Measurement of normative HDL subfraction cholesterol levels by Gaussian summation analysis of gradient gels

This report describes development of a computerized method for analyzing polyacrylamide gradient gels of high density lipoproteins (HDL) by Gaussian summation, a simple technique to obtain standardized measurements of size and amount of HDL subfractions. Conditions for reproducibility and ranges of linearity were established. By Gaussian summation analysis, five or six HDL subfractions could be found in the plasma of most normolipidemic people. The relationship of staining intensity to cholesterol level was determined for Coomassie Blue R-250, permitting determination of the cholesterol levels in the individual subfractions, with standard errors of repeated measurements of 2% or less of the total HDL area, and accuracy, limited by the standard error of the chromogenicity, of 1-2 mg/dl for the least abundant fractions and 3-4 mg/dl for the most abundant subfractions. Levels of HDL2b measured by this method were statistically the same as levels of HDL2 measured by dextran sulfate-Mg2+ precipitation. Gaussian summation analysis of gradient gels was used to measure HDL subfraction cholesterol levels in subjects from the Baltimore Longitudinal Study on Aging to obtain normative levels for men and women for the major HDL subfractions. Comparisons of these levels with each other and with triglyceride and cholesterol levels showed that triglyceride levels were inversely correlated with levels of HDL2a and HDL2b, cholesterol levels were directly correlated with levels of HDL3b and HDL3a, and that HDL3b levels were inversely correlated with levels of both HDL2a and HDL2b.