| Home |
Reflections of a comparative physiologist
and comparative morphologist in a college of engineering
As a boy I was fascinated by the amazing varieties of animals and plants everywhere I looked-- in my family's yard in Whittier, in the still-mostly-wild Puente Hills that began half a block away, in the now-gone, deep, spring-fed stretch of San Jose Creek about two miles from home, along the shore of Lake Hughes next to our summer cabin in the desert hills, in the Mojave Desert itself, in the narrow but tall San Gabriel Range between the desert and the coastal basin, in the rocky tide pools of the coast, and in the deeper water just beyond them. It was what we now call biodiversity. When I entered Stanford as a freshman, biologist was what I wanted to be; biodiversity was what I wanted to study. At Stanford, I found the latter in comparative anatomy, the study of structural diversity wrought by evolution, comparative physiology, the study of diversity of mechanisms wrought by evolution, and ecology, the study of niches and of evolutionary forces driving diversity. Because I loved STEM areas outside of biology, especially math and chemistry, I decided to concentrate on physiology. That led me to think about instrumentation for physiological studies, and that led me to what amounted to a double major-- Biology and Electrical Engineering. A full load of graduate-level EE courses in the summer at the end of the fifth year converted my EE work to an MS program. A full load of biology courses in the summer between my second and third years had helped fill the bioscience elective requirements for the AB in biology. I invested two more years on a post-MS degree (Engineer degree), continuing to combine physiology with EE and other STEM areas. Then I spent two years focusing on physics and physical electronics. My motivation for that had nothing to do with biology; I had simply become passionate about EE itself.
For his doctoral research under Ted Bullock at UCLA, Bob Josephson had worked on activity patterns in coelenterate (Cnidarian) nerve nets. Part of this work had been carried out at Librascope, the computer division of General Precision Inc. Bob had worked there with mathematician/computer-scientist Dick Reiss, carrying out digital simulations of nerve-net activity. As this work was drawing to a close, Bullock and Reiss decided to submit a proposal for another neural simulation activity-- modeling the nine-neuron network (cardiac ganglion) that serves as the pacemaker oscillator for the heart of a lobster. Thanks to the extensive work of Bullock and his colleagues, neuron by neuron this nine-cell structure was the most thoroughly studied neural network on the planet. In its cell bodies and dendrites it exhibited a vast array of subthreshold activity-- spontaneous membrane-potential oscillations of various shapes (from nearly sinusoidal to sawtoothed), single-spike synaptic responses that could be enhanced (facilitated), diminished, or neither as the input spike-bursts continued, accumulated synaptic potentials that could be followed by positive or negative rebound or no rebound at all when the input spike-bursts ceased. In some cases, invading spikes from the trigger region of a cell would partially or totally reset all of this activity. The goal of the project was to put this array of phenomena together in an electronic simulation and thus create a faithful model of the cardiac ganglion. The project was funded by the Air Force Office of Scientific Research.
Now all Ted and Dick needed was someone to do the work (Bob was off to an academic career that began at the University of Minnesota). I was traveling in the eastern U.S., visiting industrial research labs involved in semiconductor device research, considering where I might start my professional career. In Johnson City, New York, I received a call from Dick. Would I be interested in doing electronic simulation of this little piece of nervous system? The semiconductor industry would have to wait- I was heading back to biology.
I began as Dick and Ted suggested, building circuits to simulate each of the subthreshold phenomena that Ted and his colleagues had observed. Then I realized that there was no realistic way to organize these circuits into a whole-neuron or whole-ganglion simulation. I needed another approach. The electrical engineer in me realized three things-- (1) each of these subthreshold phenomena was being produced by a local patch of cell membrane, which could be modeled as a two-terminal, active circuit**, (2) if I could simulate those patches, then to simulate the cell-body/dendriditic region of each cell, I could connect the simulated patches into branching ladder networks with resistive series elements, and (3) I should be able to recreate each local patch of membrane with simulations of those two opposing ion currents (sodium and potassium) that Hodgkin and Huxley had documented plus, in some cases, Eccles-type synaptic conductances. In other words, I should be able to recreate everything Bullock and his colleagues had observed simply by combining (in parallel) simulated synaptic conductances, potassium conductances and sodium conductances in proper proportions. It worked. We didn't know about ion channels yet; but in terms of ion channels, the important parameters would have been overall channel density and the relative proportions of the channel types. Within about four years experimental neurobiologists were begining to show that some of these subthreshold phenomena were consequences of interactions between calcium conductance (rather than sodium conductance) and a form of potassium conductance that was different from that described by Hodgkin and Huxley. My theory was sound, H&H sodium and H&H potassium conductances would do the job, but the actual biology was something else. I clearly needed to move to the experimental side of neurobiology.
