The Serotonin Hypothesis in Depression: Confusion and Enlightenment

Drew Ramsey, MD; J. John Mann, MD


May 03, 2023

This transcript has been edited for clarity.

Drew Ramsey, MD: Hello, everyone. Welcome back to Medscape Psychiatry. I'm Dr Drew Ramsey. I'm an assistant clinical professor of psychiatry and part of the Medscape Psychiatry editorial board.

I'm here with a very special guest. Dr Mann is a professor of translational neuroscience in psychiatry and radiology, and he's also the director of the Columbia Depression Center. Dr Mann, welcome.

You're also someone who taught me neuroscience way back in the day when I was a resident. I wanted to speak with you because, as a clinician and someone trying to translate science in some ways like yourself, I think many of us clinicians are confused after this systematic umbrella review came out questioning one of those bedrock principles of depression treatment and mental health treatment, which is the serotonin hypothesis of depression.

Dr Mann, first, can you tell us a little bit about your work at Columbia Psychiatry? You really have a storied history in both research and clinical work.

J. John Mann, MD: We've been studying for many years the neurobiological basis of mood disorders, major depressive disorder (MDD), and bipolar disorder. We're interested in both the genetic and the environmental, or epigenetic, effects that create the pathogenesis of these illnesses.

Ramsey: For those of us in clinical practice, I think of you as one of our generals in the sense of really helping us understand where these illnesses are coming from. It's never felt satisfying as a clinician to have a general explanation, and it's been inspiring over the past decade to see a large amount of new science coming out.

Can you tell us about this umbrella meta-analysis? When you saw it, what was your first reaction? What did you think?

Mann: I read the paper. I found that it's a fairly mediocre paper. It took several selected aspects of studies of depression, mostly focused on the serotonergic system. It really didn't help the reader because it didn't separate pivotal, landmark, anchor papers from poorly designed papers with small samples. It kind of then took a vote, as meta-analysis often does, where it lumps together poor papers with good papers and gives them all an equal vote, and then says, here's the scorecard. The result is more confusion than enlightenment.

Ramsey: There were several aspects of the paper that were confusing. I think one is the notion of a chemical imbalance. As a clinician, I've always felt, believing in biology and science, that when things aren't feeling good for me upstairs, something's going on biologically, because most feelings — I would say all of them — have something to do with biology.

Is it true that the chemical imbalance hypothesis is just wrong, and we shouldn't think about that or explain depression like that to patients anymore?

Mann: First, we need to remember that there is a significant genetic component to both MDD and bipolar disorder. That means that there is a biology out there that's at least partly genetic. Then we also know that adverse childhood experiences significantly increase the risk for the development of a mood disorder in adulthood.

We also know that these illnesses are transmitted to a greater degree than just genetics in families, which reflects this combination of factors working together. When you look at it from that perspective, then you know that there are going to be biological abnormalities in people who suffer from depression. The question is, what are they and how well can we understand them?

As best as we can tell right now, there's an array of abnormalities in people with depression. One group of abnormalities are in the adrenergic system. A second group of abnormalities are around inflammation. A third group of abnormalities are around the hypothalamic-pituitary-adrenal (HPA) axis and cortisol and stress. There's a list of neurotransmitter systems in addition that also show abnormalities, which include the noradrenergic system, the dopaminergic system, the GABAergic system, and the glutamatergic system.

When you go through the array of medications that we've discovered, mostly by accident, that are good antidepressants, we find that they target all of these things. That is a remarkable collection of convergent information that points to the fact that there are biological abnormalities.

Ramsey: I wonder whether the author's goal was to inform the public of all the new science, and I felt that was missing. Maybe the evidence for serotonin isn't as strong, but no one has been saying serotonin is the only game in town for a long time.

You've been interested in these other systems, the GABAergic and glutamatergic systems, since I was a resident 20 years ago. In addition, there's a whole new interest in inflammation and neuroplasticity.

I wonder if you can share with us a little bit of how you now think of depression. There's a convergence of these systems, but are you beginning to see some genetic subtypes? Are you seeing involvement of neuroplasticity or inflammation in ways that excite you as a clinician/scientist?

Mann: We actually don't care whether those studies have found much or whether there were studies that didn't agree with earlier studies. Their relevance to the brain is pretty limited. It's the studies that focus in on the brain that are the most interesting.

In fact, recently there have been some much more comprehensive brain studies using brain scanning and postmortem brain tissue, which have shown reinforcing, complementary sets of observations that definitely implicate the serotonergic system, but also implicate other things as well, like inflammation and neurogenesis. These are not independent of each other. These are all linked together.

Just to give you some broad strokes, we know that the serotonergic system is not just a neurotransmitter system. It's a growth factor system. This is a separate and ongoing process.

