RESEARCH ARTICLE
Association between the Oxytocin Receptor Gene Polymorphism (rs53576) and Bulimia Nervosa Youl-Ri Kim1,2*, Jeong-Hyun Kim3,4, Chan-Hyung Kim5, Jae Gook Shin6 & Janet Treasure7 1
Department of Neuropsychiatry, Seoul Paik Hospital, Inje University College of Medicine, Seoul, South Korea Institute of Eating Disorders and Mental Health, Inje University, Seoul, South Korea 3 Indang Institute of Molecular Biology, Inje University, Seoul, South Korea 4 School of Biological Sciences, Inje University, Gimhae, South Korea 5 Department of Psychiatry, Yonsei University College of Medicine, Seoul, South Korea 6 Department of Pharmacology and Pharmacogenomics Research Center, Inje University College of Medicine, Busan, South Korea 7 Institute of Psychiatry, Psychology and Neuroscience, Department of Psychological Medicine, Section of Eating Disorders, King’s College London, London, United Kingdom 2
Abstract Objective: Oxytocin circuits are implicated in the regulation of appetite and weight. Variants in the oxytocin receptor (OXTR) gene have been associated with bulimic behaviour. This study aimed to investigate the association between the OXTR gene and eating disorders. Method: We genotyped six single nucleotide polymorphisms, rs53576, rs237879, rs2228485, rs13316193, rs2254298 and rs1042778, located within the OXTR gene in Korean patients with eating disorders using the single-base extension method. We studied a total of 262 women, including 69 patients with anorexia nervosa, 90 patients with bulimia nervosa (BN), and 103 healthy women. Results: We found a positive association between the G allele of OXTR rs53576 and BN. In the BN group, the G carriers showed a high score on the behavioural inhibition system. Conclusions: These findings suggest the involvement of the oxytocinergic system in the mechanism that underlies BN. Copyright © 2015 John Wiley & Sons, Ltd and Eating Disorders Association. Received 28 October 2014; Revised 21 January 2015; Accepted 13 February 2015 Keywords eating disorders; bulimia nervosa; anorexia nervosa; oxytocin; OXTR; rs53576; rs237879; rs2228485; rs13316193; rs2254298; rs1042778; genotype; allele; behavioural inhibition system; behavioural activation system *Correspondence Youl-Ri Kim, MD, PhD, Department of Neuropsychiatry, Seoul-Paik Hospital, Inje University, 85 Jeo-dong 2 Ga, Jung-gu, Seoul, South Korea, 100-032, Tel: +(82)-22270-0063, Fax: +(82)-2-2270-0344. Email: [email protected] Published online 15 March 2015 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/erv.2354
Introduction The brain neuropeptide hormone, oxytocin, has been implicated in several psychiatric disorders and is of particular interest in eating disorders, given the role that it is now known to play in appetite. Oxytocin is a neuromodulator/neurotransmitter nonapeptide produced in the paraventricular and supraoptic nuclei of the hypothalamus (Macdonald & Macdonald, 2010). It has peripheral effects through its release into the bloodstream via the posterior lobe of the pituitary gland, with roles in orgasm, lactation and parturition. It has central effects in the brain through its transport via nerve fibres and volume diffusion from hypothalamic sites. The intermittent burst firing of oxytocin cells occurs in addition to more regular firing in response to homeostatic signals (Moos et al., 1998; Rossoni et al., 2008). A positive feedback cascade can occur following a central or peripheral route of administration, as oxytocin can stimulate autoreceptors in the magnocellular, supra optic neurons (Deblon et al., 2011; Hicks et al., 2012).
In primitive species, the vasopressin/oxytocin-like peptide, nematocin, provides neuromodulatory input into the gustatory plasticity circuit, as well as reproductive systems (Beets, Temmerman, Janssen, & Schoofs, 2013). In mammals, oxytocin is an important regulator of appetite and weight (Blevins & Ho, 2013; Menzies, Skibicka, Dickson, & Leng, 2012; Sabatier, Leng, & Menzies, 2013). Although peripheral mechanisms may also be involved (Blevins & Ho, 2013), it is thought that hypothalamic mechanisms are of key importance. For example, Agouti-related protein (AGRP) neurons stimulate food intake through their inhibitory effect on paraventricular (PVN) oxytocin neurons (Atasoy, Betley, Su, & Sternson, 2012). Fasting and leptin treatment modified the transcriptional PVN oxytocin (Tung et al., 2008). Oxytocin appears to be particularly involved in inhibiting the appetite for sugar and carbohydrates. Oxytocin receptor antagonist injections in wild-type mice produced a preference for sucrose over fat (Olszewski et al., 2010). Oxytocin receptor
Eur. Eat. Disorders Rev. 23 (2015) 171–178 © 2015 John Wiley & Sons, Ltd and Eating Disorders Association.