**footnote: Imagine a box with an electric circuit inside and just two wires passing through the wall of the box, connecting the circuit to the outside world. In the case at hand, the box would contain the simulated patch of membrane; one wire would represent the local interior of the neuron; the other wire would represent the local exterior of the neuron.
By the time six years had passed, Dick Reiss had moved to England to be an Oxford don and Ted Bullock had moved to La Jolla and joined the faculty at UC San Diego. I had renewed the original 3-year AFOSR grant and augmented it with a neural-simulation facilities grant from the Bionics group at Wright-Patterson Air Force Base; and the work on both those grants was done (note). Over and over again during those six years I had found that coming to neurobiology as an engineer had given me a huge advantage. The natural science of neurobiology is filled with concepts and theoretical elements that are treated deeply and with considerable rigor in engineering. In the 1960s, for example, engineers were bringing amazing clarity to our understanding of neurosensory and neuromotor systems by applying the theories of feedback control, signal processing, and communication-- each of them a part of the electrical engineering repertoire. The theory that I had found most useful in my work had been electrical network (or circuit) theory {SEE NOTE}. My goal now was to move to the periphery, to a primary sensory system or a peripheral motor system-- to a system whose functions (selective advantages) I might infer, and which I might be able to reverse engineer.
I also had accepted a position in the EE Department at Cal. I had two goals as I came to Berkeley: (1) to encourage students to consider graduate studies and careers combining EE and biology, and (2) to establish a wet-lab for my neurophysiological studies. My research soon turned to reverse engineering of the vertebrate ear and I spent the next thirty five years combining comparative anatomy and comparative physiology of that structure. Biology was my natural science, electrical engineering provided tools and insights. When I arrived at Cal, I was comfortable as a neurobiologist and I was equally comfortable as an electrical engineer. I had encountered the label bioengineering, but was not at all clear on what it meant to be a bioengineer. My thirty five years at Cal did not help me figure that one out; I'm still not clear. My friend Larry Stark once said to me "Ted, you and I may teach courses for bioengineers, but we really aren't bioengineers." I am clear on that, Larry. When I came to biology for my career in research, I came to it as an electrical engineer, not a bioengineer. Many others before me had done the same thing. What I did in biology was reverse engineering-- again as many others had done before me (link to reference). Many had come as mechanical engineers. We all were engineers in biology. Interestingly, my colleagues Ed Keller and Frank Werblin-- two early products of the Johns Hopkins PhD program in bioengineering, also were reverse engineers. The Johns Hopkins program trained them specifically to do that. The same sort of training is available in the Joint UCB/UCSF Graduate Program in Bioengineering. In other words, one now can come to reverse engineering in biology as a bioengineer.
Comparative physiology and comparative anatomy remain alive and well
Sometime in the 1990s, the director of the NIH National Institute of General Medical Sciences (NIGMS) gave a talk in Cory Hall, the home of electrical engineering at Berkeley. Among other things, he told us that the future of biology was at the molecular level, that the fields of anatomy and physiology were anachronisms. I suppose, at the time, the most compelling rebuttal might have been to mention the career shifts of two giants of reductionism: George Zweig, co-inventor of the quark theory, and Francis Crick, co-discoverer of the genetic code. Both men had become immersed in neurophysiology and neuroanatomy, both had become reverse engineers of the nervous system.
As a boy I was most fascinated by the worlds of Alfred Russel Wallace and Alexander von Humboldt, Jacques Cousteau and Rachel Carson. It was not only biodiversity, but also nature's animation of the creatures in those worlds that drew me to them. It seems most natural to me that my research career has been focused on the structures and operations of nervous systems-- it is nervous systems that not only make purposeful, coordinated animal motor activity (nature's animation) possible, but make it possible as well for me and my fellow creatures to perceive it. As the origin of new beings was the mystery of mysteries for Darwin and Wallace in the 19th century, the origin of conscious perception is the mystery of mysteries for all of us in the 21st. It was that mystery that drew Francis Crick into neurobiology. George Zweig was drawn to it by the more tractable goal of understanding signal processing in the vertebrate ear. The vertebrate ear is where I came down as well. Of the enormous vertebrate nervous system it's a tiny, peripheral piece that I thought I might be able to understand.
