Distinct representation of cue-outcome association by D1 and D2 neurons in the ventral striatum’s olfactory tubercle

Nuné Martiros, Vikrant Kapoor, Spencer E. Kim, Venkatesh N. Murthy

TL;DR

This study reveals that two types of neurons in the olfactory tubercle (OT), D1 and D2, play distinct roles in learning and representing the value (valence) of odors. D1 neurons are crucial for encoding odor valence in a robust and persistent manner, even without behavioral responses, while D2 neurons are more involved in identifying the odor itself.

What Problem Does This Paper Address?

The brain needs to assign value to sensory experiences to guide behavior. Olfactory cues are particularly potent in forming strong, long-lasting associations, and the olfactory tubercle (OT) is a key brain region involved in this process due to its direct input from the olfactory bulb and dense dopaminergic connections. However, it was unclear how the OT transforms raw odor information into a representation of its associated value (valence) and what role the different types of neurons within the OT play. Specifically, the paper investigates the differential contributions of D1 and D2 dopamine receptor-expressing neurons in the OT to learning and representing odor-outcome associations.

Methods

The researchers used two-photon calcium imaging in awake, behaving mice to monitor the activity of specific neuronal populations in the OT. They genetically engineered mice to express a calcium indicator (GCaMP7s) in either D1-expressing neurons (using Drd1-Cre mice) or D2-expressing neurons (using Adora2A-Cre mice). These mice were trained on a task where different odors were paired with either aversive (airpuff) or rewarding (water) outcomes. To isolate the neural representation of valence from behavioral responses, they conducted probe sessions where mice were sated and outcomes were omitted, allowing them to observe neuronal activity in the absence of licking or reward. They also introduced a multimodal task where sounds were paired with outcomes to investigate if valence coding in the OT was modality-specific.

Datasets

Not explicitly discussed as separate datasets for training or evaluation, but the study generated novel in vivo calcium imaging data from D1 and D2 neurons in the olfactory tubercle of mice during odor-outcome learning. This data was collected across multiple training days and under different behavioral conditions (licking vs. no licking, presence vs. absence of outcomes).

Key Insights and Ideas

  • D1 neurons are the primary valence coders: The study provides strong evidence that D1 neurons in the OT are the main players in representing the positive or negative value of an odor, regardless of the odor's specific identity.
  • Valence representation is persistent and independent of behavior: Crucially, D1 neurons continued to encode odor valence even when the mice were sated and not licking, and the outcomes were removed. This suggests that the OT transforms sensory input into a valence representation that precedes and is independent of motor output.
  • D2 neurons are more identity-focused: In contrast to D1 neurons, D2 neurons showed a stronger tendency to represent the identity of the odor rather than its valence. Their valence coding was weaker and more dependent on the presence of behavioral responses.
  • Valence coding is modality-specific in the OT: When sounds were introduced alongside odors, D1 neurons showed distinct populations responding to odors and sounds predicting the same outcome. This indicates that the OT itself doesn't integrate multimodal valence information, but rather processes olfactory valence in a specialized manner.
  • Learning refines neuronal representations: As mice learned the associations, D1 neurons became better at encoding valence, while D2 neurons became more selective for odor identity.

Key Results

  • D1 neurons robustly and bidirectionally represented odor valence, with similar responses to odors predicting similar outcomes.
  • This valence representation in D1 neurons persisted even in the absence of licking responses and outcomes.
  • D2 neurons showed weaker valence representation, which was contingent on licking, and were more selective for odor identity.
  • In a multimodal task, D1 neurons showed distinct populations responding to odors and sounds predicting the same outcome, indicating modality-specificity.
  • With training, D1 neuron valence coding scores increased, while D2 neuron identity coding scores increased.

Limitations and Future Work

The authors acknowledge that the GRIN lens implantation, while minimized, can cause tissue damage, though they argue their key conclusions are not affected. They also note that while D1 neurons showed strong valence coding, a sizable fraction did lose this coding in the absence of licking and outcomes, suggesting heterogeneity within the D1 population and potential intertwining of valence and licking factors. Future work could explore the downstream targets of D1 and D2 neurons in the OT to understand how this distinct coding influences behavior and to further investigate the role of dopamine in shaping these representations. The modality-specific nature of valence coding in the OT also opens avenues for exploring how multimodal valence integration occurs in other brain regions.

Summary updated

Abstract

Positive and negative associations acquired through olfactory experience are thought to be especially strong and long-lasting. The conserved direct olfactory sensory input to the ventral striatal olfactory tubercle (OT) and its convergence with dense dopaminergic input to the OT could underlie this privileged form of associative memory, but how this process occurs is not well understood. We imaged the activity of the two canonical types of striatal neurons, expressing D1- or D2-type dopamine receptors, in the OT at cellular resolution while mice learned odor-outcome associations ranging from aversive to rewarding. D1 and D2 neurons both responded to rewarding and aversive odors. D1 neurons in the OT robustly and bidirectionally represented odor valence, responding similarly to odors predicting similar outcomes regardless of odor identity. This valence representation persisted even in the absence of a licking response to the odors and in the absence of the outcomes, indicating a true transformation of odor sensory information by D1 OT neurons. In contrast, D2 neuronal representation of the odor-outcome associations was weaker, contingent on a licking response by the mouse, and D2 neurons were more selective for odor identity than valence. Stimulus valence coding in the OT was modality-sensitive, with separate sets of D1 neurons responding to odors and sounds predicting the same outcomes, suggesting that integration of multimodal valence information happens downstream of the OT. Our results point to distinct representation of identity and valence of odor stimuli by D1 and D2 neurons in the OT.
BibTeX
@inproceedings{martiros2022distinct,
  author = {Nuné Martiros and Vikrant Kapoor and Spencer E. Kim and Venkatesh N. Murthy},
  title = {{Distinct representation of cue-outcome association by D1 and D2 neurons in the ventral striatum’s olfactory tubercle}},
  booktitle = {eLife},
  year = {2022},
  doi = {10.7554/elife.75463},
}
Figure