Saturday, October 01, 2005

The Tumultuous Birth of Brain Chemistry

The War of the Soups and the Sparks: The Discovery of Neurotransmitters and the Dispute Over How Nerves Communicate

By: Jerome Kagan Ph.D.,

How actions and mental states emerge from brain activity is a paramount enigma in contemporary science, but it is important to realize that as early as 1900 the nature of brain activity already held a lofty position among the fundamental questions puzzling scientists. The running scientific battle in the intervening decades, which steadily increased—and radically altered—our understanding of the nature of brain activity, is the theme of Elliot Valenstein’s masterly account of “the war of the soups and the sparks.” 

Valenstein, now retired from the department of psychology at the University of Michigan, earlier in his career earned respect for his experiments that probed the brain bases for psychological states. Emeritus status permitted him a second distinguished career as a historian, writing critical analyses of psychiatry’s fatal attraction to frontal lobotomy as a cure for schizophrenia and of today’s almost unbounded enthusiasm for drugs as the best way to treat mental illnesses. The incisive, unsentimental style of those earlier books continues in his new book, The War of the Soups and the Sparks. Valenstein’s narrative features heroes, rivalries, snobbery, territoriality, old-boy networks, dreams, predawn experiments, luck, and the careers of scientists whose loyalty to the imperative that theory must accommodate reliable new observations allowed them to spend their sunset years in a golden glow—long after admitting their mistaken opposition to the possibility that neural connections relied on chemistry. 

The story is bound together by one penetrating question: What material energy does the brain use to do its work? I use the singular, rather than the plural, form of “energy” because no scientist at the turn of the last century believed that nervous transmission required multiple mechanisms. The drama’s opening act, already colored by controversy, ended with a consensus that the brain was not a continuous net of cells, cheek-to-jowl, but consisted of individuated objects with minute spaces between them. The Spanish scientist Santiago Ramón y Cajal is usually given credit for this hard-won insight, although the Italian anatomist Camillo Golgi enabled the discovery by developing, in 1873, a silver nitrate stain for nerve cells that let Cajal see neuronal fibers more clearly. The men shared a Nobel Prize in 1906. 

Once the doctrine of separate neurons had been accepted, the debate shifted to how neurons communicated. The mechanism that enables one cell to excite another must be distinguished, of course, from the nature of activity within a neuron. Electrical energy within a nerve cell travels down its axon to sites where chemicals wait to be released into the microscopic space, or synapse, that separates one neuron from another. The excitation of the nearby cell by molecules crossing the narrow synaptic cleft initiates a repetition of the sequence: Electrical activity in that nearby neuron is followed by chemical transmission to yet another cell. The electrical phenomena within a neuron, however, are the result of ions entering and leaving through the cell’s membrane. Therefore, the mechanism of brain activity is electrochemical, with chemistry the primary foundation for changes in neuronal communication. 


The politics during the first decade of the 20th century had the neurophysiologists (the “sparks”), who worked with clean, gleaming machines, cast as the Brahmins, whereas the biochemists and pharmacologists (the “soups”), whose hands were dirty with blood and urine, were workers of a lower caste. The sparks’ arguments were persuasive. Brain processes must occur rapidly to account, for example, for the speed with which the pupil of the eye constricts in response to a bright light or the hand is reflexively withdrawn from a hot surface. Electrons travel fast; chemical reactions are often relatively slow. Thus, it seemed obvious that the brain could not rely on chemistry to accomplish its missions. No neuron ran as slowly as the pens of the early instruments used by the soups. The case for electrical transmission was bolstered because the vacuum tubes and oscilloscopes sparks used to make their measurements were electrical. Was it not obvious that the brain, too, was electrical? The sparks found it easy to ignore the new boys on the block, who were suggesting that some nervous transmission might be chemical. 

The politics during the first decade of the 20th century had the neurophysiologists (the “sparks”), who worked with clean, gleaming machines, cast as the Brahmins, whereas the biochemists and pharmacologists (the “soups”), whose hands were dirty with blood and urine, were workers of a lower caste. The sparks’ arguments were persuasive. 

