A freeze-fracture study of the perineurium in normal and protein-deprived rats.

Observations have been made using the freeze-fracture replication technique on the perineurium of normal and protein-deprived rats in which its permeability barrier function is known to be deficient. The perineurial cells of young normal rats possessed belt-like tight junctions (zonulae occludentes) at the borders and maculae occludentes at sites remote from their borders. In older rats, the zonulae occludentes were more prominent and the maculae occludentes relatively less frequent. No abnormalities were detected in the tight junctions of young rats with early induction of protein deficiency but this may have been related to sampling problems. In older severely protein-deficient animals, although many of the tight junctions were normal, some were abnormal and contained focal regions of dispersed strands. The density of caveolae in the surface membrane of the perineurial cells of older rats with severe protein deficiency was significantly greater than in the control animals. This provides support for the view that the pinocytotic-like vesicles of perineurial cells are involved in transport of substances across the cells. The increased numbers of caveolae in the protein deficient rats may reflect increased transcellular traffic. There were considerable differences in the density of P-face IMPs between the different perineurial lamellae, but the results did not allow a decision to be made as to whether there was a polarization of the cells between their endoneurial and epineurial aspects. No differences were detected in the density of P-face IMPs between the young control and protein-deprived rats. In the perineurium of the older rats with protein deficiency, IMP density was significantly greater in the E face than in the controls but not different in the P face. The delay in the development of enzymatic activity in the perineurium of protein-deficient rats that has been demonstrated histochemically is therefore not paralleled by a reduction in IMPs.