- Email: [email protected]
Phoenixin is negatively associated with anxiety in obese men
Peptides 88 (2017) 32–36
Contents lists available at ScienceDirect
Peptides journal homepage: www.elsevier.com/locate/peptides
Research Paper
Phoenixin is negatively associated with anxiety in obese men Tobias Hofmann a , Elena Weibert a , Anne Ahnis a , Ulf Elbelt a,b , Matthias Rose a , Burghard F. Klapp a , Andreas Stengel a,∗ a
Charité Center for Internal Medicine and Dermatology, Department for Psychosomatic Medicine; Charité-Universitätsmedizin Berlin, Berlin, Germany Charité Center for Internal Medicine with Gastroenterology and Nephrology, Department for Endocrinology, Diabetes and Nutrition, Charité-Universitätsmedizin Berlin, Berlin, Germany b
a r t i c l e
i n f o
Article history: Received 3 September 2016 Received in revised form 3 December 2016 Accepted 14 December 2016 Available online 15 December 2016 Keywords: Emotion Gut-brain axis Psychobiology Psychoneuroendocrine Psychosomatic
a b s t r a c t Phoenixin was recently identified in the rat hypothalamus and initially implicated in reproductive functions. A subsequent study described an anxiolytic effect of the peptide. The aim of the study was to investigate a possible association of circulating phoenixin with anxiety in humans. We therefore enrolled 68 inpatients with a broad spectrum of psychometrically measured anxiety (GAD-7). We investigated men since a menstrual cycle dependency of phoenixin has been assumed. Obese subjects were enrolled since they often report psychological comorbidities. In addition, we also assessed depressiveness (PHQ9) and perceived stress (PSQ-20). Plasma phoenixin levels were measured using a commercial ELISA. First, we validated the ELISA kit performing a spike-and-recovery experiment showing a variance of 6.7 ± 8.8% compared to the expected concentrations over the whole range of concentrations assessed, while a lower variation of 1.6 ± 0.8% was observed in the linear range of the assay (0.07–2.1 ng/ml). We detected phoenixin in the circulation of obese men at levels of 0.68 ± 0.50 ng/ml. These levels showed a negative association with anxiety scores (r = −0.259, p = 0.043), while no additional associations with other psychometric parameters were observed. In summary, phoenixin is present in the human circulation and negatively associated with anxiety in obese men, a population often to report comorbid anxiety. © 2016 Elsevier Inc. All rights reserved.
1. Introduction Phoenixin is a recently discovered 20-amino acid peptide described to have regulatory properties in the reproductive system [22] likely via the G-protein coupled receptor, the GPR173, as recently suggested [20]. Also a fragment of the peptide, phoenixin14, retained its biological activity [22]. This modulating effect was exerted via increasing the gonadotropin-releasing hormone (GnRH) receptor expression in the pituitary in vitro [22]. Pointing towards a physiological role of endogenous phoenixin, the blockade of the peptide using small interfering RNA injected intracerebroventricularly delayed the onset of estrus in female rats associated with a reduction of GnRH receptor expression [22]. Further suggesting an important role of phoenixin, the peptide sequence of phoenixin is highly conserved across non-mammalian
∗ Corresponding author at: Charité Center for Internal Medicine and Dermatology, Department for Psychosomatic Medicine, Charité-Universitätsmedizin Berlin, Hindenburgdamm 30, 12203 Berlin, Germany. E-mail address: [email protected] (A. Stengel). http://dx.doi.org/10.1016/j.peptides.2016.12.011 0196-9781/© 2016 Elsevier Inc. All rights reserved.
(fish, amphibians, chicken) and mammalian species (rodents, human) [22]. The initial landmark study identified phoenixin in several rat peripheral (e.g. heart, thymus, esophagus, stomach and spleen) and central tissues with highest peptide expression levels in the hypothalamus [22]. Immunostaining indicated the expression of phoenixin in the magnocellular and parvocellular parts of the paraventricular nucleus, supraoptic nucleus, arcuate nucleus, dorsal hypothalamus, zona incerta, ventromedial hypothalamus, lateral hypothalamus and the perifornical area [22]. Additional central expression was shown in the substantia nigra, Edinger-Westphal nucleus, nucleus of the solitary tract and the dorsal motor nucleus of the vagus nerve [22] pointing towards additional functions besides the initially proposed role in the reproductive system. A subsequent study detected even higher peptide levels in the spinal cord compared to the hypothalamus and described a predominant localization in sensory neurons of the dorsal root, nodose and trigeminal ganglia [11] giving rise to a role in pain processing. This hypothesis is supported by the observation of a reduced number of writhes following intraperitoneal injection of acetic acid in mice receiving a central injection of phoenixin [11].
T. Hofmann et al. / Peptides 88 (2017) 32–36
Following these studies, a recent report further described the phenotype of phoenixin neurons indicating a large proportion (70–86%) of phoenixin neurons co-localizing with NUCB2/nesfatin1 in rat hypothalamic nuclei (arcuate nucleus, paraventricular nucleus, ventromedial hypothalamus and lateral hypothalamus) [15]. NUCB2/nesfatin-1 was discovered a decade ago and first localized in the rat hypothalamus [14]. It is to note that early on an anxiogenic effect of NUCB2/nesfatin-1 has been described in male rats following intracerebroventricular injection of the peptide [12]. Interestingly, also in humans an association of NUCB2/nesfatin1 and anxiety has been reported with a positive correlation in women [6] and a negative association in men [4,5] indicating a sexspecific regulation of the peptide. In contrast to NUCB2/nesfatin-1, phoenixin was recently shown to exert potent anxiolytic effects in male mice [7]. However, whether phoenixin is also involved in the mediation, modulation or perception of anxiety in humans is unknown so far. Therefore, the aim of the present study was to investigate whether phoenixin is associated with anxiety in humans. Based on the involvement of the peptide in reproductive functions [22] a menstrual cycle dependency can be assumed (or at least cannot be excluded at this point). Therefore, only men were investigated in the present study. Moreover, since obese subjects often report comorbid anxiety [2] and depression [21] we investigated obese men displaying a broad range of psychopathology in the present study. We first validated the commercial phoenixin ELISA kit and then investigated the potential association between circulating phoenixin and psychometrically measured anxiety levels. Subsequently, we further characterized our study population with regards to depressiveness and perceived stress. 2. Materials and methods 2.1. Subjects For this study, we enrolled male obese inpatients (from January 2011 to December 2014) hospitalized in the Department for Psychosomatic Medicine at Charité-Universitätsmedizin Berlin that received medical treatment for obesity and its somatic and mental comorbidities. The inclusion criteria were comprised of male sex, an age of ≥18 years and a body mass index (BMI) ≥ 0 kg/m2 . Patients with a current malignant disease, untreated psychotic disorders and preceding bariatric surgery were not enrolled in the study. In addition, patients with hypercortisolism or hypothyroidism were excluded from the study. The final number of patients analyzed in this study was n = 68. All investigations in the present study were conducted according to the Declaration of Helsinki and all patients gave written informed consent. The study was approved by the institutional ethics committee of Charité-Universitätsmedizin Berlin (protocol number: EA1/114/10). 2.2. Laboratory analyses Blood samples were obtained from patients within three days after hospital admission to prevent an influence of treatment on metabolic/hormonal and psychometric parameters. All blood samples were taken after an overnight fast between 07:00 and 08:00 in the morning. Patients were allowed to drink small amounts of water but were advised not to smoke or exercise in the morning (before blood withdrawal). After withdrawal from a forearm vein, venous blood was collected in pre-cooled standard EDTA tubes containing aprotinin (1.2 Trypsin Inhibitory Unit/1 ml blood; ICN Pharmaceuticals, Costa Mesa, CA, USA) as peptidase inhibitor. The EDTA tubes were stored on ice directly after blood withdrawal and
