Is weight status associated with peripheral levels of oxytocin? A pilot study in healthy women.

Journal Pre-proof

Is weight status associated with peripheral levels of oxytocin? A pilot study in healthy women. Janelle A Skinner , Manohar L Garg , Christopher V Dayas , Tracy L Burrows PII: DOI: Reference:

S0031-9384(19)30359-2 https://doi.org/10.1016/j.physbeh.2019.112684 PHB 112684

To appear in:

Physiology & Behavior

Received date: Revised date: Accepted date:

8 April 2019 9 September 2019 16 September 2019

Please cite this article as: Janelle A Skinner , Manohar L Garg , Christopher V Dayas , Tracy L Burrows , Is weight status associated with peripheral levels of oxytocin? A pilot study in healthy women., Physiology & Behavior (2019), doi: https://doi.org/10.1016/j.physbeh.2019.112684

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.

Is weight status associated with peripheral levels of oxytocin? A pilot study in healthy women. Janelle A Skinner,a,b Manohar L Garg,c Christopher V Dayas,c Tracy L Burrowsa,b* a

Nutrition and Dietetics, School of Health Sciences, Faculty of Health and Medicine, University of

Newcastle, Callaghan NSW 2308, Australia b

Priority Research Centre for Physical Activity and Nutrition, University of Newcastle, Callaghan NSW

2308, Australia c

School of Biomedical Sciences and Pharmacy, Faculty of Health and Medicine, University of

Newcastle, Callaghan NSW 2308, Australia Author Email Addresses: Janelle A Skinner: [email protected] Manohar L Garg: [email protected] Christopher V Dayas: [email protected] Tracy L Burrows: [email protected] *Corresponding Author: Associate Professor Tracy Burrows University of Newcastle, HA12 Hunter Building, University Dr, Callaghan NSW 2308 Australia Phone: +61 2 49215514; Email: [email protected] Highlights 

Pilot data is presented on hormone levels related to food intake in female adults



Our data indicate a positive correlation between BMI and plasma oxytocin in females



Plasma oxytocin was higher and CCK lower in women with self-report food addiction

Abstract

1

The neuropeptide oxytocin is best known for its role during parturition and the milk-let down reflex. Recent evidence identifies a role for oxytocin in eating behaviour. After oxytocin administration, caloric intake is reduced with stronger inhibitory effects in individuals with obesity. Whether the experience of visual food cues affects secretion or circulating levels of oxytocin is unknown. This pilot study had three aims: 1) to measure fasting appetite hormones with a focus on plasma oxytocin concentrations; 2) determine whether healthy vs. hyperpalatable visual food cues differentially altered plasma oxytocin; and 3) assess whether appetite hormone responses to healthy vs. hyperpalatable food images depended on weight or food addiction status. Eighteen healthy women of varying weight status, with/without self-reported food addiction were recruited. Study participants completed a set of standardised questionnaires, including Yale Food Addiction Scale, and attended a one-off experimental session. Blood was collected before and after viewing two sets of food images (healthy and hyperpalatable foods). Participants were randomly allocated in a crossover design to view either healthy images or hyperpalatable foods first. A positive correlation between BMI and plasma oxytocin was found (r2=0.32, p=0.021) at baseline. Oxytocin levels were higher, and cholecystokinin levels lower, in food addicted (n=6) vs. non-food addicted females (p=0.015 and p<0.001, respectively). There were no significant changes (p>0.05) in plasma oxytocin levels in response to either healthy or hyperpalatable food images. Given that endogenous oxytocin administration tends to suppress eating behaviour; these data indicate that oxytocin receptor desensitization or oxytocin resistance may be important factors in the pathogenesis of obesity and food addiction. However, further studies in larger samples are needed to determine if peripheral oxytocin is responsive to visual food cues. Key words obesity; food addiction; appetite hormones; oxytocin; cholecystokinin 1. Introduction

2

The physiological regulation of food intake is based on an intricate feedback system controlled by shorter term ‘hunger’ and ‘satiety’ signals and longer terms signals from fat stores (e.g. leptin).1 These signals include peripheral anorexigenic (appetite-inhibiting) and orexigenic (appetitestimulating) hormones, including cholecystokinin (CCK) and ghrelin (GH) respectively (for review see2). CCK is secreted by the small intestine and acts as a satiety signal on vagal afferents to terminate feeding behaviour.2 Interestingly, some of the effects of CCK are enhanced by its interaction with leptin ,3-7 (for review see8). In contrast, ghrelin which is secreted by the stomach, stimulates rather than inhibits feeding behaviour by initiating food intake.9-12 The actions of ghrelin are complex, but involve activation of growth hormone secretagogue receptors on agouti-related peptide expressing (AgRP) neurons in the hypothalamic arcuate nucleus (for review see13). Importantly, other circulating peptides not traditionally considered to play a direct role in feeding are now known to be involved in food intake and body weight. For example, cortisol a hormone released in response to activation of the hypothalamic-pituitary-adrenal axis, can stimulate appetite for foods high in sugar and fat.14, 15 Cortisol’s effects have recently been linked with activation of AgRP neurons under conditions of hypoleptinemia.16 More recently, evidence has accumulated pointing to a role for oxytocin in the control of food intake (for review see 17, 18), energy expenditure and body weight.19-32 Oxytocin is a neuropeptide primarily produced in the supraoptic and paraventricular hypothalamic nuclei, is released via the posterior pituitary gland into the bloodstream.19 Historically, oxytocin secretion is best associated with increased uterine contractions during parturition and the milk-letdown reflex.33 However, recent studies link oxytocin secretion with appetite regulation and satiety. In support, studies have shown that ingestion of food triggers oxytocin release34 from oxytocin neurons in the hypothalamus.35-37 It can also be released by stimulation of other senses associated with common environmental food cues (smell, sight and sound).38 Oxytocin is released at synaptic terminals and dendrites in areas such as the amygdala or hypothalamus where it controls mood and

3

social behaviours.39 Exactly how oxytocin release is triggered by food intake is an area of active research, but one likely mechanism involves CCK-mediated activation of vagal afferents that project to the hindbrain, and subsequent activation of ascending projections from the hindbrain to the hypothalamus.35, 40 Research in humans and rodent models show that exogenously administered oxytocin (via intranasal spray, and central or peripheral injections/infusions in rodents) can reduce caloric intake with stronger inhibitory effects in individuals with obesity.20-22, 26-29, 41, 42 With respect to humans, exogenous oxytocin has been shown to improve cognitive control of food craving and appetite, and decrease approach bias towards highly palatable foods (3 studies, n=127).43-45 In rodent models, the role of oxytocin includes effects on food preference.18 For example, oxytocin knockout mice display an enhanced preference for carbohydrate,46-48 saccharin49 and sodium50-52 solutions, compared to their wild-type counterparts. Notably, mice lacking in oxytocin53 or oxytocin receptors54 develop late-onset obesity without significant changes in their daily food consumption. Peripheral levels of appetite regulating hormones are highly variable between individuals.55 Body weight and adiposity levels appear to influence appetite regulating hormones with dysregulated levels of ghrelin, leptin, CCK and cortisol observed in obese compared with lean individuals.55-59 Research examining plasma levels of oxytocin in relation to weight status is limited to a small number of studies in individuals with and without chronic illness (e.g. diabetes mellitus, metabolic syndrome, anorexia nervosa). Findings to date have been equivocal, with studies reporting circulating oxytocin to be either increased,60-64 decreased,65-67 or no association68, 69 in individuals with high body mass indexes (BMI) compared to healthy weight individuals. It has been proposed that oxytocin may have important implications in the risk for, and development of, clinically significant overeating, and therefore further investigations are warranted.70 The appetite hormones that control hunger, satiety and help maintain an overall energy balance, are sensitive to external influences, such as food cues.71 In response to these stimuli, the physical processes that control appetite in some vulnerable individuals may be overridden, potentially

