Neural mechanisms underlying olfactory dysfunctions in a COVID-19 mouse model

Abstract

One of the entry points of the Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) is the nasal epithelium. The epithelium harbours the two important cell types, one which can detect the odorants in the environment (called the olfactory sensory neurons) and other, the olfactory ensheathing cells (sustentacular cells) which monitor and regulate the health status of the sensory neurons. The viral entry happens via the ACE2 receptor and facilitated by the TMPRSS2 enzyme present in the sustentacular cells. Viral invasion subsequently makes the surface ACE2 receptor functionally inactive. The virus can then traverse to the olfactory bulb and other regions causing maladaptive changes in the brain. To study the effect of viral mediated ACE2 functional depletion, a strategy of creating ACE2 knockout could be of interest. Upon successful creation of ACE2 knockout, we carried out behavioural characterization of olfactory impairments as a first step towards understanding the neural mechanisms underlying the olfactory deficits in this COVID-19 mouse model. Firstly, using CT scan technique, we found out the ACE2 KO mice have air-way area defects as was quantified by using the machine learning algorithm on the nasal cavity contours. Qualitative analysis of the neuronal projections using Microtubule associated protein-2 markers in the glomerular layer of the olfactory bulb show lesser neuronal projection patterns in ACE2 KO mice. As we found the molecular and structural changes, we then systematically investigated the olfactory detection and discrimination capabilities of these mice. Both the olfactory and pheromonal detection was normal in ACE2 KO mice. However, the higher olfactory related cognitive skills were negatively affected in the ACE2 KO mice. When we carried out the odor discrimination threshold task to find out if ACE2 mice could learn to distinguish the odors at the same concentration as the controls, we found out that they carried the task with poor learning efficiencies. The discrimination time was also slower in these mice. As for the task pertaining to multimodal pheromonal learning paradigm, we found out that the ACE2 KO mice could not form pheromonal location memory. Our study thus provides a first step in determining the deficits and allowing for assessing circuitry changes in a COVID-19 mouse model.