33
centrifuged for 10 min at 3000 rpm at 4 ◦ C. Plasma was separated and stored at −80 ◦ C until further processing. Phoenixin levels were determined using a commercial enzymelinked immunosorbent assay (ELISA, linear range 0.07-2.1 ng/ml, catalog # EK-079-01, Phoenix Pharmaceuticals, Inc., Burlingame, CA, USA) per the manufacturer’s instructions. First, to validate the kit, a spike-and-recovery experiment was performed adding synthetic phoenixin-14 amide (079-01, Phoenix Pharmaceuticals) in increasing concentrations to pooled human plasma samples. Recovery was assessed and measured levels compared to the expected concentrations. All plasma samples were processed in one batch; the intra-assay variability was 9%. The antibody used in this ELISA was generated in rabbit and recognizes both human phoenixin-14 as well as phoenixin-20 with 100% cross-reactivity (cross-reacting also with rat, mouse, bovine, porcine and canine phoenixin; manufacturer’s information). No cross-reactivity was observed for adrenomedullin, alpha-atrial natriuretic polypeptide, angiotensin I and II, apelin-12, bradykinin, brain natriuretic peptide-32, gonadotropin releasing hormone, neuropeptide Y, orexin A, somatostatin and TLQP-21 (manufacturer’s information). The concentrations of circulating phoenixin described in the present manuscript were well within the linear detection range of the kit (0.07–2.1 ng/ml) and the positive control using synthetic phoenixin-14 was detectable in the expected range of 0.2–0.5 ng/ml. 2.3. Anthropometric measurements For calculation of the BMI (as kg/m2 ), body weight and height of the participants were assessed at the same day of blood withdrawal between 07:00 and 08:00 in the morning in patients wearing light underwear. 2.4. Psychometric parameters All psychometric parameters were obtained electronically. The patients were given electronic devices on the same day or the day before the blood withdrawal and were asked to fill in the following questionnaires within one day. Patients without psychometric data within six days after blood sampling were excluded. For psychometric assessment two modules of the selfadministered patient health questionnaire (PHQ) [17] were used in order to assess depression (PHQ-9) and general anxiety (GAD-7). The PHQ-9 depression module is an established screening instrument for the diagnosis of major depression and evaluation of the severity of depressive symptoms. The PHQ-9 contains nine items representing the diagnostic criteria for DSM-IV depressive disorders. Each of them can be scored as “0” (not at all) to “3” (nearly every day) with a maximum of 27 points [17]. In a meta-analysis of 36 studies (21,292 patients) the PHQ-9 showed a specificity of 0.87 in detecting major depressive disorders [13]. Our patients were handed out the German version of the PHQ-9 [10]. For the current population Cronbach’s alpha was calculated as 0.89. For the examination of anxiety, the GAD-7 scale was used. GAD-7 is an efficient tool to identify generalized anxiety disorder, posttraumatic stress disorder, panic disorder as well as social anxiety disorder and to measure the severity of symptoms [18]. It consists of seven items with scores ranging from “0” (not at all) to “3” (nearly every day) and a maximum of 21 points [18]. In our study the German version was used [9]. In a recent meta-analysis of 12 studies (5223 patients) specificity was calculated with 0.84 and sensitivity with 0.83 at a cut-off point of eight [16]. In the present sample Cronbach’s alpha was 0.90. In order to assess stress the perceived stress questionnaire (PSQ) [8] was used in its revised German version with 20 items (PSQ-20) [1]. The PSQ-20 examines the level of subjectively perceived exac-
34
T. Hofmann et al. / Peptides 88 (2017) 32–36
Table 1 Demographic and socioeconomic characteristics as well as comorbidities and medication of study population (n = 68). Parameter
Mean ± SD or total number
Missing data
Demographic characteristics Age (years) BMI (kg/m2 )
46.0 ± 13.2 48.2 ± 9.8
0 (0.0%) 0 (0.0%)
26 (38.3%)
1 (1.5%) 1 (1.5%)
Socioeconomic characteristics Living in a partnership Level of education - university entrance diploma (‘Abitur’) - vocational diploma (‘Fachabitur’) - secondary education certificate (‘Mittlere Reife’) - basic school qualification (‘Hauptschulabschluss’) - without school-leaving qualification Current employment Unemployment during past 5 years
16 (23.5%) 3 (4.4%) 24 (35.3%) 19 (27.9%) 5 (7.4%) 20 (29.4%) 32 (47.1%)
1 (1.5%) 1 (1.5%) 0 (0.0%)
Comorbidities Sleep-related breathing disorder Type 1 diabetes mellitus Type 2 diabetes mellitus Impaired fasting glucose (without DM 2) Insulin resistance (without DM 2 or IFG) Arterial hypertension Hypertriglyceridemia Hyperuricemia Coronary heart disease Polyarthrosis Chronic back pain
43 (63.2%) 1 (1.5%) 24 (35.3%) 10 (14.7%) 15 (22.1%) 53 (77.9%) 24 (35.3%) 33 (48.5%) 6 (8.8%) 8 (11.8%) 15 (22.1%)
Medication Insulin DPP-4 antagonists/GLP-1 analogues Other antidiabetics Psychopharmacological treatment Opioids Steroids
16 (23.5%) 3 (4.4%) 17 (25.0%) 14 (20.6%) 1 (1.5%) 2 (2.9%)
0 (0.0%)
Abbreviations: DM 2, type 2 diabetes mellitus; DPP-4, dipeptidylpeptidase 4; GLP-1, glucagon like peptide 1; IFG, impaired fasting glucose; SD, standard deviation.
erbating factors and experienced stress. The four factors “worries”, “tension”, “joy” as stress reactions and “demands” as the perception of external stressors are evaluated with five questions each. The sum scores were added and converted into percentage values. Calculated mean values between 0 and 1 were converted to scores ranging from 0 to 100. Cronbach’s alpha for the PSQ total score assessed in the present study population was 0.75 and ranged from 0.83 to 0.90 for the four subscales.
2.5. Statistical analysis All data are expressed as mean ± standard deviation (SD). Distribution of the data was determined by the Kolmogorov-Smirnov test and differences between groups were calculated using t-tests. Depending on the distribution of the data correlations were determined by Pearson’s or Spearman’s analyses. A multiple stepwise linear regression analysis was performed to identify potential confounding parameters (comorbidities, medication). Differences between groups and correlations were considered significant when p < 0.05. All statistical analyses were conducted using SigmaStat 3.1 (Systat Software, San Jose, CA, USA).