In retrospect, the sparks seem not to have adequately appreciated Niels Bohr’s insight from physics that one cannot separate the truth of an inference from its source of evidence. The meaning and validity of every conclusion are affected by the specific machines and procedures that generate the relevant observations. The microwave radiation detected by radio telescopes from distant galaxies is only a sign of cosmic objects, not a reflection of the features that define them. Modern investigators who use functional magnetic resonance imaging (fMRI) scanners to measure changes in deoxyhemoglobin produced by alterations in blood flow to particular brain sites appreciate that this measure, too, is only an indirect index of neuronal activity. Neurons do not rely on deoxyhemoglobin to communicate with each other. Nonetheless, the sparks of the early 20th century could not imagine any mechanism of nervous activity other than electrical and would have been stunned to read today’s journals and learn that most transmission is chemical. 


Although many scientists pioneered the chemical doctrine, Valenstein selects three to honor: Henry Hallett Dale, Otto Loewi, and Walter Bradford Cannon, all born between 1871 and 1875. Dale, an Englishman, laid the foundation for the notion that nerves secrete “humoral substances,” although this was not his original intention. In need of a steady income in order to marry, Dale took a job with Henry Wellcome’s pharmaceutical firm. Wellcome generously gave Dale time to pursue his own interests, but he did suggest that, if convenient, Dale might probe the mechanisms of action of ergot, a fungus that grows on some grains. 

Dale began by asking how ergotoxine, one of the extracts of ergot, affected a cat’s arterial blood pressure. He was surprised when, after giving some extract from a dried adrenal gland to a cat previously injected with ergotoxine, he did not observe the expected increase in pressure. Dale did not realize that ergotoxine reversed the effects of adrenaline (and resulting sympathetic nerve stimulation) on the cardiovascular system and that he had discovered, inadvertently, the first adrenaline-blocking agent. He learned later, with the help of others, that the ergot extract was acetylcholine, a molecule that produces other parasympathetic effects, including increased salivation and contractions of the gastrointestinal tract and bladder. The most cautious of the three heroes, Dale was unwilling to speculate that this form of chemical transmission might occur in other parts of the body or brain. Lord (Edgar) Adrian, who wrote the influential, The Physical Basis of Perception, said that Dale was more likely “to apply the brake than to be the first in the gold rush, but the gold he has found will keep its value.” 

Otto Loewi, a German Jew born in Frankfurt to a family of successful wine merchants, met Dale in London and later performed a celebrated experiment while professor of pharmacology at the University of Vienna. Loewi awoke about 3 a.m. on a cold February night in 1921 with the image of an experiment that might prove chemical transmission. He rushed to his laboratory and before dawn had demonstrated chemical transmission of a nervous impulse. 

In his experiment, Loewi surgically isolated the hearts of two frogs, the first heart with its nerves intact and the second with all its nerves severed. He stimulated the vagus nerve of the first heart, then transferred the solution in which it was bathed to the heart with all nerves severed. He observed a slowing of the heart rate, as though the second heart’s vagus nerve also had been stimulated. When Loewi later stimulated the accelerator nerve of the first heart and transferred the solution to the second heart, the rate of the latter again increased. Loewi gave the name “Vagusstoff” to the unknown substance that slowed the heart and “Acceleransstoff” to the substance responsible for the increase in heart rate. 

Almost a decade passed before scientists accepted Loewi’s conclusions, because many could not replicate his observations. Why? First, the Vagusstoff, which was in fact the neurotransmitter acetylcholine, is rapidly degraded by the enzyme cholinesterase. Loewi was lucky that he had performed the experiment in February, because acetylcholine is more stable during the colder months. Had Loewi performed the experiments in July, he might have found nothing. Second, frog species differ in amount of cholinesterase, and Loewi happened to choose the right species. 

Loewi, less cautious than Dale, ignored alternative interpretations when he submitted his paper for publication. But Loewi did continue to refine his experiment, and, after demonstrating the effect 18 times during a Stockholm meeting five years later, he converted many of the skeptics, who acknowledged that chemical transmission might regulate heart rate but remained convinced that it did not take place in the brain. 

Loewi and Dale shared a Nobel Prize in 1936, gaining the award largely thanks to a persuasive nomination by Dale. 