4

increasing their susceptibility for poor nutritional habits, overeating and weight gain.71 The significant rise in the prevalence of obesity over the last two decades has been attributed to changes in eating patterns associated with food cues and increased marketing of hyperpalatable foods in the modern food environment.72 During this time the concept and amount of research interest in ‘food addiction’ or ‘addictive eating’ has increased exponentially, although it remains a contentious topic.73, 74 At present, there is no recognised definition for food addition or an agreed terminology or how it is referred.75, 76 In the literature, food addiction is most often described as the chronic excessive and uncontrolled consumption of hyperpalatable foods i.e. processed foods typically high in added fat, sugar and salt.77, 78 This is often accompanied by a sense of losing control, continued use regardless of adverse consequences and an inability to reduce consumption despite the desire to do so.78 Addictive eating, as defined by the food addiction construct, is often associated with obesity and mental health conditions, such as depression.79 While food addiction can occur independently of body weight status,80 individuals with elevated BMI and individuals exhibiting higher levels of addictive eating have been shown to have heightened food cue reactivity and attentional biases to hyperpalatable foods.81, 82, This is likely a contributing factor to the onset and maintenance of addictive eating behaviour. In human research, images of food are often used in place of real foods to elicit anticipatory responses, such as prompting changes in metabolic hormones.83 A number of studies have examined peripheral hormonal responses to food images (e.g. ghrelin,84 leptin,85 and cortisol86). However, whether the experience of visual food cues affects secretion or circulating levels of oxytocin is unknown. It is hypothesised that individuals with obesity and/or food addiction may have stronger motivations to seek out hyperpalatable food in response to food-related cues. Therefore, the aims of this pilot study were to 1) measure appetite hormones, with a focus on plasma oxytocin concentrations, in adults of varying weight statuses and in those with and without self-reported food

5

addiction; 2) to investigate if visual food cues affect plasma oxytocin; and 3) determine if different food categories (healthy vs. hyperpalatable) can elicit different responses in adults of varying weight statuses, and whether these indices interact with food addiction status. 2. Methods 2.1 Participants For a homogeneous sample, only female subjects were chosen to participate as earlier reports have demonstrated significant sex differences in oxytocin concentrations87 and greater reactivity in females to visual food stimuli.88 Females aged 18-85 years, from a previous study sample (Hunter region, NSW)89 that agreed to be recontacted for future studies, were initially invited by email. Those interested in participating were screened by telephone to determine eligibility for the experimental session (n=34). The screening assessed for exclusion criteria: 1) following a vegetarian diet or health conditions requiring a specialised diet (e.g. Coeliac disease); 2) currently taking medication; 3) binge eating or substance use disorder; 4) smokers; 5) pregnant or lactating; or 6) cognitive impairment or mental health condition (e.g. bipolar disorder) requiring ongoing treatment.

2.2 Procedure The study-protocol was approved by University of Newcastle Human Research Ethics Committee (H2016-0305) and registered with the Australia New Zealand Trial Registry (ACTRN12619000351112). Each participant provided informed consent prior to completing the online questionnaire and attending a one-off morning experimental session (0700-0900h) at the University of Newcastle. Premenopausal women not using hormonal contraceptives attended during the mid-follicular phase (days 6-8) of their menstrual cycle (n=8). On arrival, following an overnight fast (≥11 hours, confirmed by verbal consent), participants had their height and weight measured by a trained assessor, and completed a self-report measure of current hunger. In a crossover design, participants were randomly allocated (block design) via computer-based program90, to view either healthy or

6

hyperpalatable foods first. An initial blood sample was taken (time point 1), participants then viewed a 10 min paradigm of food images. After viewing the images participants underwent a second blood test (time point 2), and then remained in the control environment where there was no visual stimuli for 15 min (NB. wash-out period is ~2.5-15 times the length of the half-life of oxytocin i.e. 1-6 min91, 92

). A third blood sample was taken prior to viewing the second paradigm, opposite of the first set of

food images (time point 3). Afterwards participants underwent a fourth blood test (time point 4). Participants then completed hunger rating scales before consuming a standardised breakfast. The session took approximately 60 min and participants were reimbursed AUD50. 2.3 Measures 2.3.1 Demographics Demographic information collected via fixed response questions in the online questionnaire, included age, ethnicity, highest level of education and marital status. 2.3.2 Anthropometric measures Weight and height were measured using digital scales and a wall mounted stadiometer. BMI was calculated using standardised techniques and categorised according to the World Health Organization adult cut-off points.93 2.3.3 Addictive eating behaviour Addictive eating behaviour was assessed with the Yale Food Addiction Scale (YFAS 2.0).78 This 35item self-report tool is most commonly used to assess addictive eating and asks participants to think of specific foods they have had difficulty controlling the consumption of within the past year (e.g. ice cream, chocolate, chips, hamburgers). YFAS 2.0 provides a food addiction symptom score based on similar criteria for substance use disorder of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5).94 Scores can range from zero to eleven, and symptoms include craving, loss of control, tolerance and withdrawal associated with eating behaviours. Additionally, two items assess

7

clinically significant impairment or distress from eating. A food addiction “diagnosis” is given when ≥2 symptoms are endorsed and clinically significant impairment or distress is present. Severity of the “diagnosis” can be classified as “mild” (2-3 symptoms), “moderate” (4-5 symptoms) or “severe” (≥6 symptoms). 2.3.4 Diet quality Diet quality was assessed with The Australian Eating Survey (AES).95 The Australian Recommended Food Score (ARFS), derived from a subset of 70 AES questions, reflects the overall healthiness and nutritional quality of an individual’s usual eating pattern.96 The ARFS is based on the frequency of consumption of core foods, recommended in the Australian Dietary Guidelines, with foods awarded one point for a consumption frequency of ≥once per week. The total score can range from zero to 73, with higher scores awarded for greater dietary variety. 2.3.5 Subjective assessment of appetite Subjective appetite, before and after food image viewing, was assessed using a Visual Analogue Scale (VAS) for 7 components: 1) Hunger, 2) Motivation to eat and 3-7) Food/taste preference for healthy, sweet, salty, savoury and fatty foods. Each 100mm horizontal line was anchored at either end with extremes of the subjective feelings to be quantified, for example ‘I am not hungry at all” (0mm) and “I have never been more hungry” (100mm). For each item, participants marked on the line the point they felt best represented their current state. A score was determined by measuring the distance between the lowest anchor (0mm) and the participant’s mark, providing a score from 0100. Higher scores indicated greater hunger/appetite intensity. 2.4 Visual food cue image paradigms Images were presented as 2 x 10 min paradigms (PowerPoint slideshow) on a laptop computer. Paradigms comprised 75 healthy food images and 75 hyperpalatable food images. Images were chosen from a licensed database83 and non-copyright sources.97-100 Healthy food pictures were based

8

on the five core food groups, and hyperpalatable food pictures based on foods categorised as discretionary choices, outlined in The Australian Guide to Healthy Eating.101 Food images displayed contained no food logos or product advertisements; and were similar in resolution, scale, contrast, brightness and complexity. All stimuli had previously been rated in a pilot study (n=10 adults, independent of those involved in the present study) to ensure foods were representative of each of the two categories and shown in an appetising manner for Australian adults. 2.5 Blood samples and analysis Blood samples were collected, by a trained phlebotomist, at four time periods using standard procedures, in EDTA coated tubes (total 32mL blood collected). Blood samples were centrifuged at 3000 x g for 10 min and plasma stored at -80°C until analysis. Hormones were analysed in duplicate using enzyme-linked immunoassay (ELISA) according to the manufacturer's protocol. Oxytocin (unextracted serum) was measured using a kit from Abcam (Cambridge, UK), intra-assay coefficient of variation (CV) was 12.6-13.3%, inter-assay CV was 11.8–20.9%, and sensitivity was 15.0 pg/mL. Total ghrelin was measured using a kit from ThermoFisher Scientific (Massachusetts, USA), intraassay CV was 6.0%, inter-assay CV was 8.5% and sensitivity was 11.8 pg/mL. Leptin was measured using a kit from Abcam (Cambridge, UK), intra-assay CV was 2.4%, inter-assay CV was 2.7%, and sensitivity was 2.3 pg/mL. CCK concentrations were measured using a kit from RayBiotech (Georgia, USA), intra-assay CV was <10% and inter-assay CV was <15%, and sensitivity was 0.2 pg/mL. Cortisol was measured using a kit from Abcam (Cambridge, UK), intra-assay CV was ≤9.0%, inter-assay CV was ≤9.8%, and sensitivity was 2.44 ng/mL. 2.6 Statistical analysis Statistical analysis was performed using Stata version 15.1.102 Plasma hormone concentrations were natural log transformed to achieve a normal distribution. Raw data is available in supplementary material. Between group comparisons, according to food addiction status or weight status, of baseline hormones and dietary scores were made using independent sample t-tests (NFA vs FA) and 9

one-way analysis of variance with Bonferroni corrected post‐hoc analyses (healthy weight vs overweight vs obese). Linear regression analysis was used to investigate the association between baseline oxytocin level and BMI; with multivariate least-square analysis constructed to adjust for age as a potential confounder. Within group comparisons of hormone levels at different time points (i.e. before and after image viewing) were made using paired t-tests. Hormonal and dietary scores are presented descriptively as mean ± SD. Where arithmetic means calculated from log-transformed data are back transformed, to their original scale through exponentiation, the result reported is the geometric mean (i.e. ̅ g =

y

). Wilcoxon signed-rank tests were conducted to examine differences

between pre and post image viewing VAS scores. Statistical significance was defined as p value <0.05. 3. Results 3.1 Participant characteristics Twenty-two women were eligible and consented to participate. For four individuals blood samples were not possible due to difficulties drawing sufficient blood. This left a sample of 18 participants for the current analyses. Mean age of participants was 43 ± 16.5 years (range 19-70), 12 women (66.7%) were pre-menopausal, four of which reported current oral contraceptive use. There were no significant differences (<0.05) in baseline hormone concentrations between pre- and postmenopausal women, or women taking/not taking oral contraceptives. 3.2 BMI and baseline hormone levels From the total sample of women, six were classified as healthy weight, nine as overweight and three as obese. Oxytocin levels were significantly higher in obese compared with healthy or overweight women, with mean oxytocin levels of 2751.77, 1619.71 and 1540.71pg/mL, respectively (geometric mean, back transformed data), Table 1. Mean baseline levels of ghrelin, leptin, CCK and cortisol were not significantly different across BMI categories (p>0.05).

