Abstract 430: GFP-tagging Of Rabbit ApoAI Decreases Expression And Cholesterol-efflux Activity

Abstract

Introduction: Gene therapy using a helper-dependent adenovirus (HDAd) to express rabbit apolipoprotein (apo) AI in carotids of fat-fed rabbits reduces atherosclerosis. We hypothesize that transgenic apoAI promotes reverse cholesterol transport from the artery wall; however, this could not be tested because the transgenic rabbit apoAI is indistinguishable from endogenous rabbit apoAI. We therefore constructed an HDAd expressing GFP-tagged rabbit apoAI. Methods: We constructed HDAdGFPapoAI by fusing the GFP sequence to the N-terminus of the rabbit APOA1 gene with a short linker. Bovine aortic endothelial cells (BAEC) were transduced with HDAdGFPapoAI, washed, and incubated in serum-free medium for 24h. Conditioned medium (CM) was collected and fluorescence measured. Cell extracts and CM were analyzed by western blotting for protein size, expression level, and presence of GFP and apoAI antigens. CM cholesterol-efflux activity was measured using cholesterol-loaded BHK cells expressing ABCA1. Results: CM of HDAdGFPapoAI-transduced BAEC had green fluorescence and contained a protein of the size predicted for the GFP-apoAI fusion (~56 kDa), which was detected with antibodies to both GFP and apoAI. A smaller protein (~30 kDa) with GFP and apoAI antigens was also detected. Compared to a vector expressing WT rabbit apoAI, GFP-tagging decreased expression of apoAI protein by ~50% and the percentage of secreted apoAI from 55% to 13%. In combination, this reduced secreted GFP-apoAI to only 12% of levels of secreted WT apoAI. While CM containing WT rabbit apoAI strongly stimulated cholesterol efflux from BHK cells (46%), GFP-apoAI-containing CM did not promote cholesterol efflux compared to CM from untransduced BAEC (7.8% vs 7.4% efflux), even when present in CM at a concentration similar to that of WT apoAI. Conclusion: GFP-rabbit apoAI protein has reduced expression and secretion (likely due to misfolding/degradation) and does not promote cholesterol efflux (n-terminal GFP may disrupt the phospholipid/cholesterol organization needed for efflux via ABCA1 to this dynamic apoprotein). It will not be useful to investigate mechanisms of atheroprotection by vessel wall-expressed apoAI. Alternatives include use of a smaller tag or expression of human apoAI.