CHAMPAIGN, Ill. In an attempt to decipher the communications codes used by mammalian brain cells, University of Illinois researchers are using chip lithography to "microprint" furrows that growing brain cells will follow when budding inputs (dendrites) and outputs (axons). Theoretically, the method would use off-the-shelf electrode arrays and allow engineers to characterize living nerve cells by precisely "wiring" together test circuits and measuring their performance.
Manipulating the attachment and growth patterns of individual nerve cells has the potential of creating "designer" biosensors, implants and prosthetics, the researchers said.
"There is a long history of trying to influence cell growth with patterns that are the same size as cells, [or] about the same size as transistors 1 to 10 microns," said professor Bruce Wheeler, "but it's only become feasible in the last few years. Our lab is the first to use microprinting in this biological context." Wheeler's work was done with UI professor Deborah Leckband and colleagues at Southern Illinois University School of Medicine in Springfield: professor Gregory Brewer and research assistants John Chang and Johnny Nam.
Most attempts to characterize the performance parameters of living neurons have been preoccupied with keeping them alive a nutrient bath needs to feed them as they bud out and grow. Organizations like Draper Laboratory (Cambridge, Mass.) are perfecting the use of chip lithography to pattern the growth of vital organ cells into "vascularized scaffolds" artificial organs that can someday be substituted for diseased organs in humans.
Commercially available electrode arrays coupled with Wheeler's wiring methodology now make culturing of brain cells equally feasible. The electrode array provides a means of monitoring the electrical performance of the network. Wheeler claims to have toppled the last obstacle: getting them to grow into the predictable, repeatable circuits needed for systematic performance characterization.
"The problems seemed to center on the interface between the cells and the substrate surface, and understanding that better will be very complicated we are only just getting going on it, but at least now we have a method of proceeding," he said.
Wheeler's method involves printing a pattern on a substrate using lithography in a manner similar to the way signal paths are laid down on chips. The pattern overlays the electrode array and provides pathways for the "wires" connecting the circuit into the desired configuration. The neurons sit atop the electrodes and send out their dendrites and axons the wires along the signal paths to "auto-route" themselves as they grow.
The construction of living nerve cell circuitry begins two stages before the "furrows" actually a protein that budding dendrites and axons can easily adhere to can be stamped on a substrate. First a mold must be made with which the microstamps can be cast. To make the mold a glass substrate is put in a semiconductor oven where vapor deposition is used to put down a layer of polyimide. Then, atop the polyimide, a layer of titanium is deposited.
At this point, traditional chip photolithography steps are applied dry etch,then patterning of photoresist into the shape in which the cells should grow atop the titanium. Finally, the non-patterned material is etched off. The patterned shape is then transferred to the polyimide and the titanium mask is removed by means of wet etch, leaving a polyimide mold. Polydimethylsiloxane is then poured into the mold, cured and mounted to a holder for stamping.
Stage two consists of dipping the stamp into a "friendly" protein (polylysine, an artificial polymer commonly used for cell cultures) and applying the wet pattern to the surface where the cells will be cultured. As a further enhancement, UI's Leckband placed a layer of polyethylene glycol a molecule unfriendly to living cells in the places where the pattern is not stamped, thereby successfully reducing unwanted cell adhesion and cell growth in portions of the array where there are no electrodes.
Brewer of Southern Illinois University has been removing individual brain cells from developing rat embryos and then chemically and mechanically separating the cells so that they may be poured onto the patterned polylysine. The severed neurons "find" the polylysine-patterned lines on their own, automatically attach to them and begin growing their input dendrites and output axons atop the friendly pattern. Within a few days, Wheeler said, the cultured neurons send their axons straight down the imposed pattern's lines as far as 1,000 times their cell width, or about 1 mm. A few days more, and they automatically begin communicating with one another using their distinctive electrical-pulse encoding.
"Our nerve cells remain viable for up to one month while maintaining compliance to the microstamped patterns," he said
By controlling the cell's response, Wheeler hopes to improve on today's medical implants, which tend to lose electrical sensitivity over time. "We hope to improve the long-term stability of biological activity in implants," he said. "Also, because the brain has ordered layers of cells, we believe that developing techniques for maintaining the orderly growth of the neurons in culture will lead to greater insight into brain activity itself."
Wheeler hopes to further develop the microstamping technique by depositing different types of biomolecules from aqueous solution onto the glassy substrates, eventually creating hierarchies of patterned biomolecules in a manner similar to natural neural development cycles.