10

3.3 Relationship between oxytocin and BMI There was a significant positive correlation between baseline oxytocin levels and BMI (r=0.36, p=0.005). This relationship remained significant after controlling for age (r2=0.32, p= 0.021). 3.4 YFAS-defined food addiction and diet quality Six women met the clinical significance criterion for food addiction (FA) with severity ranging from mild/moderate (n=3) to severe (n=3). Mean symptom score in non-food addicted (NFA) was 1.1 ± 1.2 (range 0-3), and in FA was 5.5 ± 2.5 (range 2-9). BMIs of NFA women ranged from 21.22 to 29.97 (healthy to overweight) with a mean of 25.64 ± 2.84kg/m2, and BMIs of FA women ranged from 25.66 to 41.48 (overweight to obese) with a mean of 33.55 ± 6.02kg/m2. There was a significant difference in BMI between groups (p=0.0014). Overall, the mean difference in diet quality score was not significantly different between NFA and FA (p=0.107), but there was a significant difference in percentage of energy intake from core (healthy foods, p=0.026) to non-core (hyperpalatable foods, p=0.026) food groups, Table 2. 3.5 Food addiction status and baseline hormone concentrations Oxytocin levels were significantly higher in women classified as FA compared with NFA, with mean oxytocin levels of 2230.54 and 1525.38pg/mL, respectively (geometric mean, back transformed data), Table 3. When overweight/obese women only were compared (Table 4), oxytocin levels were significantly higher in women classified as FA (n=3, overweight; and n=3, obese) compared with NFA (n=6, overweight), with mean oxytocin levels of 2235.18 and 1421.07pg/mL, respectively (geometric mean, back transformed data). Mean baseline levels of CCK were significantly lower in FA compared to NFA women, 437.03 and 915.99pg/mL, respectively (geometric mean, back transformed data), Table 3. In overweight/obese women (Table 4) CCK levels were significantly lower in FA compared to

11

NFA women, with mean CCK levels of 436.60 and 885.49 pg/mL, respectively (geometric mean, back transformed data). There was no significant difference in ghrelin, leptin, or cortisol levels (p>0.05). 3.6 Subjective appetite pre and post image viewing The VAS scores indicate that the viewing of food images increased appetite, with intensity greatest for: hunger, amount could eat, desire to eat sweet and fatty foods (p<0.05; Figure 1). There were no differences in pre or post scores by weight or food addiction status, all p>0.05.

] p = 0.001* ] p = 0.009* ] p = 0.695 ] p = 0.019* ] p = 0.191 ] p = 0.248 ] p = 0.045*

Wilcoxon signed-rank test; p<0.05

Figure 1. Median appetite scores pre and post food image viewing for the total sample (n=18). Subjective appetite for seven components assessed using a 100mm Visual Analogue Scale (VAS). Higher scores indicate a greater hunger/appetite intensity. 3.7 Oxytocin levels in women before and after viewing food images Following the viewing of healthy and hyperpalatable food images there were no significant changes in mean plasma oxytocin levels (all p>0.05), Table 5.

12

3.8 Ghrelin, leptin, cholecystokinin and cortisol levels in women before and after viewing food images For the total sample, there were no significant changes in mean plasma ghrelin, CCK or cortisol levels (all p>0.05; supplementary material) in response to healthy and hyperpalatable food images. When pre and post appetite hormones were compared within-subject, according to each BMI category, significant differences were observed in overweight women for leptin and CCK. Leptin levels increased in response to healthy food images (p=0.019) and CCK levels increased in response to hyperpalatable food images (p=0.048; supplementary material). There were no significant withinsubject differences between pre and post appetite hormones in either FA or NFA groups.

13

4. Discussion In this pilot study we found that basal oxytocin levels were significantly higher in obese women compared to healthy or overweight women. Despite data from only a small subgroup of overweight/obese females, we also found that individuals identified through self-report as ‘food addicted’ displayed significantly higher basal oxytocin levels and lower basal CCK levels compared to those without food addiction. Together our findings highlight a number of interesting associations between weight status and endocrine measurements and are highly relevant to the design of future studies assessing the relationship between oxytocin and satiety signals. 4.1 Relationship between oxytocin and weight status Our experiments revealed a positive correlation between fasting plasma oxytocin levels and BMI that was independent of age. On its own these findings are interesting given the lack of consistency between existing studies in reporting oxytocin and weight status. This may be due to interactions with other co-morbid diseases. For example, overweight individuals with Type 1103 and Type 2 diabetes mellitus66, 67 has been associated with lower oxytocin levels. Thus, Kujath et al. reported that oxytocin concentrations were lower in women with Type 1 Diabetes Mellitus compared with matched controls (n=162), whereas a positive association with oxytocin was found for BMI in the diabetes group only (β=0.02, p=0.01).103 In contrast, Szulc et al. found that men with metabolic syndrome had higher BMI and oxytocin levels.61 While no association between oxytocin and body weight status was found by Coiro et al. in a small sample of men (n=23)68 or by Breuil et al. (2011) in postmenopausal women with/without osteoporosis (n=36),104 in a larger sample of postmenopausal women with/without osteoporosis (n=1097), Breuil et al. (2014) reported that oxytocin was positively correlated with BMI (r=0.25, p<0.0001).63 Taken together, our findings of higher levels of plasma oxytocin in females with higher BMI (r2=0.32, p=0.021) are consistent with studies in healthy (absence of chronic disease) males and females by Stock et al. (plasma oxytocin 4-fold higher in obese vs. controls)60 and Schorr et al. (r=0.44, p=0.0004).62 Note however that Maestrini et al.65

14

found oxytocin levels were decreased in healthy women with obesity (n=109). These inconsistencies may be related to differences in study populations, sample sizes or potentially blood analysis techniques for oxytocin. It will be important for future studies to factor these variables into study design and statistical analyses. 4.2 Relationship between oxytocin and addictive eating In the current sample, six women were classified as food addicted as defined by the YFAS. As expected, the intake of non-core foods (reported as % energy intake) and BMI was higher in these women compared to non-food addicted women. The results of our study suggest that fasting plasma oxytocin levels may be higher in people exhibiting a loss of control over their eating behaviour. Unfortunately, due to limitations in sample size it is unclear whether these results can be attributed to addictive eating behaviour or weight status. It is possible the elevation in endogenous oxytocin may be an adaptive response to increased body fat, sending a signal to reduce eating and increase energy expenditure in individuals with higher BMIs. Additional research is needed to determine whether, despite high circulating levels of oxytocin, the anorexigenic effects of oxytocin may be diminished in some individuals. This suggests a role for oxytocin receptor desensitization or oxytocin resistance as a contributing factor to the development and maintenance of overeating and obesity, but further work is warranted to address this possibility. To our knowledge, only two studies in humans have examined oxytocin in and its possible relationship with addictive eating. Similar to our study, these reports predominately sampled female individuals. For example, Pedram et al.105 examined a range of hormones and neuropeptides in individuals with obesity only (n=58, 75% female), with and without food addiction. In this study, no significant difference was found in oxytocin levels or other appetite hormones (e.g. ghrelin, leptin, and cortisol). Likewise, Davis et al.106 examined the link between the oxytocin receptor gene and overeating in a sample (n=460) comprising 83% females with a broad range of body weights.

