Neurochemistry of Psychiatric Disorders

Classical neurochemical hypotheses of psychiatric illness link a disorder's symptoms to dysfunction in a specific neurotransmitter system, inferred largely by reasoning backward from what treatments do. The monoamine hypothesis proposes that depression and many anxiety disorders result from functional deficiency in serotonin (5-HT), norepinephrine (NE), or dopamine (DA) signaling in specific circuits, particularly limbic and prefrontal regions that regulate mood, motivation, and threat appraisal. The dopamine hypothesis of schizophrenia proposes that hyperactive dopamine transmission in mesolimbic pathways produces positive symptoms, while dopamine hypoactivity in mesocortical pathways contributes to negative and cognitive symptoms. More recent glutamate and GABA hypotheses propose that disturbances in excitatory (glutamatergic) and inhibitory (GABAergic) signaling — particularly NMDA receptor hypofunction and interneuron dysfunction — underlie the aspects of schizophrenia and mood disorders that monoamine and dopamine models cannot explain. These hypotheses are best understood not as competing final answers but as successive, partially overlapping models, each anchored by a specific class of evidence: pharmacological (drugs that treat or mimic the disorder), biochemical (metabolite and receptor-binding studies), genetic (association and heritability data), and more recently, functional imaging and rapid-acting drug discovery. A critical conceptual distinction runs throughout: the acute pharmacological action of a drug (e.g., blocking serotonin reuptake within minutes) is not the same as its therapeutic mechanism (relief of depressive symptoms after weeks). Bridging that gap has driven the field from simple "chemical imbalance" language toward models emphasizing receptor adaptation, intracellular signaling cascades, neurotrophic factors, and structural synaptic plasticity. Throughout, three neurotransmitter systems recur: monoamines (5-HT, NE, DA), the principal excitatory transmitter glutamate acting through NMDA and AMPA receptors, and the principal inhibitory transmitter GABA acting through GABA-A receptors. Understanding how psychiatric drugs perturb these systems — building on Drugs Acting on Neurotransmitter Systems — and how chronic perturbation reshapes circuits sharing substrates with reward and motivation circuitry, is the foundation for everything that follows.

1. The monoamine system and depression. Serotonergic neurons originate in the raphe nuclei of the brainstem and project diffusely to cortex, limbic structures (amygdala, hippocampus), and hypothalamus. Noradrenergic neurons originate mainly in the locus coeruleus; dopaminergic neurons originate in the ventral tegmental area (VTA) and substantia nigra. All three systems modulate mood, arousal, motivation, and stress reactivity rather than encoding specific sensory or motor information — they act as broadcast neuromodulators. Antidepressant drug classes act at the synapse within minutes to hours:

• SSRIs (e.g., fluoxetine, sertraline) block the serotonin transporter (SERT), increasing synaptic 5-HT.
• SNRIs (e.g., venlafaxine) block both SERT and the norepinephrine transporter (NET).
• TCAs (e.g., imipramine) block SERT and NET non-selectively, plus histamine, muscarinic, and alpha-1 receptors (producing side effects).
• MAOIs (e.g., phenelzine) inhibit monoamine oxidase, the enzyme that degrades 5-HT, NE, and DA intracellularly, raising cytoplasmic and eventually synaptic monoamine levels.

Blocking reuptake or degradation raises synaptic monoamine concentration almost immediately — SERT occupancy by an SSRI reaches near-maximal levels within hours of a single dose. Yet clinical antidepressant response typically requires 2–6 weeks. The resolution to this paradox lies downstream: acutely elevated synaptic serotonin activates somatodendritic 5-HT1A autoreceptors on raphe neurons, which initially suppresses neuronal firing and dampens serotonin release — a negative feedback loop. Over 1–3 weeks of continued SSRI exposure, these autoreceptors desensitize, disinhibiting raphe firing and producing a net, sustained increase in serotonergic tone at postsynaptic targets. In parallel, sustained monoamine signaling upregulates intracellular signaling — cyclic AMP, protein kinase A, and the transcription factor CREB — increasing expression of brain-derived neurotrophic factor (BDNF). BDNF acts on TrkB receptors to promote dendritic spine growth, synaptogenesis, and possibly adult hippocampal neurogenesis, processes plausibly necessary for symptom remission, not just receptor occupancy.

