Estrogen treatment raises plasma high density lipoprotein (HDL) levels, which may reduce cardiovascular risk. To identify the responsible mechanisms as well as the importance of the route of administration, we treated eight healthy postmenopausal women in a double-blind crossover study with three treatments for 6 weeks each: oral estradiol, 2 mg daily; transdermal estradiol, 0.1 mg twice weekly; and placebo. At the end of each treatment, apoA-I of HDL2 (d 1.063-1.125 g/ml) and HDL3 (d 1.125-1.210 g/ml) was endogenously labeled by a constant intravenous infusion of trideuterated leucine. HDL2 and HDL3 were separated by preparative ultracentrifugation. The pool sizes and enrichment curves of HDL apoA-I were used to calculate production rates and fractional catabolic rates (FCR). Oral estradiol increased the levels of HDL2 apoA-I by 37% (P < 0.005) and of HDL3 apoA-I by 11% (P < 0.05). These increased apoA-I levels resulted entirely from increased production, by 36% for HDL2 (P < 0.01), and by 19% for HDL3, (P < 0.05) as their FCRs were unchanged (0.20 pool/d with placebo and 0.21 with estradiol for HDL2, and 0.19 with placebo and 0.21 with estradiol for HDL3). The isotopic enrichment curves of HDL2 apoA-I and HDL3 apoA-I were identical, implying that apoA-I rapidly cycles between HDL particles, or that rapid interconversion of these subfractions occurs. The changes in HDL apoA-I metabolic rates were positively correlated with changes in VLDL-apoB metabolic rates measured previously. Transdermal estradiol, with systemic potency similar to that of oral estradiol, had no significant effect on HDL levels or metabolic rates. Thus, the "first pass" effect of oral estradiol on the liver and/or intestine appears to increase HDL apoA-I levels (particularly in HDL2) by increasing HDL apoA-I production, and not by reducing HDL apoA-I catabolism.