15

Interestingly, genetic variations were found to be associated with overeating, food preferences (sweet and fatty), and reward sensitivity.106 4.3 Effect of visual food cues on circulating oxytocin levels In the present study all participants reported a statistically significant increase in hunger following the presentation of food images, validating our protocol. However, we did not find a significant difference between pre and post oxytocin levels in response to either healthy or hyperpalatable food images. In this exploratory pilot study, it is interesting to note the general direction of movement in oxytocin levels. In healthy weight women oxytocin levels tended to decrease in in response to hyperpalatable food images (p=0.126). Previous studies have shown an increased bias toward hyperpalatable foods in obesity.81 It is plausible that oxytocin levels may increase in overweight/obese or food addicted individuals in response to hyperpalatable foods as a mechanism to limit the intake of these foods. Studies with larger sample sizes are required to determine if there is a relationship. 4.4 Impact of weight status and self-reported food addiction status on ghrelin, leptin, cholecystokinin and cortisol We did not find any significant differences in baseline concentrations of ghrelin, leptin, CCK and cortisol between healthy weight and overweight/obese females. In contrast to the majority of previously published data, leptin levels were not significantly higher in those with obesity compared to healthy weight females. This is likely due to the small sample size. Interestingly, when individuals were classified according to food addiction status, CCK levels were significantly lower in food addicted females (p<0.001) vs non-food addicted. Low levels of CCK may contribute to increased appetite and diminished satiety sensations during meal intake. Over the longer term this could lead to difficulties with weight gain. 4.5 Methodological considerations

16

Methods for the quantitative measurement of oxytocin in peripheral blood are not standardised and method-specific. There is considerable debate in existing research as to whether extracted or unextracted serum should be used.107 In the current study, we selected a commercially available ELISA without extraction, a method that has been used in other studies and is useful in examining relative differences in oxytocin levels. 19 An accepted methodology and validation of oxytocin measurement would significantly aid progress in the field.17, 18 Circulating ghrelin consists of two major forms, an acylated form and a des-acylated form.102 The effects of ghrelin on appetite have largely been assigned to acylated ghrelin (active ghrelin),103 with this form of ghrelin observed to change more rapidly over short time periods104 (half-life of ~913min105). In the current study, we measured plasma levels of total ghrelin (acyl plus des-acyl). It has been demonstrated that similar changes occur in both active and total ghrelin after meal consumption,106, 107 suggesting that total ghrelin is a good marker for the active hormone. Though in future studies, it may be useful to measure plasma concentrations of both active and total ghrelin to determine if there is a difference in plasma ghrelin responses to food cues. The current study was a pilot study making the small sample a limitation. Future studies in larger samples with a more homogeneous age range are required. To optimally capture the changes in oxytocin concentrations at time points two and four (i.e. post image viewing) the use of a Peripheral Venous Cannula (PVC) for blood draws may be an improved approach. Finally, although we cannot rule out the possibility that social interactions during the study visit affected oxytocin levels, procedures and exposure to research staff were similar for all experimental sessions. Conclusions In conclusion, despite a small sample size in this pilot study we identified a positive correlation between BMI and plasma oxytocin in adult females. These findings are consistent with most prior studies conducted in healthy individuals. While the mechanisms responsible for these differences were not determined and were outside the scope of this pilot study, it is possible that oxytocin 17

receptor desensitization or oxytocin resistance may be important factors to consider in the pathogenesis of obesity and addictive eating behaviour. A hypothesis arising from these findings is that in some individuals, despite higher levels of oxytocin, a decreased sensitivity to oxytocin may result in an inability to sense satiety signals making it more difficult to regulate food intake. While acknowledging the small sample size, our findings are highly relevant to the current interest in oxytocin and eating behaviour. We provide a number of relevant insights for future study design. For example, in a small subgroup of females with higher BMI and self-reported food addiction, we also observed reduced levels of the satiety promoting peptide CCK. A putative explanation and testable hypothesis is that a combination of oxytocin resistance and lower levels of CCK may interact to promote overeating in some individuals.

Acknowledgements The authors would like to thank Melissa Fry (Research Assistant, School of Biomedical Sciences and Pharmacy, University of Newcastle) for her assistance in the blood collection and analysis procedures, and the participants who completed the study. Tracy Burrows is supported by a University of Newcastle Brawn research fellowship.

Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declarations of interest: none.

18

References

1.

Valassi E, Scacchi M, Cavagnini F. Neuroendocrine control of food intake. Nutrition,

Metabolism and Cardiovascular Diseases. 2008;18(2):158-68. https://doi.org/10.1016/j.numecd.2007.06.004 2.

Austin J, Marks D. Hormonal regulators of appetite. International Journal of Pediatric

Endocrinology. 2009;2009(Article ID 141753):1-9. doi: 10.1155/2009/141753 3.

Barrachina MD, Martínez V, Wang L, Wei JY, Taché Y. Synergistic interaction between leptin

and cholecystokinin to reduce short-term food intake in lean mice. Proceedings of the National Academy of Sciences of the United States of America. 1997;94(19):10455-60. 4.

Emond M, Schwartz GJ, Ladenheim EE, Moran TH. Central leptin modulates behavioral and

neural responsivity to CCK. American Journal of Physiology. 1999;276(5):R1545-R9. doi: 10.1152/ajpregu.1999.276.5.R1545 5.

Blevins JE, Schwartz MW, Baskin DG. Evidence that paraventricular nucleus oxytocin neurons

link hypothalamic leptin action to caudal brain stem nuclei controlling meal size. American Journal of Physiology: Regulatory, Integrative and Comparative Physiology. 2004;287:R87–R96. doi:10.1152/ajpregu.00604.2003 6.

McMinn JE, Sindelar DK, Havel PJ, Schwartz MW. Leptin deficiency induced by fasting impairs

the satiety response to cholecystokinin. Endocrinology. 2000;141(12):4442-8. doi: 10.1210/endo.141.12.7815 7.

Morton GJ, Blevins JE, Williams DL, Niswender KD, Gelling RW, Rhodes CJ, et al. Leptin action

in the forebrain regulates the hindbrain response to satiety signals. Journal of Clinical Investigation. 2005;115(3):703-10. doi: 10.1172/JCI22081 8.

Baskin DG, Figlewicz Lattemann D, Seeley RJ, Woods SC, Porte D Jr, Schwartz MW. Insulin

and leptin: dual adiposity signals to the brain for the regulation of food intake and body weight. Brain Research. 1999;848(1-2):114-23. doi: 10.1016/s0006-8993(99)01974-5 19

9.

Faulconbridge LF, Cummings DE, Kaplan JM, Grill HJ. Hyperphagic Effects of Brainstem

Ghrelin Administration. Diabetes. 2003;52(9):2260-5. https://doi.org/10.2337/diabetes.52.9.2260 10.

Cummings DE, Purnell JQ, Scott Frayo R, Schmidova K, Wisse BE, Weigle DS. A Preprandial

Rise in Plasma Ghrelin Levels Suggests a Role in Meal Initiation in Humans. Diabetes. 2001;50(8):1714-9. https://doi.org/10.2337/diabetes.50.8.1714 11.

Ariyasu H, Takaya K, Tagami T, Ogawa Y, Hosoda K, Akamizu T, et al. Stomach is a major

source of circulating ghrelin, and feeding state determines plasma ghrelin-like immunoreactivity levels in humans. Journal of Clinical Endocrinology and Metabolism. 2001;86:4753–8. doi: 10.1210/jcem.86.10.7885 12.

Druce MR, Wren AM, Park AJ, Milton JE, Patterson M, Frost G, et al. Ghrelin increases food

intake in obese as well as lean subjects. International Journal of Obesity. 2005;29:1130–6. https://doi.org/10.1038/sj.ijo.0803001 13.

Andrews ZB. Central mechanisms involved in the orexigenic actions of ghrelin. Peptides.

2011;32 (11):2248-55. doi: 10.1016/j.peptides.2011.05.014 14.

Stachowicz M, Lebiedzioska A. The effect of diet components on the level of cortisol.

European Food Research and Technology. 2016;242(12):2001–9. https://doi.org/10.1007/s00217016-2772-3 15.

Torres SJ, Nowson CA. Relationship between stress, eating behavior, and obesity. Nutrition.

2007;23 (11-12):887-94. https://doi.org/10.1016/j.nut.2007.08.008 16.

Perry RJ, Resch JM, Douglass AM, Madara JC, Rabin-Court A, Kucukdereli H, et al. Leptin's

hunger-suppressing effects are mediated by the hypothalamic-pituitary-adrenocortical axis in rodents. Proceedings of the National Academy of Sciences of the United States of America. 2019;116(27):13670-9. doi: 10.1073/pnas.1901795116. 17.

Skinner J, Garg M, Dayas C, Fenton S, Burrows T. What is the relationship between dietary

intake and behaviors with oxytocin: a systematic review of studies in adults. Nutrition Reviews. 2018;76(5):303-31. https://doi.org/10.1093/nutrit/nux078 20

18.

