At a Texas meeting May 14, USC researcher Roberta Diaz Brinton will discuss her research and the hopes she has for the colonies of rat nerve cells she has cultured -- and even sculptured -- on silicon semiconductor substrate. She envisions that hybrid brain-electronic systems may soon serve to elucidate the process by which brains perform complex functions, including specific pattern recognition. Such systems may also form part of a new generation of chemical detectors.
Brinton, an associate professor of molecular pharmacology in the School of Pharmacy, also plans to use her wired colonies of cells for studies in her specialty, the study of the effects on the brain of the hormone estrogen and allied substances.
Finally and most ambitiously, Brinton and co-workers from the USC School of Engineering's department of biomedical engineering hope to someday be able to implant similar chips into living brains as both experimental instruments and eventually as prostheses.
The cells come from a part of the brain called the hippocampus, which is associated with short term memory. Brinton dissociates the cells from their intricate network of interconnections with each other, and then places them as individual cells on a specially prepared silicon testbed studded with a matrix of electrodes.
The dissociated cells, approximately 80,000 of them, affix themselves to the testbed and grow, sending out processes and reforming synaptic contacts with each other. They can live on the chips for extended periods -- as long as months, according to Brinton.
Brinton's lab is one of only a very few in the world able to culture and study such nerve cell colonies.
During the course of the nearly two years Brinton has been working with the technique of culturing on silicon, she and her co-worker, graduate student Walid Sassou have learned to make and use masks made by traditional photo-resist processes to "sculpt" the colonies into desired shapes by placement of the substrate coating to which the cells can adhere. Using this technique, the colonies can be made on certain electronic contacts, encouraging some connections while preventing others.
"We call these 'designer circuits,' said Brinton. "We arrange it so the neurons can only grow in certain directions. We can create paths for the neurons to grow in."
The electrode matrix the neurons grow on can serve as both inputs for the cells, and monitors of their activity.
With the entire array in monitoring, Brinton and collaborators like Ted Berger of the department of biomedical engineering, or neurobiologist Michel Baudry of the department of biological sciences, can both observe the spontaneous patterns of synaptic activity that occur with the neurons talking to each other.
"We can eavesdrop," said Brinton, "and try to find the algorithms that define their activity.
The researchers can also stimulate their nerve cell colonies at any desired point, and then monitor the cellular synaptic activity that results. "We are now talking to the networks of cells that we have grown," said Brinton, "and they are replying. We soon should have a much better idea of what they are saying." According to Brinton, the chemical activity that takes place at the synapses of hippocampal neurons is well understood, and can be manipulated at the individual neuron level by administration of minute, precisely-measured quantities of substances to enhance or inhibit any given neurotransmitters.
The result is a "bionic chip" that Brinton believes can eventually do information processing on its own.
In addition to helping researchers better understand how the brain functions, Brinton said such chips may form part of useful devices that might serve as chemical sensors for large complex, biochemically active substances, or even specific microorganisms.
But the most exciting prospects revolve around the possibility of actually interfacing a chip with living brain cells.
Berger and collaborators from the School of Engineering have developed special chips that duplicate activity patterns of sections of the hippocampus. These chips are composite, containing numerous separate components, each of which acts like a single cell.
"If some day, we could get living cells to interact and talk to such chips," Brinton said, "we might some day be able to implant it into a living brain." Such an implant might restore function lost by disease or injury, Berger said. "But the first step toward this is having a silicon chip live with and communicate with nerve cells," said Brinton, "and this is what we have begun to do."
Brinton will present an overview of her work at the University of Texas, Houston School of Medicine at noon May 14, at a department of neurology seminar. Her research was funded by the Defense Advanced Research Projects Agency, the National Institutes of Health and the National Science Foundation.