The last of Valenstein’s trio, Walter Bradford Cannon, was the foremost American physiologist of his era. Expansive, exuberant, generous with students, and possessed of a strict social conscience, Cannon’s initial discoveries were, like Dale’s, accidental. His exceptional intellect earned him admission to Harvard College in 1892, from which he graduated with top honors. He then enrolled in Harvard Medical School and had the good luck of being assigned to the laboratory of Henry Bowditch. 

Bowditch suggested that Cannon use a recent discovery—X-rays—to study how food moved through the digestive tract. Cannon made a critical observation: The stomach movements of a cat ceased when the animal began to struggle. Recognizing the implications of this accidental observation, Cannon exposed cats to various forms of stress, such as barking dogs, repeatedly observing that stomach motility ceased. It was clear that supposedly evanescent, nonmaterial emotions affected the material body. 

Cannon and a fellow researcher found that adrenaline was released into the blood when a cat was stressed. They were puzzled, though, by the observation that a stressed cat with no adrenal glands nevertheless produced signs of an adrenaline-like substance. Cannon did not consider the possibility that the sympathetic nerves were responsible, but eventually a young physician from Belgium who was working with Cannon discovered that an animal’s heart rate increased when its sympathetic nerves were stimulated, even though the adrenal glands had been isolated from connecting nerves and the vascular system. 

Cannon coined the term “sympathin” for the unknown chemical substance that was secreted when sympathetic nerves were stimulated. At first, he suspected that adrenaline and sympathin were the same substance. Then, the multitalented, immodest Arturo Rosenblueth joined his laboratory, and together they discovered that sympathin and adrenaline were not identical. They had no choice but to posit two types of sympathin, one excitatory and the other inhibitory. Unfortunately, the suggestion that there were two types of sympathin was not well received. Henry Dale regarded the hypothesis as overly complicated and probably wrong, and he was correct. The excitatory form of sympathin turned out to be noradrenaline, whereas the inhibitory form is adrenaline. 

Most natural scientists do not forgive overly speculative colleagues whose ideas fly too high above the evidence; they reserve their admiring glances for the cautious Henry Dales of this world. Unsubstatiated ideas that turn out to be wrong— a common occurrence in science— are dangerous because they tempt unsophisticated investigators to waste time and money on fruitless quests. 

Valenstein suggests that the prematurity of the two-sympathin theory was the primary reason Cannon did not share the Nobel Prize with Loewi and Dale. Most natural scientists do not forgive overly speculative colleagues whose ideas fly too high above the evidence; they reserve their admiring glances for the cautious Henry Dales of this world. Unsubstatiated ideas that turn out to be wrong—a common occurrence in science—are dangerous because they tempt unsophisticated investigators to waste time and money on fruitless quests. The two-sympathin theory received one fatal blow at the end of the 1940s, when Raymond Ahlquist, a pharmacologist at the Georgia School of Medicine, demonstrated the existence of two different receptors for adrenaline—alpha and beta—rather than two forms of sympathin. A second disconfirmation, from the Stockholm laboratory of Ulf von Euler, who had also worked with Dale, showed that the sympathetic nerves secrete noradrenaline. It is ironic, Valenstein notes, that Cannon’s positing of two chemical substances was correct with respect to the adrenal medulla, which produces both noradrenaline and adrenaline. But Cannon and Rosenblueth had assumed that both molecules were secreted by the sympathetic nerves. 


Despite these seminal advances, as late as 1939 some prominent sparks would not concede a role for chemical transmission in the brain. John Eccles, who shared a Nobel Prize in 1963 with Alan Hodgkin and Andrew Huxley, was particularly adamant in his opposition. Cannon addressed Eccles’s objections by noting that the sparks and soups agreed that minute quantities of acetylcholine acted at parasympathetic synapses to provoke the same reactions that characterized neural impulses. Canon did acknowledge that acetylcholine seemed unable to account for the fast reactions characteristic of skeletal muscles, but he argued that scientists still did not know enough about the velocity of chemical processes at a synapse to eliminate that possibility. In turn, Cannon asked how opponents of chemical transmission could explain the delay, about 5 to 11 times longer than the delay for electrical transmission, that occurs at a synapse. Didn’t this delay support the case for chemical transmission? 

