A novel alternative to animal skin models was developed in order to aid in the screening of transdermal penetration enhancers. The skin alternative consists of a dermal layer containing human fibroblasts dispersed in a collagen matrix and an epidermal layer of differentiated and stratified human keratinocytes. Skin alternatives were placed in modified Franz diffusion cells (receptor volume 12 mL, donor area 3.14 cm2, n = 5) and enhancer solution (0.4 M in propylene glycol (PG)) was applied. Following 1 h of pretreatment, 10 microL of saturated hydrocortisone (HC) solution in PG was applied, and the cells were occluded with Parafilm. Samples were removed from the receptor compartment over 24 h, replaced with fresh receptor solution, and analyzed for steroid content using HPLC. Skin HC content was also determined. Receptor concentration at 24 h (Q24) for full-thickness skin alternative (control) was 28.6 +/- 13.7 microM and permeability (P) was 8.3 x 10(-4) +/- 5.5 x 10(-4) cm h-1. Azone (1) produced a Q24 of 105.0 +/- 36.1 microM and a P of 11.3 x 10(-4) +/- 1.8 x 10(-4) cm h-1, while the novel penetration enhancer 1-dodecyl-2-pyrrolidinone (2) produced a Q24 of 164.8 +/- 61.2 microM and a P of 33.3 x 10(-4) +/- 6.6 x 10(-4) cm h-1. Compound 3 produced the highest values for all permeability parameters tested with a P of 48.0 +/- 36.8 cm h-1 and a Q24 of 186.1 +/- 45.1 microM. When compared to the control, compound 1 gave an enhancement ratio (ER) of 3.7 for Q24 and 1.4 for P. Compound 2 gives an ER 5.8 for Q24 and 4.0 for P. These enhancement ratios are similar to those found using HC and human skin.