In previous fMRI studies, researchers found that the medial prefrontal cortex shows deactivation during or after acupuncture stimulus1, but the underlying neurotransmitter mechanism is still unknown. There is increasing evidence that the most plausible mechanism for the BOLD activation is a neural network containing a mixture of interconnected pyramidal cells and GABAergic inhibitory interneurons. These mixed structures provide a balance of excitation-inhibition for the network, and the magnitude of the BOLD response is also sensitive to this excitation-inhibition balance. In this study, the MEGA-PRESS 1H-MRS sequence was conducted to measure GABA and Glutamine concentrations before and right after acupuncture stimulation. Additionally, the resulting concentrations were compared with BOLD response during acupuncture stimulation in normal subjects to investigate the neurotransmitter basis of acupuncture.10 healthy adult volunteers participated in the study. MR data were collected on a MAGNETOM Skyra 3T MR scanner (Siemens Healthcare, Erlangen, Germany) with 20-channel head-neck coil. The MRI exam included MPRAGE for localization and MEGA-PRESS sequence scan in the bilateral medial prefrontal before and right after acupuncture stimulus (Fig.1). The sequence parameters are as following: voxel size= 35x30x25mm3 (F>>H, R>>L, A>>P), TR=2000 ms, TE=68 ms, Averages=128, edit pulse frequency 1.9ppm, edit center frequency 4.7ppm. During acupuncture stimulus, BOLD signals were acquired with the following parameters: TR=750 ms, TE=30 ms, flip angle=90°, 43 slices, slice thickness=3 mm, distance factor=0%, FOV=210 ×210 mm2, matrix= 70×70. The acupuncture needle (Diameter 0.30mm, Huatuo, China) was inserted into Right LI4 (Point Hegu) and stimulus for 3 blocks. Each block included a 30-second task (needle rotating) period and followed with a 120-second interval (needle in place). The same licensed acupuncturist administered the acupuncture throughout the study. The MRS data were analyzed using LCModel. BOLD data was processed by SPM8 using standard task fMRI post-processing pipeline. The results of BOLD fMRI were shown in t-value (p < 0.001, cluster size>30). The activation was defined as average t-value within ROI >3 while deactivation is <-3.From the t-map shown in Figure 2, we found that the medial prefrontal cortex showed strong deactivation (t<-4) during stimulus in 7 of the 10 subjects. The GABA concentration was decreased in the same ROI (7/10). The other two of three subjects had increased GABA concentration in this area also showed activation during BOLD fMRI. Only one subject showed increased BOLD signal with decreased GABA concentration (Table 1, Fig.1). The correlation coefficient between the activation/deactivation of BOLD fMRI and pre-post GABA concentration differences is 0.75 (p=0.012);The medial prefrontal cortex is a key area for mood regulation, and also a part of the default mode network. It showed the predominant negative BOLD responses in acupuncture. Acupuncture is supposed to be able to induce inhibitiory effects in the region, as well as effectively used as a treatment for analgesia, mood modulation on insomnia, anxiety and depression. Although the study population was small, we could see a main trend of decreased GABA concentration associated with a negative BOLD response which may correspond with acupuncture’s neural inhibition effects in the brain region.Preliminary data suggests that the negative BOLD response in the medial prefrontal cortex induced by acupuncture may be associated with the neural inhibition effects. Further investigation with a larger patients population is needed.