up SearchFeedback[help] CPMCnet


“The Second Brain”

An excerpt from the widely acclaimed 1998 book, “The Second Brain. The Scientific Basis of Gut Instinct and a Groundbreaking New Understanding of Nervous Disorders of the Stomach and Intestine.”

By Michael D. Gershon, M.D.
Professor and Chairman, Department of
Anatomy and Cell Biology

The Turning Point
An excerpt from the widely acclaimed 1998 book, “The Second Brain. The Scientific Basis of Gut Instinct and a Groundbreaking New Understanding of Nervous Disorders of the Stomach and Intestine.”My own introduction to the second brain came slowly and indirectly. It began with a love affair, not with the bowel or my wife, Anne, but with serotonin. In 1958, as an undergraduate at Cornell, I learned during a course called “The Physiology of Behavior” that serotonin was likely to be a neurotransmitter and that problems with serotonin might lie at the core of schizophrenia and other mental diseases. I resolved to do some work on serotonin myself as soon as I could.

The opportunity presented itself in medical school, now no longer far above Cayuga’s waters but only a few feet above the East River. In those days, medical students were encouraged to take time off to do research, which is what I did during summers and a wonderfully productive year. When I first began, I never suspected that a neurotransmitter encountered during a course on physiology could lead me to the gut, but that is what happened.

“The Second Brain” copyright 1998 by Michael D. Gershon, M.D. All rights reserved. This excerpt printed with permission of Dr. Gershon and HarperCollins Publishers, New York.

First, Some History
My first experiments were designed to locate the sites in the body where serotonin is made. To do this I injected mice with a radioactively tagged form of serotonin’s immediate chemical precursor. This molecule has an unhappily tongue-twisting chemical name, 5-hydroxytryptophan, but, as is also true of Southerners who dislike their given names, the compound is familiarly known by its initials, 5-HTP. As expected, the mice that I had injected with radioactive 5-HTP rapidly converted the injected molecules into radioactively labeled serotonin that I could easily detect and quantify. Locating the tagged serotonin, however, turned out to be somewhat more difficult, because it did not stick around very long. I thus needed to develop a method to hold the radioactive serotonin in place so that I could find it.

Serotonin is naturally broken down in the body at a relatively fast rate. For serotonin to function, this is all to the good; zip in, zip out, and never accumulate. For me, in my efforts to try to discover, and actually visualize, the sites where the radioactive serotonin was located, the fast natural rate of breakdown was counterproductive. It was also expensive, because no one gives radioactive chemicals away. I needed to preserve the radioactive serotonin intact, so that I could locate the sites where the radioactively tagged 5-HTP I had injected was converted to labeled serotonin. I reasoned that this could be done if I also injected the mice with a drug that prevented the breakdown of serotonin.

Fortunately for me, a drug that exerted exactly this action had recently been introduced for the treatment of clinical depression. Serotonin is broken down in the body mainly by an enzyme with the obligatory long name monoamine oxidase. Drugs that inhibit this enzyme, which are classified, naturally enough, as monoamine oxidase inhibitors, tend to preserve serotonin and cause it to accumulate within the cells that make it. The first of the monoamine oxidase inhibitors to be used clinically, a drug called iproniazid, was initially given a trial as an agent for the treatment of tuberculosis. Although iproniazid was not as good against tuberculosis as its relative, isoniazid, the clinical investigators who were studying iproniazid unexpectedly observed that many of the sick and depressed patients who received the drug were no longer depressed, even if iproniazid failed to cure their tuberculosis. On the basis of these observations, iproniazid was dropped as an antitubercular agent but was tested to see if it would work as an antidepressant. Iproniazid passed that test, and although the risk of giving iproniazid (it occasionally, and for no obvious reason, destroyed the livers of unlucky patients) eventually caused it to vanish from the market (to be replaced by the antidepressant monoamine oxidase inhibitors Marplan, Nardil, and Parnate), iproniazid left behind two great legacies. One was the idea that drugs could really be used effectively to combat mental illness, and the other was that serotonin was a substance that played a critical role in the creation of happiness. A theory that the malfunctioning of serotonin as a neurotransmitter in the brain caused depression was launched. This theory is still in vogue.

The ability of the monoamine oxidase inhibitors to alleviate depression was not what interested me in them. What really excited me about the monoamine oxidase inhibitors was that they might make it possible for me to actually carry out the experiments that I wanted to do. The monoamine oxidase inhibitors, I realized, had the ability to prevent the destruction of the radioactive serotonin I had produced, at great expense, in mice. I thus injected iproniazid simultaneously with radioactive 5-HTP. The monoamine oxidase inhibitor performed as advertised, and the radioactively labeled serotonin remained intact long enough for me to look for it.