Serotonin plays an important role in early brain development, but it also plays a fundamental role in ongoing brain health throughout life. We can see now that. For example, if you look at the brain of individuals who died and who suffered from, say, MDD, the dentate gyrus, which is part of the hippocampus where neurogenesis occurs in adulthood, is half the size in depressed patients. If those depressed patients die while at the same time being treated with a selective serotonin reuptake inhibitor (SSRI), it's not half the size. It's the normal size.

If you go and look inside the dentate gyrus to find out why, you see that the untreated, depressed patients have half the number of mature granule neurons in their gyrus. They have less angiogenesis, which is like the supply of nutrition to the neurons, and the processes are shortened. Everything is diminished, which explains why, when you look at the whole volume of the dentate gyrus, it's half the size.

If you look at the ones who are taking SSRIs, you find several big differences. First, the number of mature granule neurons are the same as in controls. They don't have that deficiency. Then you see that angiogenesis is also comparable to controls. Even more striking, you see the reason for that. You see that the multiplying neuronal progenitor cells are way more than normal.

When a doctor is treating a patient with an SSRI, it's not some kind of symptomatic relief like when you take Tylenol for a fever. It's actually more like an antibiotic. The antibiotic kills the infective organism. The SSRI turns on progenitor cells and causes a catch-up in terms of turning on the pipeline of producing neurons so that the dentate gyrus comes back toward normal. You're actually doing something fundamental for the patient.

The longer they stay in this state of depression untreated, the worse the condition gets. The sooner they start treatment and the more consistently they stick with a treatment, the sooner they're going to get on the path toward normality. Meanwhile, inflammation is tied in with this. Some of the inflammation is protective, and some of the inflammation is damaging, like in many medical conditions.

We still don't fully understand that mix, but we're learning about it and we're busy studying it. You begin to see how all of these things are connected. The HPA axis is overactive; that's damaging and interfering with inflammation and protection, and maybe increasing the rate at which cells are dying.

The glutamatergic and GABAergic systems are out of balance. There's too much glutamate, which can be neurotoxic, and yet not enough GABA, which is supposed to be inhibiting the excess glutamate release protecting you. There's a whole bunch of things that aren't right. We know, at least for some of this, how antidepressants may be working to fix the problem.

Ramsey: Thank you, Dr Mann. That is very helpful for all of us who are both prescribing and thinking about the notion of serotonergic medications working toward helping growth factors or helping growth within the brain and protecting the brain. That's very striking. That also helps in terms of long-term use.

Many patients with both bipolar depression and MDD who take a serotonergic medication or another medication that helps them for depression have a question: I'm feeling better, Doctor; can I stop my medication because of long-term side effects?

You're suggesting that the most dangerous thing long-term is inadequate treatment of mental health disorders like depression because it causes the brain to shrink more quickly, and that we should really think carefully when there's a concern. There was some idea that there's a compensatory change happening when patients are on long-term SSRIs.

Could you help us understand a little bit about the long-term safety of SSRIs? It sounds like one way we should think about explaining them to patients is around the neurogenesis effect and the anti-inflammatory effects. These aren't serotonin medicines making serotonin go up. You haven't thought about them that way for a long time.

Mann: SSRIs have been prescribed to huge numbers of people for a very long time. They don't have a long-term, cumulative, toxic, bad effect. It just doesn't happen. We want people to take SSRIs to treat their episode of depression, which is usually roughly a 6-month course of treatment, and then we usually will try to taper them off.

If a person has severe depression that occurs with frequency or if the depressive episodes are dangerous, meaning that they've made a suicide attempt that they've been lucky enough to survive during a depression, we don't want them to have another episode. We want to prevent the next episode.

When they feel cheerful, yes, we do want to finish the 6-month course to make sure that they're not going to have a return of that episode of depression. We're also looking to the future and making a judgment for each patient. Is it better to go off the medicine and then, when you get depressed next time, just come and see me? Or do we think that maybe it's too much of a risk?

Certainly, the patient has to participate and buy into this decision-making process. The family and/or significant other should be often part of this conversation because they are the ones who often see the depression coming back before the patient does. This needs to be a team decision where all of the parties — stakeholder, patient, doctor — feel that they've been part of this decision-making process. Then you decide whether you are going to continue long-term or not.

What's the downside of long-term use? Some people don't like the side effects, and maybe most people don't like side effects. Of course, that's completely standard. You're trading side effects vs benefit. That's true of any medication.

The main thing with SSRIs is, if you've been taking them for a long time, you can't just stop them abruptly because you'll get what's called discontinuation syndrome, which is very unpleasant. It's kind of like getting the flu, and you have these electric feelings in the head.

Ramsey: And a little visual disturbance. My patients complain that something goes by and they track it and experience something like vertigo.

