The Malpighi lecture. Vascular permeability--the consequence of Malpighi's hypothesis.

The permeability of microvessels to water and small hydrophilic solutes varies considerably between different vascular beds and even between individual vessels in the same vascular bed. Provided the tissues are undamaged, the filtering properties of vessel walls to macromolecules are relatively constant and an analysis of the variations in hydraulic conductances and permeabilities to small ions between different endothelia suggests that the channels for hydrophilic molecules are similar in a wide range of vessels. In undamaged tissues, variations in permeability may be accounted for by variations in the frequency of channels per unit area of vessel wall. The permeability of vessel walls to water and small hydrophilic molecules and their ability to sieve macromolecules are consistent with the molecular filter being a lattice of fibrous molecules which covers and fills the channels through the endothelium. Direct and indirect evidence indicate that globular proteins are part of the molecular filter and it is suggested that they bind the fibrous molecules into a more uniform lattice (the fibre matrix) which greatly increases its selectivity. The question of how the contraction of adjacent endothelial cells might widen the intercellular clefts and thus regulate permeability is discussed. Preliminary experiments show that the permeability of frog capillaries can be varied by varying concentrations of Mg2+ (rather than Ca2+) in the presence of the ionophore A23187 and suggest an interesting line of investigation.