Liposome functionalization with copper-free "click chemistry"

The modification of liposomal surfaces is of interest for many different applications and a variety of chemistries are available that makes this possible. A major disadvantage of commonly used coupling chemistries (e.g. maleimide-thiol coupling) is the limited control over the site of conjugation in cases where multiple reactive functionalities are present, leading to heterogeneous products and in some cases dysfunctional conjugates. Bioorthogonal coupling approaches such as the well-established copper-catalyzed azide-alkyne cycloaddition (CuAAC) "click" reaction are attractive alternatives as the reaction kinetics are favorable and azide-containing reagents are widely available. In the work described here, we prepared lipids containing a reactive cyclooctyne group and, after incorporation into liposomes, demonstrated successful conjugation of both a small molecule dye (5′-TAMRA-azide) as well as a larger azide-containing model protein based upon a designed ankyrin repeat protein (azido-DARPin). By applying the strain-promoted azido-alkyne cycloaddition (SPAAC) the use of Cu(I) as a catalyst is avoided, an important advantage considering the known deleterious effects associated with copper in cell and protein studies. We demonstrate complete control over the number of ligands coupled per liposome when using a small molecule azide with conjugation occurring at a reasonable reaction rate. By comparison, the conjugation of a larger azide-modified protein occurs more slowly, however the number of protein ligands coupled was found to be sufficient for liposome targeting to cells. Importantly, these results provide a strong proof of concept for the site-specific conjugation of protein ligands to liposomal surfaces via SPAAC. Unlike conventional approaches, this strategy provides for the homogeneous coupling of proteins bearing a single site-specific azide modification and eliminates the chance of forming dysfunctional ligands on the liposome. Furthermore, the absence of copper in the reaction process should also make this approach much more compatible with cell-based and in vivo applications.