Commentary on “A Cortical Circuit for Sexually Dimorphic Oxytocin-Dependent Anxiety Behaviors” (Li et al., 2016)

NEUR 315

Heankel Oliveros

Screen Shot 2017-12-20 at 10.48.42 PM
Article: Li, K., M. Nakajima, I. Ibanez-Tallon and N. Heintz (2016). “A Cortical Circuit for Sexually Dimorphic Oxytocin-Dependent Anxiety Behaviors.” Cell 167(1): 60-72.e11.

The authors of this paper investigated a group of cortical neurons that modulate both social and emotional states differently in male and female mice. Previously, these researchers identified oxytocin receptor interneurons (OXtrINs) in the medial prefrontal cortex (mPFC) and discovered that these neurons facilitate social behavior in females during their estrus phase by acting on layer 5 pyramidal neurons. In contrast, despite males and females having an equivalent number and distribution of OXtrINs, the photostimulation of OxtrINs or administration of OXT in the mPFC had no effect on the social behavior of male mice.

In this study, the researchers focused on the regulation of anxiety-related behaviors by OXtrINs in males and females. They approached this question by photostimulating OXtrINs expressing channelrhodopsin. They transfected these neurons by doing a bilateral stereotactic injection of Cre-dependent AAV virus encoding channelrhodopsin in the mPFC of Oxtr-Cre mice. Then these mice were tested in three behavioral tasks. The first one, the three-chamber social interaction test, examined the preference of animals to spend time with a novel object or a new mouse. The other two tests, the open field (OF) and the elevated plus maze (EPM), aimed to assay the anxiety of mice. Consistent with previous experiments, females displayed social preference when OXtrINs were activated using blue light. In contrast, the stimulation of OXtrINs had no changes in social preference in male mice. On the other hand, when OXtrINs were stimulated during the EPM and OF tests, males showed a decrease in anxiety-related behavior. That is, they spent more time in the center arena of the OF and explored open arms more frequently. Females, however, did not display any changes in behavior upon OXtrINs activation.

Once the authors confirmed that OXtrINs were candidates for the modulation of anxiety-related behaviors in male mice, they proceeded to determine whether the anxiolytic effect of these neurons required OXT/OXTR signaling. To test the necessity of oxytocin signaling, the authors injected an AAV-expressing Cre to delete the OXT receptor gene (Oxtr). The controls were injected with AAV-expressing GFP. After behavioral testing, the authors observed a strong anxiogenic effect in male mice whereas controls and females did not show any changes in anxiety-related behavior.

In the second half of the article, the authors investigated the pathway by which OXtrINs activation modulates local neurons to produce anxiolytic effects. They approached this question by doing whole-cell recordings of neurons in layers 2/3 and 5 during photostimulation of OXtrINs using optogenetics. The authors discovered that males had a larger inhibitory postsynaptic current (IPSCs) in 2/3 pyramidal neurons, whereas females showed larger IPSCs in layer 5 neurons. In terms of excitatory postsynaptic currents (EPSCs), neurons in layer 2/3 were similar for both males and females, whereas layer 5 neurons displayed larger amplitudes for females. The conclusion from this experiment was that the activation of GABAergic OXtrINs results in different inhibitory effects in the local circuit activity of females and males.

Next, the authors gained more insight into the signaling pathways in these circuits by examining the translated mRNAs in OXtrINs. To analyze OXtrINs-specific mRNAs, they used a technique called TRAP (Translating Ribosome Affinity Purification). In short, TRAP works by isolating EGFP-tagged ribosomes and examining the mRNA in them. TRAP localized cell-specific ribosomes by expressing EGFP under a conditional promoter in Oxtr-cre mice. After sequencing the obtained RNA, the authors focused on the top ten most highly enriched translated mRNAs in OXtrINs. From these top candidates, the researchers decided to focus on corticotropin-releasing factor-binding protein (CRHBP) because it binds to corticotropin-releasing hormone (CRH) with high affinity and inactivates it action on CRH receptors.