Further, the sparks could not explain why drugs that can block or enhance a reaction to nervous stimulation could do so without altering the nerve impulse but do modify the effectiveness of acetylcholine. Curare, for example, which blocks activity in skeletal muscles, does not interfere with the nerve impulse, but it raises the threshold of response to acetylcholine. This observation would make sense if the nerve impulse was only effective when acetylcholine—that is, a chemical neurotransmitter—was present. 

The most serious problem for the sparks was the observation that a neural impulse was sometimes followed by excitation, sometimes by inhibition. A theory of electrical transmission had difficulty explaining inhibition. Eccles, eager to answer Cannon’s critique, proposed that a newly discovered type of neuron exerted an inhibitory influence on spinal motor neurons. Eccles, like Loewi and Cannon, claimed that a dream was the source of his idea. But, in the end, Eccles proved his own conjecture wrong. Ralph Waldo Emerson, who held that dogmatic consistency is a characteristic of small minds, should be smiling. 

Sparks evaluate inhibitory processes by ascertaining the voltage difference between the inner core of a neuron and its outer membrane. When the voltage difference is decreased, a neuron’s excitability increases. When the voltage difference is increased, the neuron’s excitability decreases and this phenomenon defines an inhibitory state. Before 1951, however, the evidence for inhibition in the spinal cord had been collected with recording electrodes placed outside the neuron. These electrodes were insensitive to tiny voltage changes that must be measured inside the cell. As is often the case, talented engineers came to the rescue by developing glass microelectrodes that could penetrate single neurons and permit precise measurements of the voltage changes that characterize inhibitory and excitatory states. Eccles, with two colleagues, performed the experiment that would prove to be critical on a cat in the summer of 1951. After a long day of recording voltage changes in spinal motor nerves in response to stimulation they conceded that their data could only be explained by some form of chemical transmission. The voltages recorded were inconsistent with the hypothesis of an inhibitory neuron, but consistent with the notion that a substance liberated by the synaptic knobs caused an increase in polarization of the membrane around the motor neuron and, as a result, inhibition at the synapse. The neurophysiologists conceded that acetylcholine, liberated from the nerve terminals, triggered the sodium mechanism responsible for the inhibition. 

Although Eccles had yielded in accepting chemical transmission for spinal motor neurons, he continued to insist that the same mechanism might not apply to the brain. As Dale put it, chemical transmission was treated “like a lady with whom the neurophysiologist was willing to live and consort in private, but with whom he was reluctant to be seen in public.” 

When Eccles sent Dale an advance copy of the paper, Dale replied that Eccles’s change of mind reminded him of Saul’s conversion on the road to Damascus, when “the sudden light shone and the scales fell from his eyes.” Although Eccles had yielded in accepting chemical transmission for spinal motor neurons, he continued to insist that the same mechanism might not apply to the brain. As Dale put it, chemical transmission was treated “like a lady with whom the neurophysiologist was willing to live and consort in private, but with whom he was reluctant to be seen in public.” As late as 1950, Clifford Morgan and Eliot Stellar, who had written a best-selling textbook on physiological psychology, described synaptic transmission as electrical and failed to mention chemical transmission as a possibility. 


History often has affirmed the theme in Bob Dylan’s song of the late 1960s, “Those who are first will soon be last, for the times they are a’changin.” We now know that more than 99 percent of transmissions between neurons are mediated by chemicals. But, nature abhors monopoly: About one percent of synapses, especially in the retina and the brain’s locus ceruleus and corpus striatum, rely on electrical transmission. Neither Dale, Loewi, nor Cannon would have imagined that contemporary scientists would posit almost 100 different molecules with distinct chemical structures that affect brain function. Amino acids, peptides, catecholamines, and gases such as nitric oxide all influence neural activity. 

What is more, the soups failed to consider the possibility that chemicals might require help to accomplish their assignments. For the chemicals to be effective, the neurons they excite or inhibit must have receptors on their surfaces, and many transmitters have more than one receptor. An army of investigators during the past half century have verified Dale’s suggestion, in his 1936 Nobel lecture, that there might be chemical transmission in the brain. 

