Novel measure of olfactory bulb function in health and disease

Abstract: Present neuroimaging techniques are capable of recording the neural activity from all over the brain but the olfactory bulb (OB). The OB is the first olfactory processing stage of the central nervous system and the site of insult in several neurological disorders, particularly Parkinson’s disease (PD). It has been suggested that the OB has a pivotal role in the olfactory system anal-ogous to primary visual cortex (V1) and thalamus in the visual system. However, due to the existing technical limitations, there has not been any non-invasive technique that can reliably measure the OB function in humans, consequently limiting its functional recording to one in-tracranial study dating back to the 60s. Initially in Study I, a non-invasive method of measuring the function of human OB is devel-oped, so-called electrobulbogram (EBG). In line with previous animal literature as well as the only intracranial study in human OB, it was demonstrated that gamma oscillations on the EBG electrodes occurred shortly after the odor onset. Subsequently, applying source recon-struction analysis provided evidence that observed oscillations were localized to the OB. Ad-ditionally, the OB recording with the EBG method showed a test-retest reliability comparable with visual event related potentials. Notably, the detected gamma oscillations were demon-strated to be insensitive to habituation, the OB’s marked characteristic which has previously been demonstrated in rodents. Last, but not least, assessing the EBG response in an individual who did not have the bilateral OB indicated that the lack of OB results in disappearance of gamma oscillations in the EBG electrodes. Given that Study I determined the possibility of reliably measuring the function of the OB using the EBG, in Study II, I assessed the functional role of OB’s oscillations in the pro-cessing of the odor valence. Odor valence has been suggested to be linked to approach–avoidance responses and therefore, processing of odor valence is thought to be one of the core aspects of odor processing in the olfactory system. Consequently, using combined EBG and EEG recording, OB activity was reconstructed on the source level during processing of odors with different valences. Gamma and beta oscillations were found to be related to va-lence perception in the human OB. Moreover, the early beta oscillations were associated with negative but not positive odors, where these beta oscillations can be linked to preparatory neural responses in the motor cortex. Subsequently, in a separate experiment, negative odors were demonstrated to trigger a whole-body motor avoidance response in the time window overlapping with the valence processes in the OB. These negative odor-elicited motor re-sponses were measured by a force plate as a leaning backward motion. Altogether, the results from Study II indicated that the human OB processes odor valence sequentially in the gamma and beta frequency bands, where the early processing of negative odors in the OB might be facilitating rapid approach-avoidance behaviors. To further evaluate the functional role of the OB in odor processing, in Study III, OB’s communication with its immediate recipient, namely piriform cortex (PC), was assessed. These two areas are critical nodes of the olfactory system which communicate with each other through neural oscillations. The activity of the OB and the PC were reconstructed using a combination of EBG, EEG, and source reconstruction techniques. Subsequently, the cross spectrogram of the OB and the PC was assessed as a measure of functional connectivity where temporal evolution from fast to slow oscillations in the OB–PC connectivity was found during the one second odor processing. Furthermore, the spectrally resolved Granger causal-ity analysis suggested that the afferent connection form the OB to the PC occurred in the gamma and beta bands whereas the efferent connection from the PC to the OB was concen-trated in the theta and delta bands. Notably, odor identity could be deciphered from the low gamma oscillatory pattern in the OB–PC connectivity as early as 100ms after the odor onset. Hence, findings from this study elucidate on our understanding of the bidirectional infor-mation flow in the human olfactory system. Olfactory dysfunction, due to neurodegeneration in the OB, commonly appears several years earlier than the occurrence of the PD-related characteristic motor symptoms. Consequently, a functional measure of the OB may serve as a potential early biomarker of PD. In Study IV, OB function was assessed in PD to answer whether the EBG method can be used to dissociate individuals with a PD diagnosis from healthy age-matched controls. The spectrogram of the EBG signals indicated that there were different values in gamma, beta, and theta for PDs compared with healthy controls. Specifically, six components were found in the EBG re-sponse during early and late time points which together dissociate PDs from controls with a 90% sensitivity and a 100% specificity. Furthermore, these components were linked to med-ication, disease duration and severity, as well as clinical odor identification performance. Overall, these findings support the notion that EBG has a diagnostic value and can be further developed to serve as an early biomarker for PD. In the last study, Study V, the prevalence of COVID-19 was determined using odor intensity ratings as an indication of olfactory dysfunction. Using a large sample data (n = 2440) from a Swedish population, odor intensity ratings of common household items over time were found to be closely associated with prevalence prediction of COVID-19 in the Stockholm region over the same time-period (r = -.83). Impairment in odor intensity rating was further correlated with the number of reported COVID-19 symptoms. Relatedly, individuals who progressed from having no symptoms to having at least one symptom had a marked decline in their odor intensity ratings. The results from this study, given the relatively large sample size, provided a concrete basis for the future studies to further assess the potential association between the deficits in the OB function and olfactory dysfunction in COVID-19. In conclusion, our proposed method for non-invasive measurement of the OB function was shown to provide a reliable recording with a potential as a diagnostic tool for PD. Combining EBG and EEG allowed for reconstruction of the OB signal at the source level, where specific oscillations were found to be critical for odor valence processing and rapid avoidance re-sponse. Moreover, oscillations in different frequency bands were found to be critical for the OB reciprocal communications and transfer of odor identity information to higher order ol-factory subsystems. Finally, COVID-19 was found to be associated with a decline in olfactory acuity which might originate from damage to the patient’s OB. In conclusion, the results from the studies within this thesis provide a new perspective on the functional role of oscillations in the human OB.