To study the role of the glutamatergic system in alcohol abuse disorders, we conducted a longitudinal study of alcohol exposure in rats. Rats were followed with 1H-magnetic resonance spectroscopy during a baseline, exposure and abstinence phase to measure prefrontal glutamate and glycine concentrations. Additionally, 18F-FPEB PET was used to assess metabotropic glutamate receptor 5 (mGluR5) availability before and after exposure to alcohol or saccharin. Finally, individual anxiety levels were assessed with an open field test at baseline. We found that the total distance traveled in the open field test and mGluR5 availability in the nucleus accumbens were associated with future alcohol preference (P= 0.02). Furthermore, prefrontal glutamate concentration decreased significantly during alcohol exposure (> �1. 25�0.55 mmol/L, P= 0.038), but normalized during abstinence. Finally, 18F-FPEB PET showed that selfadministration of alcohol resulted in decreased mGluR5 availability in the hippocampus (> 14�3 percent, pCluster= 0.047) and amygdala (> 16�2 percent, pCluster= 0.004), whereas saccharin induced decreases in the prefrontal cortex (> 11�1 percent, pCluster= 0.035) and hippocampus (> 15�2 percent, pCluster= 0.003). A direct comparison of both groups showed differences in mGluR5 availability in the bilateral striatum (�2�4 percent versus+ 2�2 percent, pCluster=< 0.0001) and hippocampus (�16�4 percent versus �5�4 percent, pCluster=< 0.0001). In conclusion, this study corroborates the anxiolytic effect of alcohol and provides evidence for mGluR5 downregulation in the amygdala as potential underlying mechanism. Saccharin and alcohol differentially affect prefrontal glutamate concentrations, but both induce an mGluR5 decrease, potentially to modulate the dopamine release elicited by these drugs.