The sparks and soups made peace when the electron microscope enabled scientists to see the small vesicles in neurons that secrete the chemical substances. Only then did Eccles finally concede that there were chemical synapses in the brain. By the late 1950s, most scientists were prepared to believe that some mental illnesses were caused by abnormalities in brain chemistry and ready to assent to Ralph Gerard’s startling principle: “No twisted thought without a twisted molecule.” Neuroscientists had embraced the speculations of the first soups, Hippocrates and Galen, who supposed that melancholia (depression) was produced by an excess of black bile.


One implication of Valenstein’s masterly account is that biological systems are extremely specific. Almost every time a scientist suggests some relation between two aspects of brain function, or between a brain profile and a psychological outcome, other investigators soon discover that the claim was too general. I remember reading a paper on owls that illustrates the specificity in biological systems. The average diameter of the black spots on the underside feathers of the female owl is related to the competence of the immune system of her offspring. But, surprisingly, the number of black spots is not so related. Many similar discoveries have made most biologists “splitters” who attend to the details and remain ready to replace one concept with two, when the evidence demands such a conclusion. 

But “lumpers” are more prevalent among social scientists. A good many contemporary psychologists, anthropologists, sociologists, and psychiatrists love broad concepts that ignore the details. Compare the specificity of the consequences of the diameter of the spots on the female owl with permissive psychological concepts such as arousal, reward, intelligence, self-esteem, attachment, or anxiety. Nature loves the splitters, not the lumpers, for she is compulsive in her attention to detail, even as she is exuberant in her variety. Further, the heroes in the life sciences, as in physics, have been the engineers and technologists who designed more powerful amplifiers, tinier glass electrodes, electron microscopes, and fMRI scanners that enable measurements of phenomena at a level of detail that Dale, Loewi, and Cannon could not have dreamed. 

One lesson affirmed in The War of the Soups and the Sparks is the need to appreciate the effect of a particular experimental condition on the observations scientists make and later use to support a particular idea. Remember Loewi’s good fortune in working with the appropriate species of frog, and with frogs rather than rats, as well as performing his experiment in February rather than July. A more accurate understanding of the relation between brain and mind will require accepting the principle that the validity of every relation depends on the context in which the agent—animal or human—is acting. 

Most current concepts in psychology and psychiatry, too pretty to be true, remain appealing because, like the loyalty to the theory of electrical transmission by the sparks, they satisfy the desire for a simple, parsimonious explanation.

It is puzzling why lumpers are so enchanted with the theoretical usefulness of broad psychological states, such as fear, stress, or consciousness. I suspect that the appeal is aesthetic: Lumping makes experimentation easier and permits scientists to be indifferent to the species and the specific methods in their experiments. It is not a coincidence that beauty and convenience are the primary criteria mathematicians use to decide which equations deserve celebration. The mathematical physicist Paul Dirac believed that the beauty of a mathematical solution should always take precedence over correspondence with evidence when the two were inconsistent. But, the aspects of nature scientists are trying to understand existed long before our species evolved with an attraction to beauty and a wish for convenience. Most current concepts in psychology and psychiatry, too pretty to be true, remain appealing because, like the loyalty to the theory of electrical transmission by the sparks, they satisfy the desire for a simple, parsimonious explanation. 


Valenstein’s story has a second lesson for social scientists and psychiatrists. Biologists take for granted that investigators must begin their inquiry with a specific phenomenon they have imagined or observed and with a method that might isolate the causes of that phenomenon. Once the appropriate method has been developed, confirmation of a bold hypothesis becomes possible, and the long-suffering investigator can experience the profound pleasure of an understanding that was anticipated years earlier. 

Unfortunately, many social scientists and psychiatrists are impatient and reluctant to develop new methods. They want to study big ideas, such as anxiety or positive emotions, and are so eager to get on with the work they often select the method readily available. For example, some scientists who want to study the effect of a drug on human anxiety place a mouse in the popular elevated maze and conclude that a mouse that does not enter a brightly lit area of the maze is anxious. Developmental psychologists eager to prove that infants younger than one year can “add” numbers measure how long the infants look at dolls on a stage. If these investigators had first studied all the conditions that influenced a mouse’s entrance into a lit portion of a maze, or an infant’s staring at objects, they would have appreciated that they were not ready to use these methods to evaluate anxiety or an infant’s ability to add. 