Skinner JA, Campbell EJ, Dayas C, Garg ML, Burrows TL. What is the relationship between

oxytocin, dietary intake and feeding: a systematic review and meta-analysis of studies in mice and rats. Frontiers in Neuroendocrinology. 2019(52):65-78. https://doi.org/10.1016/j.yfrne.2018.09.002 19.

Lawson EA. The effects of oxytocin on eating behaviour and metabolism in humans. Nature

Reviews Endocrinology. 2017;13(12):700–9. doi: 10.1038/nrendo.2017.115 20.

Deblon N, Veyrat-Durebex C, Bourgoin L, Caillon A, Bussier AL, Petrosino S, et al.

Mechanisms of the Anti-Obesity Effects of Oxytocin in Diet-Induced Obese Rats. PLOS ONE 2011;6(9):e25565. https://doi.org/10.1371/journal.pone.0025565 21.

Roberts ZS, Wolden-Hanson T, Matsen ME, Ryu V, Vaughan CH, Graham JL, et al. Chronic

hindbrain administration of oxytocin is sufficient to elicit weight loss in diet-induced obese rats. American Journal of Physiology - Regulatory Integrative and Comparative Physiology. 2017;313(4):R357-R71. https://dx.doi.org/10.1152/ajpregu.00169.2017 22.

Blevins JE, Thompson BW, Anekonda VT, Ho JM, Graham JL, Roberts ZS, et al. Chronic CNS

oxytocin signaling preferentially induces fat loss in high-fat diet-fed rats by enhancing satiety responses and increasing lipid utilization. American Journal of Physiology - Regulatory Integrative and Comparative Physiology. 2016;310(7):R640-58. http://dx.doi.org/10.1152/ajpregu.00220.2015 23.

Blevins JE, Graham JL, Morton GJ, Bales KL, Schwartz MW, Baskin DG, et al. Chronic oxytocin

administration inhibits food intake, increases energy expenditure, and produces weight loss in fructose-fed obese rhesus monkeys. American Journal of Physiology Regulatory, Integrative and Comparative Physiology. 2015;308:R431-8. doi:10.1152/ajpregu.00441.2014 24.

Zhang H, Wu C, Chen Q, Chen X, Xu Z, Wu J, et al. Treatment of Obesity and Diabetes Using

Oxytocin or Analogs in Patients and Mouse Models. PLOS ONE. 2013;8(5):e61477. https://doi.org/10.1371/journal.pone.0061477 25.

Zhang G, Bai H, Zhang H, Dean C, Wu Q, Li J, et al. Neuropeptide Exocytosis Involving

Synaptotagmin-4 and Oxytocin in Hypothalamic Programming of Body Weight and Energy Balance. Neuron. 2011;69(3):523-35. https://doi.org/10.1016/j.neuron.2010.12.036 21

26.

Zhang G, Cai D. Circadian intervention of obesity development via resting-stage feeding

manipulation or oxytocin treatment. American Journal of Physiology - Endocrinology and Metabolism. 2011;301(5):E1004-12. http://dx.doi.org/10.1152/ajpendo.00196.2011 27.

Maejima Y, Iwasaki Y, Yamahara Y, Kodaira M, Sedbazar U, Yada T. Peripheral oxytocin

treatment ameliorates obesity by reducing food intake and visceral fat mass. Aging-Us. 2011;3(12):1169-77. doi: 10.18632/aging.100408 28.

Morton GJ, Thatcher BS, Reidelberger RD, Ogimoto K, Wolden-Hanson T, Baskin DG, et al.

Peripheral oxytocin suppresses food intake and causes weight loss in diet-induced obese rats. American Journal of Physiology - Endocrinology and Metabolism. 2012;302(1):E134-44. http://dx.doi.org/10.1152/ajpendo.00296.2011 29.

Altirriba J, Poher AL, Caillon A, Arsenijevic D, Veyrat-Durebex C, Lyautey J, et al. Divergent

effects of oxytocin treatment of obese diabetic mice on adiposity and diabetes. Endocrinology. 2014;155(11):4189-201. http://dx.doi.org/10.1210/en.2014-1466 30.

Seelke AM, Rhine MA, Khun K, Shweyk AN, Scott AM, Bond JM, et al. Intranasal oxytocin

reduces weight gain in diet-induced obese prairie voles. Physiology and Behavior. 2018;1(196):66-77. doi: 10.1016/j.physbeh.2018.08.007 31.

Plante E, Menaouar A, Danalache BA, Yip D, Broderick TL, Chiasson JL, et al. Oxytocin

treatment prevents the cardiomyopathy observed in obese diabetic male db/db mice. Endocrinology. 2015;156(4):1416-28. doi: 10.1210/en.2014-1718 32.

Labyb M, Chrétien C, Caillon A, Rohner-Jeanrenaud F, Altirriba J. Oxytocin Administration

Alleviates Acute but Not Chronic Leptin Resistance of Diet-Induced Obese Mice. International Journal of Molecular Sciences. 2019;20(1):1-8. doi: 10.3390/ijms20010088 33.

Lee HJ, Macbeth AH, Pagani JH, Scott Young W. Oxytocin: The great facilitator of life.

Progress in Neurobiology. 2009;88(2):127-51. https://doi.org/10.1016/j.pneurobio.2009.04.001

22

34.

Ohlsson B, Forsling ML, Rehfeld JF, Sjölund K. Cholecystokinin stimulation leads to increased

oxytocin secretion in women. European Journal of Surgery. 2002;168(2):114-8. doi: 10.1080/11024150252884340 35.

Verbalis JG, McCann MJ, McHale CM, Stricker EM. Oxytocin secretion in response to

cholecystokinin and food: differentiation of nausea from satiety. Science. 1986;232(4756):1417‐9. doi: 10.1126/science.3715453 36.

Yamashita M, Takayanagi Y, Yoshida M, Nishimori K, Kusama M, Onaka T. Involvement of

prolactin‐releasing peptide in the activation of oxytocin neurones in response to food intake. Journal of Neuroendocrinology. 2013;25(5):455‐65. doi: 10.1111/jne.12019 37.

Johnstone LE, Fong TM, Leng G. Neuronal activation in the hypothalamus and brainstem

during feeding in rats. Cell Metabolism. 2006;4(4):313‐21. doi: 10.1016/j.cmet.2006.08.003 38.

vn s-Moberg K, etersson M. xytocin, a mediator of anti-stress, well-being, social

interaction, growth and healing. eitschrift f r sychosomatische Medizin und Psychotherapie. 2005;51(1):57-80. 39.

Neumann ID, Landgraf R. Balance of brain oxytocin and vasopressin: implications for anxiety,

depression, and social behaviors. Trends in Neurosciences. 2012;35(11):649-59. https://doi.org/10.1016/j.tins.2012.08.004 40.

Verbalis JG, Stricker EM, Robinson AG, Hoffman GE. Cholecystokinin activates C-fos

expression in hypothalamic oxytocin and corticotropin-releasing hormone neurons. Journal of Neuroendocrinology. 1991;3(2). 41.

Thienel M, Fritsche A, Heinrichs M, Peter A, Ewers M, Lehnert H, et al. Oxytocin's inhibitory

effect on food intake is stronger in obese than normal-weight men. International Journal of Obesity. 2016;40(11):1707-14. http://dx.doi.org/10.1038/ijo.2016.149 42.

Maejima Y, Aoyama M, Sakamoto K, Jojima T, Aso Y, Takasu K, et al. Impact of sex, fat

distribution and initial body weight on oxytocin's body weight regulation. Scientific Reports. 2017;7(1):8599. https://dx.doi.org/10.1038/s41598-017-09318-7 23

43.

Mitchell JM, Arcuni PA, Weinstein D, Woolley JD. Intranasal Oxytocin Selectively Modulates

Social Perception, Craving, and Approach Behavior in Subjects With Alcohol Use Disorder. Journal of Addiction Medicine. 2016;10(3):182-9. http://dx.doi.org/10.1097/ADM.0000000000000213 44.

Striepens N, Schroter F, Stoffel-Wagner B, Maier W, Hurlemann R, Scheele D. Oxytocin

enhances cognitive control of food craving in women. Human Brain Mapping. 2016;37(12):4276-85. http://dx.doi.org/10.1002/hbm.23308 45.

Kim YR, Kim CH, Cardi V, Eom JS, Seong Y, Treasure J. Intranasal oxytocin attenuates

attentional bias for eating and fat shape stimuli in patients with anorexia nervosa. Psychoneuroendocrinology. 2014;44:133-42. doi: 10.1016/j.psyneuen.2014.02.019 46.

Amico JA, Vollmer RR, Cai HM, Miedlar JA, Rinaman L. Enhanced initial and sustained intake

of sucrose solution in mice with an oxytocin gene deletion. American Journal of Physiology Regulatory Integrative and Comparative Physiology. 2005;289(6):R1798-806. doi:10.1152/ajpregu.00558.2005 47.