Mann: People feel pretty horrible, but it usually only lasts a few days or a week or two, and then it just progressively fades away. You can attenuate it or avoid it by very gradually tapering the SSRI over a few weeks, and then the person doesn't get those symptoms and they're off it. This isn't a sign that the person is now addicted to SSRIs. There is no such thing as addiction to SSRIs. They have no addictive effect.

They do not have the risk for tardive dyskinesia that you see with antipsychotics. Their main problem is that they have some side effects — of which the most notable, I think if people are honest, are the sexual side effects. Then there are more minor side effects, like some people getting a tremor; there may maybe some gastrointestinal (GI) effects, and so forth. That's weighed against the other factors and that's part of the conversation.

Ramsey: It's such a hard thing, I find, for many patients who benefit from SSRIs because there's such a bias against them. I hope some time we can hear more of your thoughts on where that comes from. It's interesting to me that, as you speak about antidepressants, it gives a reassurance of helping on a biological level and there being clarity in the research of why that's happening, how patients are benefiting, and although there are some side effects, the real risk to patients is having untreated depression.

We know from the data that this is quite true. We have the most disabling illness on the planet, and in the United States, and it's horribly undertreated. In men, I've seen statistics that 80% of men with depression don't seek any treatment at all. I think 50% of kids with mental health disorders don't ever get treatment.

It's wonderful to hear some of the details of your research and how you respond to this systematic analysis. It sounds, just to summarize, that SSRIs continue to be helpful to patients. The data are fine and good long-term. You have no concerns about long-term use, but also as you described them, you think about them as neuroplastic agents.

Mann: Yes, that's correct. I see them as fundamentally changing the pathogenesis of the illness to help people get better. I think there's a huge stigma out there about SSRIs and all psychotropic medication. We have made significant progress in trying to reduce that stigma, but it is still enormous. You can see Black Lives Matter and all this structural discrimination and racism. These beliefs have been hammered into people's minds for decades. They don't go away rapidly, and they're still there.

I remember a reporter once said to me, "Dr Mann, there are millions and millions of prescriptions of SSRIs. Don't you think that's too many?" I said to this reporter, "You know, unfortunately, there are also millions of prescriptions of drugs for cancer, for diabetes, and for hyperlipidemia. Are you telling me that too many of these drugs are being prescribed, or is it just a question of are we adequately responding to a medical problem that's really out there that needs to be fixed?"

Ramsey: Dr Mann, thank you. I think also invoking that there is a structural discrimination against mental health patients and against mental health treatment. There is an incredible bias. Spending a large amount of time in the integrative health world, it's surprising to me that I've become someone defending Zoloft often and reminding people that, along with lifestyle and proper nutrition, many individuals need more help. They have a biological illness that needs treatment, and there's tremendous science behind that.

I forgot one question, which is around the tryptophan depletion studies. I wanted to understand: Is this because the brain is able to pull tryptophan from muscle and that these studies are so short-term that we don't actually need dietary tryptophan in an acute sense to make serotonin?

Mann: The brain needs a steady supply of tryptophan. Even after a short-term, like an hour or two of cutting off the supply of tryptophan, the serotonin supply drops, and the amount of serotonin release drops. It's very quick. Just by eating a normal diet, you have plenty of tryptophan.

The only way we can produce this diminution in tryptophan in the brain is by flooding the person with this horrible drink of large neutral amino acids that compete for the transport site across the blood-brain barrier and cut off the supply of tryptophan. We are not depleting tryptophan in their bodies. We're only interfering with its transport from the blood into the brain for a short time. A normal person doesn't get any benefit from having tryptophan supplements.

Actually, our work shows that there is an excess of serotonin and then an overabundance of serotonin neurons in the brains of depressed people. The problem is not having enough serotonin. The problem is related to the autoreceptor, which regulates the firing of serotonin neurons. That autoreceptor, for either epigenetic or genetic reasons, in depressed people is excessively expressed so that they shut off the firing too quickly and they can't release the serotonin that they have. People are not suffering from a deficiency of serotonin. They're suffering from a deficiency of serotonin firing and release.

Ramsey: Thank you. I'm here with J. John Mann. Dr Mann is a professor of translational neuroscience and radiology at Columbia University, and he directs the Center for the Prevention and Treatment of Depression at Columbia Psychiatry.

What a masterclass on understanding the cutting-edge research and all the exciting brain research going on so that we're really beginning to understand these illnesses. I always feel an incredible sense of optimism speaking with you because there is so much stigma and so much bias.

I think people have a hard time sorting through an article like this — as you say, a mediocre-quality paper in your opinion — and not really seeing the true science of the multiple systems that are working together to create our mood and create consciousness, and how disruptions in these are very clearly at the foundation and pathogenesis of illnesses and mood disorders like depression and bipolar disorder.

Everyone, thank you so much for joining us. I'm Dr Drew Ramsey, and we'll be back again soon.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


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.