Effect of hypoxia on endothelial morphology and interendothelial junctions in the isolated perfused rat heart

Shedding of the coronary endothelial glycocalyx: effects of hypoxia/reoxygenation vs ischaemia/reperfusion

BACKGROUND

Vascular endothelium is covered by a glycocalyx. Damage to the glycocalyx after systemic inflammation or ischaemia/reperfusion contributes to increased vascular permeability and leucocyte adhesion. The underlying mechanisms leading to ischaemia/reperfusion-induced glycocalyx shedding are incompletely understood, in terms of lack of oxygen, absence of flow, or return of oxygen.

METHODS

Isolated guinea pig hearts perfused with Krebs-Henseleit buffer at 37°C underwent 20 min of either stopped-flow ischaemia or hypoxic perfusion with subsequent reperfusion/reoxygenation (n = 6 each). Hearts perfused with normoxic buffer served as time controls. Epicardial transudate was collected to assess coronary net fluid filtration, colloid extravasation, and histamine release by mast cells. Syndecan-1 and heparan sulphate were measured in coronary effluent, together with lactate, purines, and the release of mast-cell tryptase β. Additional hearts were perfusion-fixed to visualize the glycocalyx.

RESULTS

Both ischaemia and hypoxia with reperfusion/reoxygenation resulted in significant increases in net fluid filtration (P < 0.05) and release of syndecan-1 and heparan sulphate in coronary effluent. These effects were already seen with the onset of hypoxic perfusion. Histamine was released during hypoxia and reoxygenation and also reperfusion, as was tryptase β, and high concentrations of adenosine (>1 µmol litre⁻¹, hypoxia group) and inosine (> 7 µmol litre⁻¹, ischaemia group) were measured in effluent (P < 0.05). Damage to the coronary glycocalyx was evident upon electron microscopy.

CONCLUSIONS

Both ischaemic and hypoxic hypoxia initiate glycocalyx degradation, promoting an increase in permeability. A contributing mechanism could be purine-mediated degranulation of resident mast cells, with liberated tryptase β acting as potential 'sheddase'.

Calcium/calmodulin-dependent protein kinase IIdelta 2 and gamma isoforms regulate potassium currents of rat brain capillary endothelial cells under hypoxic conditions

Endothelial K+ and Ca2+ homeostasis plays an important role in the regulation of tissue supply and metabolism under normal and pathological conditions. However, the exact molecular mechanism of how Ca2+ is involved in the regulation of K+ homeostasis in capillary endothelial cells, especially under oxidative stress, is not clear. To reveal Ca2+-triggered pathways, which modulate K+ homeostasis, Ca2+/calmodulin-dependent protein kinase II and voltage-gated outward K+ currents were studied in rat brain capillary endothelial cells under hypoxia. Whole cell voltage-clamp measurements showed voltage-gated outward K+ current with transient and sustained components. mRNA and protein of Ca2+/calmodulin-dependent protein kinase II delta2 and two gamma isoenzymes were identified. Activation of the isoforms (autophosphorylation) was typically achieved by the Ca2+ ionophore ionomycin, which was prevented by the Ca2+/calmodulin-dependent protein kinase II-specific inhibitor KN-93. Hypoxia resulted in autophosphorylation of the delta2 and gammaB isoforms, augmented the current amplitude, increased the inactivation time constant, and decreased the extent of inactivation of the transient current. KN-93 prevented both the activation of the isoforms and the alterations in the K+ current characteristics. It is concluded that the activation of Ca2+/calmodulin-dependent protein kinase II decreases inactivation of the voltage-gated outward K+ current, thereby counteracting depolarization of the hypoxic endothelium.

The influence of hypoxia on transvascular leakage in the isolated rat heart: quantitative and ultrastructural studies

1. The multiple indicator dilution method was used to study the transvascular movement of gamma-globulin, bovine serum albumin, insulin and cyanocobalamin in the isolated rat heart. 2. Perfusion of the heart with well-oxygenated solution for 75 min (constant flow) did not produce a significant change either in the total area under the dilution curve or the 'leakage index' (an arbitrary measure of transvascular flux) for all the tracers. 3. Perfusion of the heart with hypoxic solution produced a significant increase in leakage of gamma-globulin of 38.6 +/- 18, 48.5 +/- 17.6, 60.5 +/- 24 and 58 +/- 20% after 15, 30, 45 and 60 min, respectively, compared with the well-oxygenated equilibration period. Permeability- surface area products (PS) for the smaller diffusible solutes, therefore, could not be estimated. 4. The flux of albumin, insulin and cyanocobalamin in response to hypoxia was similar to that of gamma-globulin. 5. Ultrastructural examination of well-oxygenated hearts revealed that Monastral Blue-labelled albumin remained within the lumen and that endothelial integrity remained intact. 6. Conversely biopsies from hypoxic hearts showed that the labelled albumin had passed to the interstitium through gaps (approximately 3 microns) in venular endothelium. 7. The results showed that, in intact hearts, hypoxia produced gaps in the endothelium of venules and that these gaps could be the possible route for transvascular leakage of macromolecules.

Interstitial lactate and glucose concentrations of the isolated perfused rat heart before, during and after anoxia

In isolated rat hearts perfused according to the Langendorff technique lactate and glucose concentrations were determined in the interstitial transudate and the venous effluent before, during and after periods of 15 or 30 min anoxia. The interstitial transudate emerged at the surface of the heart as a result of albumin-free perfusion. During normoxic perfusion the interstitial lactate concentration was 0.144 +/- 0.025 mmol/l (n = 6); the venous lactate concentration was 0.033 +/- 0.005 mmol/l. From the interstitial and the mean vascular concentration, together with the lactate release and the glucose uptake, the apparent permeability surface area products (P.S product) were calculated using Fick's law. The apparent P.S products for lactate and glucose were 4.6 and 3.9 ml/(min x g), respectively. During anoxia we measured a four- to sixfold increase of the interstitial lactate concentration. At the end of the anoxic periods the apparent P.S product was two- to threefold higher than during normoxia; the apparent glucose P.S product increased about fourfold. After 15 min anoxia the increases of permeability were completely reversed in the reoxygenation period. However, after a period of 30 min anoxia the apparent P.S products for lactate and glucose remained raised, which means that there was a prolonged or even irreversible increase of capillary permeability. Besides the marked transcapillary concentration difference for lactate, these data show prolonged functional alteration of the capillary wall after 30 min anoxia.

Fine structure of the paracellular junctions of terminal villous capillaries in the perfused human placenta

Selected lobules of human term placentae were extracorporeally perfused for a recovery period of 20 min, fixed by perfusion and mordanted with ferrocyanide prior to processing for transmission electron microscopy. The lateral membranes of the endothelial cells of the terminal villous capillaries were found to be separated by paracellular clefts of mean width 15.6 nm. At tight junctional regions (1-4 sites per cleft) the two membranes approached each other more closely and frequently appeared to fuse. However, tilting of the sections in the electron microscope stage showed that the membranes were separated by a gap of mean width 4.1 nm in at least 94% of tight junctional profiles. When individual tight junctions were studied by a combination of serial sectioning and goniometric tilting, they were seen to widen abruptly within a distance of three to seven consecutive thin sections, indicating they were not continuous throughout the axial length of the capillaries. The wide regions of the clefts usually showed linkers, strands of glycocalyx-like material spanning the gap. Linkers may contribute to cell adhesion and possibly form part of a filter within the tortuous paracellular pathway provided by the discontinuous network of tight junctional strands. Human term placental capillaries appear to resemble closely other continuous non-brain capillaries.