Structural alterations of tight junctions are associated with loss of polarity in stroke-prone spontaneously hypertensive rat blood–brain barrier endothelial cells

Abstract

The mechanisms leading to stroke in stroke-prone spontaneously hypertensive rats (SHRSP) are not well understood. We tested the hypothesis that the endothelial tight junctions of the blood–brain barrier are altered in SHRSP prior to stroke. We investigated tight junctions in 13-week-old SHRSP, spontaneously hypertensive stroke-resistant rats (SHR) and age-matched Wistar–Kyoto rats (WKY) by electron microscopy and immunocytochemistry. Ultrathin sections showed no difference in junction structure of cerebral capillaries from SHRSP, SHR and WKY, respectively. However, using freeze-fracturing, we observed that the blood–brain barrier specific distribution of tight junction particles between P- and E-face in WKY (58.7±3.6%, P-face; 41.2±5.59%, E-face) and SHR (53.2±19.3%, P-face; 55.6±13.25%, E-face) was changed to an 89.4±9.9% predominant E-face association in cerebral capillaries from SHRSP. However, the expression of the tight junction molecules ZO-1, occludin, claudin-1 and claudin-5 was not changed in capillaries of SHRSP. Permeability of brain capillaries from SHRSP was not different compared to SHR and WKY using lanthanum nitrate as a tracer. In contrast, analysis of endothelial cell polarity by distribution of the glucose-1 transporter (Glut-1) revealed that its abluminal:luminal ratio was reduced from 4:1 in SHR and WKY to 1:1 in endothelial cells of cerebral capillaries of SHRSP. In summary, we demonstrate that early changes exist in cerebral capillaries from a genetic model of hypertension-associated stroke. We suggest that a disturbed fence function of the tight junctions in SHRSP blood–brain barrier endothelial cells may lead to subtle changes in polarity. These changes may contribute to the pathogenesis of stroke.

Introduction

Hypertension plays an important role in the pathogenesis of stroke. An increase in blood pressure predisposes to stroke and about 70% of stroke patients are hypertensive [4], [5], [36]. However, the molecular mechanisms linking hypertension to stroke are incompletely understood. Earlier reports [9], [13], [37] have indicated that blood–brain barrier permeability is increased in hypertension, suggesting that permeability may be important for the development of stroke. A utilitarian animal model of stroke is the stroke-prone spontaneously hypertensive rat (SHRSP) [48]. In SHRSP, 80% of the animals develop stroke spontaneously during aging or in a defined time when given the appropriate diet [15], [30], [46], [47]. During the development of high blood pressure, changes in endothelial cell function occur in the cerebral vessels of SHRSP [29], [35], [38], [41], [49]. Prior to stroke a reduction in global cerebral blood flow and cortical protein synthesis is observed [24], [30]. These observations have led to a hypothetical pathologic sequel where endothelial dysfunction and subsequent formation of edema precede the development of stroke [2], [17], [47]. The aim of the present study was to analyze these early changes in endothelial cell function.

Specific characteristics of the blood–brain barrier endothelial cells are the tight junctions. Using freeze-fracture electron microscopy it has been found that the tight junction particles have a special distribution within the fracture faces [20], that is about 57% of the particles within the internal membrane leaflet (P-face) and about 44% of the particles within the external leaflet (E-face) of the endothelial cell membrane [19]. This distribution is a prerequisite of the maintenance of the blood–brain barrier tightness [32], [44] as well as of the functional polarization of the blood–brain barrier endothelial cells [42]. These cells exhibit a specific distribution pattern of enzymes and transporter molecules to the abluminal and luminal plasma membranes, respectively. Maturation of the blood–brain barrier during embryonal development and also soon after birth leading to tightening of the barrier is accompanied by a redistribution of enzyme activities and transporter molecules to their final target membranes and it is suggested that the polarity of the endothelial cells reflects the maintenance of the fence function of their tight junctions [42]. There have been preliminary reports that enzymatic activities are shifted to the other membrane which coincide with increased permeability under pathological circumstances [42], [43].

We analyzed blood–brain barrier endothelial cell tight junctions prior to the development of stroke in SHRSP to test the hypothesis that the morphology of tight junctions is altered prior to stroke. We were also interested in the expression pattern of junction-forming proteins. We analyzed endothelial cell permeability using lanthanum nitrate and endothelial cell polarity by assessing the subcellular distribution of the glucose-1 transporter (Glut-1) [1], [7]. We found that an altered E-face association of tight junction particles as well as their disturbed fence function, measured by Glut-1 distribution in SHRSP blood–brain barrier endothelial cells, may lead to subtle changes in the cerebral microvessel endothelial cells thereby contributing to the susceptibility to stroke.