(37b) Experimental Verification of Endothelial Cell Aquaporin-1 Expression's Influence on Sub-Endothelial Intima Thickness and Relevance to Early Atherosclerosis

Experimental Verification of Endothelial Cell Aquaporin-1 Expression's Influence on Sub-Endothelial Intima Thickness and Its Relevance to Early Atherosclerosis

Roman Yakobov, Klaudio Mitri, Deyvi Tenemaza, Ruven Pinkhasov, Diana Yakobov, Jiah Lee, Kung-ming Jan and David S. Rumschitzki

Atherosclerosis is perennially among the top two killers in the US and in all Western countries. It begins decades before symptoms arise by the infiltration of low density lipoprotein (LDL) cholesterol from the blood across the arterial endothelium into the arterial wall, where it can bind to extracellular matrix (ECM) in the subendothelial intima (SI) wall region. This SI lipid accumulation appears to trigger a cascade of events that can lead to early lesion formation. A large (~22nm diameter) aggregate, LDL crosses the endothelium via the junctions around rare (~1/(2000-5000) cells in rat aorta) endothelial cells (EC) that have temporarily (~1h) widened junctions with their neighbors. Our group has shown that this transport is advection-dominated, via a slow (~30 nm/s) transmural pressure difference-driven transmural flow of the liquid portion of the blood through these leaky junctions. Naturally this transmural pressure difference and flow also act across the normal endothelium and its tight junctions. We have shown, both theoretically and experimentally, that this flow passes not just paracellularly through the EC-EC tight junctions, but a portion (~1/3 of the total, but in this case pure H₂O) also passes transcellularly via endothelial cell aquaporin-1 (AQP1) water channel membrane proteins. This water-and-plasma flow is of great interest because it can wash unbound LDL from the SI, dilute the local SI LDL concentration and, since binding to ECM is a chemical reaction, potentially strongly affect LDL-ECM binding kinetics. Thus more EC AQP1 could mean slower LDL-ECM binding and slower progression to lesions.

Interestingly our group has observed that the aortic SI region is mostly (>95%) void, in contrast to the aortic media (~42% void) and that transmural pressure can compress the SI, causing the ECs to block SI fenestrae and significantly increase transmural flow resistance; this drastically lowers transmural flow. Our theory predicts that increased/decreased EC AQP1 expression and activity raises/lowers the critical transmural pressure needed to accomplish this SI compression and concomitant increase in flow resistance. Since under normal conditions, the SI is compressed, if this theory is correct, increased ECAQP1 expression has the potential to decompress the SI, decrease flow resistance and thus increase SI LDL dilution in the physiological range. We have already shown that an increase/decrease in functioning EC AQP1 indeed increases/decreases the transmural pressure at which flow resistance drops. What is missing is to show that changes in functioning EC AQP1s accomplish these changes in flow resistance via a corresponding shift in the critical pressure for SI compression. This is the subject of this paper, i.e., to directly test these predictions experimentally.

We anaesthetize two rats each for each set of conditions. These sets of conditions include treatment with either (a sub-toxic level of) HgCl₂, which reversibly blocks EC AQP1 channels, Forskolin (which increases EC AQP1 expression) or a blank control solution each at one of four different transmural pressures. We then cannulate the rat’s carotid artery and attach it to a constant pressure reservoir at one of four different pressures. We stop the rat’s heart, fix the vessel in situ, extract, further fix and embed the aorta in Epon, section (60-100 nm) for transmission electron microscopy (TEM), stain to increase contrast, view the sections under TEM and analyze the sections for SI thickness using a custom-written Matlab code. We analyze 100-200 random sections per set of conditions (i.e., treatment and pressure). We plot the results in the form of SI thickness vs pressure for each of the three treatments and compare with theoretical predictions. The results are a very strong test of our theory and tie together theory and experiments on both SI thickness, and on aortic wall hydraulic conductivity (the inverse of its flow resistance) as functions of transmural pressure and functioning EC AQP1 numbers.