- American Physical Society Sites
- Meetings & Events
- Policy & Advocacy
- Careers In Physics
- About APS
- Become a Member
By Henry D. I. Abarbanel and Allen I. Selverston
Figure 1: The blue trace is for the electrical neuron, the red trace is the current flowing between the neurons, and the black trace is the membrane voltage activity of the biological neuron.
Physics and medicine have a rich joint history. Ben Franklin wore spectacles to correct his short sightedness. No doubt today he would have had laser surgery to reshape his cornea. And can you imagine a doctor diagnosing your injured ankle, knee, or shoulder without x-rays and MRI or treating your problem without arthroscopy? You expect rods and screws, made from space-age materials, to be implanted in your broken bone to ensure its rapid and strong recovery. And soon you will have a remotely powered micro-electromechanical strain gauge implanted to provide a real-time measure of the healing process and to permit your physician to prescribe an optimum course of physical therapy. Cancer is still an enormous problem, but you can expect a cancerous tumor to be successfully treated by carefully focused radiation, monitored in real time by a large area amorphous silicon imaging system to ensure that you receive the correct dosage.
As innovative and important as these contributions are, they do not help in the treatment of problems within the nervous systems. Many people are paralyzed for life by injuries to their spinal cords. Just recently, President Clinton in his State of the Union address challenged scientists, engineers, and physicians to develop a chip that when implanted would relay the severed signal to the isolated limbs and restore their function. And many sufferers of Alzheimer's disease might benefit from bio-circuits that serve as brain pacemakers.Not surprisingly, American research universities are attracted by the intellectual challenges of the problem. Stanford's Bio-x Program and the University of Chicago's Institute for Biophysical Dynamics are two prominent examples. Stanford has raised $350 million to support Bio-x with significant leadership from physicist Steven Chu and an individual donation of $150 million from Netscape founder Jim Clark. Other equally staggering amounts have been raised elsewhere. Such ambitious R&D projects will certainly lead to major new directions and understanding in each of the basic sciences involved but should also lead to the next wave of physical science in medical practice.
Neuroscience is now in a state where the maturity of the biological experimental base proves fecund for the skills of a physicist monitoring the activity of individual neurons as they participate in the functioning of a network. We have studied the properties of a biological neural network comprised of fourteen neurons, which act as a control system in the California spiny lobster. The network directs muscles surrounding the so-called pyloric chamber to contract and dilate in a complex pattern that moves shredded food from the stomach to the digestive tract of the lobster. When the physics/biology interaction was initiated through a casual conversation, the biologists knew the interconnections among the component neurons but were seeking ways to understand the functioning of the network. The physicists, working in nonlinear dynamics, saw an opportunity to model a functioning network of nonlinear oscillators and learn something of the biology in the process.
After analyzing the cross membrane voltage of many of the component neurons in the pyloric circuit when they were isolated from the rest of the circuit, we discovered that the degrees of freedom expressed in the voltage activity was typically only three or four. As the voltage activity is the signaling method for communication among neurons and responsible for the functioning of the network, we can simulate and even replicate the entire network in simple electrical circuitry. In fact, we realized that we could purposely damage the network by selectively removing key neurons, and then restore the damaged activity to good health by electrical circuitry.
After an intense period of modeling, we created an electronic neuron in the spring of 1999 and inserted it in place of a deliberately damaged neuron in the lobster's neurological circuit. By themselves, the remaining intact biological neurons oscillated quite irregularly producing electrical signals that could not be interpreted by the muscles but the mutual interaction of the electronic neuron with the biological neurons restored the natural activity of the circuit. Figure 1 shows a sample of the experimental data.
Small neural networks like these are found in humans and other animals where they have the functional task of driving rhythmic activity of muscles. While the circuitry is not as well mapped as that of the lobster and developing a successful surgical procedure is a major challenge, we believe we are seeing the earliest stages of a new neural therapy.
Not all areas of biological sciences are ready for the mathematical, analytic, and modeling skills and inclinations of physicists, but many clearly are. With the sequencing of the human genome imminent, predictive models of the proteins expressed by genes take on an immediacy and provide an arena in which the skills of physicists will be immediately productive. Other areas ready for interaction with the quantitative methods of the physical sciences include the dynamics of folding proteins, "bio-inspired nanomaterials," signaling between cells, and similar questions.
The opportunities for physicists in many fields of biology, neuroscience in particular, are open to those with physics undergraduate degrees. There are now a few programs where physicists can receive the training in wet lab neurophysiology utilized in our example, and biologists can be trained in the skills of the physicist. Starting in such a program one may be part of a remarkable next wave of physical science and medicine, and the beginning of a creative career that ultimately benefits the whole of society.
Henry D. I. Abarbanel is a professor of physics and research physicist, Marine Physical Laboratory, Scripps Institution of Oceanography. He is a member of the Panel on Public Affairs of the APS. Allen I. Selverston is a research professor of biology. Both are at the University of California, San Diego.
|The electronic neuron implements three ordinary differential equations describing the observed dynamics of the neurons in the lobster circuit. Our colleague Alexander Volkovskii designed and built a simple analog circuit which integrates three dynamical equations, basically Kirchoff's law, |
and implements a model of the I/V characteristics of one 'slow' z(t) and one 'fast' y(t) transmembrane ionic current. The circuit equations
produce membrane voltages x(t) which accurately represent the wave forms created by the biological neurons. The analog circuitry was then implemented on a PC board and instrumented so its parameters could be easily and reproducibly set. The experimental electronic neuron is shown in the first figure. Next to the electronic neuron we show a composite picture of a neuron from another crustacean, the crab, taken by D. Baldwin and K. Graubard of the University of Washington (J. Comp. Neuro. 356, 355-367 (1995)). The dark region under the number 7 is the neuron body; the rest of the dark regions are the axons projected from this neuron in its communication network. The electronic neuron reproduces the electrical activity of the neuron body; other analog circuits were built to represent the neural interconnections. The work was reported in Neuroreport, 83, Feb. 28, 2000). For below figures: (left) Analog Circuit of an Electronic Neuron; (right) Neuron from the Crab stomatogastric ganglion
©1995 - 2022, AMERICAN PHYSICAL SOCIETY
APS encourages the redistribution of the materials included in this newspaper provided that attribution to the source is noted and the materials are not truncated or changed.