Cystic fibrosis (CF) lung disease has been linked to multiple primary defects in airway epithelia caused by a dysfunctional cystic fibrosis transmembrane conductance regulator (CFTR) gene. These defects include altered Cl- and Na+ permeability as well as intracellular defects in glycoprotein processing. This apparent diversity in CFTR function is reflected in the complex patterning of CFTR expression in airway epithelia. Such complexities present challenges in the design of CF gene therapies that are capable of reconstituting the endogenous patterns of CFTR gene expression in appropriate target cells. Using a human bronchial xenograft model of the CF airway, we have evaluated the efficacy of recombinant adenoviral and cationic liposome-mediated gene transfer to correct Cl- permeability and mucous sulfation defects found in CF lung disease. Results from these studies demonstrated a clear vector-specific complementation profile for these two defects that was dependent on the type of cell transduced and the level of transgene expression. Single-dose administration of recombinant adenovirus effectively transduced high levels of CFTR transgene expression in 11 +/- 1% of epithelial cells and was capable of correcting cAMP-induced changes in Cl- permeability to 91 +/- 14% that seen in non-CF airways. However, this level of transgene expression was incapable of reversing defects in mucous sulfation due to the lack of efficient targeting to goblet cells. In contrast, cationic liposome-mediated delivery of CFTR encoding plasmids to CF airways achieved extremely low levels of transgene expression with insignificant correction (7.4 +/- 2.4%) of cAMP-induced Cl- permeability. This low level of transgene expression, however, efficiently reduced mucous sulfation to levels seen in non-CF airways. Differences in the complementation profiles of these two vectors in correcting Cl- permeability and mucous sulfation defects mirror the ability of recombinant adenovirus and liposomes to reconstitute only certain features of the endogenous distribution and abundance of CFTR protein expression. Such findings suggest that the level of intracellular CFTR required to facilitate proper glycoprotein processing may be much lower than that needed to mediate bulk Cl- flow across the airway epithelium. In summary, these data present the first example by which two different vector systems can efficiently complement independent primary defects associated with a single dysfunctional gene.