Figure 32 Bert Sakmann at a much older age but much as he appeared in the early 1970s Courtesy of Bert Sakmann photo by Sven Erik Dahl

The issue that captured our attention, and that of many other neuroscientists at the time, was how the neural activity generated by everyday experience affects synaptic interactions and neuronal connectivity. It had long been apparent that understanding how the effects of experience are encoded in the nervous system presented another major challenge in neuroscience. Successfully addressing this issue would explain the way we and other animals learn. And unlike the mechanisms of neural signaling, this problem was far from being solved. Humans and many other species are obviously changed by what happens to us in life, and the lessons learned are an important determinant of evolutionary success. For all but the simplest organisms, modification of the nervous system through learning contributes importantly to the efficacy of behavior, and ultimately to the likelihood of reproducing.

Because the currency of experience in neural terms is the action potentials that stimuli generate through the agency of sense organs, it had been widely assumed for decades that learning involves activity-dependent changes at synapses. These changes—referred to more generally as synaptic plasticity—would therefore encode new information by actually or effectively altering neural connectivity. Transient changes that reflected alterations in the efficacy of synaptic transmission would presumably explain short-term learning and memory (such as a telephone number read from the phonebook and remembered for only as long as needed to dial it). More permanent anatomical changes in synaptic connectivity seemed likely to be the basis for long-term memories that could last for years (such as remembering your telephone number in childhood).

While I was still at Harvard, a handful of us had organized a series of evening meetings to identify pathways in neuroscience that seemed promising avenues we might follow when eventually we had independent labs. A recurrent theme in these inconclusive conversations was figuring out how activity changed neural circuits. Pursuing some aspect of this challenge seemed a worthy goal to both Bert and me. This had been one of the major purposes in much of the work on the leech and other simple central nervous systems (relating changes in the connectivity of single identified neurons to changes in observed behavior), and thinking along these lines was not much of a stretch. Katz had pioneered studies of this general problem in the 1950s by asking how the release of neurotransmitter at the neuromuscular junction, the model synapse illustrated in Figure 1.4A, was affected by prior activity. The result showed that the efficacy of neurotransmission could be either increased or decreased, depending on the nature of the preceding activity: Small amounts of activity facilitated transmitter release, and large amounts depressed it, a phenomenon that was followed by a later rebound increase that lasted minutes or longer.

This general perspective had already motivated a lot of related work in other labs in the 1960s, and one of these was the lab of Per Andersen in Oslo. Andersen, along with Kuffler and Katz, was a trainee of John Eccles, albeit many years later (the roster of important people Eccles trained during several decades is a remarkable testament to his impact on the field, whatever one might think of his sometimes odd and stridently expressed views). A student of

Andersen's named Terje Lomo had discovered a particularly long-lasting form of potentiation in the brains of rabbits in 1966, a topic he pursued with Tim Bliss, another fellow who had arrived in Andersen's lab in Oslo in 1968. The phenomenon that Lomo and Bliss described was in the hippocampus, a region of the brain known to be involved in a particular form of human memory, and was rightly taken to be especially important. Long-term potentiation in the hippocampus eventually spawned hundreds of research papers, and its role in memory continues to be a topic of intense interest (and ongoing controversy) today.

Lomo was a fellow in Katz's department at the time Sakmann and I were considering what we might do, but he was about to leave to work further with Bliss at Mill Hill in north London, where they pursued hippocampal potentiation and firmly established the importance of this line of investigation. However, Lomo had worked on a different project at University College with Jean Rosenthal, another fellow who had just left. Together they had shown that prolonged stimulation of a muscle changed the sensitivity of the muscle cells to the neurotransmitter (acetylcholine) that normally activates these cells by release at the neuromuscular junction. This effect was also of obvious interest because it suggested another way of exploring how activity could change the behavior of excitable cells, and therefore how the nervous system might encode information derived from experience. After some further discussion, Bert and I decided that following up on what Lomo and Rosenthal had done would be a fine way to better understand how activity could alter the properties of nerve and muscle cells and, in principle, store information.

After hashing it over, we discussed this general goal with Miledi ( Figure 3.3). Miledi's work in the 1960s done independently of Katz had shown that when the innervation of a muscle was removed by cutting the nerve to it, the sensitivity of the muscle fibers to neurotransmitter acetylcholine, which was normally limited to the immediate region of the synapse, spread over the whole muscle fiber surface. This observation had set the stage for Lomo and Rosenthal's demonstration that directly stimulating the muscle could reverse this "supersensitivity." Not surprisingly, Miledi was keen when Sakmann and I indicated that further work along these lines piqued our interest.