Sclafani A, Rinaman L, Vollmer RR, Amico JA. Oxytocin knockout mice demonstrate enhanced

intake of sweet and nonsweet carbohydrate solutions. American Journal of Physiology - Regulatory Integrative and Comparative Physiology. 2007;292(5):R1828-R33. http://dx.doi.org/10.1152/ajpregu.00826.2006 48.

Miedlar JA, Rinaman L, Vollmer RR, Amico JA. Oxytocin gene deletion mice overconsume

palatable sucrose solution but not palatable lipid emulsions. American Journal of Physiology Regulatory Integrative and Comparative Physiology. 2007;293(3):R1063-8. doi: 10.1152/ajpregu.00228.2007 49.

Billings LB, Spero JA, Vollmer RR, Amico JA. Oxytocin null mice ingest enhanced amounts of

sweet solutions during light and dark cycles and during repeated shaker stress. Behavioural Brain Research. 2006;171(1):134-41. doi:10.1016/j.bbr.2006.03.028 50.

Amico JA, Morris M, Vollmer RR. Mice deficient in oxytocin manifest increased saline

consumption following overnight fluid deprivation. American Journal of Physiology - Regulatory 24

Integrative and Comparative Physiology. 2001;281(5):R1368-73. doi:10.1152/ajpregu.2001.281.5.R1368 51.

Amico JA, Mantella RC, Vollmer RR. Consumption of solutions containing sodium chloride is

enhanced in female oxytocin-deficient mice. Behavioral Neuroscience. 2003;117(1):32-7. http://dx.doi.org/10.1037/0735-7044.117.1.32 52.

Puryear R, Rigatto KV, Amico JA, Morris M. Enhanced salt intake in oxytocin deficient mice.

Experimental Neurology. 2001;171(2):323-8. doi: 10.1006/exnr.2001.7776 53.

Camerino C. Low sympathetic tone and obese phenotype in oxytocin-deficient mice.

Obesity. 2009;17(5):980-4. http://dx.doi.org/10.1038/oby.2009.12 54.

Takayanagi Y, Kasahara Y, Onaka T, Takahashi N, Kawada T, Nishimori K. Oxytocin receptor-

deficient mice developed late-onset obesity. NeuroReport. 2008;19(9):951–5. https://doi.org/10.1097/WNR.0b013e3283021ca9 55.

Lean MEJ, Malkova D. Altered gut and adipose tissue hormones in overweight and obese

individuals: cause or consequence? International Journal of Obesity. 2016;40:622–32. doi: 10.1038/ijo.2015.220 56.

Carlson JJ, Turpin AA, Wiebke G, Hunt SC, Adams TD. Pre- and post- prandial appetite

hormone levels in normal weight and severely obese women. Nutrition and Metabolism. 2009;6(32):1-21. doi: 10.1186/1743-7075-6-32 57.

Lawson EA, Eddy KT, Donoho D, Misra M, Miller KK, Meenaghan E, et al. Appetite-regulating

hormones cortisol and peptide YY are associated with disordered eating psychopathology, independent of body mass index. European Journal of Endocrinology. 2011;164(2):253–61. doi: 10.1530/EJE-10-0523 58.

Incollingo Rodriguez AC, Epel ES, White ML, Standen EC, Seckl JR, Tomiyama AJ.

Hypothalamic-pituitary-adrenal axis dysregulation and cortisol activity in obesity: A systematic review. Psychoneuroendocrinology. 2015;62:301-18. https://doi.org/10.1016/j.psyneuen.2015.08.014 25

59.

Mills JG, Thomasa SJ, LarkinaTA, Paia NB, Denga C. Problematic eating behaviours, changes

in appetite, and weight gain in Major Depressive Disorder: The role of leptin. Journal of Affective Disorders 2018;240:137–45. https://doi.org/10.1016/j.jad.2018.07.069 60.

Stock S, Granström L, Backman L, Matthiesen AS, Uvnäs-Moberg K. Elevated plasma levels of

oxytocin in obese subjects before and after gastric banding. International Journal of Obesity. 1989;13(2):213-22. 61.

Szulc P, Amri EZ, Varennes A, Panaia-Ferrari P, Fontas E, Goudable J, et al. High serum

oxytocin is associated with metabolic syndrome in older men – The MINOS study. Diabetes Research and Clinical Practice. 2016;122:17-27. http://dx.doi.org/10.1016/j.diabres.2016.09.022 62.

Schorr M, Marengi DA, Pulumo RL, Yu E, Eddy KT, Klibanski A, et al. Oxytocin and Its

Relationship to Body Composition, Bone Mineral Density, and Hip Geometry Across the Weight Spectrum. Journal of Clinical Endocrinology and Metabolism. 2017;102(8):2814-24. doi: 10.1210/jc.2016-3963 63.

Breuil V, Panaia-Ferrari P, Fontas E, Roux C, Kolta S, Eastell R, et al. Oxytocin, a new

determinant of bone mineral density in post-menopausal women: analysis of the OPUS Cohort. Journal of Clinical Endocrinology and Metabolism. 2014;99(4):E634–E41. https://doi.org/10.1210/jc.2013-4126 64.

Pataky Z, Guessous I, Caillon A, Golay A, Rohner-Jeanrenaud F, Altirriba J. Variable oxytocin

levels in humans with different degrees of obesityand impact of gastric bypass surgery. International Journal of Obesity. 2019;43:1120–4. https://doi.org/10.1038/s41366-018-0150-x 65.

Maestrini S, Mele C, Mai S, Vietti R, Di Blasio A, Castello L, et al. Plasma Oxytocin

Concentration in Pre- and Postmenopausal Women: Its Relationship with Obesity, Body Composition and Metabolic Variables. Obesity Facts. 2018;11(5):429-39. doi: 10.1159/000492001 66.

Qian W, Zhu T, Tang B, Yu S, Hu H, Sun W, et al. Decreased circulating levels of oxytocin in

obesity and newly diagnosed type 2 diabetic patients. The Journal of Clinical Endocrinology and Metabolism. 2014;99(12):2014-206. doi: 10.1210/jc.2014-2206 26

67.

Yuan G, Qian W, Pan R, Jia J, Jiang D, Yang Q, et al. Reduced circulating oxytocin and High-

Molecular-Weight adiponectin are risk factors for metabolic syndrome. Endocrine Journal. 2016;63(7):655-62. doi: 10.1507/endocrj.EJ16-0078 68.

Coiro V, Passeri M, Davoli C, d'Amato L, Gelmini G, Fagnoni F, et al. Oxytocin response to

insulin-induced hypoglycemia in obese subjects before and after weight loss. Journal of Endocrinological Investigation. 1988;11(2):125-8. doi: 10.1007/BF03350119 69.

Breuil V, Amric E, Panaia-Ferrarid P, Testae J, Elabdc C, Albert-Sabonnadièrea C, et al.

Oxytocin and bone remodelling: Relationships with neuropituitary hormones, bone status and body composition. Joint Bone Spine. 2011;78:611–5. doi: 10.1016/j.jbspin.2011.02.002 70.

Davis C, Moghimi E. ‘ xytocin Deficiency’: Implications for the Food-Addiction Construct.

Current Addiction Reports. 2017;4(2):158–64. doi: 10.1007/s40429-017-0135-1 71.

Bilman E, van Kleef E, van Trijp H. External cues challenging the internal appetite control

system-Overview and practical implications. Critical Reviews in Food Science and Nutrition. 2017;57(13):2825-34. doi: 10.1080/10408398.2015.1073140 72.

Hopkins M, Blundell J, Halford J, King N, Finlayson G. The Regulation of Food Intake in

Humans. [Updated 2016 Mar 30]. In: De Groot LJ, Chrousos G, Dungan K, et al., editors. Endotext [Internet] South Dartmouth (MA): MDTextcom, Inc; 2000-. Available from: https://www.ncbi.nlm.nih.gov/books/NBK278931/ 73.

Ruddock HK, Hardman CA. Food Addiction Beliefs Amongst the Lay Public: What Are the

Consequences for Eating Behaviour? . Current Addiction Reports. 2017;4 110–5 doi: 10.1007/s40429-017-0136-0 74.

Finlayson G. Food addiction and obesity: unnecessary medicalization of hedonic overeating.

Nature Reviews Endocrinology. 2017;13(8):493-8. doi: 10.1038/nrendo.2017.61 75.

Gordon EL, Ariel-Donges AH, Bauman V, Merlo LJ. What Is the Evidence for "Food

Addiction?" A Systematic Review. Nutrients. 2018;10(4):477. doi: 10.3390/nu10040477

27

76.

