Insights into the molecular triggers of parosmia based on gas chromatography olfactometry

Abstract

Background: Parosmia is a debilitating condition in which familiar smells become distorted and disgusting, with consequences for diet and mental health. It is a feature of post-infectious olfactory loss, particularly resulting from COVID-19. There is currently little understanding of its pathophysiology, and the prevailing hypothesis for the underlying mechanism is aberrant growth of regenerating olfactory sensory neurons after damage.

Methods: We use gas-chromatograph olfactometry to individually present components of a complex olfactory mixture as a rapid screening tool for assessment of both quantitative and qualitative olfactory dysfunction in those with and without parosmia. This allows them to report the associated sensory effects and to identify those molecules which are altered or parosmic in nature.

Results: Here we show 15 different molecular triggers of this symptom. These trigger molecules are common to many in the parosmic volunteer group and share certain characteristics such as extremely low olfactory threshold and common molecular structure.

Conclusions: We posit that specific highly odour-active molecules are the cause of the parosmic symptom in most cases and initiate the sense of disgust, suggesting that parosmia is, at least in part, a receptor-level phenomenon.

Keywords: Brain injuries; Central nervous system infections; Chemical biology; Health sciences; Olfactory receptors.

Fig. 1. Correlations between olfactory function and GC-O.

a Correlation between TDI score (Threshold, Discrimination, and Identification Sniffin’ Sticks test score) and number of aromas detected at the GC odour-port. b Relationship between number of triggers detected in the coffee extract and TDI score. In both figures, non-parosmic participants = blue, pre-COVID-19 parosmic participants = orange, post-COVID-19 parosmic participants = green. Source data in Supplementary Data 2.

Fig. 2. Structures of the most frequently detected compounds.

The most common trigger molecules are grouped into four distinct categories based on structure: thiols, trisubstituted pyrazines, methoxypyrazines, and disulfides; although some less common triggers did not fall into any one of these categories.

Fig. 3. Principal component plot (PC1 vs. PC2) for intensity of 20 most frequently detected compounds.

Principal component analysis demonstrating the clustering of the various compounds. Source data are provided in Supplementary Information Table 1.