Analysis of absorptive cell occluding junction structure-function relationships in a state of enhanced junctional permeability.

Junctional strand count, the number of individual junctional strands which intersect a perpendicular, has been widely used as a morphological estimate of the ability of an occluding junction to resist passive transjunctional molecular flow. Such junctional analysis, which requires the study of freeze fracture replicas, has been useful in studies of occluding junction structure function relationships in unperturbed epithelia and in intestinal epithelia under conditions which lead to increases in junctional resistance. It is unclear if the above junctional structure-function correlates also exist in intestinal or in other epithelia under conditions which result in a state of enhanced junctional permeability. To gain further insight into occluding junction structure-function relationships under such conditions, we utilized an in vivo hypertonic perfusion model previously shown to result in the transfer of luminal macromolecules into the paracellular space of small intestinal villus epithelium. After a 1-hour perfusion with either 600 mOsmoles or 750 mOsmoles mannitol solutions, the macromolecule horseradish peroxidase diffusely filled the paracellular spaces of the upper half of villi. However, analysis of thin sections showed that only junctions at the tip of villi were leaking this tracer. Freeze fracture analysis revealed occluding junction structural abnormalities most marked at the villus tip thus corresponding to the site of transjunctional horseradish peroxidase leak. The most frequent abnormality noted was loss of strand-strand crosslinking and dilation of the interstrand compartments. At such sites, wide unobstructed channels could be traced through much of the junction. However, the apical junctional strand was never noted to display unequivocal discontinuities. Subsequent in vitro studies of perfused tissues revealed that dose dependent decreases in transepithelial resistance and junctional charge selectivity were induced by hypertonic mannitol perfusions. These studies indicate that: paracellular localization of a luminally applied tracer may result from a transjunctional leak at a distant site with subsequent lateral diffusion of the marker; traditional junctional strand count-function relationships may not hold in states which lead to the rearrangement of junctional architecture; and, at least the apical junctional strand may become permeable to macromolecules without the introduction of morphologically detectable strand discontinuities as assessed by routine freeze fracture techniques.(ABSTRACT TRUNCATED AT 400 WORDS)