Hebebrand J, Albayrak O, Adan R, Antel J, Dieguez C, de Jong J, et al. "Eating addiction",

rather than "food addiction", better captures addictive-like eating behavior. Neuroscience and Biobehavioral Reviews. 2014;47:295-306. https://dx.doi.org/10.1016/j.neubiorev.2014.08.016 77.

Gearhardt AN GC, DiLeone RJ, Brownell KD, Potenza MN. . Can food be addictive? Public

health and policy implications. Addiction. 2011;106(7):1208-2 doi: 10.1111/j.13600443.2010.03301.x 78.

Gearhardt AN, Corbin WR, Brownell KD. Development of the Yale Food Addiction Scale

Version 2.0. Psychology of Addictive Behaviors. 2016;30(1):113-21. doi: 10.1037/adb0000136 79.

Burrows T, Kay‐Lambkin F, Pursey K, Skinner J, Dayas C. Food addiction and associations with

mental health symptoms: a systematic review with meta‐analysis. Journal of Human Nutrition and Dietetics. 2018;31(4):544-72. https://doi.org/10.1111/jhn.12532 80.

Pursey K, Stanwell P, Gearhardt A, Collins CE, Burrows TL. The prevalence of food addiction

as assessed by the Yale Food Addiction Scale: a systematic review. Nutrients. 2014;6(10):4552–90. doi: 10.3390/nu6104552 81.

Hendrikse JJ, Cachia RL, Kothe EJ, McPhie S, Skouteris H, Hayden MJ. Attentional biases for

food cues in overweight and individuals with obesity: a systematic review of the literature. Obesity Reviews. 2015;16(5):424-32. doi: 10.1111/obr.12265 82.

Gearhardt AN, Yokum S, Orr PT, Stice E, William R. Corbin WR, Brownell KD. The Neural

Correlates of “Food Addiction”. Archives f General sychiatry. 2011;68 (8):808–16. doi: 10.1001/archgenpsychiatry.2011.32 83.

Blechert J, Meule A, Busch NA, Ohla K. Food-pics: an image database for experimental

research on eating and appetite. Frontiers in Psychology. 5:617. 2014;5(617). https://doi.org/10.1016/j.appet.2015.08.041 84.

Schüssler P, Kluge M, Yassouridis A, Dresler M, Uhr M, Steiger A. Ghrelin levels increase after

pictures showing food. Obesity (Silver Spring). 2012;20(6):1212-7. doi: 10.1038/oby.2011.385

28

85.

Grosshans M, Vollmert C, Vollstädt-Klein S, Tost H, Leber S, Bach P, et al. Association of

Leptin With Food Cue–Induced Activation in Human Reward Pathways. Archives Of General Psychiatry. 2012;69(5):529–37 doi: 10.1001/archgenpsychiatry.2011.1586 86.

Legenbauera T, Vögeleb C, Rüddel H. Anticipatory effects of food exposure in women

diagnosed with bulimia nervosa. Appetite. 2004;42(1):33-40. https://doi.org/10.1016/S01956663(03)00114-4 87.

Weisman O, Zagoory-Sharon O, Schneiderman I, Gordon I, Feldman R. Plasma oxytocin

distributions in a large cohort of women and men and their gender-specific associations with anxiety. Psychoneuroendocrinology. 2013;38(5):694—701. http://dx.doi.org/10.1016/j.psyneuen.2012.08.011 88.

Chao AM, Loughead J, Bakizada ZM, Hopkins CM, Geliebter A, Gur RC, et al. Sex/gender

differences in neural correlates of food stimuli: a systematic review of functional neuroimaging studies. Obesity Reviews. 2017;18:687–99. doi: 10.1111/obr.12527 89.

Burrows T, Hides L, Brown R, Dayas CV, Kay-Lambkin F. Differences in Dietary Preferences,

Personality and Mental Health in Australian Adults with and without Food Addiction. Nutrients. 2017;9 (3):E285. doi: 10.3390/nu9030285 90.

Dallal GE. Randomization.com 2013 [Available from: http://www.randomization.com/.

91.

Romano A, Tempesta B, Micioni Di Bonaventura MV, Gaetani S. From Autism to Eating

Disorders and More: The Role of Oxytocin in Neuropsychiatric Disorders. Frontiers in Neuroscience. 2016;9:1-19. doi: 10.3389/fnins.2015.00497 92.

Szeto A, McCabe PM, Nation DA, Tabak BA, Rossetti MA, McCullough ME, et al. Evaluation of

enzyme immunoassay and radioimmunoassay methods for the measurement of plasma oxytocin. Psychosomatic Medicine. 2011;73 (5):393-400. doi: 10.1097/PSY.0b013e31821df0c2 93.

World Health Organization (WHO). Body mass index - BMI: WHO Regional Office for Europe;

c2018 [cited 2018 September 22]. Available from: http://www.euro.who.int/en/healthtopics/disease-prevention/nutrition/a-healthy-lifestyle/body-mass-index-bmi. 29

94.

American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 5th

ed. Arlington: VA American Psychiatric Publishing; 2013. 95.

Collins CE, Boggess MM, Watson JF, Guest M, Duncanson K, KristinePezdirc K, et al.

Reproducibility and comparative validity of a food frequency questionnaire for Australian adults. Clinical Nutrition. 2014;33(5):906-14. doi: 10.1016/j.clnu.2013.09.015 96.

Ashton L, Williams R, Wood L, Schumacher T, Burrows T, Rollo M, et al. Comparison of

Australian Recommended Food Score (ARFS) and Plasma Carotenoid Concentrations: A Validation Study in Adults. Nutrients. 2017;9(8):888. http://doi.org/10.3390/nu9080888 97.

Free digital photos [internet]. AgileBit Limited: Hong Kong; c2017 [cited 2016, November.

Available from: http://www.freedigitalphotos.net/. 98.

Pixabay [internet]. Germany: Hans Braxmeier & Simon Steinberger GbR; c2017 [cited 2016,

November. Available from: https://pixabay.com/. 99.

Wikimedia Commons [internet]. USA: Wikimedia Foundation Inc.; c2017 [cited 2016,

November. Available from: https://commons.wikimedia.org. 100.

Bing [internet]. USA: Microsoft; c2017 [cited 2016, November]. Available from:

https://www.bing.com/. 101.

National Health and Medical Research Council (NHMRC). Eat for Health Educator Guide

2013. Available from: https://www.eatforhealth.gov.au/sites/default/files/files/the_guidelines/n55b_eat_for_health_edu cators_guide.pdf. 102.

StataCorp. Stata Statistical Software: Release 15. College Station, TX: StataCorp LLC. 2017.

103.

Kujath AS, Quinn L, Elliott ME, Varady KA, LeCaire TJ, Carter CS, et al. Oxytocin levels are

lower in premenopausal women with type 1 diabetes mellitus compared with matched controls. Diabetes/metabolism research and reviews. 2015;31:102–12. doi: 10.1002/dmrr.2577

30

104.

Breuila V, Amric E, Panaia-Ferrarid P, Testae J, Elabdc C, Albert-Sabonnadièrea C, et al.

Oxytocin and bone remodelling: Relationships with neuropituitary hormones, bone status and body composition. Joint Bone Spine. 2011;78:611–5. doi:10.1016/j.jbspin.2011.02.002 105.

Pedram P, Sun G. Hormonal and dietary characteristics in obese human subjects with and

without food addiction. Nutrients. 2015;7 223–38. doi: 10.3390/nu7010223 106.

Davis C, Patte K, Zai C, Kennedy JL. Polymorphisms of the oxytocin receptor gene and

overeating: the intermediary role of endophenotypic risk factors. Nutrition and Diabetes. 2017;7 (5):e279. doi: 10.1038/nutd.2017.24 107.

McCullough ME, Churchland PS, Mendezc AJ. Problems with measuring peripheral oxytocin:

can the data on oxytocin and human behavior be trusted? Neuroscience and Biobehavioral Reviews. 2013;37:1485–92. http://dx.doi.org/10.1016/j.neubiorev.2013.04.018

Table 1. Fasting baseline plasma concentrations of appetite hormones by BMI categories Healthy BMI (n=6)

Overweight BMI (n=9)

Obese BMI (n=3)

P values

H vs O/W

H vs O

O/W vs O

Overall

Log oxytocin

7.39 ± 0.21

7.34 ± 0.27

7.92 ± 0.39

0.999

0.029*

0.017*

0.019*

Log ghrelin

7.82 ± 0.42

7.74 ± 0.72

7.63 ± 0.29

-

-

-

0.903

Log leptin

6.32 ± 0.45

6.45 ± 0.45

6.53 ± 0.94

-

-

-

0.843

Log CCK

6.86 ± 0.39

6.50 ± 0.52

6.23 ± 0.17

-

-

-

0.143

Log cortisol

4.35 ± 2.27

4.57 ± 1.01

5.94 ± 0.47

-

-

-

0.331

Mean ± SD. ANOVA and Bonferroni post hoc analysis; *p<0.05; - post hoc analysis not conducted on nonsignificant (p>0.05) ANOVA. BMI, Body Mass Index; CCK, cholecystokinin.