2. Dopamine and schizophrenia. The mesolimbic dopamine pathway runs from the VTA to the nucleus accumbens and is implicated in salience attribution and reward — the same circuitry central to Addiction and the Mesolimbic Reward Circuit. The dopamine hypothesis proposes that excess dopamine release or D2-receptor sensitivity in this pathway causes aberrant assignment of salience to irrelevant stimuli, generating delusions and hallucinations (positive symptoms). A separate mesocortical pathway, projecting from VTA to prefrontal cortex, is instead hypothesized to be hypoactive in schizophrenia, contributing to negative symptoms (flattened affect, avolition) and cognitive deficits (working memory impairment) — this is the "dual dopamine" refinement of the original hypothesis. Antipsychotic drugs act primarily by blocking D2 dopamine receptors:

• First-generation (typical) antipsychotics (e.g., haloperidol, chlorpromazine) are potent D2 antagonists; their clinical potency correlates almost linearly with D2-receptor binding affinity.
• Second-generation (atypical) antipsychotics (e.g., clozapine, risperidone, olanzapine) combine D2 antagonism with strong serotonin 5-HT2A antagonism, which appears to reduce extrapyramidal side effects and may modestly improve negative symptoms.

Clinically effective D2 occupancy in the striatum typically falls in a defined range:

This "therapeutic window" of D2 occupancy is one of the most robust quantitative relationships in psychiatric pharmacology, established using positron emission tomography (PET) with radiolabeled D2 ligands in medicated patients.

3. Glutamate, GABA, and the limits of dopamine. Roughly 30% of patients with schizophrenia show minimal response to D2-blocking antipsychotics, and the dopamine hypothesis does not naturally explain negative or cognitive symptoms. Phencyclidine (PCP) and ketamine, both antagonists at the NMDA-type glutamate receptor, reproduce not just hallucinations but also negative-symptom-like blunting and cognitive deficits when given to healthy volunteers — a much fuller symptom profile than amphetamine (a dopamine releaser), which mainly produces positive symptoms. This led to the NMDA receptor hypofunction hypothesis: reduced NMDA receptor signaling, particularly on fast-spiking GABAergic interneurons that normally provide inhibitory control over cortical pyramidal neurons, causes disinhibition and downstream dopamine dysregulation. Loss of interneuron-mediated inhibition reduces gamma-band cortical oscillations (30–80 Hz), which are thought to be necessary for working memory and sensory binding — a finding consistent with EEG studies in schizophrenia patients. GABA itself is central to anxiety. GABA-A receptors are ligand-gated chloride channels; when GABA binds, chloride influx hyperpolarizes the neuron, reducing excitability. Benzodiazepines (e.g., diazepam, alprazolam) bind an allosteric site on the GABA-A receptor distinct from the GABA-binding site, increasing the frequency of chloride channel opening in the presence of GABA — they are positive allosteric modulators, not direct agonists, which is why they require endogenous GABA to act and why overdose alone is rarely fatal (unlike barbiturates, which prolong channel opening directly and can act GABA-independently at high doses, suppressing respiration). Reduced cortical GABA levels and interneuron dysfunction have been found in some anxiety disorders and in schizophrenia, tying inhibitory tone directly to symptom generation in both diagnostic categories.

The reserpine and iproniazid observations (1950s) — birth of the monoamine hypothesis. Reserpine, used as an antihypertensive, depletes presynaptic stores of serotonin, norepinephrine, and dopamine by blocking the vesicular monoamine transporter. Clinicians observed that a meaningful subset of patients receiving reserpine developed depressive symptoms. Around the same time, iproniazid, developed as a tuberculosis drug, was observed to elevate mood in patients — it was later shown to be a monoamine oxidase inhibitor, raising synaptic monoamine levels. These two clinical observations, one drug lowering monoamines and causing depression-like states, the other raising monoamines and improving mood, gave rise to the monoamine hypothesis, formalized explicitly by Joseph Schildkraut in 1965. It is essential to note this was correlational clinical observation, not a controlled causal experiment — reserpine-induced depression is now understood to be much less common and less specific than originally reported, one of the hypothesis's acknowledged weaknesses.