171
Y.-R. Kim et al.
OXTR and Eating Disorders
knockout animals consume greater amounts of sweet solutions than wild types (Amico, Vollmer, Cai, Miedlar, & Rinaman, 2005) and develop late onset obesity (Nishimori et al., 2008; Takayanagi et al., 2008). Animals engineered not to express oxytocin overconsume sweetened food (Miedlar, Rinaman, Vollmer, & Amico, 2007) and carbohydrates (Sclafani, Rinaman, Vollmer, & Amico, 2007). Oxytocin expression was downregulated upon long-term intermittent exposure to sugar, which may represent a form of neuroadaptation to a high sugar diet (Mitra et al., 2010). The functional relevance of oxytocin in appetite and weight control is seen in animal models of obesity that result from the global or selective hypothalamic loss of oxytocin neurons (Leng et al., 2008; Wu et al., 2012). Also, animals with deletions in the single minded 1 gene are obese, and this phenotype is associated with reduced hypothalamic oxytocin (Kublaoui, Gemelli, Tolson, Wang, & Zinn, 2008). Mice with dietary induced obesity also have functional abnormalities in their oxytocin systems (Zhang et al., 2011). The relevance of oxytocin in animal models of obesity is also seen in humans. Variants of the single minded 1 genes in humans are also associated with obesity (Holder, Butte, & Zinn, 2000; Ramachandrappa et al., 2013; Swarbrick et al., 2011). Rare variants of genes that encode for G-protein coupled receptors (GPCRs) (one of which is the oxytocin receptor gene) are associated with childhood obesity (Wheeler et al., 2013). Patients with Prader–Willi syndrome have reductions in the number and size of PVN oxytocin neurons (Swaab, Purba, & Hofman, 1995). There is preliminary evidence that oytocin plays a role in eating disorders. Plasma oxytocin levels are correlated with eating symptoms and insula activation in patients with anorexia nervosa (AN) (Lawson et al., 2012). Furthermore, high levels of methylation on some of the CpG sites of the oxytocin receptor (OXTR) gene have been found in patients with AN, and are associated with the body mass index (Kim, Kim, Kim, & Treasure, 2014). Few studies are available describing oxytocin functioning in patients with BN, and the only such study reported that cerebrospinal fluid oxytocin is normal in subjects who have recovered from BN (Frank, Kaye, Altemus, & Greeno, 2000). Recently, Connelly et al. (2014) found a relationship between the rs53576 GG allele and bulimic behaviour in an exploratory analysis of a community cohort. Thus, these possible links between the oxytocin genotype and the eating disorder phenotype need further study. The observed important variations in sensitivity to oxytocin are assumed to result from genetic and epigenetic variations in the OXTR gene (Ebstein, Israel, Chew, Zhong, & Knafo, 2010; Insel, 2010; Kogan et al., 2011). The OXTR gene is located on chromosome 3p25 and has four exons and three introns. We selected six candidate single nucleotide polymorphisms (SNPs) in OXTR gene suspected to be associated with eating disorders based on a review of the literature. The SNP rs53576, in the third intron is thought to be a particularly promising candidate to explain the differences in oxytocinergic functioning (Meyer-Lindenberg, Domes, Kirsch, & Heinrichs, 2011). Some, but not all, association studies suggested that OXTR rs53576 GG carriers showed more positive outcomes than A allele carriers did in respect to mood and emotional response. The rs53576 GG genotype was associated with general social phenotypes, psychological resources (Saphire-Bernstein, Way, Kim, Sherman, & Taylor, 2011), higher empathy and lower 172
stress reactivity (Rodrigues, Saslow, Garcia, John, & Keltner, 2009) and prosocial temperament (Tost et al., 2010). In addition, there is evidence that rs53576 G allele may contribute to the risk of emotional and behavioural problems through geneenvironment interaction. The OXTR rs53576 GG genotype carrier participants who experienced childhood maltreatment had higher tendency to show disorganized attachment styles and higher levels of emotional dysregulation in one study (Bradley et al., 2011). Another study reported that rs53576 G carrier participants who had experienced high levels of childhood maltreatment had greater depressive symptomology than that of AA genotype carrier participants (McQuaid, McInnis, Stead, Matheson, & Anisman, 2013). Genetic variations of rs2228485, which is located in the exon region, were shown to be associated with autistic traits (Lucht et al., 2013). An rs 2254298 variant was reported to be associated with