Endogenous Versus Exogenous: Still Not the Issue

Thomas AM Kramer, MD



Many readers may recall a time during the early 1980s when it was believed to be important to classify depression as either endogenous or exogenous. The idea was that there was a difference between depression precipitated by life events, called exogenous depression, and depression that was inherent to the patients' physiology, referred to as endogenous depression. The theory was that patients with exogenous depression did not respond to antidepressants -- ie, tricyclic antidepressants or monoamine oxidase inhibitors (MAOIs) -- because, presumably, their depression was not a function of their physiology but rather a reaction to their life situation. As such, they required treatment with some form of talking therapy. This theory, as it was promoted at the time, not only made the distinction between endogenous and exogenous depression based on symptoms (ie, did they or did they not have vegetative symptoms of depression), but also by assumed etiology. Thus, it was believed that depression precipitated by the loss of a loved one or any other grief-inducing event would not respond to antidepressants because it was exogenous, ie, not physiological.

In retrospect, this rather dualist approach to depression seemed to imply that only some behaviors had anything to do with the chemistry of the brain, but other behaviors were somehow exempt. All of this thinking came, to a certain extent, from the discovery of monoamine neurotransmitters and their role in depression. Since antidepressants seemed to increase the amount of norepinephrine, serotonin, and perhaps dopamine by making more neurotransmitters seemingly available, it made sense at the time to understand depression as a deficit of these neurotransmitters. Ignoring the fact that the effect of these drugs on the neurotransmitters was virtually immediate, their effect on the patient took considerably longer. The assumption was that the depressed patients clearly needed more of something, and the neurotransmitters were the best candidate at the time. It was hard to believe, then, that life events could change fundamental biochemistry. We now are fairly certain -- armed with new knowledge from various studies about dietary manipulation, blood and CSF level monitoring, and other sophisticated methodology -- that the deficit model of neurotransmitters is considerably more simplistic than whatever the reality of the pathophysiology of depression is.

Much more recently, we came up with a new application of the semantic distinction between endogenous and exogenous. In spite of the data that question the validity of a deficit paradigm, we continue to think of psychopharmacology as somehow having an effect on some sort of balance. One often hears patients parroting this idea by referring to themselves as having a chemical imbalance. As we strive to somehow rebalance that imbalance, we struggle to conceive of exactly what it is that is out of balance. Throughout most of the history of the treatment of depression, we have done this with reuptake blockers, which ostensibly increase the amount of neurotransmitter available to the outside of the neuron by blocking reuptake. These drugs, by the semantic distinction described above, would be exogenous, ie, they are not something that the body naturally produces but are ingested to achieve an impact on the balance of neurotransmitters.

More recently, we have begun to get interested in the use of endogenous compounds, ie, hormones or other substances that are naturally produced by the body, in the treatment of depression. The idea is that if we administer substances that the body already has, but perhaps doesn't have enough of, this may treat the depression. Recent studies have shown that estrogen supplementation, growth hormone, and even secretin, which is used in the treatment of autism, may have beneficial effects in depressed patients. The idea here is once again to rebalance an imbalance by giving the actual substance that the body may be in deficit of. This brings about interesting discussions concerning the actual definition of a drug and whether it is somehow better or safer to give, as treatment, substances that are already found within the body.

This kind of work can be misleading or deceiving. Virtually every medical disorder that results from having too little of a hormone has a companion disorder that is a result of too much of that same hormone. In addition, it is often impossible to deliver a naturally occurring neurotransmitter or hormone to its target in all cases.

One intriguing example of this was the development of gabapentin. Gabapentin is a biologically derived compound that was developed with a very simple idea. Many of the drugs that we use to treat epilepsy are active in the gamma-aminobutyric acid (GABA) system. The idea was that if we could somehow give the patient GABA, instead of drugs that accentuate the GABA system, we would somehow have a better, purer response. There was only one problem with this idea: GABA does not cross the blood-brain barrier. All of the GABA that is in the brain was manufactured there. In order to give the brain a dose of GABA, you would either have to inject it directly into the brain, a procedure that most patients would object to if it were done on a regular basis, or modify the GABA molecule in some way that would maintain its action but allow it to cross the blood-brain barrier. That clever thing was done; a pentin ring was attached to GABA, and thus gabapentin was born. It works quite well, and everyone was happy until someone actually conducted studies of gabapentin receptor binding. What they discovered was that gabapentin had absolutely no interest in GABA receptors or any GABA circuitry, but seemed to be very interested in the glutamate system, where it turns out all of its actions take place. In other words, the drug's efficacy had nothing to do with the ideas behind its development. I have been told this story informally, and I have no idea if it is actually true, but it illustrates the point I am trying to make quite nicely.

I propose a reframing of the paradigms that we use for psychopharmacology and its relationship to neurotransmitters. Neurotransmitters slosh around the body, and specifically inside the brain, in relatively constant amounts. The drugs that we give, even MAOIs, do not really affect the number of neurotransmitters in the body very much. What these drugs do is affect receptors. Instead of being concerned about the effect of norepinephrine and serotonin, we really need to redefine our concept of psychopharmacology as receptor drugs. Saying, for example, that selective serotonin reuptake inhibitors (SSRIs) treat depression by increasing serotonin is like saying that a boat sinking on the ocean needs to have reduced water levels. The water is there and all around. To fix the boat you need to plug the holes. That will be a great deal more effective than worrying about decreasing the overall amount of water in the system.

SSRIs affect the serotonin transporters in cell membranes; they do not necessarily affect the overall level of serotonin. There is no deficit or surplus of serotonin; there are cells with impaired ability to have certain levels of serotonin on either side of their membranes. If serotonergic drugs really did affect serotonin overall, they would cause absolutely horrible GI side effects, since the gut has considerably more serotonin and serotonergic neurons than the brain has. This is also why different patients get better on different SSRIs. These agents may all block serotonin reuptake, but each is structurally distinct and thus may bind differently to serotonin transporters, depending on the patient. Similarly, dopamine blockers that are used to treat psychosis do not affect dopamine as much as they lower the sensitivity of certain cells to dopamine by blocking some of their receptors. This has no effect on the total volume of dopamine. It is even more likely that some drugs that appear to work by serotonin receptor blockade actually work by shunting the serotonin moving around from one receptor group to another. In other words, if a certain class of serotonin receptors is completely blocked, the serotonin has no choice but to bind to other receptors.

For years we have struggled to attribute fluctuations in neurotransmitters to drugs' mechanisms of action. Once we begin to conceive of drugs as affecting receptors, things generally seem to make more sense. Even the hormonal treatments described above have their effects at the actual receptors on the cells. Newer exciting treatments, such as the use of corticotrophin-releasing factor antagonists to treat depression and anxiety, and the most recent work, involving the noncontroversial use of a controversial compound, RU486, for the treatment of depression, are aimed at antagonizing hormonal receptors. Why should we concern ourselves with receptors? Because of Willie Sutton's law. Sutton, the noted bank robber, when asked why he robbed all those banks, replied, "Because that's where the money is."


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.