"For many years I've worked on new techniques for treating malignant brain tumours, and particularly on how therapeutic agents can be delivered directly to the bulk brain tissues via infusion through the interstitial space in the brain," explained George Gillies of the University of Virginia. "About 5 years ago we worked out a technique for using an inexpensive agarose gelatin as a surrogate for brain tissue to test various kinds of infusion-based drug-delivery techniques."

According to Gillies, such gels promised to reproduce certain features of in vivo infusions into mammalian brains. But the scientists were curious as to what it was about the gel's structure that made it such a useful surrogate for living brain tissue. "We decided that we needed to explore [the gel's] nanoscale porosity to help answer that question," said Gillies.

To do this, the team infused a 0.6% agarose gel with a slurry of rare-earth-doped phosphor nanoparticles typically 8-12 nm in size. The nanoparticles were Y2O3:Eu3+, which fluoresces visible light when excited with UV radiation. Tracking the dispersion of the nanoparticles using UV irradiation revealed that they spread through the interstitial structure of the gel.

"Our infusions of the fluorescent nanoparticles through the internal structure of the gel open an interesting window on the scale of the inter-crosslink voids that exist within this polymeric material," said Gillies. "The results indicate that the characteristic size of these voids in the gel is similar to those found in the extracellular space in the brain, thus helping to explain why the volumes of distribution of infusions into both substances tend to be very similar. This, in turn, gives us added confidence in our use of the gel model as a surrogate brain when carrying out tests of the function of drug-delivery neurocatheters, etc."

Previously, the scientists found that lambda phage viruses - which are typically at least ten times as large as the nanoparticles used here - would not diffuse into the agarose gel. As a result, the team proposed that the equivalent of a pore-size cut-off for low-concentration agarose gels lies in the 10-100 nm range.

Now the team wants to infuse the gel with larger nanoparticles to determine a more precise value for the transport cut-off, and to find out how the cut-off varies with gel concentration.

"Some other colleagues and I are putting the finishing touches to a paper in which we refine the gel model to mimic some gross features of the structure of the spinal cord," added Gillies. "The goal of that work is similar - that is, to help provide a test bed for the rapid and inexpensive evaluation of certain aspects of new therapeutic strategies, in this case for the delivery of therapeutic agents into the spinal cord to treat paralysis caused by spinal-cord injuries. We are very excited about the possibilities that could arise from this approach."