Stress during pregnancy has profound and long term detrimental consequences on maternal and fetal health, including on the maintenance of gestation itself. However, while intrauterine mechanisms underlying adverse effects on the developing fetus are becoming clear, the maternal neuroendocrine responses that mediate stress perception and signal to the periphery are often not reported. Many neuroendocrine hormones respond to stress and affect the feto-maternal interface, including the pregnancy-protective cytokine balance and cytokines act on the brain and pituitary to disrupt neuroendocrine secretion; e.g. mimicking infection with lipopolysaccharide (LPS) is a well-known stressor. Generally, stress robustly increases hormonal secretion from the hypothalamo-pituitary-adrenal (HPA) axis and the sympathetic system in pregnancy. However, the role of hormones such as glucocorticoids in stress-induced pregnancy failure is unclear. For example, evidence does not consistently show that elevated circulating glucocorticoids accompany pregnancy failure, so any role for them must depend upon local mechanisms in the feto-maternal interface such as the 11beta hydroxysteroid dehydrogenase enzymes that regulate local glucocorticoid concentration. In the absence of obvious maternal HPA axis signals mediating pregnancy failure, we investigated what other neuroendocrine signals link maternal stress perception with stress-induced peripheral effects using a pregnant mouse model. Various stressors (e.g. sound stress, hunger or immune stress) decrease secretion of key pregnancy hormones such as progesterone. One major neuroendocrine hormone system is involved in driving high progesterone secretion in rodent models and humans: anterior pituitary prolactin, whose secretion is elevated from early gestation. As well as enhancing progesterone secretion prolactin also directly facilitates implantation in the decidua and is an immune regulator. We have shown that unlike in virgins, stress strongly decreases prolactin secretion, so stress-induced lack of prolactin may compromise pregnancy maintenance. Indeed preliminary data in women indicates that low circulating prolactin correlates with pregnancy failure (Douglas 2010), and the literature shows that lack of prolactin or progesterone in early pregnancy leads to pregnancy failure, adverse fetal programming and suboptimal maternal behavior (Larsen and Grattan 2010; Pincus et al 2010). We have investigated hypothalamic control of prolactin secretion in early pregnancy. Dopamine is the main neuroendocrine hormone controlling pituitary lactotrophs, and typically strongly inhibits lactotroph prolactin secretion. In early pregnancy dopamine control of prolactin is attenuated, allowing the increasing basal prolactin secretion. However, when the mother is exposed to stress there is hyperactivity of the neuroendocrine tuberoinfundibular dopamine (TIDA) neurones located in the arcuate nucleus and increased dopamine release in the median eminence, explaining the stress-inhibited prolactin secretion. Evidence shows that this response is similar regardless of the stressor used and provides a direct explanation for how stress can rapidly elicit detrimental effects on the important maternal hormonal mechanisms sustaining pregnancy. Other hypothalamic neuroendocrine neurones additionally respond to the same stressors and some of these neuronal populations also control pituitary prolactin secretion; for example oxytocin and thyroid releasing hormone from parvocellular paraventricular nucleus neurones are prolactin releasing factors. In pregnancy responses of neurones in this region to some stressors (such as peripheral injection of LPS) are attenuated, suggesting reduced drive of the prolactin releasing factors in addition to the increased dopaminergic inhibition. LPS induces increased circulating cytokines which signal to the hypothalamus by driving prostaglandin secretion from blood vessel walls in the brain ependyma and activate brainstem neurones. The responses of neurones in the afferent stress-responsive noradrenergic brainstem pathways (such as the nucleus tractus solitarius) that project to the arcuate nucleus are enhanced in pregnancy, perhaps contributing to hyperactivation of the TIDA neurones. Therefore, there are multi-level adaptations in the maternal brain that exhibit altered responsiveness to stress that may result in attenuated pregnancy hormone secretion. In conclusion, maternal neuroendocrine hormone systems play a prominent role in mediating the adverse consequences that occur when a pregnant mother is exposed to stress. The resulting lack of key pregnancy hormones such as prolactin and progesterone in the susceptible peri-implantation period inevitably causes an inhospitable intrauterine environment that risks ongoing pregnancy maintenance.