3. Results 3.1. Demographic, socioeconomic and medical characteristics Demographic, socioeconomic characteristics as well as comorbidities and medication of the study population are described in Table 1.
3.2. Psychometric characteristics The study population of obese men (n = 68) displayed a broad range of psychopathology. With regards to anxiety, scores ranged from 0 to 19 (maximum score of the test: 21, Table 2). On average, patients reported mild anxiety symptoms (mean ± SD: 6.1 ± 5.3, Table 2). Also depressiveness (range: 0–22, maximum score of the test: 27, mean ± SD: 8.1 ± 6.0) and perceived stress (range of total score: 3–95, maximum score of the test: 100, mean ± SD: 43.2 ± 22.0) showed a broad distribution within the study population (Table 2).
Table 2 Mean scores and ranges of psychometric variables and their correlation with circulating phoenixin levels in a population of obese men (n = 68). Parameter
Mean ± SD
Range
GAD-7 PHQ-9 PSQ total - worries - tension - joy - demands
6.1 ± 5.3 8.1 ± 6.0 43.2 ± 22.0 38.7 ± 25.5 45.6 ± 27.6 46.8 ± 24.9 35.4 ± 24.7
0–19 0–22 3–95 0–100 0–100 0–100 0–100
Correlation with phoenixin r
p
−0.259 −0.140 −0.169 −0.123 −0.230 0.096 −0.107
0.043 0.291 0.194 0.344 0.075 0.460 0.414
Distribution of the data was determined by the Kolmogorov-Smirnov test. Depending on the distribution of the data, correlations were determined by Pearson’s or Spearman’s analyses. Significant associations are shown in bold. Abbreviations: GAD-7, Generalized Anxiety Disorder questionnaire; PHQ-9, Perceived Health Questionnaire; PSQ-20, Perceived Stress Questionnaire.
T. Hofmann et al. / Peptides 88 (2017) 32–36
Phoenixin (ng/ml)
1.5
r = -0.259 p = 0.043
1.0
0.5
0.0 0
10 GAD-7 score
20
Fig. 1. GAD-7 scores in obese men and the correlation with circulating phoenixin levels. Plasma phoenixin levels showed a negative correlation with the GAD-7 scores. Values for r and p are indicated in the correlation graph. Abbreviations: GAD-7, generalized anxiety disorder questionnaire (7 items).
3.3. Phoenixin levels are negatively associated with anxiety in obese men The spike-and-recovery experiment displayed a variance in recovery of 6.7 ± 8.8% over the whole range of concentrations assessed (0.1–20 ng/ml), while a lower variation of 1.6 ± 0.8% was observed in the linear range of the assay (0.07–2.1 ng/ml; data not shown). Therefore, the kit was considered valid for the assessment of phoenixin-14 in human plasma. In the study population of obese men, circulating phoenixin levels showed a negative correlation with reported anxiety (GAD-7 levels, r = −0.259, p = 0.043; Fig. 1, Table 2). A stepwise multiple linear regression analysis with obesity-associated comorbidities as possible confounders showed that GAD scores were able to predict phoenixin plasma levels (p = 0.028), whereas sleep-related breathing disorder (p = 0.411), type 1 diabetes mellitus (p = 0.623), type 2 diabetes mellitus (p = 0.139), impaired fasting glucose (p = 0.943), insulin resistance (p = 0.505), arterial hypertension (p = 0.272), hypertriglyceridemia (p = 0.498), hyperuricemia (p = 0.655), coronary heart disease (p = 0.809), polyarthrosis (p = 0.387) and chronic back pain (p = 0.718) were not. Similarly, a stepwise multiple linear regression analysis testing medication as possible confounders showed a significant association between phoenixin levels by GAD (p = 0.031), while none with insulin (p = 0.078), DPP-4 antagonists/GLP-1 analogues (p = 0.442), other antidiabetics (p = 0.844), psychopharmacological treatment (p = 0.539), opioids (p = 0.325) and steroids (p = 0.766). Using median split of the data we divided the study sample per GAD-7 scores and obtained a group with minimal anxiety (2.0 ± 1.7) and one with moderate anxiety (10.2 ± 4.4, p < 0.001; n = 34/group). Phoenixin levels were 25% higher in the minimal anxiety group (0.78 ± 0.59 ng/ml) compared to the moderate anxiety group (0.59 ± 0.41 ng/ml); however, this difference missed statistical significance (p = 0.068). Calculation of the effect size indicated a Cohen’s d of 0.37. All other psychometric parameters assessed (depressiveness and perceived stress) did not show statistically significant associations with plasma phoenixin concentrations (Table 2). Lastly, no association was observed between circulating phoenixin levels and age (r = −0.211, p = 0.103) or body mass index (r = 0.131, p = 0.319). 4. Discussion In the present study we detected the novel peptide phoenixin, that was recently identified in the rat brain [22], in the human
35
circulation with levels ∼0.7 ng/ml. Furthermore, we showed that phoenixin levels were negatively associated with psychometrically measured anxiety in obese men with higher levels under conditions of low anxiety. Phoenixin has been reported to exert an anxiolytic effect in mice as indicated by more time spent in the open arms of the elevated plus maze and more entries into the center of the open field following lateral brain ventricle injection or microinjection of phoenixin into the anterior hypothalamic area, an effect blocked by a GnRH receptor antagonist [7] pointing towards an involvement of downstream GnRH signaling in the mediation of phoenixin’s anxiolytic effect. Based on these animal data one might speculate that the higher phoenixin levels observed in subjects with minimal anxiety contribute to these low anxiety levels. Interestingly, no additional correlations of phoenixin with other psychometric characteristics of the present obese patient population, namely depressiveness and perceived stress, were observed here, giving rise to the assumption of a specific association of phoenixin and anxiety. The recent observation of an extensive co-localization (>70%) of phoenixin with the pleiotropic peptide NUCB2/nesfatin-1 in the rat hypothalamus [15] attracted attention, especially in light of the finding that both peptides were implicated in the modulation of anxiety in rodents. While NUCB2/nesfatin-1 induced anxiogenic behavior in male rats [12], phoenixin was reported to exert an anxiolytic effect in male mice [7]. NUCB2/nesfatin-1 was associated with anxiety in humans as well with a positive association in women [6], while a negative association was observed in men [5]. In the present study, we also showed a negative association of phoenixin and anxiety in men. Whether NUCB2/nesfatin-1 has anxiogenic effects, while phoenixin exerts an anxiolytic effect in human subjects as well will have to be further investigated. Since this is the first investigation of phoenixin levels in humans, several limitations should be considered. First, the study was only performed in men as menstrual cycle-dependent alterations of phoenixin have been assumed. Moreover, a sex-specific regulation of phoenixin can be hypothesized as well. These points − along with a possible sex steroid dependency – should be further investigated in future studies. Second, in the present study only obese subjects were included due to the high rate of psychological comorbidities [2,21] and in order to follow up on our previous study in obese men [5]. Although suited for the current