A less obvious thread running through Valenstein’s story is the profound change in scientific practice during the past century. Dale, Loewi, and Cannon, like Galileo and Darwin, were both the sources of particular hypotheses and the minds that observed the results of the experiments designed to test them. Today, these two functions—originator and interpreter—are often separate in the laboratories of research universities, where postdoctoral fellows and graduate students perform the experiments and bring the processed evidence to the laboratory director, who may have formulated, or at least approved, the hypothesis. My intuition tells me that this practice may be retarding the pace of original discovery. 

Respect for evidence and openness to controversial ideas, which characterized the actors in The War of the Soups and the Sparks, are also salient traits in Valenstein’s makeup. Elliot was my colleague in the early 1960s at the Fels Research Institute, and I suspect that he identified with the heroes of this story when he conducted his elegant experiments that challenged popular hypotheses. Valenstein and his colleagues demonstrated that the behavior of a rat following stimulation of a particular site in the hypothalamus, then believed to be the locus of hunger, depended on what objects were in the cage—food, water, or wood. Therefore, this location in the hypothalamus is not the place where hunger resides. Instead, the stimulation produced a relatively open state of arousal that could be followed by different behaviors, depending on the objects in the rat’s setting. A satisfying understanding of the relation between brain activity and psychological phenomena, which lies in the future, will require specification of the species and experiential history of the agent as well as the contexts in which the brain and behavioral information are quantified. The deep message in Valenstein’s beautifully crafted monograph is that in order to illuminate the relation between brain and mind scholars must remain open to new ideas. We have far to go before we sleep.   


From The War of the Soups and the Sparks: The Discovery of Neurotransmitters and the Dispute Over How Nerves Communicate by Elliot S. Valenstein. © 2005 by Elliot S. Valenstein. Reprinted with permission of Columbia University Press. 

Loewi’s version of these events seems to be somewhat more dramatic than what actually occurred. Dale, who remarked later that he was one of the first to hear Loewi’s account of “the remarkable story of the dream,” recalled that Loewi had originally told him that when he awoke on the second night at 3:00 A.M. he made careful notes so that he would have no trouble deciphering his thoughts the next morning. Loewi’s more dramatic later version is that he went directly to the laboratory at 3:00 o’clock in the morning and that by “five o’clock the chemical transmission of the nervous impulse was conclusively proved.”

Moreover, Loewi’s recollection that the dream occurred “on the night before Easter Sunday” does not seem to be accurate. He performed the experiment in late February, perhaps extending into early March, and he sent the manuscript describing the experiment to the journal later in March. The article is marked as having been received in the editorial office of Pflügers Archiv on March 20, 1921. However, in 1921 Easter Sunday was on March 27, a week after the manuscript was received by the editor. 

While these discrepancies do not subtract from the importance of the experiment, they would seem to reflect the artistic bent of Loewi’s temperament, which might have made it difficult for him to resist making the description of what transpired even more dramatic than it actually was. Dale, in contrast, would probably have never strayed from the facts, no matter how inconsequential. It may also be true that Loewi’s different temperament played a role in allowing him to risk drawing a conclusion from quite preliminary and inconclusive experimental results. 

The experiment inspired by the dream was also described by Loewi on many occasions: 

The hearts of two frogs were isolated, the first with its nerves, the second without. Both hearts were attached to Straub cannula with a little Ringer solution. The vagus nerve of the first heart was stimulated for a few minutes. Then the Ringer solution that has been in the first heart during the stimulation was transferred to the second heart. It slowed and its beats diminished just as if its vagus nerve had been stimulated. Similarly, when the accelerator nerve was stimulated and the Ringer from this period transferred, the second heart speeded up and its beats increased.

About Cerebrum

Bill Glovin, editor
Carolyn Asbury, Ph.D., consultant

Scientific Advisory Board
Joseph T. Coyle, M.D., Harvard Medical School
Kay Redfield Jamison, Ph.D., The Johns Hopkins University School of Medicine
Pierre J. Magistretti, M.D., Ph.D., University of Lausanne Medical School and Hospital
Robert Malenka, M.D., Ph.D., Stanford University School of Medicine
Bruce S. McEwen, Ph.D., The Rockefeller University
Donald Price, M.D., The Johns Hopkins University School of Medicine

Do you have a comment or question about something you've read in CerebrumContact Cerebrum Now.