Major depression encompasses heterogeneous disorders in humans that are associated with abnormalities in reward-related brain structures such as the nucleus accumbens, prefrontal cortex, amygdala and hippocampus. Changes in the activity and functional connectivity of these sites leads to abnormalities in the perception and interpretation of reward valence, in the motivation for rewards, and in subsequent decision-making.Recent drug development efforts and other new treatment approaches such as deep brain stimulation offer the potential to more effectively treat depression. However, the field still faces major difficulties. The heterogeneity of depression symptoms suggests that its aetiology is diverse, there are still no known or accepted biomarkers to diagnose major depression — let alone its many subtypes — and promising new treatments have yet to gain approval by the US Food and Drug Administration (FDA).Increasing evidence indicates that precipitating factors such as chronic stress induce changes in the functional connectivity within the brain's reward regions, and that such changes mediate reward-related depression-like behavioural symptoms in animal models, including social avoidance and anhedonia. The molecular and cellular bases of these behavioural abnormalities include changes in glutamatergic and GABAergic synaptic plasticity, dopamine neuron excitability, epigenetic and transcriptional mechanisms, and neurotrophic factors.The nucleus accumbens is central in processing and responding to rewarding and aversive stimuli. It has been extensively implicated in reward-related behavioural abnormalities that characterize depression and associated syndromes. Chronic exposure to stress alters gene expression patterns in and the morphology (and ultimately the functional activity and connectivity) of nucleus accumbens neurons — neuroadaptations that contribute importantly to depression-like behaviours.Advanced experimental tools, such as inducible mutations in mice, virus-mediated gene transfer and optogenetics, have made it possible for the first time to directly delineate the role of specific proteins acting within specific cell types within reward-related brain structures in mediating depression-like behavioural abnormalities in animal models. For example, medium spiny neurons (MSNs) that predominantly express D1 dopamine receptors have a very different effect on reward from MSNs that predominantly express D2 dopamine receptors.It will be important for future studies to examine the molecular and cellular underpinnings of depression-like behaviours in females. Depression is twice as likely to occur in women than in men, but animal studies have mostly been conducted in males. There is evidence that females use different cognitive strategies, exhibit increased stress sensitivity and show variations in reward-related behaviours throughout the oestrus cycle that may render them more sensitive to the deleterious effects of stress.