approach, further studies should also describe phoenixin levels in a healthy population and in normal weight subjects with anxiety disorders. Third, the negative association between phoenixin and anxiety described here does not allow to draw conclusions about cause and consequence. For this purpose, a follow-up study using a longitudinal design will have to investigate whether phoenixin levels increase during/following successful treatment of anxiety symptoms. Fourth, the association between phoenixin and anxiety levels was rather moderate; whether this holds true in larger samples will have to be investigated. Moreover, since the major aim of the study was to investigate the association of phoenixin and anxiety, and all other psychometric parameters (depressiveness and perceived stress) have been mainly included to further characterize the study population, no Bonferroni correction has been performed. Fifth, we did not examine circulating phoenixin levels in relation to fixed diagnoses such as generalized anxiety disorder and instead used the results of selfassessment questionnaires which might be impaired by incorrect self-report caused by recall bias, social acceptability or difficulties in self-observation. Nevertheless, these psychometric assessment tools used in the present study are well validated and established screening instruments [3,8,17–19]. Lastly, the ELISA kit used to detect phoenixin plasma levels was new and has not been used before in humans. Therefore, we performed a spike-and-recovery experiment showing a variation of less than 2% from the expected recovery in the linear range of the kit. Based on these data, the
36
T. Hofmann et al. / Peptides 88 (2017) 32–36
kit can accurately detect phoenixin-14 in human plasma. Whether the phoenixin levels detected here are derived from a central (e.g. hypothalamus) or peripheral (e.g. stomach) origin should be further determined in experiments investigating e.g. portal vein as well as cerebrospinal fluid concentrations of the peptide. In the present study, we report circulating phoenixin levels of ∼0.7 ng/ml in obese human subjects. These levels show a negative association with anxiety with higher levels in subjects reporting low or minimal anxiety and lower levels in individuals reporting higher scores of anxiety. This finding might corroborate the assumption derived from a recent animal study that phoenixin has an anxiolytic effect. Conflict of interest The authors have nothing to disclose. No conflicts of interest exist. Funding This work was supported by the German Research FoundationSTE 1765/3-1 (A.S.) and Charité University Funding UFF 88-226-168 (A.S., T.H.). Acknowledgements We thank Karin Johansson and Christina Hentzschel for their help with organization and realization of clinical measurements and blood withdrawals, Petra Buße and Reinhard Lommel for their laboratory work, and Florian Bruckbauer and Friederice Schröder for data collection, updating and maintaining the database. References [1] H. Fliege, M. Rose, P. Arck, O.B. Walter, R.D. Kocalevent, C. Weber, et al., The Perceived Stress Questionnaire (PSQ) reconsidered: validation and reference values from different clinical and healthy adult samples, Psychosom. Med. 67 (2005) 78–88. [2] G. Gariepy, D. Nitka, N. Schmitz, The association between obesity and anxiety disorders in the population: a systematic review and meta-analysis, Int. J. Obes. (Lond). 34 (2010) 407–419. [3] D.M. Garner, M.P. Olmsted, J. Polivy, Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia, Int. J. Eat. Disord. 2 (1983) 15–34. [4] H. Gunay, R. Tutuncu, S. Aydin, E. Dag, D. Abasli, Decreased plasma nesfatin-1 levels in patients with generalized anxiety disorder, Psychoneuroendocrinology 37 (2012) 1949–1953.
[5] T. Hofmann, U. Elbelt, A. Ahnis, M. Rose, B.F. Klapp, A. Stengel, Sex-specific regulation of NUCB2/nesfatin-1: Differential implication in anxiety in obese men and women, Psychoneuroendocrinology 60 (2015) 130–137. [6] T. Hofmann, A. Stengel, A. Ahnis, P. Busse, U. Elbelt, B.F. Klapp, NUCB2/nesfatin-1 is associated with elevated scores of anxiety in female obese patients, Psychoneuroendocrinology 38 (2013) 2502–2510. [7] J.H. Jiang, Z. He, Y.L. Peng, W.D. Jin, J. Mu, H.X. Xue, et al., Effects of Phoenixin-14 on anxiolytic-like behavior in mice, Behav. Brain Res. 286 (2015) 39–48. [8] S. Levenstein, C. Prantera, V. Varvo, M.L. Scribano, E. Berto, C. Luzi, et al., Development of the Perceived Stress Questionnaire: a new tool for psychosomatic research, J. Psychosom. Res. 37 (1993) 19–32. [9] B. Löwe, O. Decker, S. Muller, E. Brahler, D. Schellberg, W. Herzog, et al., Validation and standardization of the Generalized Anxiety Disorder Screener (GAD-7) in the general population, Med. Care 46 (2008) 266–274. [10] B. Löwe, R. Spitzer, S. Zipfel, W. Herzog, Patient Health Questionnaire (PHQ)—German Version, Manual and materials, Karlsruhe, Pfizer, 2002. [11] R.M. Lyu, X.F. Huang, Y. Zhang, S.L. Dun, J.J. Luo, J.K. Chang, et al., Phoenixin: a novel peptide in rodent sensory ganglia, Neuroscience 250 (2013) 622–631. [12] Z. Merali, C. Cayer, P. Kent, Anisman H: Nesfatin-1 increases anxiety- and fear-related behaviors in the rat, Psychopharmacology (Berl). 201 (2008) 115–123. [13] A.S. Moriarty, S. Gilbody, D. McMillan, L. Manea, Screening and case finding for major depressive disorder using the Patient Health Questionnaire (PHQ-9): a meta-analysis, Gen. Hosp. Psychiatry 37 (2015) 567–576. [14] S. Oh-I, H. Shimizu, T. Satoh, S. Okada, S. Adachi, K. Inoue, et al., Identification of nesfatin-1 as a satiety molecule in the hypothalamus, Nature 443 (2006) 709–712. [15] A. Palasz, E. Rojczyk, K. Bogus, J.J. Worthington, R. Wiaderkiewicz, The novel neuropeptide phoenixin is highly co-expressed with nesfatin-1 in the rat hypothalamus, an immunohistochemical study, Neurosci. Lett. 592 (2015) 17–21. [16] F. Plummer, L. Manea, D. Trepel, D. McMillan, Screening for anxiety disorders with the GAD-7 and GAD-2: a systematic review and diagnostic metaanalysis, Gen. Hosp. Psychiatry 39 (2016) 24–31. [17] R.L. Spitzer, K. Kroenke, J.B. Williams, Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study: primary care evaluation of mental disorders, Patient Health Questionnaire JAMA. 282 (1999) 1737–1744. [18] R.L. Spitzer, K. Kroenke, J.B. Williams, B. Löwe, A brief measure for assessing generalized anxiety disorder: the GAD-7, Arch. Intern. Med. 166 (2006) 1092–1097. [19] A.J. Stunkard, S. Messick, The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger, J. Psychosom. Res. 29 (1985) 71–83. [20] A.K. Treen, V. Luo, D.D. Belsham, Phoenixin activates immortalized GnRH and kisspeptin neurons through the novel receptor GPR173, Mol. Endocrinol. 30 (2016) 872–888. [21] Q. Xu, D. Anderson, J. Lurie-Beck, The relationship between abdominal obesity and depression in the general population: a systematic review and meta-analysis, Obes. Res. Clin. Pract. 5 (2011) e267–360. [22] G.L. Yosten, R.M. Lyu, A.J. Hsueh, O. Avsian-Kretchmer, J.K. Chang, C.W. Tullock, et al., A novel reproductive peptide, phoenixin, J. Neuroendocrinol. 25 (2013) 206–215.