sociability, including autism spectrum disease (Jacob, Brune, Carter, Leventhal, & Lord, 2007; Wu et al., 2005) and the volume of the bilateral amygdala (Inoue et al., 1994), although the link was not replicated in other reports (Apicella et al., 2010). The TT homozygote of the rs1042778 SNP was reported to be associated with less empathic communication (Schneiderman, Kanat-Maymon, Ebstein, & Feldman, 2014). The OXTR SNPs rs237897 and rs13316193 are also candidates for the potential functional variation of the OXTR gene (Dixon et al., 2007; Tansey et al., 2010). In this study, we investigated the genotype distributions and allelic frequencies of the six OXTR SNPs (rs53576, rs237897, rs1042778, rs2228485, rs2254298 and rs13316193) in patients with AN or BN, as well as normal weight healthy women. We also investigated whether the OXTR gene was associated with differences in eating disorder symptoms or personality styles such as the approach/avoidance system in patients with eating disorders.
Methods and materials Participants A total of 262 Korean women (69 patients with AN, 90 patients with BN and 103 healthy university students) participated in the genetic association study. The patients with eating disorders were recruited from both the outpatient clinic and the inpatient ward at the Eating Disorders Clinic of Seoul Paik Hospital, South Korea. The diagnosis of eating disorders was confirmed according to the DSM-5 diagnostic criteria from a clinical interview. Exclusion criteria included active substance use disorder and psychotic disorder (schizophrenia, schizoaffective and psychosis not otherwise specified). All of the subjects provided written-informed consent to participate in the study. The study was approved by the local ethics committee. OXTR genotyping Genomic DNA was extracted from peripheral blood cells using a QIAamp DNA Blood Mini Kit (Qiagen, Chatsworth, CA, USA). The SNPs of the OXTR were determined using single-base extension methods (Turner, Choudhury, Reynard, Railton, & Navarrete, 2002). The amplification primers were designed using Primer Express® Software Version 3.0 (Applied Biosystems, Foster City,
Eur. Eat. Disorders Rev. 23 (2015) 171–178 © 2015 John Wiley & Sons, Ltd and Eating Disorders Association.
Y.-R. Kim et al.
OXTR and Eating Disorders
CA, USA). The primer and single-base extension primer sequences of the SNPs are shown in Table 1. PCR was performed using a 9700 Thermal Cycler (Applied Biosystems, Foster City, CA, USA) with the following conditions: initial denaturation at 95°C for 2 min, followed by 35 cycles of 94°C for 30 s, 63.5°C for 30 s, and 72°C for 30 s, with a final elongation step at 72°C for 5 min. The PCR products were treated with 1.5 μL of ExoSAP-IT per 3 μL of PCR product at 37°C for 30 min, and then incubated at 80°C for 15 min to inactivate the enzyme. The PCR product, purified by ExoSAP-IT, was then used as a template to examine the six SNPs in the OXTR. Each single-base extension primer was designed to avoid overlapping peak signals. Multiplex single-base extension was performed using SNaPshot® in accordance with the manufacturer’s protocol (Applied Biosystems). The SNaPshot products were finally analysed using capillary electrophoresis, for which 0.7 μL of the SNaPshot product was mixed with 9.0 μL of Hi-Di formamide and 0.3 μL of the Genescan-120 LIZ™ size marker. The samples were denatured at 95°C for 5 min and run on an ABI-Prism 3100 genetic analyser using a 36-cm capillary array and POP-7 polymer (Applied Biosystems). The SNaPshot data were analysed using GeneMapper (ver. 3.7; Applied Biosystems) software. Psychological measures The eating habits of the participants were assessed in a clinical interview. The degrees of eating psychopathology (weight, shape, eating concerns and dietary restraint) were evaluated using the Korean version of the Eating Disorder Examination self-report version Questionnaire (EDE-Q) (Fairburn & Beglin, 1994) for all patients. The participants also completed selfreport measures including the standardized Korean version of the Beck Depression Inventory (Beck, Ward, Mendelson, Mock, & Erbaugh, 1961) to assess depression, the standardized Korean version of the Spielberger State and Trait Anxiety Inventory (Spielberger, Gorsuch, Lushene, Vagg, & Facobs, 1983) to assess the state and trait anxiety, and the Standardized Korean version of Behavioural Inhibition System/Behavioural Approach System (BIS/BAS)(Carver & White, 1994; Kim & Kim, 2001) to measure those personality dimensions that reflect the sensitivity of the motivational systems.
