Rickettsiae induce microvascular hyperpermeability via phosphorylation of VE-cadherins: evidence from atomic force microscopy and biochemical studies

Abstract

The most prominent pathophysiological effect of spotted fever group (SFG) rickettsial infection of microvascular endothelial cells (ECs) is an enhanced vascular permeability, promoting vasogenic cerebral edema and non-cardiogenic pulmonary edema, which are responsible for most of the morbidity and mortality in severe cases. To date, the cellular and molecular mechanisms by which SFG Rickettsia increase EC permeability are largely unknown. In the present study we used atomic force microscopy (AFM) to study the interactive forces between vascular endothelial (VE)-cadherin and human cerebral microvascular EC infected with R. montanensis, which is genetically similar to R. rickettsii and R. conorii, and displays a similar ability to invade cells, but is non-pathogenic and can be experimentally manipulated under Biosafety Level 2 (BSL2) conditions. We found that infected ECs show a significant decrease in VE-cadherin-EC interactions. In addition, we applied immunofluorescent staining, immunoprecipitation phosphorylation assay, and an in vitro endothelial permeability assay to study the biochemical mechanisms that may participate in the enhanced vascular permeability as an underlying pathologic alteration of SFG rickettsial infection. A major finding is that infection of R. montanensis significantly activated tyrosine phosphorylation of VE-cadherin beginning at 48 hr and reaching a peak at 72 hr p.i. In vitro permeability assay showed an enhanced microvascular permeability at 72 hr p.i. On the other hand, AFM experiments showed a dramatic reduction in VE-cadherin-EC interactive forces at 48 hr p.i. We conclude that upon infection by SFG rickettsiae, phosphorylation of VE-cadherin directly attenuates homophilic protein-protein interactions at the endothelial adherens junctions, and may lead to endothelial paracellular barrier dysfunction causing microvascular hyperpermeability. These new approaches should prove useful in characterizing the antigenically related SFG rickettsiae R. conorii and R. rickettsii in a BSL3 environment. Future studies may lead to the development of new therapeutic strategies to inhibit the VE-cadherin-associated microvascular hyperpermeability in SFG rickettsioses.

Figure 1. Immunofluorescence studies show rickettsiae (red) located in human cerebral microvascular endothelial cells at 24, 48 and 72 hr after infection.

Dual immunofluorescence staining of SFG rickettsiae (red) and VE-cadherin (green) using dual wave lengths filter system reveals that, compared to normal controls, R. montanensis infection (10 MOI) resulted in degradation of the density of VE-cadherin, suggesting disruption in the continuity of VE-cadherin at neighbouring areas at 72 hr post-infection.

Figure 2. R. montanensis infection enhanced human cerebral microvascular endothelial cell permeability.

Endothelial cells were seeded on type I rat-tail collagen-coated polycarbonate transwell filters and infected with R. montanensis at an MOI of 10 in triplicate, or mock infected. After 24, 48, and 72 hr, FITC-dextran was added to the upper chamber medium, and the presence of FITC dextran in the lower chamber was quantified after 3 hr. The results are expressed as the fold-increase in monolayer permeability over basal permeability levels (* p<0.05).

Figure 3. R. montanensis enhanced VE-cadherin tyrosine phosphorylation.

Human cerebral microvascular endothelial cells were mock-infected (control) or infected with R. montanensis at a MOI of 10. At 24, 48, or 72 hr post-infection, cells were harvested for Western immunoblot or immunoprecipitated with anti-VE-cadherin antibody using a magnetic bead system. 3A). The Western immunoblot for α-tubulin served as a control to verify equal loading and transfer. There was no significant difference in VE-cadherin expression detected by Western immunoblot. 3B and 3C). Loading of VE-cadherin was detected by anti-VE-cadherin antibody. The normalized relative densities from IP phosphorylation assay showed a 2.16- and 4.48-fold increase in phosphorylation of VE-cadherin (P-VE-cadherin) at 48 hr and 72 hr post-infection, respectively, compared to control cells (* p<0.05). These data are representative of three independent experiments.

Figure 4. AFM measurements of the de-adhesion forces between VE-cadherin and endothelial cells.

4A). Typical force-extension curves obtained between cells and VE-cadherin coated AFM tips. The dashed lines indicate zero force. The experiments were carried out in uninfected and infected cells at different time points post-R. montanensis infection (48 hr and 72 hr). 4B). Work of de-adhesion between VE-cadherin beads and endothelial cells at different time points post-infection. R. montanensis-infected cells required a significantly lower level of average work to break the interaction compared with uninfected cells. The addition of a monoclonal antibody against VE-cadherin significantly blocked the VE-cadherin-endothelial cell interaction. Ab: anti-VE-cadherin antibody.