Contents lists available at ScienceDirect
Peptides journal homepage: www.elsevier.com/locate/peptides
Research Paper
Phoenixin is negatively associated with anxiety in obese men Tobias Hofmann a , Elena Weibert a , Anne Ahnis a , Ulf Elbelt a,b , Matthias Rose a , Burghard F. Klapp a , Andreas Stengel a,∗ a
Charité Center for Internal Medicine and Dermatology, Department for Psychosomatic Medicine; Charité-Universitätsmedizin Berlin, Berlin, Germany Charité Center for Internal Medicine with Gastroenterology and Nephrology, Department for Endocrinology, Diabetes and Nutrition, Charité-Universitätsmedizin Berlin, Berlin, Germany b
a r t i c l e
i n f o
Article history: Received 3 September 2016 Received in revised form 3 December 2016 Accepted 14 December 2016 Available online 15 December 2016 Keywords: Emotion Gut-brain axis Psychobiology Psychoneuroendocrine Psychosomatic
a b s t r a c t Phoenixin was recently identified in the rat hypothalamus and initially implicated in reproductive functions. A subsequent study described an anxiolytic effect of the peptide. The aim of the study was to investigate a possible association of circulating phoenixin with anxiety in humans. We therefore enrolled 68 inpatients with a broad spectrum of psychometrically measured anxiety (GAD-7). We investigated men since a menstrual cycle dependency of phoenixin has been assumed. Obese subjects were enrolled since they often report psychological comorbidities. In addition, we also assessed depressiveness (PHQ9) and perceived stress (PSQ-20). Plasma phoenixin levels were measured using a commercial ELISA. First, we validated the ELISA kit performing a spike-and-recovery experiment showing a variance of 6.7 ± 8.8% compared to the expected concentrations over the whole range of concentrations assessed, while a lower variation of 1.6 ± 0.8% was observed in the linear range of the assay (0.07–2.1 ng/ml). We detected phoenixin in the circulation of obese men at levels of 0.68 ± 0.50 ng/ml. These levels showed a negative association with anxiety scores (r = −0.259, p = 0.043), while no additional associations with other psychometric parameters were observed. In summary, phoenixin is present in the human circulation and negatively associated with anxiety in obese men, a population often to report comorbid anxiety. © 2016 Elsevier Inc. All rights reserved.
1. Introduction Phoenixin is a recently discovered 20-amino acid peptide described to have regulatory properties in the reproductive system [22] likely via the G-protein coupled receptor, the GPR173, as recently suggested [20]. Also a fragment of the peptide, phoenixin14, retained its biological activity [22]. This modulating effect was exerted via increasing the gonadotropin-releasing hormone (GnRH) receptor expression in the pituitary in vitro [22]. Pointing towards a physiological role of endogenous phoenixin, the blockade of the peptide using small interfering RNA injected intracerebroventricularly delayed the onset of estrus in female rats associated with a reduction of GnRH receptor expression [22]. Further suggesting an important role of phoenixin, the peptide sequence of phoenixin is highly conserved across non-mammalian
∗ Corresponding author at: Charité Center for Internal Medicine and Dermatology, Department for Psychosomatic Medicine, Charité-Universitätsmedizin Berlin, Hindenburgdamm 30, 12203 Berlin, Germany. E-mail address: [email protected] (A. Stengel). http://dx.doi.org/10.1016/j.peptides.2016.12.011 0196-9781/© 2016 Elsevier Inc. All rights reserved.
(fish, amphibians, chicken) and mammalian species (rodents, human) [22]. The initial landmark study identified phoenixin in several rat peripheral (e.g. heart, thymus, esophagus, stomach and spleen) and central tissues with highest peptide expression levels in the hypothalamus [22]. Immunostaining indicated the expression of phoenixin in the magnocellular and parvocellular parts of the paraventricular nucleus, supraoptic nucleus, arcuate nucleus, dorsal hypothalamus, zona incerta, ventromedial hypothalamus, lateral hypothalamus and the perifornical area [22]. Additional central expression was shown in the substantia nigra, Edinger-Westphal nucleus, nucleus of the solitary tract and the dorsal motor nucleus of the vagus nerve [22] pointing towards additional functions besides the initially proposed role in the reproductive system. A subsequent study detected even higher peptide levels in the spinal cord compared to the hypothalamus and described a predominant localization in sensory neurons of the dorsal root, nodose and trigeminal ganglia [11] giving rise to a role in pain processing. This hypothesis is supported by the observation of a reduced number of writhes following intraperitoneal injection of acetic acid in mice receiving a central injection of phoenixin [11].
T. Hofmann et al. / Peptides 88 (2017) 32–36
Following these studies, a recent report further described the phenotype of phoenixin neurons indicating a large proportion (70–86%) of phoenixin neurons co-localizing with NUCB2/nesfatin1 in rat hypothalamic nuclei (arcuate nucleus, paraventricular nucleus, ventromedial hypothalamus and lateral hypothalamus) [15]. NUCB2/nesfatin-1 was discovered a decade ago and first localized in the rat hypothalamus [14]. It is to note that early on an anxiogenic effect of NUCB2/nesfatin-1 has been described in male rats following intracerebroventricular injection of the peptide [12]. Interestingly, also in humans an association of NUCB2/nesfatin1 and anxiety has been reported with a positive correlation in women [6] and a negative association in men [4,5] indicating a sexspecific regulation of the peptide. In contrast to NUCB2/nesfatin-1, phoenixin was recently shown to exert potent anxiolytic effects in male mice [7]. However, whether phoenixin is also involved in the mediation, modulation or perception of anxiety in humans is unknown so far. Therefore, the aim of the present study was to investigate whether phoenixin is associated with anxiety in humans. Based on the involvement of the peptide in reproductive functions [22] a menstrual cycle dependency can be assumed (or at least cannot be excluded at this point). Therefore, only men were investigated in the present study. Moreover, since obese subjects often report comorbid anxiety [2] and depression [21] we investigated obese men displaying a broad range of psychopathology in the present study. We first validated the commercial phoenixin ELISA kit and then investigated the potential association between circulating phoenixin and psychometrically measured anxiety levels. Subsequently, we further characterized our study population with regards to depressiveness and perceived stress. 2. Materials and methods 2.1. Subjects For this study, we enrolled male obese inpatients (from January 2011 to December 2014) hospitalized in the Department for Psychosomatic Medicine at Charité-Universitätsmedizin Berlin that received medical treatment for obesity and its somatic and mental comorbidities. The inclusion criteria were comprised of male sex, an age of ≥18 years and a body mass index (BMI) ≥ 0 kg/m2 . Patients with a current malignant disease, untreated psychotic disorders and preceding bariatric surgery were not enrolled in the study. In addition, patients with hypercortisolism or hypothyroidism were excluded from the study. The final number of patients analyzed in this study was n = 68. All investigations in the present study were conducted according to the Declaration of Helsinki and all patients gave written informed consent. The study was approved by the institutional ethics committee of Charité-Universitätsmedizin Berlin (protocol number: EA1/114/10). 2.2. Laboratory analyses Blood samples were obtained from patients within three days after hospital admission to prevent an influence of treatment on metabolic/hormonal and psychometric parameters. All blood samples were taken after an overnight fast between 07:00 and 08:00 in the morning. Patients were allowed to drink small amounts of water but were advised not to smoke or exercise in the morning (before blood withdrawal). After withdrawal from a forearm vein, venous blood was collected in pre-cooled standard EDTA tubes containing aprotinin (1.2 Trypsin Inhibitory Unit/1 ml blood; ICN Pharmaceuticals, Costa Mesa, CA, USA) as peptidase inhibitor. The EDTA tubes were stored on ice directly after blood withdrawal and
