Thyrotropin-releasing hormone (TRH) was first described for its neuroendocrine role in controlling the hypothalamus-pituitary-thyroid axis (HPT). Anatomical and pharmacological data evidence its participation as a neuromodulator in the central nervous system. Administration of TRH induces various behavioural effects including arousal, locomotion, analepsy, and in certain paradigms, it reduces fear behaviours. In this work we studied the possible involvement of TRHergic neurons in anxiety tests. We first tested whether an ICV injection of TRH had behavioural effects on anxiety in the defensive burying test (DBT). Corticosterone serum levels were quantified to evaluate the stress response and, the activity of the HPT axis to distinguish the endocrine response of TRH injection. Compared to a saline injection, TRH reduced cumulative burying, and decreased serum corticosterone levels, supporting anxiolytic-like effects of TRH administration. The response of TRH neurons was evaluated in brain regions involved in the stress circuitry of animals submitted to the DBT and to the elevated plus maze (EPM), tests that allow to correlate biochemical parameters with anxiety-like behaviour. In the DBT, the response of Wistar rats was compared with that of the stress-hypersensitive Wistar Kyoto (WKY) strain. Behavioural parameters were analysed in recorded videos. Animals were sacrificed 30 or 60min after test completion. In various limbic areas, the relative mRNA levels of TRH, its receptors TRH-R1 and -R2, and its inactivating ectoenzyme pyroglutamyl peptidase II (PPII), were determined by RT-PCR, TRH tissue content by radioimmunoassay (RIA). The extent of the stress response was evaluated by measuring the expression profile of CRH, CRH-R1 and GR mRNA in the paraventricular nucleus (PVN) of the hypothalamus and in amygdala, corticosterone levels in serum. As these tests demand increased physical activity, the response of the HPT axis was also evaluated. Both tasks increased the levels of serum corticosterone. WKY rats showed higher anxiety-like behaviour in the DBT than Wistar, as well as increased PVN mRNA levels of CRH and GR. TRH mRNA levels increased in the PVN and TSH values remained unchanged in both strains although TRH content decreased in the medial basal hypothalamus of Wistar rats only. TRH content was measured in several limbic regions but only amygdala showed specific task-related changes after DBT exposure of both strains: increased TRH content. Expression of TRH mRNA decreased in the amygdala of Wistar, suggesting inhibition of TRHergic neuronal activity in this region. The participation of amygdalar TRH neurons in anxiety was confirmed in the EPM where TRH expression and release correlated with the number of entries, and the % of time spent in open arms, supporting an anxiolytic role of these TRH-neurons. These results contribute to the understanding of the involvement of TRH during emotionally charged situations and shed light on the participation of particular circuits in related behaviours.