Statistical analysis The genotypes were analysed for their fit with the Hardy— Weinberg equilibrium using the Chi-squared goodness-of-fit test. The frequencies were first calculated separately in a co-dominant model. Subsequently, the allelic and genotypic distributions between the patients and controls were examined using the Pearson χ 2 test on 2 × 3 contingency tables. Based on the database (http://www.ncbi.nlm.nih.gov/projects/SNP) and previous studies, for rs53576, rs237897, rs2228485 and rs13316193, heterozygotes and dominant homozygotes (GG) were collapsed into one group (GA + GG = G-carriers) for comparison with the AA genotype following statistical analyses in a presumably dominant genetic model. For rs1042778 and rs2254298, the allele frequency of the AA genotype was combined with the heterozygotes for comparison with the GG genotype for the odds ratios and the 95% confidence intervals. In order to avoid type II errors, we did not correct for multiple testing at this stage, in order to explore any pattern of association. p-values < 0.05 were considered significant and two-tailed tests were used. Statistical analyses were performed using SPSS 19.0 (IBM Inc., Chicago, IL, USA).
Results Demographic and clinical characteristics The demographic and group characteristics of the participants are shown in Table 2. The average age in the groups was similar but, as expected, the body mass index (BMI, kg/m2) differed (F2,250 = 130.779, p
Association between the Oxytocin Receptor Gene Polymorphism (rs53576) and Bulimia Nervosa Youl-Ri Kim1,2*, Jeong-Hyun Kim3,4, Chan-Hyung Kim5, Jae Gook Shin6 & Janet Treasure7 1
Department of Neuropsychiatry, Seoul Paik Hospital, Inje University College of Medicine, Seoul, South Korea Institute of Eating Disorders and Mental Health, Inje University, Seoul, South Korea 3 Indang Institute of Molecular Biology, Inje University, Seoul, South Korea 4 School of Biological Sciences, Inje University, Gimhae, South Korea 5 Department of Psychiatry, Yonsei University College of Medicine, Seoul, South Korea 6 Department of Pharmacology and Pharmacogenomics Research Center, Inje University College of Medicine, Busan, South Korea 7 Institute of Psychiatry, Psychology and Neuroscience, Department of Psychological Medicine, Section of Eating Disorders, King’s College London, London, United Kingdom 2
Abstract Objective: Oxytocin circuits are implicated in the regulation of appetite and weight. Variants in the oxytocin receptor (OXTR) gene have been associated with bulimic behaviour. This study aimed to investigate the association between the OXTR gene and eating disorders. Method: We genotyped six single nucleotide polymorphisms, rs53576, rs237879, rs2228485, rs13316193, rs2254298 and rs1042778, located within the OXTR gene in Korean patients with eating disorders using the single-base extension method. We studied a total of 262 women, including 69 patients with anorexia nervosa, 90 patients with bulimia nervosa (BN), and 103 healthy women. Results: We found a positive association between the G allele of OXTR rs53576 and BN. In the BN group, the G carriers showed a high score on the behavioural inhibition system. Conclusions: These findings suggest the involvement of the oxytocinergic system in the mechanism that underlies BN. Copyright © 2015 John Wiley & Sons, Ltd and Eating Disorders Association. Received 28 October 2014; Revised 21 January 2015; Accepted 13 February 2015 Keywords eating disorders; bulimia nervosa; anorexia nervosa; oxytocin; OXTR; rs53576; rs237879; rs2228485; rs13316193; rs2254298; rs1042778; genotype; allele; behavioural inhibition system; behavioural activation system *Correspondence Youl-Ri Kim, MD, PhD, Department of Neuropsychiatry, Seoul-Paik Hospital, Inje University, 85 Jeo-dong 2 Ga, Jung-gu, Seoul, South Korea, 100-032, Tel: +(82)-22270-0063, Fax: +(82)-2-2270-0344. Email: [email protected] Published online 15 March 2015 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/erv.2354
