The association between metabolic health status and smell perception in obesity: behavioral and brain anatomical correlates

Obesity is a major health concern that is accompanied by a high risk for several disorders such as type 2 diabetes, cardiovascular disease and certain forms of cancer (Stevens et al., 2012; Lahey and Khan, 2018). Since the prevalence of obesity has nearly tripled within the last 45 years and is still on the rise (WHO, 2018), it is imperative to understand mechanisms that might underlie the emergence and maintenance of obesity to develop new prevention- and intervention strategies. The high availability of energy-rich food is one of the main contributing factors for the increasing prevalence of obesity. Hence, the mechanisms underlying eating without physiological needs come into focus. In that regard, the olfactory system plays a major role: it is equally involved in homeostatic signaling of hunger and hedonic eating (Palouzier-Paulignan et al., 2012). Given that the sense of smell is altered in obesity (for an overview see Peng et al., 2019), the overall goal of this thesis is to contribute to further understanding of the mechanisms that might underlie this phenomenon. It has been previously shown that people with obesity evaluate food odors as more pleasant (Stafford and Whittle, 2015) and show higher reactivity towards them (Proserpio et al., 2019), however, they persistently have lower olfactory function. This low function is most evident in olfactory sensitivity, i.e. picking up odors from the environment, and is therefore related to appetite and food search. Odor sensitivity evaluation is usually a lengthy procedure and is standardly performed with non-food odors. However, Stafford and Whittle (2015), revealed a different result for sensitivity to food odors: obese outperformed normal-weight participants for chocolate odor. Strikingly, further scrutiny reveals that metabolic and endocrine health factors could provide a possible explanation for divergent results of olfactory sensitivity to food and non-food odors in obesity: whereas the chocolate-study included only class 1 obese participants (BMI 30-35 kg/m2), all other studies included class 2-3 obese participants (BMI > 35 kg/m2). It is likely that obese class 2-3 participants are more affected by hormonal changes, such as higher insulin resistance and higher leptin levels than their less obese counterparts. On a recent note, it has been shown that the olfactory and endocrine systems are closely linked (Palouzier-Paulignan et al., 2012). As such there are many receptors for hunger-related hormones located in brain structures that are highly relevant for odor processing as well as for the regulation of homeostatic needs (Baly et al., 2007; Lacroix et al., 2008; Henkin, 2010). Especially, the olfactory bulbs, where olfactory information is firstly processed in the brain, have a high density of insulin and leptin receptors (Baskin et al., 1983; Thanarajah et al., 2019; Havrankova et al., 1981; Marks et al., 1990). Further, animal studies have reliably demonstrated that obesity leads to structural and functional changes in the olfactory system (Thiebaud et al., 2014; Fadool et al.2011; Riviére et al., 2016). However, brain anatomical changes in the olfactory system of humans have not been studied yet. To conclude, we firstly aimed to develop an olfactory test that is easy to administer and of short duration to apply in a complex research design, because available tests are time consuming and highly variable in duration (10-25 min). Secondly, in order to elucidate the potential link between olfactory impairments in obesity and metabolic health factors, we investigated food and non-food odor sensitivity in a wide body weight range and related it to metabolic and endocrine factors such as insulin resistance. Third, we aimed to investigate the possible relationship between obesity and brain anatomical changes in the olfactory bulbs. Study 1: In our first study, we measured olfactory sensitivity in a within-subject repeated-measures design in 20 young and healthy participants. Using the odor detection threshold subtest from the “Sniffin’ Sticks” test battery, we applied three different presentation methods: (1) gold standard, (2) shorter single staircase method and (3) ascending procedure. Compared to the gold-standard, the shorter single staircase procedure was 26% and the ascending procedure was 51% shorter in duration. Both short procedure thresholds correlated highly with the gold standard threshold. All three tests showed similar test-retest reliability. To conclude, we have developed a test that takes on average 5-7 minutes less time and is as reliable as the gold