33
centrifuged for 10 min at 3000 rpm at 4 ◦ C. Plasma was separated and stored at −80 ◦ C until further processing. Phoenixin levels were determined using a commercial enzymelinked immunosorbent assay (ELISA, linear range 0.07-2.1 ng/ml, catalog # EK-079-01, Phoenix Pharmaceuticals, Inc., Burlingame, CA, USA) per the manufacturer’s instructions. First, to validate the kit, a spike-and-recovery experiment was performed adding synthetic phoenixin-14 amide (079-01, Phoenix Pharmaceuticals) in increasing concentrations to pooled human plasma samples. Recovery was assessed and measured levels compared to the expected concentrations. All plasma samples were processed in one batch; the intra-assay variability was 9%. The antibody used in this ELISA was generated in rabbit and recognizes both human phoenixin-14 as well as phoenixin-20 with 100% cross-reactivity (cross-reacting also with rat, mouse, bovine, porcine and canine phoenixin; manufacturer’s information). No cross-reactivity was observed for adrenomedullin, alpha-atrial natriuretic polypeptide, angiotensin I and II, apelin-12, bradykinin, brain natriuretic peptide-32, gonadotropin releasing hormone, neuropeptide Y, orexin A, somatostatin and TLQP-21 (manufacturer’s information). The concentrations of circulating phoenixin described in the present manuscript were well within the linear detection range of the kit (0.07–2.1 ng/ml) and the positive control using synthetic phoenixin-14 was detectable in the expected range of 0.2–0.5 ng/ml. 2.3. Anthropometric measurements For calculation of the BMI (as kg/m2 ), body weight and height of the participants were assessed at the same day of blood withdrawal between 07:00 and 08:00 in the morning in patients wearing light underwear. 2.4. Psychometric parameters All psychometric parameters were obtained electronically. The patients were given electronic devices on the same day or the day before the blood withdrawal and were asked to fill in the following questionnaires within one day. Patients without psychometric data within six days after blood sampling were excluded. For psychometric assessment two modules of the selfadministered patient health questionnaire (PHQ) [17] were used in order to assess depression (PHQ-9) and general anxiety (GAD-7). The PHQ-9 depression module is an established screening instrument for the diagnosis of major depression and evaluation of the severity of depressive symptoms. The PHQ-9 contains nine items representing the diagnostic criteria for DSM-IV depressive disorders. Each of them can be scored as “0” (not at all) to “3” (nearly every day) with a maximum of 27 points [17]. In a meta-analysis of 36 studies (21,292 patients) the PHQ-9 showed a specificity of 0.87 in detecting major depressive disorders [13]. Our patients were handed out the German version of the PHQ-9 [10]. For the current population Cronbach’s alpha was calculated as 0.89. For the examination of anxiety, the GAD-7 scale was used. GAD-7 is an efficient tool to identify generalized anxiety disorder, posttraumatic stress disorder, panic disorder as well as social anxiety disorder and to measure the severity of symptoms [18]. It consists of seven items with scores ranging from “0” (not at all) to “3” (nearly every day) and a maximum of 21 points [18]. In our study the German version was used [9]. In a recent meta-analysis of 12 studies (5223 patients) specificity was calculated with 0.84 and sensitivity with 0.83 at a cut-off point of eight [16]. In the present sample Cronbach’s alpha was 0.90. In order to assess stress the perceived stress questionnaire (PSQ) [8] was used in its revised German version with 20 items (PSQ-20) [1]. The PSQ-20 examines the level of subjectively perceived exac-
34
T. Hofmann et al. / Peptides 88 (2017) 32–36
Table 1 Demographic and socioeconomic characteristics as well as comorbidities and medication of study population (n = 68). Parameter
Mean ± SD or total number
Missing data
Demographic characteristics Age (years) BMI (kg/m2 )
46.0 ± 13.2 48.2 ± 9.8
0 (0.0%) 0 (0.0%)
26 (38.3%)
1 (1.5%) 1 (1.5%)
Socioeconomic characteristics Living in a partnership Level of education - university entrance diploma (‘Abitur’) - vocational diploma (‘Fachabitur’) - secondary education certificate (‘Mittlere Reife’) - basic school qualification (‘Hauptschulabschluss’) - without school-leaving qualification Current employment Unemployment during past 5 years
16 (23.5%) 3 (4.4%) 24 (35.3%) 19 (27.9%) 5 (7.4%) 20 (29.4%) 32 (47.1%)
1 (1.5%) 1 (1.5%) 0 (0.0%)
Comorbidities Sleep-related breathing disorder Type 1 diabetes mellitus Type 2 diabetes mellitus Impaired fasting glucose (without DM 2) Insulin resistance (without DM 2 or IFG) Arterial hypertension Hypertriglyceridemia Hyperuricemia Coronary heart disease Polyarthrosis Chronic back pain
43 (63.2%) 1 (1.5%) 24 (35.3%) 10 (14.7%) 15 (22.1%) 53 (77.9%) 24 (35.3%) 33 (48.5%) 6 (8.8%) 8 (11.8%) 15 (22.1%)
Medication Insulin DPP-4 antagonists/GLP-1 analogues Other antidiabetics Psychopharmacological treatment Opioids Steroids
16 (23.5%) 3 (4.4%) 17 (25.0%) 14 (20.6%) 1 (1.5%) 2 (2.9%)
0 (0.0%)
Abbreviations: DM 2, type 2 diabetes mellitus; DPP-4, dipeptidylpeptidase 4; GLP-1, glucagon like peptide 1; IFG, impaired fasting glucose; SD, standard deviation.
erbating factors and experienced stress. The four factors “worries”, “tension”, “joy” as stress reactions and “demands” as the perception of external stressors are evaluated with five questions each. The sum scores were added and converted into percentage values. Calculated mean values between 0 and 1 were converted to scores ranging from 0 to 100. Cronbach’s alpha for the PSQ total score assessed in the present study population was 0.75 and ranged from 0.83 to 0.90 for the four subscales.
2.5. Statistical analysis All data are expressed as mean ± standard deviation (SD). Distribution of the data was determined by the Kolmogorov-Smirnov test and differences between groups were calculated using t-tests. Depending on the distribution of the data correlations were determined by Pearson’s or Spearman’s analyses. A multiple stepwise linear regression analysis was performed to identify potential confounding parameters (comorbidities, medication). Differences between groups and correlations were considered significant when p < 0.05. All statistical analyses were conducted using SigmaStat 3.1 (Systat Software, San Jose, CA, USA).
3. Results 3.1. Demographic, socioeconomic and medical characteristics Demographic, socioeconomic characteristics as well as comorbidities and medication of the study population are described in Table 1.