31

Table 2. Diet quality in YFAS-defined food addicted vs non-food addicted women Group by FA Diagnosis NFA (n=12) FA (n=6) p value

Diet quality score (ARFS) 39.00 ± 6.51 (34.86, 43.14) 33.33 ± 6.92 (26.07, 40.59) 0.107

Core foods (%E) 70.25 ± 11.05 (63.23, 77.27) 51.5 ± 21.75 (28.67, 74.33) 0.026*

Non-core foods (%E) 29.75 ± 11.05 (22.73, 36.77) 48.5 ± 21.75 (25.67, 71.33) 0.026*

Mean ± SD (95% Confidence Interval). Two sample independent t-test, *p>0.05. ARFS, Australian recommended dietary score; NFA, non-food addicted; FA, food addicted.

Table 3. Fasting baseline plasma concentrations of appetite hormones by food addiction status

Oxytocin (pg/mL, log transformed) Ghrelin (pg/mL, log transformed) Leptin (pg/mL, log transformed) Cholecystokinin (pg/mL, log transformed) Cortisol (ng/mL, log transformed)

FA (n=6) 7.71 ± 0.34 ( 7.16, 7.49) 7.87 ± 0.35 ( 7.28, 8.09) 6.56 ± 0.62 (6.05, 6.65) 6.08 ± 0.21 (6.59, 7.05) 5.50 ± 0.63 (3.25, 5.43)

NFA (n=12) 7.33 ± 0.26 (7.36, 8.06) 7.69 ± 0.64 ( 7.50, 8.23) 6.35 ± 0.47 ( 5.92, 7.21) 6.82 ± 0.36 (5.86, 6.30) 4.34 ± 1.72 (4.84, 6.16)

p values 0.015* 0.528 0.423 <0.001** 0.134

Mean ± SD (95% confidence interval). Two sample independent t test; *p<0.05, **p<0.001. FA, food addicted; NFA, non-food addicted.

32

Table 4. Median appetite scores pre and post food image viewing for the total sample (n=18) VAS component Hunger Amount could eat Desire to eat healthy food Desire to eat sweet food Desire to eat salty food Desire to eat savoury food Desire to eat fatty food

Pre VAS score 42.5 (21, 60) 49.5 (34, 65) 77 (57, 92) 37 (17, 55) 40.5 (26, 65) 70 (50, 81) 18.5 (10, 27)

Post VAS score 72.5 (55, 78) 72 (51, 82) 79 (70, 91) 51 (37, 80) 62.5 (37, 75) 78 (61, 90) 37 (9, 72)

% increase 70.6 45.5 2.6 37.8 54.3 11.4 100

p-value 0.001* 0.009* 0.695 0.019* 0.191 0.248 0.045*

Median and interquartile range. One sample Wilcoxon signed-rank test on the pair differences; *p<0.05. VAS, visual analogue scale.

33

Healthy food images

Group

Total sample (n=18) BMI category Healthy weight (n=6) Overweight (n=9)

Obese (n=3)

Food Addiction status NFA (n=12)

FA (n=6)

Hyperpalatable food images

Log Log Mean P oxytocin oxytocin difference value (pg/mL) (pg/mL)post pre viewing viewing 7.38 ± 7.29 ± 0.43 -0.09 ± 0.333 0.38 (7.08, 7.51) 0.36 ((7.19, 0.27, 7.57) 0.10)

7.10 ± 0.24 (6.85, 7.36) 7.39 ± 0.28 (7.18, 7.60) 7.91 ± 0.39 (6.96, 8.87)

7.10 ± 0.19 (6.90, 7.30)

7.22 ± 0.30 (7.03, 7.40) 7.71 ± 0.34 (7.36, 8.06)

7.07 ± 0.31 (6.87, 7.27)

7.20 ± 0.41 (6.88, 7.51)

7.96 ± 0.04 (7.87, 8.05)

7.74 ± 0.25 (7.48, 8.00)

Log oxytoci n (pg/mL) pre viewing

Log oxytoci n (pg/mL) post viewing

7.39 ± 0.35 (7.22, 7.57)

-0.002 ± 0.21 (0.22, 0.22) -0.19 ± 0.45 (0.53, 0.15) 0.05 ± 0.35 (0.83, 0.93)

0.986

7.30 ± 0.28 (7.01, 7.59)

0.241

7.27 ± 0.23 (7.09, 7.45)

0.833

7.96 ± 0.26 (7.31, 8.60)

-0.14 ± 0.41 (0.41, 0.12) 0.03 ± 0.23 (0.21, 0.27)

0.253

7.22 ± 0.24 (7.07, 7.38)

0.758

7.74 ± 0.30 (7.42, 8.05)

Log Mean P oxytocin difference value (pg/mL) post viewing 7.38 ± -0.01 ± 0.856 0.45 0.20 ((7.16, 0.11 7.61) 0.01)

7.17 ± 0.34 (6.82, 7.53) 7.32 ± 0.37 (7.03, 7.61) 8.01 ± 0.37 (7.10, 8.92)

-0.13 ± 0.17 (0.30, 0.05) 0.05 ± 0.21 (0.12, 0.21) 0.05 ± 0.12 (0.23, 0.34)

0.126

7.18 ± 0.32 (6.97, 7.38) 7.80 ± 0.38 (7.40, 8.20)

-0.05 ± 0.20 (0.17, 0.08) 0.07 ± 0.18 (0.12, 0.26)

0.435

0.510

0.503

0.402

Hyperpalatable food images

Healthy food images

Group

Log oxytocin (pg/mL) pre viewing

Mean differenc e

P value

Log oxytoci n (pg/mL) pre viewing

Log oxytoci n (pg/mL) post viewing

Mean differenc e

P value

34

7.38 ± 0.38 (7.19, 7.57)

7.29 ± 0.43 (7.08, 7.51)

-0.09 ± 0.36 (0.27, 0.10)

7.10 ± 0.24 (6.85, 7.36) 7.39 ± 0.28 (7.18, 7.60) 7.91 ± 0.39 (6.96, 8.87)

7.10 ± 0.19 (6.90, 7.30) 7.20 ± 0.41 (6.88, 7.51) 7.96 ± 0.04 (7.87, 8.05)

-0.002 ± 0.21 (0.22, 0.22) -0.19 ± 0.45 (0.53, 0.15) 0.05 ± 0.35 (0.83, 0.93)

Food Addiction status 7.22 ± 0.30 NFA (n=12) (7.03, 7.40) 7.71 ± 0.34 FA (n=6) (7.36, 8.06)

7.07 ± 0.31 (6.87, 7.27) 7.74 ± 0.25 (7.48, 8.00)

-0.14 ± 0.41 (0.41, 0.12) 0.03 ± 0.23 (0.21, 0.27)

Total sample (n=18)

0.33 3

7.39 ± 0.35 (7.22, 7.57)

7.38 ± 0.45 (7.16, 7.61)

-0.01 ± 0.20 (0.11 0.01)

7.30 ± 0.28 (7.01, 7.59) 7.27 ± 0.23 (7.09, 7.45) 7.96 ± 0.26 (7.31, 8.60)

7.17 ± 0.34 (6.82, 7.53) 7.32 ± 0.37 (7.03, 7.61) 8.01 ± 0.37 (7.10, 8.92)

-0.13 ± 0.17 (0.30, 0.05) 0.05 ± 0.21 (0.12, 0.21) 0.05 ± 0.12 (0.23, 0.34)

7.22 ± 0.24 (7.07, 7.38) 7.74 ± 0.30 (7.42, 8.05)

7.18 ± 0.32 (6.97, 7.38) 7.80 ± 0.38 (7.40, 8.20)

-0.05 ± 0.20 (0.17, 0.08) 0.07 ± 0.18 (0.12, 0.26)

0.85 6

BMI category Healthy weight (n=6) Overweigh t (n=9)

Obese (n=3)

0.98 6 0.24 1

0.83 3

0.25 3

0.75 8

0.12 6

0.51 0

0.50 3

0.43 5

0.40 2

Mean ± SD (95% confidence interval). Paired t test. BMI, Body Mass Index. FA, food addicted. NFA, non-food addicted.

Table 5. Mean changes in plasma oxytocin concentrations in response to viewing healthy and hyperpalatable food images

35