Michael D. Gershon, M.D.
-By Michael D. Gershon, M.D. Professor and Chairman, Department of Anatomy and Cell Biology
While iproniazid took care of one of the biological obstacles that stood in the way of my experiments, there were others. Radioactivity is easily detectable, but simply finding radioactivity in a sample does not, by itself, provide a clue as to what substances in that sample are radioactive. The analysis of radioactive material is a little like the analysis of the use of a bank’s ATM machine. It is easy to look at the bottom line and determine how much money was withdrawn in a given period of time, but to discover the identity of the people who obtained the money, one would have to ascertain the PIN code of each user and then decode them to produce the names. I had injected radioactive 5-HTP into mice, and I found radioactive serotonin, as I thought I would, but I still needed to know what other radioactive compounds the mice had made and how to distinguish radioactive serotonin from them. Equally necessary, I needed to distinguish radioactive serotonin from any residual radioactive 5-HTP that had not yet been converted to serotonin. Essentially, a molecular PIN code was required. I also needed to find a method to hold the radioactive serotonin in place while I went about looking for it. Inhibition of monoamine oxidase would be helpful in this regard, in that it would prevent the formation of radioactive breakdown products of serotonin. The innovation, however, that made my experiment feasible came unexpectedly and easily from my studies of how to preserve or “fix” radioactive serotonin in place.

I was testing the adequacy of a variety of preservatives or “fixatives” of the kind that are generally used for the microscopic examination of tissues. Formaldehyde seemed almost to work. When I added the formaldehyde, there was a brief moment during which the outflow of radioactive serotonin increased, but then it stopped totally. Except for the initial moment of disaster, this outcome would have been perfect. The formaldehyde must have chemically coupled the radioactive serotonin to protein successfully, because nothing short of incineration could extract it. The initial brief increase in the outflow of radioactive serotonin, however, was a problem that had to be prevented. If radioactive serotonin was allowed to dance in whatever direction the tides of molecular motion took it, there would be no point in finding out where it ultimately came to rest. I wanted to know where serotonin is made during life, not where its artifactual motion happens to stop.

For some time, I was stalled by the problem of the initial fixative-induced loss of radioactive serotonin. The solution turned out to be delightfully simple. The aldehydes that I was using did not properly balance the salts and other molecules in the tissues and were causing cells to swell. When I corrected the salt balance, the outflow of radioactive serotonin ceased as soon as the tissue entered the fixative, and after fixation, no solvent could extract it. As an extra bonus, I also found that none of the other radioactive compounds that were present in the tissues of animals injected with radioactive 5-HTP were similarly fixed. After fixation, serotonin was the only radioactive compound left in the tissue, and I had evidence that the process of fixation did not change serotonin’s location. Formaldehyde had made a molecular PIN code unnecessary. If an ATM has only one user—in this case, serotonin—it does not take a PIN code to figure out its identity.

My next experiments, made possible by iproniazid and aldehyde fixation, were uncomplicated. At various times after the injection of radioactive 5-HTP (serotonin’s precursor), I set out to determine the location of radioactive serotonin in the mice. My goal was not just to find out what organ, or even which layer of an organ, contained the labeled serotonin but to find first the cells and then the parts of the cells (called organelles) that made it.

To locate the “hot” serotonin at this level of resolution, I used a technique known as radioautography. Radioautography is another long word, like so many employed by scientists, but conceptually the technique is simplicity itself, and in contrast to many of the other long words of science, radioautography is a logically constructed name. To locate a radioactive substance by radioautography, radioactive sections of tissue are coated with a photographic emulsion and put away in the dark for several weeks. During this time, the subatomic particles of radioactive decay bombard the emulsion immediately above the tissue sections and a latent image forms. After exposure, the coated sections are developed as if they were film. Silver is precipitated in the region of the latent image, just as if the latent image had been formed by light. In essence, therefore, the radioactive material in the tissue takes its own picture. The picture is thus a “radioactive autograph,” which is shortened to radioautograph, and the method of making radioautographs is therefore called radioautography.

The radioautographic result of my experiment, which riveted my attention on the gut, was that every time I injected radioactive 5-HTP into mice, I found radioactive serotonin in their enteric nervous system. Moreover, and just as important, I did not find radioactive serotonin in any other nerves outside the brain. This demonstrated that nerves in the bowel have an affinity for serotonin that other peripheral nerves do not share.

While my demonstration that enteric nerves have a unique affinity for serotonin did not entitle me to conclude that serotonin is a neurotransmitter in the gut, that seemed to be the simplest explanation. Cognizant of the dictum that “when you hear the sound of hoofbeats, don’t think of zebras,” I conducted one more experiment to see if serotonin would indeed behave like an enteric neurotransmitter. This time, I stimulated reflex activity in the gut to make its nerves work. Working nerves release their neurotransmitter. Sure enough, when the nerves of the gut from the mice that I had injected with radioactive 5-HTP were stimulated, they released radioactive serotonin.

