Complex social connections are essential for health and survival, and memory-impacting disorders like schizophrenia and Alzheimer’s disease can be debilitating for the relationships between patients and loved ones. To form and sustain relationships requires the ability to, first, identify strangers versus familiar individuals (identification) and, second, revise one’s representations of them based on past experience (learning). This ability is called social memory.
A range of evidence confirms that the CA2 subregion of the hippocampus is crucial for social memory, and CA2-specific abnormalities are linked to social memory deficits in disease mouse models. However, the specific social cues that CA2 processes to inform social memory—as well as how CA2 adapts its responses to representations of other individuals through learning and experience—remains unclear.Since mice rely most heavily on olfaction to investigate conspecifics, odor sensory cues likely inform the basis of social identification processes in the murine brain. Furthermore, the hippocampus receives information from the olfactory bulb through the entorhinal cortex, suggesting that CA2 may be capable of processing odor sensory information for memory storage. It is already known the neighboring hippocampal subregion CA1 processes nonsocial odor cues and encodes the relationship between nonsocial odors and positive valence through learned experience. Therefore, since CA2 is necessary for social recognition overall, and since it is possible CA2 receives odor information through the same circuits as CA1, I hypothesized that CA2 processes social odor cues for social identification and combines this information with contextual information to develop and maintain social memory.
In my thesis, I used two-photon calcium imaging to confirm that CA2 indeed encodes and distinguishes social odors belonging to unique individuals, as well as nonsocial odors. I also found that CA2 neurons adapt their responses to odor stimuli when a reward contingency is introduced—pairing some odors and not others with an artificial reward. Intensive decoding analyses further revealed that CA2 is capable of forming a generalized or abstract representation of social versus nonsocial and rewarded versus unrewarded social odor stimuli. Finally, with archaerhodopsin-mediated CA2 silencing, I confirmed that CA2 is necessary for social—but not nonsocial—odor-reward associative learning, further promoting the specificity of this brain region in the encoding of socially-relevant episodic memory.
A link exists between CA2-specific dysfunction (namely, poor CA2 neuronal excitability) and social recognition deficits in the Df(16)A+/- microdeletion mouse model of the human 22q11.2 Deletion Syndrome—in which nearly a third of patients develop schizophrenia. I next hypothesized that CA2 in this model has a deficit in processing social sensory cues and forming the appropriate association between those cues and learned valence. Indeed, I discovered behavioral deficits in both social and nonsocial odor-reward associative learning in the Df(16)A+/- model. I further showed that CA2 is important in this impairment because selective expression of a dominant negative TREK-1 potassium channel subunit, which has been shown to improve CA2 function in these mice, rescued the deficits in social and nonsocial odor-reward learning.
With two-photon imaging, I found that CA2 neurons in Df(16)A+/- mice were able to discriminate between social and nonsocial odors with an accuracy that was similar to that seen in wild-type mice, which was surprising given the CA2-dependent deficit in odor-reward learning in the Df(16)A+/- mice. However, the Df(16)A+/- mice did show a reduced fraction of neurons that were selectively activated by the rewarded odor compared to the wild-type mice. Perhaps the most salient finding is that CA2 representations in Df(16)A+/- mice showed a reduced generalized or abstract coding of odor-reward across the social and nonsocial odor categories. This suggests that the Df(16)A+/- mice failed to generalize the task variable of reward, but rather learned separate rules for social and nonsocial odor-reward association. This is reminiscent of a reduction in abstract thought in individuals with schizophrenia.
Overall, my thesis provides evidence for the first time that CA2 encodes social odors and odor-reward learned experiences, that these identification and learning-related adaptation mechanisms are impaired in a disease model harboring social memory deficits, and that specific manipulations to restore CA2 function can rescue abnormal learning in this model. These results reinforce the notion that CA2 may provide a novel target for therapeutic intervention in restoring cognitive function associated with neuropsychiatric disease.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/qbhk-zq71 |
| Date | January 2024 |
| Creators | Bigler, Shivani Karen |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0015 seconds