3.2. Psychometric characteristics The study population of obese men (n = 68) displayed a broad range of psychopathology. With regards to anxiety, scores ranged from 0 to 19 (maximum score of the test: 21, Table 2). On average, patients reported mild anxiety symptoms (mean ± SD: 6.1 ± 5.3, Table 2). Also depressiveness (range: 0–22, maximum score of the test: 27, mean ± SD: 8.1 ± 6.0) and perceived stress (range of total score: 3–95, maximum score of the test: 100, mean ± SD: 43.2 ± 22.0) showed a broad distribution within the study population (Table 2).
Table 2 Mean scores and ranges of psychometric variables and their correlation with circulating phoenixin levels in a population of obese men (n = 68). Parameter
Mean ± SD
Range
GAD-7 PHQ-9 PSQ total - worries - tension - joy - demands
6.1 ± 5.3 8.1 ± 6.0 43.2 ± 22.0 38.7 ± 25.5 45.6 ± 27.6 46.8 ± 24.9 35.4 ± 24.7
0–19 0–22 3–95 0–100 0–100 0–100 0–100
Correlation with phoenixin r
p
−0.259 −0.140 −0.169 −0.123 −0.230 0.096 −0.107
0.043 0.291 0.194 0.344 0.075 0.460 0.414
Distribution of the data was determined by the Kolmogorov-Smirnov test. Depending on the distribution of the data, correlations were determined by Pearson’s or Spearman’s analyses. Significant associations are shown in bold. Abbreviations: GAD-7, Generalized Anxiety Disorder questionnaire; PHQ-9, Perceived Health Questionnaire; PSQ-20, Perceived Stress Questionnaire.
T. Hofmann et al. / Peptides 88 (2017) 32–36
Phoenixin (ng/ml)
1.5
r = -0.259 p = 0.043
1.0
0.5
0.0 0
10 GAD-7 score
20
Fig. 1. GAD-7 scores in obese men and the correlation with circulating phoenixin levels. Plasma phoenixin levels showed a negative correlation with the GAD-7 scores. Values for r and p are indicated in the correlation graph. Abbreviations: GAD-7, generalized anxiety disorder questionnaire (7 items).
3.3. Phoenixin levels are negatively associated with anxiety in obese men The spike-and-recovery experiment displayed a variance in recovery of 6.7 ± 8.8% over the whole range of concentrations assessed (0.1–20 ng/ml), while a lower variation of 1.6 ± 0.8% was observed in the linear range of the assay (0.07–2.1 ng/ml; data not shown). Therefore, the kit was considered valid for the assessment of phoenixin-14 in human plasma. In the study population of obese men, circulating phoenixin levels showed a negative correlation with reported anxiety (GAD-7 levels, r = −0.259, p = 0.043; Fig. 1, Table 2). A stepwise multiple linear regression analysis with obesity-associated comorbidities as possible confounders showed that GAD scores were able to predict phoenixin plasma levels (p = 0.028), whereas sleep-related breathing disorder (p = 0.411), type 1 diabetes mellitus (p = 0.623), type 2 diabetes mellitus (p = 0.139), impaired fasting glucose (p = 0.943), insulin resistance (p = 0.505), arterial hypertension (p = 0.272), hypertriglyceridemia (p = 0.498), hyperuricemia (p = 0.655), coronary heart disease (p = 0.809), polyarthrosis (p = 0.387) and chronic back pain (p = 0.718) were not. Similarly, a stepwise multiple linear regression analysis testing medication as possible confounders showed a significant association between phoenixin levels by GAD (p = 0.031), while none with insulin (p = 0.078), DPP-4 antagonists/GLP-1 analogues (p = 0.442), other antidiabetics (p = 0.844), psychopharmacological treatment (p = 0.539), opioids (p = 0.325) and steroids (p = 0.766). Using median split of the data we divided the study sample per GAD-7 scores and obtained a group with minimal anxiety (2.0 ± 1.7) and one with moderate anxiety (10.2 ± 4.4, p < 0.001; n = 34/group). Phoenixin levels were 25% higher in the minimal anxiety group (0.78 ± 0.59 ng/ml) compared to the moderate anxiety group (0.59 ± 0.41 ng/ml); however, this difference missed statistical significance (p = 0.068). Calculation of the effect size indicated a Cohen’s d of 0.37. All other psychometric parameters assessed (depressiveness and perceived stress) did not show statistically significant associations with plasma phoenixin concentrations (Table 2). Lastly, no association was observed between circulating phoenixin levels and age (r = −0.211, p = 0.103) or body mass index (r = 0.131, p = 0.319). 4. Discussion In the present study we detected the novel peptide phoenixin, that was recently identified in the rat brain [22], in the human
35
circulation with levels ∼0.7 ng/ml. Furthermore, we showed that phoenixin levels were negatively associated with psychometrically measured anxiety in obese men with higher levels under conditions of low anxiety. Phoenixin has been reported to exert an anxiolytic effect in mice as indicated by more time spent in the open arms of the elevated plus maze and more entries into the center of the open field following lateral brain ventricle injection or microinjection of phoenixin into the anterior hypothalamic area, an effect blocked by a GnRH receptor antagonist [7] pointing towards an involvement of downstream GnRH signaling in the mediation of phoenixin’s anxiolytic effect. Based on these animal data one might speculate that the higher phoenixin levels observed in subjects with minimal anxiety contribute to these low anxiety levels. Interestingly, no additional correlations of phoenixin with other psychometric characteristics of the present obese patient population, namely depressiveness and perceived stress, were observed here, giving rise to the assumption of a specific association of phoenixin and anxiety. The recent observation of an extensive co-localization (>70%) of phoenixin with the pleiotropic peptide NUCB2/nesfatin-1 in the rat hypothalamus [15] attracted attention, especially in light of the finding that both peptides were implicated in the modulation of anxiety in rodents. While NUCB2/nesfatin-1 induced anxiogenic behavior in male rats [12], phoenixin was reported to exert an anxiolytic effect in male mice [7]. NUCB2/nesfatin-1 was associated with anxiety in humans as well with a positive association in women [6], while a negative association was observed in men [5]. In the present study, we also showed a negative association of phoenixin and anxiety in men. Whether NUCB2/nesfatin-1 has anxiogenic effects, while phoenixin exerts an anxiolytic effect in human subjects as well will have to be further investigated. Since this is the first investigation of phoenixin levels in humans, several limitations should be considered. First, the study was only performed in men as menstrual cycle-dependent alterations of phoenixin have been assumed. Moreover, a sex-specific regulation of phoenixin can be hypothesized as well. These points − along with a possible sex steroid dependency – should be further investigated in future studies. Second, in the present study only obese subjects were included due to the high rate of psychological comorbidities [2,21] and in order to follow up on our previous study in obese men [5]. Although suited for the current approach, further studies should also describe phoenixin levels in a healthy population and in normal