Plasma from preeclamptic women increases blood-brain barrier permeability: role of vascular endothelial growth factor signaling

Abstract

Circulating factors in preeclamptic women are thought to cause endothelial dysfunction and thereby contribute to the progression of this hypertensive condition. Despite the involvement of neurological complications in preeclampsia, there is a paucity of data regarding the effect of circulating factors on cerebrovascular function. Using a rat model of pregnancy, we investigated blood-brain barrier permeability, myogenic activity, and the influence of endothelial vasodilator mechanisms in cerebral vessels exposed intraluminally to plasma from normal pregnant or preeclamptic women. In addition, the role of vascular endothelial growth factor signaling in mediating changes in permeability in response to plasma was investigated. A 3-hour exposure to 20% normal pregnant or preeclamptic plasma increased blood-brain barrier permeability by ≈6.5- and 18.0-fold, respectively, compared with no plasma exposure (P<0.01). Inhibition of vascular endothelial growth factor receptor kinase activity prevented the increase in permeability in response to preeclamptic plasma but had no effect on changes in permeability of vessels exposed to normal pregnant plasma. Circulating factors in preeclamptic plasma did not affect myogenic activity or the influence of endothelium on vascular tone. These findings demonstrate that acute exposure to preeclamptic plasma has little effect on reactivity of cerebral arteries but significantly increases blood-brain barrier permeability. Prevention of increased permeability by inhibition of vascular endothelial growth factor signaling suggests that activation of this pathway may be responsible for increased blood-brain barrier permeability after exposure to preeclamptic plasma.

Figure 1

Graph showing hydraulic conductivity (L_p) as a function of time in cerebral veins perfused with either HEPES physiologic solution only (No Plasma), normal pregnant (P) plasma, or preeclamptic (Prex) plasma. Veins exposed to normal pregnant and preeclamptic plasma had increased L_p compared to no plasma exposure. In addition, veins exposed to preeclamptic plasma had L_p that was significantly increased vs. normal pregnant plasma. (**p<0.01 vs. no plasma; ††p<0.01 vs. P plasma)

Figure 2

Graph showing hydraulic conductivity (L_p) at 36 minutes of cerebral veins exposed to normal pregnant (P) or preeclamptic (Prex) plasma with and without the addition of VEGFR-II to inhibit VEGF receptor tyrosine kinase activity in the plasma perfusate, or 60 pg/mL VEGF without plasma. Veins exposed to P plasma were unaffected by VEFGR-II. However, VEGFR-II prevented an increase of L_p in veins exposed to Prex plasma, suggesting VEGF receptor tyrosine kinase activity is involved in increased BBB permeability in response to Prex plasma. VEGF alone produced modest permeability that was significantly decreased from PREX plasma only. (**p<0.01 vs. all)

Figure 3

Graph showing active and passive diameter vs. pressure curves for posterior cerebral arteries (PCAs) perfused with either normal pregnant (P) or preeclamptic (Prex) plasma. Arteries perfused with both types of plasma constricted and exhibited similar myogenic reactivity at pressures > 50 mmHg. Passive diameters of PCAs perfused with P and Prex plasma were similar at all intravascular pressures studied.

Figure 4

Graph showing percent tone vs. pressure of posterior cerebral arteries (PCAs) perfused with either normal pregnant (P) or preeclamptic (Prex) plasma. Acute exposure to plasma did not affect tone in either group.

Figure 5

Graph showing percent constriction of posterior cerebral arteries perfused with normal pregnant (P) plasma or preeclamptic (Prex) plasma to the addition of L-NNA and indomethacin to inhibit nitric oxide synthase and cyclooxygenase, respectively. All vessel and plasma combinations constricted similarly with the addition of L-NNA. The addition of indomethacin caused minimal changes in vessel diameter.

Figure 6

Graph showing percent dilation of posterior cerebral arteries perfused with normal pregnant (P) plasma or preeclamptic (Prex) plasma with the addition of A23187 in the presence of nitric oxide synthase and cyclooxygenase inhibition. All vessel and plasma combinations dilated similarly with the addition of A23187.