Background: In recent years, deep brain stimulation (DBS) has emerged as a promising treatment option for patients suffering from treatment-resistant depression (TRD). Several stimulation targets have successfully been tested in clinical settings, including the subgenual cingulum (Cg25) and the medial forebrain bundle (MFB). MFB-DBS has led to remarkable results, surpassing the effect of previous targets in terms of response latency and number of responders. However, the question remains as to which mechanisms underlie this difference. Objective/hypothesis: The aim of the present study was to thoroughly study the anti-depressant effect of MFB-DBS in the Flinders sensitive line (FSL) rat model of depression as well as to investigate whether MFB-DBS and Cg25-DBS operate through the same neurobiological circuits. Methods: FSL and control rats received bilateral high-frequency stimulation to the MFB at the level of the lateral hypothalamus, while being subjected to a variety of depression-and anxiety-related behavioral paradigms. To further compare the effects of MFB-DBS and Cg25-DBS on reward-related behavior, animals were stimulated in either the MFB or ventromedial prefrontal cortex (vmPFC, rodent analog to Cg25), while being tested in the intra-cranial self-stimulation paradigm. Results: A marked symptom-specific anti-depressant effect of MFB-DBS was demonstrated. The ICSS-paradigm revealed that MFB-DBS, as opposed to vmPFC-DBS interacts with the reward system. Conclusion: Our data suggest that MFB-DBS and Cg25-DBS do not operate via the same neurobiological circuits. This differentiation might be of interest when selecting patients for either Cg25- or MFB-DBS. (C) 2015 Elsevier Inc. All rights reserved.