Computational exploration of molecular receptive fields in the olfactory bulb reveals a glomerulus-centric chemical map

Abstract

Progress in olfactory research is currently hampered by incomplete knowledge about chemical receptive ranges of primary receptors. Moreover, the chemical logic underlying the arrangement of computational units in the olfactory bulb has still not been resolved. We undertook a large-scale approach at characterising molecular receptive ranges (MRRs) of glomeruli in the dorsal olfactory bulb (dOB) innervated by the MOR18-2 olfactory receptor, also known as Olfr78, with human ortholog OR51E2. Guided by an iterative approach that combined biological screening and machine learning, we selected 214 odorants to characterise the response of MOR18-2 and its neighbouring glomeruli. We found that a combination of conventional physico-chemical and vibrational molecular descriptors performed best in predicting glomerular responses using nonlinear Support-Vector Regression. We also discovered several previously unknown odorants activating MOR18-2 glomeruli, and obtained detailed MRRs of MOR18-2 glomeruli and their neighbours. Our results confirm earlier findings that demonstrated tunotopy, that is, glomeruli with similar tuning curves tend to be located in spatial proximity in the dOB. In addition, our results indicate chemotopy, that is, a preference for glomeruli with similar physico-chemical MRR descriptions being located in spatial proximity. Together, these findings suggest the existence of a partial chemical map underlying glomerular arrangement in the dOB. Our methodology that combines machine learning and physiological measurements lights the way towards future high-throughput studies to deorphanise and characterise structure-activity relationships in olfaction.

Mapping odorant sensitivities reveals a sparse but structured representation of olfactory chemical space by sensory input to the mouse olfactory bulb

“…We first focused on the 19 functionally-identified glomeruli that were responsive to more than one odorant, sorting the median response spectra of each glomerulus according to odorant distance in physicochemical descriptor space, relative to the primary odorant for that glomerulus. We initially used a subset of 1377 descriptors from the E-Dragon web app (Todeschini and Consonni, 2003; Tetko et al, 2005) chosen previously from a large-scale characterization of mammalian OR binding properties (Haddad et al, 2008; Saito et al, 2009; Chae et al, 2019; Soelter et al, 2020; Gerkin, 2021). This descriptor set appeared moderately effective at predicting glomerulus co-tuning given the primary odorant identity, although some glomeruli were co-tuned to odorants distributed across large distances in the descriptor space, and all glomeruli failed to respond to numerous odorants located more closely within this space ( Fig.…”

“…Earlier studies have used sets of physicochemical descriptors of odorants to infer relationships between the chemical space of odorants and their neural representations, OR tuning, or odor perception (Haddad et al, 2008; Saito et al, 2009; Chae et al, 2019; Soelter et al, 2020; Gerkin, 2021). Given the exceptionally narrow tuning of glomerular inputs in our dataset, we used this approach to explore the chemical relationships between the small number of odorants to which OSN populations were most sensitive.…”

“…Mapping high-sensitivity odorants to individual glomeruli allowed for the straightforward functional identification of glomeruli across the dorsal OB using only a single diagnostic odorant at a low concentration. Previous studies have functionally identified glomeruli using a combination of location and response spectrum across multiple odorants (Wachowiak and Cohen, 2001; Soucy et al, 2009; Soelter et al, 2020), and have used these to link glomeruli to their cognate ORs (Oka et al, 2006; Shirasu et al, 2014). However, broader use of functionally-identified glomeruli has been limited thus far.…”

“…A related confound is the strong dependence of OSN specificity on odorant concentration and a lack of consensus on meaningful concentration ranges at which to characterize odor coding strategies (Meister and Bonhoeffer, 2001; Wachowiak and Cohen, 2001). With few exceptions (Si et al, 2019) previous studies have characterized OSN or glomerular responses to odorants presented at one or a few concentrations, typically far above threshold for evoking neural activity (Wachowiak and Cohen, 2001; Nara et al, 2011; Ma et al, 2012; Chae et al, 2019; Pashkovski et al, 2020; Soelter et al, 2020). The consensus from these and other studies is that odorant identity is encoded by combinatorial patterns of OSN and glomerular activity; however, details of such a coding strategy, including the logic of OSN tuning properties, the nature of glomerular maps across the OB surface, and the dimensionality of odorant representations remain unclear.…”

“…This approach avoids confounds from arbitrarily-chosen stimulus intensities, facilitates comparison across levels and approaches, and can more clearly reveal organizational features such as topographic mapping across neural space and transformations by neural circuits (M H Goldstein et al, 1970; Stiebler et al, 1997; Kandler et al, 2009). In olfaction, high-affinity odorant-glomerulus or odorant-OR interactions have been identified for only a handful of glomeruli or ORs (Oka et al, 2006; Zhang et al, 2012; Zhang et al, 2013; Peterlin et al, 2014; Sato-Akuhara et al, 2016; Horio et al, 2019; Soelter et al, 2020).…”