Mechanotransductional basis of endothelial cell response to intravascular bubbles

Abstract

Vascular air embolism resulting from too rapid decompression is a well-known risk in deep-sea diving, aviation and space travel. It is also a common complication during surgery or other medical procedures when air or other endogenously administered gas is entrained in the circulation. Preventive and post-event treatment options are extremely limited for this dangerous condition, and none of them address the poorly understood pathophysiology of endothelial response to intravascular bubble presence. Using a novel apparatus allowing precise manipulation of microbubbles in real time fluorescence microscopy studies, we directly measure human umbilical vein endothelial cell responses to bubble contact. Strong intracellular calcium transients requiring extracellular calcium are observed upon cell-bubble interaction. The transient is eliminated both by the presence of the stretch activated channel inhibitor, gadolinium, and the transient receptor potential vanilliod family inhibitor, ruthenium red. No bubble induced calcium upsurge occurs if the cells are pretreated with an inhibitor of actin polymerization, cytochalasin-D. This study explores the biomechanical mechanisms at play in bubble interfacial interactions with endothelial surface layer (ESL) macromolecules, reassessing cell response after selective digestion of glycocalyx glycosoaminoglycans, hyaluran (HA) and heparin sulfate (HS). HA digestion causes reduction of cell-bubble adherence and a more rapid induction of calcium influx after contact. HS depletion significantly decreases calcium transient amplitudes, as does pharmacologically induced sydencan ectodomain shedding. The surfactant perfluorocarbon Oxycyte abolishes any bubble induced calcium transient, presumably through direct competition with ESL macromolecules for interfacial occupancy, thus attenuating the interactions that trigger potentially deleterious biochemical pathways.

Fig. 1

Schematic of bubble probing apparatus (described in Methods) to bring bubble, on the end of a pulled and ground glass capillary tip, in contact with cells for time-lapse fluorescent microscopy imaging studies.

Fig. 2

Normalized Fluo-4 fluorescence time courses with contact occurrence indicated at t = 0 for (solid black line) control HUVEC contacted with a bubble and (dotted gray trace) HUVEC contacted with glass bead. Representative false-colored Fluo-4 intensity map images acquired and used to calculate the control bubble touch trace are shown with line segments to indicate occurrence during time course. Additional bubble contact F_N(t) traces for HUVECs in conditions which inhibited response: (green) pretreated with 100 nM cytochalasin D; (purple) in 10 μM Gd³⁺; (blue) in low external Ca²⁺; (red) with 1 μM RR; and (dotted purple) in 10% oxycyte.

Fig. 3

[Ca²⁺]_i response of one cell to successive bubble contacts. Each bubble touch event is indicated with an arrow. Inset shows fractional maximum amplitude (F_N/F_max) as a function of bubble contact event number and the best fit to a one site competition curve with variable slope.

Fig. 4

Bar chart showing F_N(t) amplitudes for control (2 mM), low (~2 μM) and high (20 mM) extracellular Ca²⁺ concentrations (*) p = 0.0001; (#) p = 0.0286.

Fig. 5

Bar chart with BSA and oxycyte effects on the mean bubble-induced amplitudes after bubble contact (*) p < 0.0001 (paired with bubble control) and oxycyte effect on the mean glass-bead induced F_N(t) amplitudes (#) p = 0.017 (paired with glass-bead control).

Fig. 6

Exemplar traces of Ca²⁺ response after bubble contact for HUVEC (solid) control; pretreated with (dashed) Hyaluronidase and (solid) Heparinase I. (Inset) Bar chart with mean amplitudes of Ca²⁺ signaling for control and glycocalyx component digested samples (*) p = 0.0004.

Fig. 7

Barcharts indicating effects of targeted HSPG depletion from the cell surface prior to bubble probe experiments. (Left panel) Mean bubble-induced transient amplitudes and ± SEM of responding cells after pretreatment as indicated below (*) p=0.0194. (Right panel) percent of cells tested that responded to bubble touch in the same experiments.

Fig. 8

Simplified schematic of glycocalyx and cell plasma membrane components in left panel, and proposed mechanism of bubble induced transient [Ca²⁺]_i increase in the right panel: HA is shown as grey lines with brown ovals representing associated plasma proteins. Syndecan HS side chains (green) are pulled into the bubble air/liquid interface (represented by a blue arc) when HS bound protein ligands (purple ovals) adhere to its surface. The resultant tensile force physically displaces a transmembrane syndecan (shown as dimer), altering cytoplasmic domain actin cytoskeleton associations (purple rectangles) and activating a TRPV Ca²⁺ channel.