Understanding Coronavirus-Related Delirium: Neurovirulent Cytokines and Their Impact on Neuronal Excitability

ABSTRACT

Rationale Delirium affects a majority of critically-ill patients, increasing mortality and dementia risk. The absence of effective therapy reversing neuronal changes in delirium emphasizes the need for greater understanding of delirium pathophysiology. Neuroinflammation represents a common pathway through which delirium-triggering diseases act. Glial cells sense systemic inflammation across the blood-brain barrier and become activated, releasing cytokines within the brain. In one disease model, systemic infection with neurovirulent coronaviruses caused delirium and specifically increased levels of TNF-α, IL-1β, IL-6, IL-12 and IL-15 in the mouse brain. Methods Here, we tested how neuronal function was affected in a coronavirus-induced neurovirulent cytokine (NVC) model of delirium. Using whole-cell patch clamp methods, we examined how single neuron excitability in murine primary neocortical cultures was impacted by 22-28 hour incubation in NVC. Results NVC treatment depolarized the resting membrane potential (RMP) compared to control (-65 ± 1.6 mV versus -73 ± 1 mV;P < 0.0001, n = 37 and 31 respectively) without affecting action potential characteristics. Delirium is often diagnosed due to altered responses to external stimuli. NVC exposure altered the sensitivity of neurons to changes in external Ca2+ and Mg2+ from physiological (1.1 mM, T1.1) to low (0.2 mM, T0.2) levels. The frequency of spontaneous firing was substantially increased following T0.2 application in control but not in NVC-treated neurons (p=0.026, ANOVA, control 0.02 ± 0.01 Hz to 2.1 ± 1.2 Hz, n=16, p=0.046;NVC 0.5 ± 0.4 Hz to 0.9 ± 0.3 Hz, n = 15, p=0.16). Consistent with this, evoked spiking following current injection was also observed in control but not NVC-treated neurons following the switch from T1.1 to T0.2 (P = 0.006, 2WRM ANOVA, Control 3.9 ± 1.2 Hz vs. 8.5 ± 1.3 Hz n=37, p<0.0001;NVC 4.8 ± 1.3 Hz vs. 4.5 ± 1.0 Hz, n=31, p=0.78). The attenuated excitability observed in NVC-treated neurons was reversed by hyperpolarization of the RMP. Evoked firing was substantially improved in NVC-treated cells after correcting the RMP (p =0.049, ANOVA, Control 7.5 ± 1.8 Hz vs. 10 ± 2.5 Hz, n=15, p=0.25;NVC 6.7 ± 2.2 Hz vs. 11.7 ± 2.8 Hz, n=16 p=0.01). Conclusion Our studies indicate that NVC-treated neurons have attenuated sensitivity to microenvironment changes. As these changes are reversible by correction of the RMP, further characterization of the underlying pathophysiological mechanism is essential to identify biologically plausible targets for delirium.