HepG2 cells secrete an unidentified factor that reshapes nascent HDL into HDL2b particles, with cholesteryl ester transfer detected but not attributable to CETP (J Lipid Res 1993)

We have previously shown that lecithin:cholesterol acyltransferase (LCAT) can transform ultracentrifugally isolated HepG2 lipoproteins (d < 1.235 g/ml) into particles that differ substantially from their nascent precursors. Transformed high density lipoprotein (HDL) subpopulations, as judged by nondenaturing gradient gel electrophoresis (GGE), resemble plasma HDL, i.e., HDL2a- and HDL3a-sized particles predominate. In HepG2 conditioned medium (CM), 60-70% of apoA-I is in the d > 1.235 g/ml fraction (lipid-poor apoA-I); hence we investigated whether inclusion of d > 1.235 g/ml fraction in LCAT incubations altered HDL subpopulations. After 18 h incubation of CM (containing lipoproteins and d > 1.235 g/ml fraction) with purified LCAT, the major transformation product on GGE was a large 9.7-nm particle (HDL2b pattern); a minor component appeared at 7.4 nm (HDL3c). Differences in particle size distribution between CM and isolated lipoprotein incubations were not the result of differences in LCAT activity; mass ratios of unesterified cholesterol:cholesteryl ester and phospholipid:cholesteryl ester were similar. Removal of apoA-I from the d > 1.235 g/ml fraction by immunoaffinity chromatography prior to incubation with the d < 1.235 g/ml fraction produced the same products (i.e., HDL2b pattern) as incubations performed with the unaltered d > 1.235 g/ml fraction; therefore, lipid-poor apoA-I does not influence nascent HDL transformation. Cholesteryl ester was transferred from HepG2 HDL to LDL in CM incubations; however, cholesteryl ester transfer protein was not immunochemically identified. Removal of HepG2 LDL from CM prior to incubation with LCAT still resulted in the HDL2b pattern. We conclude that HepG2 cells secrete a factor(s) that modifies nascent HDL transformation products into a predominantly HDL2b subpopulation.