Introduction The brain neuropeptide hormone, oxytocin, has been implicated in several psychiatric disorders and is of particular interest in eating disorders, given the role that it is now known to play in appetite. Oxytocin is a neuromodulator/neurotransmitter nonapeptide produced in the paraventricular and supraoptic nuclei of the hypothalamus (Macdonald & Macdonald, 2010). It has peripheral effects through its release into the bloodstream via the posterior lobe of the pituitary gland, with roles in orgasm, lactation and parturition. It has central effects in the brain through its transport via nerve fibres and volume diffusion from hypothalamic sites. The intermittent burst firing of oxytocin cells occurs in addition to more regular firing in response to homeostatic signals (Moos et al., 1998; Rossoni et al., 2008). A positive feedback cascade can occur following a central or peripheral route of administration, as oxytocin can stimulate autoreceptors in the magnocellular, supra optic neurons (Deblon et al., 2011; Hicks et al., 2012).
In primitive species, the vasopressin/oxytocin-like peptide, nematocin, provides neuromodulatory input into the gustatory plasticity circuit, as well as reproductive systems (Beets, Temmerman, Janssen, & Schoofs, 2013). In mammals, oxytocin is an important regulator of appetite and weight (Blevins & Ho, 2013; Menzies, Skibicka, Dickson, & Leng, 2012; Sabatier, Leng, & Menzies, 2013). Although peripheral mechanisms may also be involved (Blevins & Ho, 2013), it is thought that hypothalamic mechanisms are of key importance. For example, Agouti-related protein (AGRP) neurons stimulate food intake through their inhibitory effect on paraventricular (PVN) oxytocin neurons (Atasoy, Betley, Su, & Sternson, 2012). Fasting and leptin treatment modified the transcriptional PVN oxytocin (Tung et al., 2008). Oxytocin appears to be particularly involved in inhibiting the appetite for sugar and carbohydrates. Oxytocin receptor antagonist injections in wild-type mice produced a preference for sucrose over fat (Olszewski et al., 2010). Oxytocin receptor
Eur. Eat. Disorders Rev. 23 (2015) 171–178 © 2015 John Wiley & Sons, Ltd and Eating Disorders Association.
171
Y.-R. Kim et al.
OXTR and Eating Disorders
knockout animals consume greater amounts of sweet solutions than wild types (Amico, Vollmer, Cai, Miedlar, & Rinaman, 2005) and develop late onset obesity (Nishimori et al., 2008; Takayanagi et al., 2008). Animals engineered not to express oxytocin overconsume sweetened food (Miedlar, Rinaman, Vollmer, & Amico, 2007) and carbohydrates (Sclafani, Rinaman, Vollmer, & Amico, 2007). Oxytocin expression was downregulated upon long-term intermittent exposure to sugar, which may represent a form of neuroadaptation to a high sugar diet (Mitra et al., 2010). The functional relevance of oxytocin in appetite and weight control is seen in animal models of obesity that result from the global or selective hypothalamic loss of oxytocin neurons (Leng et al., 2008; Wu et al., 2012). Also, animals with deletions in the single minded 1 gene are obese, and this phenotype is associated with reduced hypothalamic oxytocin (Kublaoui, Gemelli, Tolson, Wang, & Zinn, 2008). Mice with dietary induced obesity also have functional abnormalities in their oxytocin systems (Zhang et al., 2011). The relevance of oxytocin in animal models of obesity is also seen in humans. Variants of the single minded 1 genes in humans are also associated with obesity (Holder, Butte, & Zinn, 2000; Ramachandrappa et al., 2013; Swarbrick et al., 2011). Rare variants of genes that encode for G-protein coupled receptors (GPCRs) (one of which is the oxytocin receptor gene) are associated with childhood obesity (Wheeler et al., 2013). Patients with Prader–Willi syndrome have reductions in the number and size of PVN oxytocin neurons (Swaab, Purba, & Hofman, 1995). There is preliminary evidence that oytocin plays a role in eating disorders. Plasma oxytocin levels are correlated with eating symptoms and insula activation in patients with anorexia nervosa (AN) (Lawson et al., 2012). Furthermore, high levels of methylation on some of the CpG sites of the oxytocin receptor (OXTR) gene have been found in patients with AN, and are associated with the body mass index (Kim, Kim, Kim, & Treasure, 2014). Few studies are available describing oxytocin functioning in patients with BN, and the only such study reported that cerebrospinal fluid oxytocin is normal in subjects who have recovered from BN (Frank, Kaye, Altemus, & Greeno, 2000). Recently, Connelly et al. (2014) found a relationship between the rs53576 GG allele and bulimic behaviour in an exploratory analysis of a community cohort. Thus, these possible links between the oxytocin genotype and the eating disorder phenotype need further study. The observed important variations in sensitivity to oxytocin are assumed to result from genetic and epigenetic variations in the OXTR gene (Ebstein, Israel, Chew, Zhong, & Knafo, 2010; Insel, 2010; Kogan et al., 2011). The OXTR gene is located on chromosome 3p25 and has four exons and three introns. We selected six candidate single nucleotide polymorphisms (SNPs) in OXTR gene suspected to be associated with eating disorders based on a review of the literature. The SNP rs53576, in the third intron is thought to be a particularly promising candidate to explain the differences in oxytocinergic functioning (Meyer-Lindenberg, Domes, Kirsch, & Heinrichs, 2011). Some, but not all, association studies suggested that OXTR rs53576 GG carriers showed more positive outcomes than A allele carriers did in respect to mood and emotional response. The rs53576 GG genotype was associated with general social phenotypes, psychological resources (Saphire-Bernstein, Way, Kim, Sherman, & Taylor, 2011), higher empathy and lower 172
stress reactivity (Rodrigues, Saslow, Garcia, John, & Keltner, 2009) and prosocial temperament (Tost et al., 2010). In addition, there is evidence that rs53576 G allele may contribute to the risk of emotional and behavioural problems through geneenvironment interaction. The OXTR rs53576 GG genotype carrier participants who experienced childhood maltreatment had higher tendency to show disorganized attachment styles and higher levels of emotional dysregulation in one study (Bradley et al., 2011). Another study reported that rs53576 G carrier participants who had experienced high levels of childhood maltreatment had greater depressive symptomology than that of AA genotype carrier participants (McQuaid, McInnis, Stead, Matheson, & Anisman, 2013). Genetic variations of rs2228485, which is located in the exon region, were shown to be associated with autistic traits (Lucht et al., 2013). An rs 2254298 variant was reported to be associated with sociability, including autism spectrum disease (Jacob, Brune, Carter, Leventhal, & Lord, 2007; Wu et al., 2005) and the volume of the bilateral amygdala (Inoue et al., 1994), although the link was not replicated in other reports (Apicella et al., 2010). The TT homozygote of the rs1042778 SNP was reported to be associated with less empathic communication (Schneiderman, Kanat-Maymon, Ebstein, & Feldman, 2014). The OXTR SNPs rs237897 and rs13316193 are also candidates for the potential functional variation of the OXTR gene (Dixon et al., 2007; Tansey et al., 2010). In this study, we investigated the genotype distributions and allelic frequencies of the six OXTR SNPs (rs53576, rs237897, rs1042778, rs2228485, rs2254298 and rs13316193) in patients with AN or BN, as well as normal weight healthy women. We also investigated whether the OXTR gene was associated with differences in eating disorder symptoms or personality styles such as the approach/avoidance system in patients with eating disorders.
