An ultrastructural study of vascular endothelial junctions in normal and tensioned rat periodontal ligament

Abstract

Ultrastructural changes in the microvascular bed of the periodontal ligament (PDL) incident to the applied forces have rarely been reported. Clark (1986) and Lew (1986) demonstrated changes in the fenestrae populations and Cooper (1988) established there was an increase of tissue channels when a tensile load was applied to the PDL. However, there are no published data on the effects on the endothelial junctions or their dimensions where there is an applied loading to the PDL. The purpose of this investigation was to morphologically study and quantitatively compare the dimensions of endothelial junctions in the microvascular bed of the normal PDL with those of the tensioned PDL. The rat maxillary molar specimen blocks were from Cooper's study (1988). Twelve week old male Porton rats were used as the experimental animal model. A 1.0 N. extrusive load was applied continuously, for a period of 30 minutes, to the right first molar. The contralateral left first molar served as the experimental control. The tissue r4as demineralized in a solution of 0.1 M. EDTA in O.06 M. cacodylate buffer and 2.5% glutaraldehyde for 8-12 weeks. Then the tissue was post-fixed in 2% osmium tetroxide, dehydrated through a graded ethanol series, infiltrated with propylene oxide and Agar® 100, and embedded in Agar® 100. Four matching pairs of experimental and control blocks from four animals were selected for this investigation. The silver sections were prepared using a Reichert-Jung OM-U4 ultramicrotome. The sections were stained with uranyl acetate and Reynolds' lead solution. The area to be examined was the PDL at the most mesial side of the first molar. The tissue was examined at levels 250 microns apart from the cervical to apical region. A minimal number of 150 complete junctions from the normal PDL and 150 complete junctions from tensioned PDL were selected randomly for the examination. A transmission electron microscope, JEM -2OOO FX, was used to examine and photograph the sections. The goniometer was used to explore the junctions so that the narrowest opposing membranes could be studied and measured. Measurements were made of the various widths and depths along the junctions. Five types of brood vessels were encountered in the investigation, i.e., venous capillaries, arterial capillaries, postcapillary-sized venules, terminal arterioles and small collecting venules. The morphology of the junctions was found to be related to the type of blood vessel. There were three types of junction found in the PDL microvasculature, i.e., close junctions, tight junctions and open junctions. No gap junctions were found. There were no significant differences in the numbers of each junction type between normal and tensioned PDL. A chi-square analysis also indicated that the proportion of tight junctions was not significantly different among different types of blood vessels. However, a significant change in the dimensions of junctions was found. Close junctions were in the majority forming 80% of all the junctions investigated. The width of the close region was approximately 5-6 nm. The junction width of the close junctions at the non-close/tight region increased (9.4%) as a result of extrusive -loads (p˂0.01). Since the close junctions comprise the majority of junctions in the PDL, the increase in the junction width could imply that the fluid and substrate exchange through the junctions also increased incident to the extrusive load. It is therefore possible that close junctions are one of the major exchange pathways i n the PDL microvasculature. Tight .junctions were found in approximately 16% of junction population. The tight regions where the opposing membranes fused were commonly found in the luminal third (87.5%). The tight regions appeared to be ‘spot-fusion’, rather than 'zone-fusion'. There was no significant change in the junction width of the tight junctions. Open .junctions were rarely found (4%). They were found only in the venous capillaries and postcapillary-sized venules where the walls were relatively thin. The pore sizes ranged from 136 nm. to 1231 nm. The junction depth was found to have a strong positive correlation to the endothelial walI thickness across the middle of the junction (p<0.001 ). The organization of the endothelial junctions in three dimensions could not be determined by thin section technique in this study. Since the three-dimensional reconstruction of tight junctions requires a series of ultrathin sections. Further investigation of junctions by freeze-fracture replicas is suggested.