standard. Study 2: Within the second study, we focused on metabolic health parameters that might explain the relationship between odor sensitivity and obesity. We investigated food and non-food odor sensitivity in the hungry and sated state in 75 young healthy participants with normal weight, overweight and obesity in a within-subject, repeated-measures design. We assessed metabolic health status with BMI, WHR, pre- and postprandial levels of insulin, leptin, glucose, and ghrelin. We showed that odor sensitivity did not directly depend on body weight status or BMI. However, we found a strong negative mediating effect of insulin resistance as assessed by HOMA-IR score on the relationship between BMI and olfactory sensitivity for the food odor. Post-hoc regression models revealed that insulin resistance rather than obesity is responsible for this effect. To conclude, our findings indicate a strong negative association between insulin resistance and sensitivity to food odors. Study 3: In the third study, we examined neuroanatomical correlates of smell perception in obesity and its relationship with metabolic health factors. Olfactory bulb volume was assessed with magnetic resonance imaging in 67 healthy normal weight, overweight and obese participants. To examine recently proposed mechanistic explanatory models of altered smell perception in obesity, we collected parameters that are associated with metabolic health in obesity, such as insulin resistance, leptin, body fat percentage and fat mass index. We showed that in our sample, people with obesity had significantly lower olfactory bulb volume when compared to people with normal weight. Further, we found that olfactory bulb volume was negatively associated with other measures of metabolic health, especially insulin resistance, leptin, and body fat percentage. Our results imply that, similar to other diseases such as depression and Parkinson’s disease, obesity also involves a neuroanatomical change in the olfactory bulbs compared to healthy participants with normal weight. Hence, our study provides first indications that obesity is associated with brain anatomical changes in the olfactory bulbs. Conclusion The overall aim of this thesis was to shed light on the complex relationship between obesity and olfaction. Study 1 provides two easy-to-use odor threshold test procedures for clinical use or for complex research designs with limited time frames. Importantly, this thesis emphasizes the major role of metabolic health status and especially insulin resistance in the altered smell perception in obesity. Most notably, poor metabolic health mediates the relationship between obesity and olfactory sensitivity (study 2). Metabolic health parameters rather than obesity per se might be responsible for low olfactory function and should be further scrutinized in future studies. In particular, a group-design with elevated vs. normal HOMA-IR participants instead of BMI groups could provide more insights. Intriguingly, a high BMI and related metabolic health factors, such as high insulin resistance and high body fat percentage are associated with neuroanatomical changes in the olfactory system, i.e., lower olfactory bulb volume (study 3). These findings contribute to a further understanding of explanatory models introduced by Peng et al. (2019). In accordance with this metabolic and hormonal model our results support the theoretical framework that metabolic and hormonal shifts in obesity might be crucial for changes in olfactory perception. Thereby, these results provide a deeper understanding of the pathophysiological mechanisms underlying altered olfactory function in obesity. Subsequently, olfaction might represent a new target for prevention or therapy.:LIST OF ABBREVIATIONS I LIST OF FIGURES II LIST OF TABLES III I. INTRODUCTION 1 1. THE OBESITY PANDEMIC 1 2. HORMONES INVOLVED IN OBESITY AND OLFACTION 4 2.1 HORMONES IN THE REGULATION OF EATING BEHAVIOR AND OBESITY 4 2.2 HORMONES IN THE CONTEXT OF SMELL PERCEPTION 7 3. THE OLFACTORY SYSTEM 9 3.1 ANATOMY AND PHYSIOLOGY 9 3.2 MEASURING SMELL ABILITY: THREE DIMENSIONS OF OLFACTORY FUNCTION 13 3.3 THE ROLE OF OLFACTION IN THE CONTROL OF EATING BEHAVIOR 14 3.4 SMELL PERCEPTION IN OBESITY 15 4. THE LINK: WHY TARGET THE OLFACTORY SYSTEM IN OBESITY? 19 II. RATIONALE OF THE EXPERIMENTAL WORK 20 III. EXPERIMENTAL WORK 21 STUDY 1: SHORT PROCEDURE TO ASSESS ODOR DETECTION THRESHOLDS 21 STUDY 2: ODOR SENSITIVITY FOR FOOD AND NON-FOOD ODORS IN OBESITY 30 STUDY 3: BRAIN ANATOMICAL CORRELATES OF SMELL PERCEPTION IN OBESITY 47 IV. SUMMARY 60 V. REFERENCES 65 VI. APPENDIX 75 A. DECLARATION OF AUTHENTICITY 75 B. AUTHOR CONTRIBUTIONS 76 C. CURRICULUM VITAE 80 D. ACKNOWLEDGEMENTS 8