Biochemical confirmation: metabolite studies. Researchers measured 5-hydroxyindoleacetic acid (5-HIAA, the main serotonin metabolite) in cerebrospinal fluid of depressed patients versus controls. Subsets of depressed patients, particularly those with a history of violent suicide attempts, showed reduced CSF 5-HIAA, consistent with reduced central serotonin turnover. This strengthened the monoamine hypothesis but also revealed its heterogeneity: not all depressed patients show this biomarker, indicating monoamine deficiency is neither necessary nor sufficient in all cases.

PET receptor-occupancy studies establishing the dopamine hypothesis's quantitative core. In the 1970s, Philip Seeman and colleagues, and independently Solomon Snyder's group, measured the binding affinity of a wide range of antipsychotic drugs at dopamine D2 receptors in vitro. When plotted against each drug's average clinically effective daily dose, the correlation was striking and nearly linear across several orders of magnitude of dose — drugs requiring higher doses had proportionally lower D2 affinity. This single dataset is among the most cited pieces of evidence in psychiatric pharmacology, directly linking D2 blockade to antipsychotic efficacy. Later PET imaging in living patients (using radioligands such as raclopride) confirmed the ~60–70% striatal D2 occupancy threshold for clinical response described above.

PCP/ketamine psychotomimetic studies — evidence for glutamate hypofunction. In controlled studies during the 1990s (notably by John Krystal and colleagues at Yale), healthy volunteers given subanesthetic intravenous ketamine developed transient positive symptoms (perceptual distortions, paranoid ideation), negative-symptom-like blunting, and working memory impairment — a broader symptom profile than dopamine-releasing stimulants produce. Because ketamine and PCP are selective NMDA receptor antagonists, this provided direct pharmacological evidence that reduced NMDA signaling can generate a schizophrenia-like state, motivating the interneuron-disinhibition model described above.

The serendipitous discovery of ketamine's rapid antidepressant effect. In a landmark 2000 study, Berman and colleagues gave a single subanesthetic ketamine infusion to patients with treatment-resistant depression in a randomized, placebo-controlled, double-blind crossover design. Depressive symptoms improved within hours, often peaking around 24 hours, in patients who had failed multiple standard antidepressant trials. This result — measured objectively using standardized depression rating scales (Hamilton Depression Rating Scale) — was replicated in larger trials over the following decade and led to FDA approval of esketamine (2019) for treatment-resistant depression. It is the single most important piece of evidence forcing revision of monoamine-centric models: a drug with no direct action on serotonin or norepinephrine transporters relieves depression far faster than any monoamine-based treatment, implicating rapid glutamatergic and synaptic-plasticity mechanisms (increased AMPA receptor signaling and BDNF release from a burst of glutamate following NMDA blockade on inhibitory interneurons) as more proximate to symptom relief than monoamine levels themselves.

Example 1 — Dosing and receptor occupancy in antipsychotic treatment. A patient begins risperidone for a first psychotic episode. PET studies indicate that striatal D2 occupancy of 65% is typically required for antipsychotic response, and occupancy above 80% sharply increases extrapyramidal side effect (EPS) risk. Suppose risperidone's D2 receptor binding follows a simple occupancy relationship where occupancy rises steeply between 1 mg/day and 4 mg/day, such that:

A clinician wants the lowest dose achieving the 65–78% "therapeutic window" (efficacy without high EPS risk). Reading the table: 1 mg/day (45%) is subtherapeutic; 2 mg/day (62%) is just below threshold and may be a reasonable starting dose with room to titrate; 3 mg/day (74%) sits squarely in the therapeutic window; 4 mg/day (83%) already exceeds the window and raises EPS risk without proportionally added benefit. The clinician should titrate toward 3 mg/day rather than reflexively increasing to 4–6 mg/day, since occupancy — and clinical benefit — plateaus while side-effect risk keeps climbing. This mirrors the actual clinical pharmacology of risperidone, where doses above roughly 6 mg/day rarely add efficacy but substantially raise EPS incidence. The broader lesson: because the dopamine hypothesis is quantitatively precise (occupancy correlates with both efficacy and toxicity), dosing decisions can be reasoned through mechanistically rather than by trial and error alone.

