A chemically defined medium was developed to study liposome-mediated delivery of methotrexate-gamma-aspartate to cells under conditions where dilute suspensions of negatively charged liposomes to not leak extensively. The defined medium induced 14% leakage of methotrexate-gamma-aspartate from egg phosphatidylglycerol/cholesterol (67:33) liposomes diluted to 53 nM lipid. In contrast, commercially available serum replacements induced up to 91% leakage from the same liposomes. The growth inhibitory properties of non-loaded phosphatidylglycerol liposomes were greater in the chemically defined medium that they were in medium supplemented with 10% serum. Egg phosphatidylglycerol, dioleoylphosphatidylglycerol and dilaurylphosphatidylglycerol liposomes inhibited cell growth more than dimyristoylphosphatidylglycerol and dipalmitoylphosphatidylglycerol liposomes. In 10% serum, phosphatidylglycerol liposomes with widely varying phase-transition temperatures were nearly equally effective to deliver drug to CV1-P and L929 cells, despite great differences in liposome stability. Liposome encapsulated methotrexate-gamma-aspartate was more potent when the cells were grown in the defined medium, and the increase in drug delivery was observed from phosphatidylglycerol liposomes of different phase-transition temperatures. The minimum fraction of negatively charged phospholipid required for optimal liposome-mediated drug delivery varied between cell types and among growth media. The growth inhibitory effects of liposome-encapsulated methotrexate-gamma-aspartate was also determined under conditions where the cells were exposed to drug for periods shorter than the entire growth assay. Reduction of the exposure time decreased the potency of both encapsulated and free drug in medium containing 10% serum, and decreased the potency of free drug in the defined medium. However, the potency of encapsulated drug in the defined medium was similar for all exposure lengths between 1 and 48 hours.