Dumbell drugs
8 Apr 2005
DNA could be the key to the fast, efficient production of cancer drugs based on nanoparticles, according to researchers at the University of Michigan.
The team, led by James Baker of the Centre for Biologic Nanotechnology at the university, uses DNA as a linking tool to attach two different types of nanoparticles — one to seek out cancer cells, another for imaging or to destroy the cell.
‘With this approach, you can target a wide variety of molecules — drugs, contrast agents — to almost any cell,’ Baker says.
Baker’s team is investigating the potential of dendrimers as anticancer agents. These molecules have many branches, so it is theoretically possible to attach different groups to various parts of the molecule to target specific cells and to have a therapeutic effect.
The dendrimers are nano-scale, and can enter cells to destroy them from within.
But making them is a complicated and inefficient process, with many steps and a very low final yield. Baker’s team has found a way to reduce the complexity of the process by using two different dendrimers. Each molecule carries a single strand of DNA, specially designed so that it doesn’t code for a protein, but can recognise and bind to the complementary base pairs on the other dendrimer.
When the two strands of DNA knit together, the two dendrimers form a dumbell-shaped molecule.
A graduate student on Baker’s team, Youngseon Choi, synthesised a version which carried folate, a molecule which binds to cancer cells, at one end, and a fluorescent group at the other, as a proof of concept. Baker envisions using the system as a toolkit to build custom therapeutic molecules.
‘If you wanted to make a therapeutic that targeted five drugs to five different cells, it would take 25 synthetic steps the traditional way,’ Baker says.
And with each synthesis taking two to three months, and with yields declining at each step, the process would not be practical. Instead, the group intends to make a library of dendrimers with a single function, which can be synthesised in parallel, and linked together as needed. This would reduce the number of steps from 25 to 10, Baker says.