Methods and materials Participants A total of 262 Korean women (69 patients with AN, 90 patients with BN and 103 healthy university students) participated in the genetic association study. The patients with eating disorders were recruited from both the outpatient clinic and the inpatient ward at the Eating Disorders Clinic of Seoul Paik Hospital, South Korea. The diagnosis of eating disorders was confirmed according to the DSM-5 diagnostic criteria from a clinical interview. Exclusion criteria included active substance use disorder and psychotic disorder (schizophrenia, schizoaffective and psychosis not otherwise specified). All of the subjects provided written-informed consent to participate in the study. The study was approved by the local ethics committee. OXTR genotyping Genomic DNA was extracted from peripheral blood cells using a QIAamp DNA Blood Mini Kit (Qiagen, Chatsworth, CA, USA). The SNPs of the OXTR were determined using single-base extension methods (Turner, Choudhury, Reynard, Railton, & Navarrete, 2002). The amplification primers were designed using Primer Express® Software Version 3.0 (Applied Biosystems, Foster City,
Eur. Eat. Disorders Rev. 23 (2015) 171–178 © 2015 John Wiley & Sons, Ltd and Eating Disorders Association.
Y.-R. Kim et al.
OXTR and Eating Disorders
CA, USA). The primer and single-base extension primer sequences of the SNPs are shown in Table 1. PCR was performed using a 9700 Thermal Cycler (Applied Biosystems, Foster City, CA, USA) with the following conditions: initial denaturation at 95°C for 2 min, followed by 35 cycles of 94°C for 30 s, 63.5°C for 30 s, and 72°C for 30 s, with a final elongation step at 72°C for 5 min. The PCR products were treated with 1.5 μL of ExoSAP-IT per 3 μL of PCR product at 37°C for 30 min, and then incubated at 80°C for 15 min to inactivate the enzyme. The PCR product, purified by ExoSAP-IT, was then used as a template to examine the six SNPs in the OXTR. Each single-base extension primer was designed to avoid overlapping peak signals. Multiplex single-base extension was performed using SNaPshot® in accordance with the manufacturer’s protocol (Applied Biosystems). The SNaPshot products were finally analysed using capillary electrophoresis, for which 0.7 μL of the SNaPshot product was mixed with 9.0 μL of Hi-Di formamide and 0.3 μL of the Genescan-120 LIZ™ size marker. The samples were denatured at 95°C for 5 min and run on an ABI-Prism 3100 genetic analyser using a 36-cm capillary array and POP-7 polymer (Applied Biosystems). The SNaPshot data were analysed using GeneMapper (ver. 3.7; Applied Biosystems) software. Psychological measures The eating habits of the participants were assessed in a clinical interview. The degrees of eating psychopathology (weight, shape, eating concerns and dietary restraint) were evaluated using the Korean version of the Eating Disorder Examination self-report version Questionnaire (EDE-Q) (Fairburn & Beglin, 1994) for all patients. The participants also completed selfreport measures including the standardized Korean version of the Beck Depression Inventory (Beck, Ward, Mendelson, Mock, & Erbaugh, 1961) to assess depression, the standardized Korean version of the Spielberger State and Trait Anxiety Inventory (Spielberger, Gorsuch, Lushene, Vagg, & Facobs, 1983) to assess the state and trait anxiety, and the Standardized Korean version of Behavioural Inhibition System/Behavioural Approach System (BIS/BAS)(Carver & White, 1994; Kim & Kim, 2001) to measure those personality dimensions that reflect the sensitivity of the motivational systems.
Statistical analysis The genotypes were analysed for their fit with the Hardy— Weinberg equilibrium using the Chi-squared goodness-of-fit test. The frequencies were first calculated separately in a co-dominant model. Subsequently, the allelic and genotypic distributions between the patients and controls were examined using the Pearson χ 2 test on 2 × 3 contingency tables. Based on the database (http://www.ncbi.nlm.nih.gov/projects/SNP) and previous studies, for rs53576, rs237897, rs2228485 and rs13316193, heterozygotes and dominant homozygotes (GG) were collapsed into one group (GA + GG = G-carriers) for comparison with the AA genotype following statistical analyses in a presumably dominant genetic model. For rs1042778 and rs2254298, the allele frequency of the AA genotype was combined with the heterozygotes for comparison with the GG genotype for the odds ratios and the 95% confidence intervals. In order to avoid type II errors, we did not correct for multiple testing at this stage, in order to explore any pattern of association. p-values < 0.05 were considered significant and two-tailed tests were used. Statistical analyses were performed using SPSS 19.0 (IBM Inc., Chicago, IL, USA).
Results Demographic and clinical characteristics The demographic and group characteristics of the participants are shown in Table 2. The average age in the groups was similar but, as expected, the body mass index (BMI, kg/m2) differed (F2,250 = 130.779, p
