How are circadian rhythms of orexin and serotonin affected by the dorsal raphe nucleus and lateral hypothalamus area interactions?

Circadian rhythm abnormalities, sleep and mood disorders are commonly observed features in depression [1]. Sleep disorders are known to be associated with abnormal orexin levels (Ox) in the lateral hypothalamus (LHA) area of the brain while mood disorders are associated with abnormal serotonin (5-HT) levels in the dorsal raphe nucleus (DRN). There are studies that suggest that feedback loop circuitry between LHA and DRN plays an important role in the pathogenesis of sleep disorder and depression (e.g. [2]). There are also anatomical evidences that suggest that both these areas receive the circadian inputs from the central pacemaker, suprachiasmatic nucleus (SCN) through indirect pathways [3], [4]. In our previous work the interaction between LHA and DRN has been modelled and studied [5]. However, it remains unknown how the circadian rhythms of 5-HT and Ox levels are affected by the LHA-DRN interactions.

In this work, we built on our previous nonlinear firing-rate model of the coupled LHA-DRN circuit [5] by incorporating oscillatory SCN inputs to both the DRN and LHA (Figure 1). We then investigate how the rhythms of the 5-HT and Ox levels and their phase difference (ΔФ_out) can be altered by the model parameters (Ox decay rate and rise factor, slope and shift factor in LHA input-output function and amplitude values of the two rhythms). We first set our model parameters such that the basal firing rates and concentration levels are similar to those found in experiments. Assuming, for simplicity, no initial phase difference from the SCN inputs to DRN and LHA (ΔФ_in= 0), the model provided the phase difference of approximately (ΔФ_out ~ 1 minute). This value is largely controlled by the slowest timescale in our model i.e. the effective change in Ox neural firing rate due to 5-HT level (as τ_Ox ~ 1 minute). Increasing the decay rate or the production (rise factor) of Ox generally increases and decreases ΔФ_out, respectively.  If decay rate is low, both Ox and 5-HT rhythms have an in-phase relationship. Increasing the decay rate to intermediate values can result in the coexistence of in-phase and anti-phase within the Ox rhythm. Further increase will eventually allow only an anti-phase relationship. Similar trend is observed for the lower values of the rise factor. However, for the higher values both the rhythms maintain the in-phase relationship. Interestingly, as we increase the slope of the Ox firing rate vs 5-HT level (input-output) function, we found a maximum ΔФ_out value (~ 2 minute). Increasing the amplitudes of SCN input to DRN first decreases the phase difference ΔФ_out (for the lower values) and for the intermediate and higher values of amplitude the ΔФ_out increases. However, increasing the amplitudes of SCN input to LHA decreases the ΔФ_out. Interestingly for the lower values of amplitude of Ox rhythms we found similar trends from in-phase to anti-phase (as described for increasing decay rate). But no such behaviour has been observed if only the SCN-DRN input amplitude is varied.  Finally, we incorporate the drug effects of 5-HT antagonists by shifting the respective input-output functions (assuming that there is no change in the release dynamics of 5-HT at LHA). We found that larger shifts generally lead to an anti-phase relationship between 5-HT and Ox rhythms. 

Our modelling results suggest that reciprocal connections between the DRN and LHA, mediated by 5-HT and Ox, can result in significant changes in their circadian rhythms, which in turn can disrupt the temporal relationships among sleep, mood, and other important physiological properties. The model predicts interesting rhythmic phenomena that can be tested in experiments, e.g. by perturbing the circuit with an agonist/antagonist while simultaneously recording Ox and 5-HT levels.

[1] A. Germain and D. J. Kupfer, "Circadian rhythm disturbances in depression," Human Psychopharmacology, vol. 23, pp. 571, 2008.

[2] P. Feng, D. Vurbic, Z. Wu, Y. Hu and K. P. Strohl, "Changes in brain orexin levels in a rat model of depression induced by neonatal administration of clomipramine," Journal of Psychopharmacology, vol. 22, pp. 784, 2008.

[3] S. Deurveilher and K. Semba, "Reciprocal connections between the suprachiasmatic nucleus and the midbrain raphe nuclei: A putative role in the circadian control of behavioral states," Serotonin and Sleep: Molecular, Functional and Clinical Aspects, pp. 103-131.

[4] N. Tsujino and T. Sakurai, "Orexin/hypocretin: a neuropeptide at the interface of sleep, energy homeostasis, and reward system," Pharmacol. Rev., vol. 61, pp. 162, 2009.

[5] Joshi, A., Wong-Lin, K.F., McGinnity, T.M., and Prasad, G. "A mathematical model to explore the interdependence between the serotonin and orexin/hypocretin systems" (Submitted).

This work was supported under the CNRT award by the Northern Ireland Department for Employment and Learning through its “Strengthening the All-Island Research Base” initiative.