Abstract
This issue contains a speculative hypothesis by Professor Max Bennett on the neurobiology of depression. This hypothesis links alterations in a regenerative loop at central glutamatergic synapses involving both neurons and glial cells, especially in the hippocampus, to alterations in the functional state of neural networks related to mood disorders. To assist readers of the journal evaluate this hypothesis, this editorial summarises historical neurobiological theories of depression, and developments over the past decade which provide a background to this hypothesis.
The original amine hypothesis of depression is now over forty years old [1] and arose from three clinical-pharmacological observations. Drugs, such as reserpine, which deplete pre-synaptic stores of the amine neurotransmitters serotonin and/or noradrenaline, and reduce availability of these amine neurotransmitters in synapses induce depression in some people. Tricyclic antidepressant drugs relieve depression by increasing the availability of amine neurotransmitters in the synapse by inhibiting their reuptake into pre-synaptic nerve terminals. Monoamine oxidase inhibitors relieve depression by increasing the availability of amine neurotransmitters in the synapse by inhibiting the breakdown of these chemicals in the synapse. This amine depletion theory of depression guided neurobiological research on depression for a number of decades, and in a simplistic form has been accepted by some members of the public who hold a view that “depression is a serotonin deficiency”.
Even from the early days of neurobiological depression research there were many doubts about a simplistic amine deficiency theory including concerns that the pharmacological effects of antidepressant drugs were almost immediate, but clinical response typically took many days or even weeks for therapeutic benefit. During the 1980s research then focussed on alterations in pre- and/or post-synaptic monoamine receptor function, rather than just synaptic cleft depletion of neurotransmitters.
Over the past decade there has been a revolution in how we are thinking about the neurobiological basis of depression, with a major shift away from a focus on the synapse alone. Changing paradigms include thinking about the neural circuits involved in depression, the waxing and waning of neurogenesis and synaptogenesis, and realising that there may be important abnormalities of glial cells rather than just abnormalities of neuron to neuron synapses. These shifts in paradigm have occurred within broader developments in neuroscience, which now appreciate that the adult brain does not have a fixed number of neurons which slowly die, but adult brains are growing new neurons, and connections between neurons are in a state of dynamic flux (plasticity). With these changing paradigms, we can no longer just think of depression as an abnormality of the connections (synapses) between statically wired neurons, but the wires (neurons) themselves, their connections (synapses), and their supporting structures (glial cells), are in a state of dynamic change.
Neural Circuitry and Glial Cells
Developments in neuroimaging have shown that there exist highly interconnected neural circuits linking cortical, limbic and subcortical structures, including the prefrontal cortex, thalamus, amygdala, hippocampus, striatum and hypothalamus [2]. Abnormalities of these neural circuits are likely associated with mood disorders. In 1997, Drevets et al.[3] reported decreased blood flow in the subgenal prefrontal cortex, which is an area in the prefrontal cortex involved in mediating interactions of cognitive and emotional responses. Further neuropathological research reported that this prefrontal subgenal area (Brodmann area 24) had a decrease in the number of glial cells, but not of neurons [4]. Subsequent work has revealed that there are reduced numbers of oligodendrocytes, rather than reductions in astrocytes or microglial cells. The protein S100b (beta) is a glial cell protein under the regulation of the serotonin 1A receptor which at high levels stimulates cell death and at low levels promotes cell survival, is abnormal in bipolar disorder, but not schizophrenia [5].
Neuroimaging techniques that measure blood flow in various brain regions, as an indicator of brain activity, are measuring the metabolic activity within glial cells, not the activity of neurons. Furthermore, synapses are not just composed of a pre- and post-synaptic neuron, but inevitably there are glial cells at each synapse which are vital aspects of synaptic activity.
Cellular and Molecular Abnormalities in Mood Disorders
The neurobiological abnormalities in mood disorders are not limited to alterations in serotonin and catecholamines. A wide range of abnormalities in neuroendocrine functions including the hypothalamic pituitary adrenal axis, in excitatory amino acid neurotransmitters such as glutamate, in signal transduction pathways and G protein-coupled receptors, in neurotrophic factors such as cAMP response element binding protein (CREB) and brain derived neurotrophic factor (BDNF) have been reported [6]. Stress and stress hormones contribute to neuronal atrophy, endangerment and death; while antidepressant drugs and mood stabilisers appear to promote synaptogenesis and neurogenesis. Jacobs et al.[7] postulate that depression is associated with the waxing and waning of neurogenesis. Fluoxetine's antidepressant activity may be dependent upon neurogenesis, rather than direct serotonin reuptake inhibition at synapses [8]. With these shifts in thinking about antidepressant action including effects on synaptogenesis and neurogenesis, rather than just on direct synaptic activity, it is no longer a mystery as to why antidepressant action may take weeks, rather than hours.
Summary
The hypothesis by Bennett in this issue continues these recent changes in thinking about the neurobiology of mood disorders, which involves thinking about synaptic activities which involve both glial cells and neurons, within the context of neural networks which modify psychological functioning.