One of the neat tools in molecular biology is the ability to recombine parts of two proteins to create fusion or chimeras. They're often extremely useful for simple experiments, some of the time for targeting protein domains to subcellular sites, or to isolate a structural component of a protein. The functional information is often quite interesting.
One manuscript recently in Nature Precedings used the chimeric molecule approach slightly differently. John Rossi and coworkers describe a aptamer conjugate for delivering anti-HIV siRNA specificially to infected cells. The aptamer in question is a gp120-interacting molecule. The envelope glycoprotein gp120 is expressed on the surface of HIV-1 infected cells, allowing binding and interalization of the whatever is conjugated to the aptamer, in concept. In this case, Rossi et al. have conjugated the chimera to an siRNA that releases an anti-tat/rev siRNA, which in turn inhibits HIV replication.
Not too complicated. Similar simple strategies for specifically targeting cancer cells, or any of a number of other medical conditions have been a fad since at least the mid- to late- 1990's. Rarely do they appear to live up to their expectations however. There are just too many molecules in the body with cross-specificity. And even if you get your therapeutic payload to the target cells, it may not be efficiently presented. Especially for delivery of genes to be expressed as proteins or siRNA in vivo – how do you ensure efficient transcription?
And this is what Rossi et al. present – a new and promising approach that they demonstrate in cell culture, but it is a guessing game as to whether this proof-of-concept approach will actually work in the clinical setting. I sympathize, I just got done writing a research proposal that aims to use biomolecular conjugates for drug targeting.