inflammation and other disease conditions. Targeting SOD to endothelial surface vs. intracellular compartments is desirable to prevent pathological effects of external vs. endogenous superoxide, respectively. Thus, SOD conjugated with antibodies to cell adhesion molecule PECAM (Ab/SOD) inhibits pro-inflammatory signaling mediated by endogenous superoxide produced in the endothelial endosomes in response to cytokines. Here we defined control of surface vs. endosomal delivery and effect of Ab/SOD, focusing on conjugate size and targeting to PECAM vs. ICAM. Ab/SOD enlargement from about 100 to 300nm enhanced amount of cell-bound SOD and protection against extracellular superoxide. In contrast, enlargement inhibited endocytosis of Ab/SOD and diminished mitigation of inflammatory signaling of endothelial superoxide. In addition to size, shape is important: endocytosis of antibody-coated spheres was more effective than that of polymorphous antibody conjugates. Further, targeting to ICAM provides higher endocytic efficacy than targeting to PECAM. ICAM-targeted Ab/SOD more effectively mitigated inflammatory signaling by intracellular superoxide in vitro and in animal models, although total uptake was inferior to that of PECAM-targeted Ab/SOD. Therefore, both geometry and targeting features of Ab/SOD conjugates control delivery to cell surface vs. endosomes for optimal protection against extracellular vs. endosomal oxidative stress, respectively.

Keywords: Cell adhesion molecules; Endocytosis; Endothelial cells; Intracellular delivery; Vascular immunotargeting.

Figure 1. Fractionation of PECAM-targeted Ab/SOD conjugates, endothelial uptake and effect of non-fractionated vs. fractionated conjugates

(A-B). TEM of total unfractionated (A) and supernatant fraction (B) of PECAM-targeted Ab/SOD conjugate preparation with mean size of 250 nm. Scale bar is 200 nm. (C). Representative curves of size distribution of non-fractionated Ab/SOD (red) and supernatant fraction of Ab/SOD (green).. (D-E). Endothelial binding and internalization of non-fractionated Ab/SOD and supernatant fraction of Ab/SOD. Green and yellow colors indicate internalized and surface-bound conjugates, respectively. Nuclei are stained with DAPI (shown in blue). (F). Western blot analysis shows that supernatant fraction of Ab/SOD inhibits TNF-induced VCAM expression in HUVEC more effectively than non-fractionated conjugates. (G). Comparative summary of endothelial binding, internalization and relative inhibition of TNF-induced VCAM-1 expression calculated as inhibition rate to binding ratio (Effect, I/B; supernatant set as 100 %) for non-fractionated Ab/SOD vs. supernatant Ab/SOD (red and blue bars, respectively). *P<0.05 supernatant vs. non-fractionated.

Figure 2. Synthesis of Ab/SOD conjugates with controlled sizes

(A). TEM (A) of representative Ab/SOD preparations of 53, 89 and 250 nm, respectively. Scale bar is 200 nm. (B). Size distribution as measured by DLS. (C). Representative curve of conjugate size dependence on molar ratio. Molar ration 1:1 produce the conjugates less than 30 nm and was not used in this study. See Methods for the details.

Figure 3. Comparative binding, internalization, and protective chracteristics of Ab/SOD conjugates of different sizes towards internal and external ROS

(A). Effects of conjugate size on binding and internalization of Ab/SOD. Binding was measured using ¹²⁵I-labeled SOD. Internalization was estimated by analysis of microscopic images. (B-C) Comparison of protection against internal vs. external ROS. Cells were preincubated with large or small Ab/SOD and treated with ROS generating by TNF treatment (internal ROS, B; representative experiment) or xanthine/xanthine oxidase system (external ROS, C). Ab/SOD size was 800 and 300 nm in xanthine/xanthine oxidase model and 300 and 200 nm in TNF model. Allo, allopurinol, xanthine oxidase inhibitor. Inhibition of VCAM expression in TNF-activated HUVEC was estimated by Western blot analysis. Cells were pretreated with 100 μg/ml of Ab/SOD of indicated size for 1 h before TNF treatment foe 5 h. VCAM expression was analyzed by Western blot (B, inset) and quantified. Cell viability of xanthine/xanthine oxidase treated cells was measured by ⁵¹Cr release assay. *P<0.05 allopurinol vs. untreated; # P<0.05 large vs. small.

Figure 4. Endothelial internalization of Ab/SOD conjugates targeted to PECAM vs. ICAM by HUVEC

The conjugates of different sizes were incubated with cells for 1 h. (A). Green and yellow colors are internalized and surface-bound particles, respectively. Nuclei are stained with DAPI (blue). (B). Quantification of the internalized fractions, mean±SD is shown.

Figure 5. Endothelial internalization of Ab/particles targeted to PECAM vs. ICAM by HUVEC

Particles were incubated with cells for 1 h. (A). Green and yellow colors are internalized and surface-bound particles, respectively. Nuclei are stained with DAPI (blue). (B). Quantification of the internalized fractions, mean±SD is shown. *P<0.05 anti-PECAM vs. anti-ICAM.

Figure 6. Clustering of CAM by anti-CAM/NP

(A) Internalization of polymorphous vs. spherical anti-ICAM/NP of various sizes. (B) EC were incubated with either FITC-labeled (green) spherical anti-ICAM/NP (beads, right) vs. polymorphous conjugates (left) of various sizes for 1 h at 37°C. ICAM was visualized after cell permeabilization using TR-labeled LB2 anti-ICAM antibody that binds to domain distinct of that of R6.5 antibody used for targeting (red). Notice continuous ICAM clusters enveloping both small and large spherical anti-ICAM/NP (yellow-reddish color). In contrast, only small polymorphous anti-ICAM conjugates are continuously surrounded by ICAM, while large conjugates form irregular clusters of ICAM. (C) Hypothesis of internalization control by anti-ICAM/NP geometry: discordant formation of CAM clusters by large polymorphous NP does not favor proper signaling, cytoskeleton rearrangements and plasmalemma zipping.

Figure 7. Comparison of Ab/SOD targeted to PECAM and ICAM and effects in vivo

(A). Lung targeting of SOD conjugated with corresponding antibody. Distribution of Ab/¹²⁵I-SOD in lung (blue) vs. blood (red) 1 h after intravenous injection in mice. The data are shown as mean ± S.D, n = 3. (B). Comparative protection against pulmonary VCAM expression in LPS-challenged animals. Ab/SOD (50 or 100 μg/mouse) were injected in tail vein followed by LPS challenge (50 μg/kg). After 5 h, lungs were harvested; VCAM expression was analyzed by Western blot and normalized per actin level, presented as a percentage of maximal LPS-induced VCAM level (analysis of ≥3 independent experiments); inset, Protection as normalized on lung uptake estimated for corresponding Ab/SOD (shown in panel A). n ≥ 3; mean±SD is shown. *P<0.05 anti-CAM vs. IgG; # P<0.05 anti-ICAM vs. anti-PECAM.

Figure 8. Comparison of Ab/SOD targeted to PECAM and ICAM and protection in vitro

(A). Binding of Ab/SOD to HUVEC. Cells were incubated with Ab/¹²⁵I-SOD for 1 h, washed and total cell-associated fraction was measured using gamma-counter. n ≥ 3; mean±SD is shown. (B). Comparative protection against TNF-induced increase of VCAM expression by HUVEC. Cells were pre-treated with Ab/SOD for 1 h and then activated by TNF for 4 h. VCAM expression was analyzed by Western blot and normalized per actin. (C). Protection as normalized on cell binding estimated for Ab/SOD (shown in panel A).