Figure 3.3 Ricardo Miledi with Katz by their experimental rig in the early 1960s. Although room temperatures in the department were always on the cool side, they had purposely turned off the heat for the experiments they were doing that day. (Courtesy of Ricardo Miledi)

Figure 3.3 Ricardo Miledi with Katz by their experimental rig in the early 1960s. Although room temperatures in the department were always on the cool side, they had purposely turned off the heat for the experiments they were doing that day. (Courtesy of Ricardo Miledi)

Our idea for a way to attack this issue was based on an odd fact that neurologists had known and used as a diagnostic tool for decades. When muscle fibers are no longer innervated (a common enough occurrence in human injuries or diseases such as polio or amyotrophic lateral sclerosis), the muscle fibers begin generating action potentials on their own, a phenomenon called "fibrillation." The origin and consequences of this spontaneous activity raised some further ways to explore the control of nerve and muscle cell membrane properties by activity (the occurrence of action potentials), and these could be studied in muscles taken out of an animal such as a rat and kept alive for a week or more in a Petri dish. In these circumstances, we could directly monitor the levels of spontaneous activity in individual muscle cells with a recording electrode, experimentally alter the levels of activity by electrical stimulation or blockage with a drug, and test the sensitivity of the fibers to neurotransmitter. Katz and Miledi agreed that this would be a sensible project. So for the next two years, we happily set about exploring these issues.

Because neither we nor anyone else in the department knew exactly how to go about this, we fiddled with various chambers, muscles, methods of stimulation, recording electrodes, and culture conditions until we got things to work. Eventually, we could record for several days from single muscle fibers and watch their activity wax and wane as spontaneous action potentials or their absence in the fibrillating fibers altered their membrane properties and, consequently, their sensitivity to neurotransmitter that we applied. We could also stimulate the muscle artificially, showing that denervated fibers that were kept active never started to fibrillate, and could be made to stop fibrillating if this spontaneous activity had been allowed to start. Although the results were another modest contribution (a couple of good but rarely cited papers in the Journal of Physiology), we had a fine time being on our own and doing what we thought was interesting.

Katz would drop by every few days to see what was going on, but rarely offered detailed advice and was usually more interested in chatting about neuroscience generally, politics, and the grand scheme of things (which we were delighted to do). He was then Secretary of the Royal Society and thoughtfully arranged for us to go to occasional evening meetings, pompous but interesting events where members presented demonstrations in much the same way they had since the Society's founding in the seventeenth century. (We had to wear rented tuxedos to attend.) Miledi also dropped by our lab from time to time, usually offering specific advice about complicated variants of our experiments that he thought we should try, advice that I don't think we ever followed. His oversight was well meaning, but I soon came to appreciate the extraordinary clarity and focus of Katz's thinking compared to the rest of his faculty. The most important feedback we got was from the other fellows who were there at the time, particularly Nick Spitzer, Mike Dennis, and John Heuser, who were all transplants from Harvard like me.

Katz gave the two manuscripts that Bert and I wrote up at the end of our time in the department his seal of approval and suggested we show the papers to Huxley, who was in the Department of Physiology a few floors and corridors away in the warren of University College buildings. He had been studying muscle contraction since the 1950s in work that was as impressive in its own way as what he had done with Alan Hodgkin on the action potential in the 1940s. Because our work concerned muscle cells, Katz thought Huxley would be interested, and that he might have some useful criticisms. Huxley deemed the papers more or less fine, but he chastised us for having blacked out some fuzziness around the oscilloscope traces with a marker—an innocuous bit of pre-Photoshop-era improvement to our figures that, for a purist like Huxley, was a cardinal sin.