weight subjects with anxiety disorders. Third, the negative association between phoenixin and anxiety described here does not allow to draw conclusions about cause and consequence. For this purpose, a follow-up study using a longitudinal design will have to investigate whether phoenixin levels increase during/following successful treatment of anxiety symptoms. Fourth, the association between phoenixin and anxiety levels was rather moderate; whether this holds true in larger samples will have to be investigated. Moreover, since the major aim of the study was to investigate the association of phoenixin and anxiety, and all other psychometric parameters (depressiveness and perceived stress) have been mainly included to further characterize the study population, no Bonferroni correction has been performed. Fifth, we did not examine circulating phoenixin levels in relation to fixed diagnoses such as generalized anxiety disorder and instead used the results of selfassessment questionnaires which might be impaired by incorrect self-report caused by recall bias, social acceptability or difficulties in self-observation. Nevertheless, these psychometric assessment tools used in the present study are well validated and established screening instruments [3,8,17–19]. Lastly, the ELISA kit used to detect phoenixin plasma levels was new and has not been used before in humans. Therefore, we performed a spike-and-recovery experiment showing a variation of less than 2% from the expected recovery in the linear range of the kit. Based on these data, the
36
T. Hofmann et al. / Peptides 88 (2017) 32–36
kit can accurately detect phoenixin-14 in human plasma. Whether the phoenixin levels detected here are derived from a central (e.g. hypothalamus) or peripheral (e.g. stomach) origin should be further determined in experiments investigating e.g. portal vein as well as cerebrospinal fluid concentrations of the peptide. In the present study, we report circulating phoenixin levels of ∼0.7 ng/ml in obese human subjects. These levels show a negative association with anxiety with higher levels in subjects reporting low or minimal anxiety and lower levels in individuals reporting higher scores of anxiety. This finding might corroborate the assumption derived from a recent animal study that phoenixin has an anxiolytic effect. Conflict of interest The authors have nothing to disclose. No conflicts of interest exist. Funding This work was supported by the German Research FoundationSTE 1765/3-1 (A.S.) and Charité University Funding UFF 88-226-168 (A.S., T.H.). Acknowledgements We thank Karin Johansson and Christina Hentzschel for their help with organization and realization of clinical measurements and blood withdrawals, Petra Buße and Reinhard Lommel for their laboratory work, and Florian Bruckbauer and Friederice Schröder for data collection, updating and maintaining the database. References [1] H. Fliege, M. Rose, P. Arck, O.B. Walter, R.D. Kocalevent, C. Weber, et al., The Perceived Stress Questionnaire (PSQ) reconsidered: validation and reference values from different clinical and healthy adult samples, Psychosom. Med. 67 (2005) 78–88. [2] G. Gariepy, D. Nitka, N. Schmitz, The association between obesity and anxiety disorders in the population: a systematic review and meta-analysis, Int. J. Obes. (Lond). 34 (2010) 407–419. [3] D.M. Garner, M.P. Olmsted, J. Polivy, Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia, Int. J. Eat. Disord. 2 (1983) 15–34. [4] H. Gunay, R. Tutuncu, S. Aydin, E. Dag, D. Abasli, Decreased plasma nesfatin-1 levels in patients with generalized anxiety disorder, Psychoneuroendocrinology 37 (2012) 1949–1953.
[5] T. Hofmann, U. Elbelt, A. Ahnis, M. Rose, B.F. Klapp, A. Stengel, Sex-specific regulation of NUCB2/nesfatin-1: Differential implication in anxiety in obese men and women, Psychoneuroendocrinology 60 (2015) 130–137. [6] T. Hofmann, A. Stengel, A. Ahnis, P. Busse, U. Elbelt, B.F. Klapp, NUCB2/nesfatin-1 is associated with elevated scores of anxiety in female obese patients, Psychoneuroendocrinology 38 (2013) 2502–2510. [7] J.H. Jiang, Z. He, Y.L. Peng, W.D. Jin, J. Mu, H.X. Xue, et al., Effects of Phoenixin-14 on anxiolytic-like behavior in mice, Behav. Brain Res. 286 (2015) 39–48. [8] S. Levenstein, C. Prantera, V. Varvo, M.L. Scribano, E. Berto, C. Luzi, et al., Development of the Perceived Stress Questionnaire: a new tool for psychosomatic research, J. Psychosom. Res. 37 (1993) 19–32. [9] B. Löwe, O. Decker, S. Muller, E. Brahler, D. Schellberg, W. Herzog, et al., Validation and standardization of the Generalized Anxiety Disorder Screener (GAD-7) in the general population, Med. Care 46 (2008) 266–274. [10] B. Löwe, R. Spitzer, S. Zipfel, W. Herzog, Patient Health Questionnaire (PHQ)—German Version, Manual and materials, Karlsruhe, Pfizer, 2002. [11] R.M. Lyu, X.F. Huang, Y. Zhang, S.L. Dun, J.J. Luo, J.K. Chang, et al., Phoenixin: a novel peptide in rodent sensory ganglia, Neuroscience 250 (2013) 622–631. [12] Z. Merali, C. Cayer, P. Kent, Anisman H: Nesfatin-1 increases anxiety- and fear-related behaviors in the rat, Psychopharmacology (Berl). 201 (2008) 115–123. [13] A.S. Moriarty, S. Gilbody, D. McMillan, L. Manea, Screening and case finding for major depressive disorder using the Patient Health Questionnaire (PHQ-9): a meta-analysis, Gen. Hosp. Psychiatry 37 (2015) 567–576. [14] S. Oh-I, H. Shimizu, T. Satoh, S. Okada, S. Adachi, K. Inoue, et al., Identification of nesfatin-1 as a satiety molecule in the hypothalamus, Nature 443 (2006) 709–712. [15] A. Palasz, E. Rojczyk, K. Bogus, J.J. Worthington, R. Wiaderkiewicz, The novel neuropeptide phoenixin is highly co-expressed with nesfatin-1 in the rat hypothalamus, an immunohistochemical study, Neurosci. Lett. 592 (2015) 17–21. [16] F. Plummer, L. Manea, D. Trepel, D. McMillan, Screening for anxiety disorders with the GAD-7 and GAD-2: a systematic review and diagnostic metaanalysis, Gen. Hosp. Psychiatry 39 (2016) 24–31. [17] R.L. Spitzer, K. Kroenke, J.B. Williams, Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study: primary care evaluation of mental disorders, Patient Health Questionnaire JAMA. 282 (1999) 1737–1744. [18] R.L. Spitzer, K. Kroenke, J.B. Williams, B. Löwe, A brief measure for assessing generalized anxiety disorder: the GAD-7, Arch. Intern. Med. 166 (2006) 1092–1097. [19] A.J. Stunkard, S. Messick, The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger, J. Psychosom. Res. 29 (1985) 71–83. [20] A.K. Treen, V. Luo, D.D. Belsham, Phoenixin activates immortalized GnRH and kisspeptin neurons through the novel receptor GPR173, Mol. Endocrinol. 30 (2016) 872–888. [21] Q. Xu, D. Anderson, J. Lurie-Beck, The relationship between abdominal obesity and depression in the general population: a systematic review and meta-analysis, Obes. Res. Clin. Pract. 5 (2011) e267–360. [22] G.L. Yosten, R.M. Lyu, A.J. Hsueh, O. Avsian-Kretchmer, J.K. Chang, C.W. Tullock, et al., A novel reproductive peptide, phoenixin, J. Neuroendocrinol. 25 (2013) 206–215.