Boosting a bad signal

Dopamine (DA) has been suggested to increase the signal-to-noise ratio (SNR) of neural activity in the medial prefrontal cortex (mPFC); however, direct evidence of this model is lacking. Vander Weele et al. now show in rodents that DA released in the mPFC increases the SNR of aversive signals to the dorsal periaqueductal grey (dPAG).

The authors first investigated the activity of ventral tegmental area (VTA) neurons that release dopamine in the mPFC (VTA^DA→mPFC neurons), using fast-scan cyclic voltammetry and optogenetics. Photoinhibition of these neurons’ terminals in layers 5 and 6 of the rat mPFC reduced their release of DA in response to an aversive stimulus (pinch). Photoactivation of VTA^DA→mPFC terminals was not aversive (as assessed in real time or conditioned place avoidance assays). However, in rats that were conditioned to associate auditory or visual cues with an aversive stimulus (a shock) or a reward (sucrose) and that were then presented with both cues simultaneously, photoactivation of VTA^DA→mPFC terminals induced more freezing and less reward-approaching behaviour. Thus, in the presence of conflicting cues, DA in the mPFC may bias responses towards aversion.

Credit: Jennie Vallis/Springer Nature Limited

mPFC neurons are known to project to other structures in the brain, including the PAG. Selective activation of either the somata or terminals of mPFC→PAG neurons in rats led to avoidance in the place aversion assays and increased marble burying and time spent digging (behaviours that are thought to be defensive). Calcium imaging of mPFC→PAG neurons in mice revealed that a greater proportion of this cell population responded to shock than to sucrose, consistent with a role for these neurons in aversion signalling.

In mouse brain slices, optogenetic stimulation of VTA^DA→mPFC neurons did not affect the excitability of mPFC→dPAG neurons, and retrograde labelling showed that mPFC→dPAG neurons do not express DA receptors. Thus, the authors reasoned that the VTA^DA→mPFC projections may not directly excite mPFC→dPAG neurons but may instead increase the SNR of incoming sensory inputs relating to aversive stimuli. Consistent with this model, 10 minutes of VTA^DA→mPFC stimulation in mice in vivo reduced and increased the frequency and amplitude of calcium events in mPFC→dPAG neurons, respectively. Electrophysiological recordings revealed that although this VTA^DA→mPFC stimulation did not affect the firing rates of mPFC→dPAG neurons under basal conditions or in response to sucrose, it did increase mPFC→dPAG firing frequency in response to an aversive airpuff.

photoactivation of VTA^DA→mPFC terminals induced more freezing and less reward-approaching behaviour

Together, these results imply that DA released in the mPFC increases the SNR of information transmitted to the dPAG.