After discovering a high expression of Crhbp in OXtrINs, the authors inferred that the anxiogenic action of CRH on the mPFC could be regulated by the production of CRHBP when OXtrINs are activated in males. To test this hypothesis, the authors performed electrophysiological recordings in layer 2/3 pyramidal neurons and applied a CRH bath to female and male slices. The results showed that male slices treated with CRH had an increase in action potentials. These CRH-generated spikes were suppressed by co-application of CRHR1 antagonist. In contrast, the spiking activity of neurons in female slices had a very small increase that was insensitive to CRHR1 antagonist. In addition, layer 5 neurons were insensitive to CRH treatment in both males and females. This experiment demonstrated that male layer 2/3 neurons were more sensitive to CRH compared to females and that the activation of Crhr1 receptor was responsible for this neural activity. Next, the authors performed electrophysiology again from layer 2/3 neurons, but this time they stimulated OXtrINs using optogenetics. The recordings showed that the CRH-induced activity of male slices strongly decreased during OXtrINs photostimulation. Taken together, these results demonstrated that the activation of OXtrINs may decrease the response of layer 2/3 neurons to CRH.

Finally, the authors used a conditional knockdown of crhbp in OXtrINs by injecting a lentiviral construct to express shRNAs for Crhbp in Cre-positive OXtrINs only. They had controls expressing EGFP instead of the shRNA cassette. Then the animals were tested in the OF and EPM tests. The results demonstrated that anxiety-like behaviors were not affected in transfected female mice whereas males did show an increase in anxiety-like behavior compared to controls. In addition, female sociosexual behavior was not changed by this shRNAs knockdown. These results suggest that CRHBP synthesis from OXtrINs regulates anxiety in males only. Considering their electrophysiological recordings, the anxiogenic action of CRH in layer 2/3 neurons is likely to be decreased by the production of CRHBP from OXtrINs in male mice. The authors conclude by suggesting that the higher CRH levels in females compared to males might be responsible for the insensitivity of these cortical neurons to the production of CRHBP from OXtrINs.

In general, I think that this article includes good controls and behavioral tests that support their cortical model for oxytocin-dependent anxiety behaviors. For instance, they tested females during different phases of their estrous cycle to make sure that the anxiety-related behaviors regulated by this cortical circuitry were not dependent on other hormonal levels. In addition, they used two tests for anxiety-related behaviors to support their conclusions and they also examined the sociosexual behavior of animals whenever they did a genetic manipulation.  Moreover, their CRH hypothesis is supported by their experiments (electrophysiology and TRAP/RT-PCR) because they showed that 1) CRH did not change the neural activity of pyramidal neurons in the mPFC of females, and 2) CRH levels are higher in females. The only limitation that I see in this study is that the anxiolytic effects of this circuitry were not tested in fear conditioning. One could argue that the findings in this study are not related to anxiety, but rather to an impairment in risk assessment, so the animals are more likely to explore the open arms and the center of the open field because they do not process the risk that such actions imply. In fact, the mPFC is also involved in risk assessment (Xue et al., 2009). By testing this circuitry in fear conditioning, we could eliminate the confounding variable of risk assessment because animals have already learned to fear a stimulus. However, the involvement of CRH in this circuit is a strong indicator that this is indeed an anxiety-related neural circuit.

In future experiments, the authors could investigate how the levels of CRH might mediate sex differences in this circuit. One approach would be to regulate CRH levels by targeting Crh in the paraventricular hypothalamus, which was reported to have a very high expression of CRH in females. The authors could perform a conditional knockdown of this gene by using a similar siRNA technique as they did on this paper. They could put their CRH-siRNA under a tissue-specific promoter for the targeted cells in the hypothalamus. They could test females (OF and EPM) early in development and in adulthood in order to see if there is a critical period for the formation of this sexually dimorphic anxiolytic circuit. This type of experiment would allow us to explore how CRHBP is able to have different regulatory actions on cortical circuits that do not have any evident differences in neuroanatomy and mRNA profiles related to CRH.



Li, K., M. Nakajima, I. Ibanez-Tallon and N. Heintz (2016). “A Cortical Circuit for Sexually Dimorphic Oxytocin-Dependent Anxiety Behaviors.” Cell 167(1): 60-72.e11.

Xue G, Lu Z, Levin IP, Weller JA, Li X, Bechara A (2009) Functional dissociations of risk and reward processing in the medial prefrontal cortex. Cereb Cortex 19:1019-1027.