The Serotonin Hypothesis
The widespread theory of depression is that depressed people have a chemical imbalance manifesting as low serotonin. Serotonin is perceived to be ‘the happiness neurotransmitter’ so low levels would make one depressed. There is some evidence to support this:
- People with depression and people who suicide have low markers of serotonin function and metabolism .
- People with depression have a higher incidence of chronic pain and serotonin dampens feelings of pain .
- Feeding people a tryptophan deficient diet produces serotonin depletion and may result in depression-like symptoms .
- Antidepressants that increase serotonin (SSRIs, SNRIs and MAOIs) can treat depression.
However, there are a few problems:
- Serotonin depletion only lowers mood in people with either a family history of depression or who have previously had depression .
- SSRIs take about 3-6 weeks to have much effect and up to 6-12 weeks for the patient to be in remission from depression .
SSRIs have relatively immediate effects on serotonin levels, but take many weeks for noticeable effects suggesting any acute effect on mood isn’t what’s driving recovery. Perhaps another function of serotonin has been overlooked.
An interesting observation is that people with depression and anxiety disorders have a lower brain volume  and years spent with depression, but not age, is associated with greater reductions in hippocampal volume  and more low signal foci, which is suggestive of dysfunctional neural networks .
One of the other functions of serotonin is to stimulate the production and release of a number of growth factors such as BDNF, VEGF, IGF-1, FGF-2 and NGF which all increase neurogenesis. Neurogenesis increases cell proliferation and survival, followed by the production of new neurons or glial cells, then increasing synaptic connections throughout the brain and neural plasticity, which can lead to a more functional reorganisation of neural networks .
Evidence that antidepressants work by increasing neurogenesis:
- Mice without BDNF receptors don’t respond to antidepressants .
- Other treatments for depression such as NSRIs, MAOIs and electroconvulsive seizures (shock therapy) all promote neurogenesis. Of these, the treatments that increase BDNF the most are the most clinically effective .
- Antidepressant treatment increased BDNF up to normal levels within the first month, which correlated with mood improvement 
- Antidepressants take 6-12 weeks likely because that’s how long it takes for the average depressed brain to increase in volume and improve its function.
- SSRIs and growth factors like BDNF aren’t ‘happiness molecules’ as they have limited effects on mood in healthy people who have normal sized brains and functional neural networks .
Other things that increase neurogenesis are therapeutic for depression:
Exercise is roughly equally as effective as SSRIs for depression and is improves anxiety disorders and bipolar disorder . This effect is often explained by an increase in endorphins leading to better mood. However, just like ideas surrounding serotonin, any acute effects on mood do not have a lasting effect on a chronic disease like depression. Any therapeutic effect should be long-lasting and cumulative. Just like antidepressants, exercise also increases the same growth factors that promote neurogenesis* .
Zinc supplementation to correct low zinc levels, which are often seen in depression , increases BDNF  and IGF-1 , is anti-inflammatory and is therapeutic for depression .
DHA increases BDNF , is anti-inflammatory and it is therapeutic for depression , although this effect may be attributed to EPA instead  (either way it’s LCO3). Low plasma levels of DHA are associated with more depressive symptoms  and a high ratio of long chain omega 6:3s results in an excessive neuroinflammatory response . During gestation and lactation, mothers who have an inadequate DHA intake may become depleted in this nutrient due to the developmental needs of the foetus/infant and DHA depletion is a likely cause of postpartum depression .
While people with depression have low serum levels of BDNF , their depression isn’t due to a Prozac deficiency, and the reduced brain volume is unaccounted for.
* Given that much of the neural atrophy in depression is in the hippocampus, the role of the hippocampus and neurogenesis in learning and memory. My guess is that the kinds of exercise that would most promote neurogenesis would be complex movements, engaging with the environment and sport, rather than stationary bikes and machines.
(1) The chemical imbalance myth
(2) If low serotonin levels aren't responsible for depression, what is?