Example 2 — A treatment-resistant depression case reasoned through mechanism. A 42-year-old patient has failed adequate 8-week trials of two SSRIs (sertraline, then escitalopram) and one SNRI (venlafaxine), each producing SERT/NET occupancy sufficient for pharmacological effect but no meaningful symptom improvement (defined as treatment-resistant depression, or failure of ≥2 adequate trials). The patient's psychiatrist considers three mechanistically distinct options: (a) switching to an MAOI, (b) augmenting with an atypical antipsychotic (which adds 5-HT2A and modest D2 antagonism to boost prefrontal dopamine/norepinephrine tone), or (c) a course of ketamine or esketamine. Reasoning it through: options (a) and (b) still operate primarily through monoaminergic mechanisms — an MAOI simply blocks a different degradative enzyme, and the patient has already failed three drugs targeting reuptake at the same class of synapses, suggesting the monoaminergic bottleneck (if present at all) is not being solved by yet another monoamine-pathway drug. Option (c) targets a mechanistically distinct system — NMDA receptor antagonism triggering rapid AMPA-receptor-mediated glutamatergic and BDNF-dependent synaptic plasticity — bypassing the weeks-long autoreceptor-desensitization pathway altogether. Given the repeated monoaminergic treatment failures and the urgency of persistent depressive symptoms (with associated suicide risk that argues for the fastest available onset of action), option (c) is the most defensible next step, consistent with actual clinical guidelines that recommend ketamine/esketamine specifically for treatment-resistant depression after failure of at least two adequate antidepressant trials. This case illustrates why understanding mechanism — not just "try another antidepressant" — changes real clinical decision-making.

Arvid Carlsson's work on dopamine reshaped both the addiction and psychiatric neurochemistry fields. Working in Sweden in the 1950s, Carlsson demonstrated that dopamine is a neurotransmitter in its own right (not merely a precursor to norepinephrine), showed that reserpine depletes brain dopamine and causes akinesia reversible by L-DOPA, and later helped establish that antipsychotic drugs work by blocking dopamine receptors — foundational steps toward both the dopamine hypothesis of schizophrenia and modern Parkinson's disease treatment. His meticulous biochemical assays, developed before modern receptor-binding technology existed, allowed him to trace dopamine's distribution and turnover in the brain with a precision that made later PET-based confirmation possible.

Mechanism-based reasoning now shapes real clinical protocols. PET-guided D2 occupancy targets inform antipsychotic dosing to minimize extrapyramidal side effects while preserving efficacy, exactly as reasoned through in the worked example above. Esketamine (Spravato), an FDA-approved nasal spray, is used specifically for treatment-resistant depression and for depression with acute suicidal ideation, precisely because its glutamatergic mechanism produces relief within hours rather than weeks — a direct clinical translation of the ketamine studies described above. Clozapine remains the only antipsychotic with demonstrated superior efficacy in treatment-resistant schizophrenia, and its unusual receptor profile (relatively weak D2 affinity, strong D4, 5-HT2A, and muscarinic activity) continues to motivate research into non-dopaminergic antipsychotic mechanisms, including recently approved drugs (such as xanomeline-trospium) that act on muscarinic rather than dopamine receptors entirely. GABAergic mechanisms underlie both benzodiazepine anxiolytics and newer neurosteroid-based treatments (e.g., brexanolone and zuranolone for postpartum depression), which act as positive allosteric modulators at GABA-A receptors with a distinct subunit-selectivity profile from benzodiazepines. Genetic and imaging research increasingly integrates these neurochemical hypotheses with circuit-level functional data, a synthesis that depends directly on the imaging methods introduced next.

Ask the tutor: How would combining a glutamatergic rapid-acting agent with a traditional monoaminergic antidepressant be expected to affect the overall time course of depression treatment?