By Grant H. Brenner
Over the years, the serotonin hypothesis of depression, which suggests that low or impaired serotonin neurotransmission is related to the symptoms and possible causes of depression and related psychiatric disorders, has taken many hits.
Recently, a widely publicized review paper in Nature (Moncrieff et al., 2022) found no direct evidence that serotonin is involved in the pathophysiology of depression. This led to an explosion of reports and even attacks which reverberated across the internet about the serotonin hypothesis, the long-debunked “chemical imbalance” explanation, and, to an extent, psychiatry as a field.
The uproar is understandable–depression affects an estimated 20 million US citizens and nearly 300 million worldwide, and current treatments are only partially effective. There is a growing awareness that mental health problems are epidemic–a recent survey by the American Psychiatric Association found 79 percent of people see mental health as a public health emergency.
Moreover, rates of anxiety and depression are skyrocketing, especially among younger people, suicide has become a leading cause of death, and we are more aware of every day of the hazardous effect of endemic stress and trauma on mental and physical health.
Despite various studies over the years and the recognition that antidepressant and psychedelic medications,1 which improve depression, often increase serotonin levels and are associated with positive brain changes (such as increased neuronal complexity via “sprouting” and possible restoration of brain volume in areas like the hippocampus), there has been a startling lack of evidence showing a direct role for serotonin, perhaps because more sophisticated research methods have not been available.
Absence of Evidence Is Not Evidence of Absence
That is, until now. For the first time, study authors have found a clear association between altered serotonin activity and depression in a robust experimental design.
Researchers Erritzoe and colleagues (2022) report their findings in a recent paper in Biological Psychiatry entitled “Brain Serotonin Release Is Reduced in Patients With Depression.” In this study, they compared two groups of people, 17 people with clinical depression (MDD group for major depressive disorder) rigorously diagnosed with depression and free from other conditions, not taking antidepressants at the time of the study (most never had, some had in the past), and 17 people without any mental illness (the HC or healthy control group).
The statistics analyzed clinical data and brain-imaging findings to determine if MDD and HC differ significantly in the serotonin neurotransmission, and if so, in what direction? If MDD showed lower serotonin, or 5HT, activity, this would directly support the serotonin hypothesis of depression.
Study participants were administered a mildly radioactive chemical developed to directly measure serotonin activity in the brain. Rather than the previous studies in the Nature article, which included indirect measures, using [11C]Cimbi-36 (a “radioligand”) allows researchers to probe actual 5HT activity on PET (Positron Emission Tomography) and MRI (Magnetic Resonance Imaging) scan.
PET is an imaging approach more sophisticated than MRI. Unlike MRI, which produces a lower resolution scan indirectly estimating brain activity (often based on inferred blood flow), PET images positrons (a form of antimatter, antielectrons) to get a fine-grained, direct window into the brain’s inner workings.
Why [11C]Cimbi-36? Prior PET studies have shown that when serotonin goes up, [11C]Cimbi-36 predictably and reliably goes down. It’s called a radioligand because the molecule contains a radioactive form of carbon that lights up on PET.
Study participants, in addition to careful clinical diagnosis, completed depression ratings and underwent PET-MRI imaging after being given a dose of d-amphetamine (a drug familiar for its use in treating ADHD and also for its abuse potential as a street drug and misused prescriptions).
Prior research has shown that if we give a dose of d-amphetamine to non-depressed individuals (HCs), [11C]Cimbi-36 levels drop significantly, reflecting an increase in 5HT activity. Is this response the same in people with depression? If not, it would lend new support to the serotonin hypothesis of depression.
The main study result was that in patients with depression, serotonin activity is indeed diminished compared with people without depression. Following d-amphetamine challenge, [11C]Cimbi-36 tracer binding across brain cortex (surface) areas was as expected for healthy individuals, consistent with prior experiments.
However, for depressed participants, a very different pattern emerged. When this group received d-amphetamine, [11C]Cimbi-36 binding was significantly lower than in the non-depressed group, reflecting underlying decreased serotonin activity in depressed patients compared with healthy controls. This activity was seen across many brain regions, reaching statistical significance in the temporal cortex (areas located on the sides of the brain, in rough proximity to the ears2).
The effect was stronger in MDD participants who had never taken antidepressants (“medication naive”) versus those who had but were not taking them during the study, but this finding is hard to interpret due to the small number of participants in the prior medication group .
Implications and Future Directions
Overall, this study found diminished serotonin activity in patients with diagnoses of clinical depression. This does not mean that depression is caused by serotonin deficiency or that medications that increase serotonin will treat depression. If only it were that easy.
The picture is much more complex than that, with an array of not only many different serotonin receptors and systems throughout the brain but also many other factors, including but not limited to an array of implicated non-serotonin neurotransmitters (norepinephrine, dopamine, glutamate, BA, and others), the role of other biological factors and psychosocial factors, and the recognized role of supporting role of glial cells in brain function–cells which are not neurons but which support and modulate what neurons do.3
A New Hope
This study is not the end of the discussion but rather a new beginning. It is the first study to show diminished serotonin activity in depression.
However, it is relatively only one study with a small number of participants, and as with any study requires replication to determine if it is really true and to build in more nuance to the study design to account for potential confounding factors (like time of day , clinical presentation in a more diverse population, the role of serotonin in disorders which overlap with depression, and so on).
In addition to the main finding that people with depression do have lower serotonin activity compared with those without depression, at least here, this work is important because it is a proof of concept for a study approach to directly assess brain serotonin activity. This approach can and will be refined–for example, using challenges other than with d-amphetamine to perturb serotonin tracer activity.
While, it is a powerful tool and one which, along with others under development such as computational tools to develop better ways of understanding psychiatric conditions (“transdiagnostic approaches”),4 It will help to usher in a new era, with the expectation that growing understanding will enable more effective treatment and ultimate reduce the considerable suffering associated with mental illness.