In the end, our apprenticeships in neuroscience at University College had not provided either Bert or me with a compelling problem to follow up. Although the work we had done was perfectly good, it didn't present us a clear path to the future, in the way Lomo's work on potentiation in the hippocampus led him toward a specific goal and early notoriety. There is no formula for figuring out what to do in science, and everyone who eventually finds a good problem does so in a different way, if they find one at all. Bert left London in 1973 a month or so before I did to begin an assistant professorship at the University of Goettingen. Within the year, he began a collaboration with Erwin Neher that eventually led to the Nobel Prize for Physiology or Medicine in 1991. The award was for making possible a further key step in understanding the basis of neural signaling that Hodkgin, Huxley, Katz, and Kuffler had done so much to advance in the preceding 30 years. Although it had long been clear that both the action potential and synaptic transmission depended on changes in the movement of specific ions though channels in nerve cell and muscle fiber membranes, the details of how this actually occurred had remained uncertain. The major obstacle was the absence of a way to measure what everyone presumed was the opening and closing of these channels in the cell membranes of nerve and muscle cells, activated by either voltage changes across the membrane in the case of the action potential, or by the action of a neurotransmitter at synapses. This was the issue that Katz and Miledi had been working on in their studies of synaptic noise that were going on down the hall. Sakmann and Neher solved this problem by the ingenious trick of pulling a small patch of membrane into the mouth of a highly polished electrode, which enabled them to record the tiny electrical events associated with the opening and closing of single ion channels ( Figure 3.4). The information obtained in this way confirmed that ion channels were quite real and showed how they operated in response to the voltage changes that triggered action potentials and to the binding of neurotransmitters at synapses. When coupled with molecular genetic techniques, the method provided a way of eventually understanding the structure and function of many of these channels in a variety of cells. Within a few years, dozens of labs around the world were using this approach, and ion channels are now routinely studied in this way. When I was moderator of a retrospective many years later, I asked Katz whether there was anything in his remarkably long string of accomplishments that he wished he done differently. He answered without hesitation that he very much regretted not having invented the patch-clamp electrode, a relatively simple method that he could have pioneered, had he thought of it. But he was obviously pleased that one of his protégés had been partly responsible for the discovery, and he and Sakmann remained close until Katz's death in 2003.

Figure 3.4 The technique developed by Bert Sakmann and Erwin Neher for measuring the activity of single ion channels (shown in pink). The method entails the application of a polished electrode to the membrane surface of nerve or muscle cells, enabling the recording of electrical events orders of magnitude smaller than the potential changes recorded with conventional electrodes that penetrate the cell membrane (see Figure 1.3). The diagram shows a small patch of membrane being gently sucked against the pipette, which is why the method is called "patch clamping." (After Purves, Augustine, et al., 2008)

After Katz stepped down as department head in 1978, Miledi was appointed his successor. Political problems apparently ensued both within the department and with the administration, and Miledi moved to the University of California at Irvine in 1985. The Department of Biophysics was disbanded a few years later, after 70 years of extraordinary productivity and a major impact on neuroscience.

During my last year at University College I also had to worry about getting a permanent job and what research I would pursue when I did. While still at Harvard in 1970, I had met Carlton Hunt when, as with Katz, he had come by to visit Kuffler (Hunt's middle name was Cuyler, and everyone called him Cuy) (Figure 3.5). Hunt, who was then in his early fifties, had been Kuffler's first fellow after Kuffler had arrived at Johns Hopkins from Australia in 1947, and had spent four years collaborating with him. Together they had worked on stretch receptors in muscle fibers (see Figure 3.5), a seemingly odd project but typical of Kuffler's nose for important and solvable problems in neuroscience. Most people take it for granted that the five senses (vision, audition, touch/pressure/pain, taste, and smell) provide all the fundamental information about the human environment that we need to survive in the world. In reality we possess dozens of other types of sensors that monitor what is happening within and around us, some of them more critical than the five obvious ones. One of the most important of these is the stretch sensors in muscles that continually inform the central nervous system about the position and status of the body's muscles, providing the feedback needed to maintain appropriate postures and perform successful motor acts. People who are blind or deaf get along reasonably well, but the absence of information arising from sensory receptors in muscles would be incompatible with life. No one had done much to explore how this sensory system works (the visual and auditory systems were, of course, the usual targets of such studies), and Hunt and Kuffler's collaboration had done a lot to put this key aspect of sensation in the prominent place it deserved.

When I first ran into Hunt at Harvard, he had recently moved to Washington University from Yale. Although I was already planning to go to England for another two years of training, I took note that Hunt was then in the process of building a Department of Physiology and Biophysics in St. Louis, having already put together excellent departments at the University of Utah and then at Yale (where, as Chairman of Physiology, he had hired John Nicholls). Hunt was then—as always—a distinguished figure, and it was obvious that Kuffler and the rest of the faculty at Harvard liked him and admired the two departments that he had already created. Whatever conversation we had then about future plans must have been quite tentative. However, I took note that if history and first impressions were any guide, Hunt would be an excellent person to work for.

Figure 3.5 Cuy Hunt circa 1980, studying the properties of stretch receptors in muscles in his lab at Washington University. The diagram shows one of these complex sensors found in nearly all human muscles (the "extrafusal" fibers are the regular muscle fibers that generate body movements; the "intrafusal" fibers control the stretch receptor length to ensure ongoing sensitivity when muscles contract). Like Katz, Hunt fell into the bulldog category of scientists, and he continued to work on muscle spindles for more than 50 years. (Courtesy of Marion Hunt; the spindle diagram is from Purves, Augustine, et al., 2008)

Was this article helpful?

0 0

Post a comment