The experiments I had conducted to this point gave me a feeling of confidence that my work could withstand anyone’s scrutiny, which I assumed (foolishly, it turned out) would be both logical and reasonable. I also thought that my data would be considered to be important by other neuroscientists. I wrote up my results in a series of three articles that appeared in Science and the Journal of Physiology. My suggestion that serotonin might be an enteric neurotransmitter was based on the following pieces of information: (i) Serotonin is manufactured and stored in the bowel. (ii) Following its biosynthesis from its immediate precursor, serotonin is preferentially located in enteric nerves. (iii) These nerves release serotonin when they are stimulated. (iv) Others had previously shown that serotonin exerts the same effect on the bowel as does the stimulation of enteric nerves. If serotonin was not a neurotransmitter, therefore, it was certainly giving a pretty good imitation of one.

My Mother Never Told Me It Would Be Like This
Since I had not anticipated that my suggestion that serotonin might be a neurotransmitter in the gut would be viewed by the scientific world as outrageous, I was upset by the reaction I actually encountered. My first impulse was to feel empathy with those of my ancestors who faced the Inquisition. Later, after I became numb and ceased to feel pain, I understood the reaction that I had inadvertently caused. According to the scientific gospel that was prevalent at the time, only two transmitters, acetylcholine and norepinephrine, took care of all of the neurotransmission that went on in the peripheral nervous system. The thought that an additional molecule might be a peripheral neurotransmitter was considered not just wrong but perverse and immoral. Scientists, more than most people, admire order, and the order that had been established in the peripheral nervous system left no room for another neurotransmitter.

Disorder is so widespread in nature that when scientists believe that order has been encountered, they immediately think that some great force has been at work to overcome the sinister effects of chance. All fledgling scientists learn in Physics 101, if they have not been taught it earlier in Introductory Chemistry, that disorder in the universe is always increasing. This ever-escalating disorder is called entropy. To overcome entropy, the Darth Vader of reality, serious work has to be done. The molecules that assemble to form the human body would never do so on their own if they were simply mixed together. Countless thousands of unlikely chemical reactions have to occur in just the right place and at just the right time. Those with a deeply religious inclination contemplate the sheer improbability of these events and turn to God for an explanation. Scientists, however, have surrendered this option, even if they, like me, believe in God.

When we scientists see order, we tend to think that we have found biological reality. Biological processes exert the kind of work and provide the energy necessary to overcome entropy. They impose order on the otherwise reluctant molecules of life, getting them to react with one another to establish the form that we have come to love. For me to upset the order that people thought had been found in the peripheral nervous system was not to be tolerated lightly. My idea that serotonin might be an enteric neurotransmitter was incompatible with the orderly belief that had been held for a long time and thus was much cherished. Back in 1965, the entire peripheral nervous system could easily be described in a simple chart:

  Skeletal Autonomic
  (vontarylu) (involuntary)
    Sympathetic Parasympathetic
Final transmitter: acetylcholine norepinephrine acetylcholine
Targets: skeletal glands, blood vessels, heart and visceral muscles

Note that the autonomic nervous system only had two subdivisions, because the third was in eclipse. The two acceptable autonomic divisions, the sympathetic and the parasympathetic, were believed always to oppose one another, like the free world and the communist empire of recent history. There was a concept of a beautiful duality, two autonomic systems, two neurotransmitters, ever in battle, ever in opposition. My innocent suggestion that serotonin was an autonomic neurotransmitter thus was unsettling. If I was right, the perceived duality was wrong. I was a heretic, and was being treated accordingly.

I decided to be obstinate. I was young in 1965, as well as combative and idealistic. I had confidence that truth, as I saw it, would inevitably win out. Even the most orthodox of my detractors, I believed, could not prevent facts from emerging to demolish the dogma of the fundamentalists. Besides, in those days the National Institutes of Health was tolerant of ideas that opposed received wisdom. Its peer-review panels provided funding even for experiments that were not guaranteed, in advance, to work. Funds were thus available to me to pursue the issue of serotonin as an enteric neurotransmitter.

I could, and did, proceed to put serotonin through a set of tests (modeled after “Koch’s postulates,” which establish the cause of an infectious disease), which are accepted by virtually all neuroscientists as the criteria that must be met by any substance if it is to be accepted into the pantheon of established neurotransmitters. To satisfy this biochemical equivalent of the labors of Hercules for enteric serotonin, it was necessary to prove:

  1. that serotonin is really present in the nerve endings at the sites where I had proposed that serotonin might be a neurotransmitter;
  2. that serotonin exactly mimics the effects of the natural neurotransmitter;
  3. that serotonin is actually released when the nerves that contain it are stimulated;
  4. that blocking the action of serotonin (or depleting serotonin) abolishes the effects of nerve stimulation; and
  5. that there is an effective inactivating mechanism, which can turn off the response of nerve cells to serotonin once neurotransmission has been accomplished.
Return to Table of Contents