My late friend and office mate, Marty Graham, was an extraordinarily talented electronics engineer. In the late 1950s he supervised the design and construction of one of the last state-of-the-art digital computers that employed vacuum tubes (it was called R1, The Rice Computer, and in 1960 it was reputed to be the biggest, fastest digital computer on any campus). As an old engineer whose introduction to physical electronics was with vacuum tubes, I sometimes try to imagine the extent to which my knowledge of vacuum-tube physics alone would help me understand the capabilities of Marty's computer. I've played that game before, in print; and so have many others. I'll not play it again here. I'll simply say that serious neurobiologists would not look to molecular or cellular processes alone to understand higher-level neural function-- e.g., how a human listener, using just one ear, can mentally separate the sound of the oboe from that of the other instruments playing at the same time in an ensemble, and then track it in time as the oboe pours out its musical contribution along with the others.
There are many reasons for the modern STEM-oriented biology student to focus on cells and molecules. It is there, presumably, that we will be able to draw strong inferences about the Big Bang of life on earth, and it is there that we find so many promising avenues for medical research (as those two new NAM members, Desai and Discher did). Even systems that are quintessentially integrated may be susceptible to failure owing to a flaw at the cellular or molecular level. But, eventually, in order to understand and engineer those systems when they are healthy, the biological scientist must turn from the level of the molecule and cell and confront the complexity of life as it really appears. When he or she does that, physiology and morphology will be the unavoidable paths to biology's integrative side. And much of what fascinates most of us about the nervous system lies there-- on biology's integrative side (reductionism on biology's integrative side).
Then there is ecology, the biology of our environment
As vitamins, vaccinations, anesthesia, antiseptics and sterile surgery were being introduced in the 19th century, civil engineers were busy designing systems for safe sewage disposal and treatment and secure isolation of waste water from the public and, especially, from drinking water supplies. They also were busy draining swamps and introducing Mosquito Fish. It is said that, at that time, the environmental infrastructure changes wrought by civil engineering saved vastly more lives than did the many advances in medical science.
In large part because of these advances, medical and environmental, between 1900 and 2015 the human population of the world increased approximately 4.5 fold; since 1980, it has added a billion approximately every twelve years and it currently continues at that pace. This growth, itself, has produced new and difficult environmental challenges, some of which are quintessentially biological (i.e., ecological). Ecology definitely lies on biology's integrative side. As anyone who has studied ecology well knows, it also brings a broad world of STEM into biology: mathematics (e.g., population biology and epidemiology), chemistry (e.g., global geochemistry of carbon and other atomic species), physics (e.g., the physical properties of water, and the impacts of coriolis and fluid-density gradients, as well as those of polyatomic atmospheric gases, on climate and biome distributions), geophysics (e.g., the need for pioneer organisms and biological succession). And engineering comes into ecology at virtually every interface between human and the rest of the natural world. As teams are formed to design and build new structures and devices for those interfaces, or to establish new policies for them, inclusion of traditional engineers trained in ecology and/or ecologists trained in traditional engineering would be most prudent. Such professionals surely define an important flavor of engineering in biology.
And neuroethology, the neurobiology of natural behavior-- nature's animation
Here is a discipline that has attracted engineers to biology for decades. It was the discipline of Ted Bullock, a 1936 Cal zoology grad and one of my two postdoc mentors. It was the discipline of Bob Capranica, long-time Cornell professor of neurobiology and behavior, a 1958 Cal EE grad (and MIT PhD in EE), my friend and colleague. It is the discipline of his former doctoral student, Peter Narins, UCLA professor of physiological science, a 1965 Cornell EE grad (MS EE, 1966), my frequent collaborator in bioacoustic field studies-- carried out largely at night in hyper-arid deserts and tropical rain forests. It was the principal subject of the 1959 classic Sensory Communication, edited by Walter Rosenblith (MIT Professor of EE, Provost). It was the discipline of Hansjochem Autrum, physicist turned neurobiologist and founding editor of the Journal of Comparative Physiology A. It is the field that has brought communication theory, signal processing theory, systems theory, network theory, and cutting-edge instrumentation into sensory neurophysiology and studies of behavior related to the senses. It was the subject of one of the two neurobiological EE courses I taught when I first arrived at Cal. And it most definitely is on biology's integrative side.
During the year I came to Cal, 1967, Kenneth D. Roeder published a neuroethology classic-- Nerve Cells and Insect Behavior. It tells of bats using sonar to catch moths in the dark, and of moths evolving neurosensory counter-measures. As are so many tales from the field of neuroethology, it is enthralling-- the stuff of a PBS special, such as one might find on the program Nature. It is my impression that hiring in Berkeley's Bioengineering Department has largely emphasized the cutting edges of molecular and cell biology and microelectronics-- as the Whitaker Foundation mandated. But, among their most recent faculty hires is a young, 21st-century neuroethologist, studying the acoustic and visual senses of bats in action. A neuroethologist in Berkeley's College of Engineering-- all's well with the world.