For decades, researchers have pursued methods to efficiently bring impermeant macromolecules into the cytosol and the nucleus of living cells. The list of established techniques is long. It goes from chemical loading (temporarily permeabilizing cells with a chemical agent), to vehicle-based loading (fusing cells to a cargo-carrying liposome or a red blood cell ghost), to mechanical loading (microinjection or wounding of the cell membrane), to the use of cell-penetrating peptides fused to cargo, to electroporation.

Speedy cargo delivery. Credit: Marina Corral

Surprisingly, none of these methods work universally well for large molecules other than DNA. The problems are manifold: the size of the delivered cargo is restricted, uptake can be inefficient, cargo may be trapped in endocytic vesicles or aggregate inside the cell, delivery can be restricted to a certain cell type, cells are often damaged in the process, and some methods are time- and labor-intensive, requiring special skill and equipment.

Constant improvements to delivery techniques aim to overcome these hurdles, and the potential applications are legion. To pick one example: the effective delivery of antibodies labeled with dyes or quantum dots would enable studies of their endogenous targets' subcellular distribution in live cells in real time. One could visualize chromatin modifications with antibodies to histone modifications as well as other chromatin-associated proteins and follow their dynamic changes. Hiroshi Kamura and colleagues recently showed the potential of such an approach (J. Cell Biol. 187, 781–790; 2009) when they used mechanical loading to deliver dye-labeled Fab fragments against histone H3 phosphorylated on serine 10, a histone mark essential for chromosome segregation, to live cells and preimplantation mouse embryos.

The delivery of fluorescent antibodies would mean that it is no longer necessaryto express fluorescent fusion proteins or fuse the protein of interest to a peptide tag that binds a fluorophore-conjugated chemical. Instead, any endogenous proteins for which a good antibody exists could be localized, even at a resolution below the diffraction limit in the case of super-resolution microscopy. What stands between the universal realization of these and other applications are lipid bilayer membranes that prove surprisingly hard to penetrate.