AUTHOR=Neyazi Alexandra , Buchholz Vanessa , Burkert Alexandra , Hillemacher Thomas , de Zwaan Martina , Herzog Wolfgang , Jahn Kirsten , Giel Katrin , Herpertz Stephan , Buchholz Christian A. , Dinkel Andreas , Burgmer Markus , Zeeck Almut , Bleich Stefan , Zipfel Stephan , Frieling Helge 

TITLE=Association of Leptin Gene DNA Methylation With Diagnosis and Treatment Outcome of Anorexia Nervosa

JOURNAL=Frontiers in Psychiatry

VOLUME=10

YEAR=2019

URL=https://www.frontiersin.org/journals/psychiatry/articles/10.3389/fpsyt.2019.00197

DOI=10.3389/fpsyt.2019.00197

ISSN=1664-0640

ABSTRACT=<p>Epigenetic alterations are increasingly implicated in the pathophysiology of anorexia nervosa (AN) but are as yet poorly understood. We investigated possible associations between the leptin gene (<italic>LEP</italic>) and the leptin receptor gene (<italic>LEPR</italic>) DNA promoter methylation and (1) a diagnosis of AN and (2) outcome after a 10 months psychotherapeutic outpatient treatment. 129 (<italic>LEPR</italic>: <italic>n</italic> = 135) patients with AN were investigated during the large scale psychotherapeutic Anorexia Nervosa Treatment Outpatient Study (ANTOP) trial, compared to 117 (<italic>LEPR</italic>: <italic>n</italic> = 119) age and height matched, normal-weight healthy controls. Blood samples were taken at baseline, the end of therapy (40 weeks) and the 12-months follow-up and compared to controls. Methylation was measured in whole blood via bisulfite sequencing. Within the promoter region 32 (<italic>LEP</italic>) and 39 CpG sites (<italic>LEPR</italic>) were analyzed. Two key findings were observed. First, <italic>LEP</italic> and <italic>LEPR</italic> methylation at baseline were lower in patients compared to controls (<italic>LEP</italic>: [%] AN: 30.94 ± 13.2 vs. controls: 34.53 ± 14.6); LEPR ([%] AN: 3.73 ± 5.4 vs. controls: 5.22 ± 8.3, mixed linear models: both <italic>P</italic> &lt; 0.001). Second, lower DNA methylation of the <italic>LEP</italic> promoter, with a dynamic upregulation during treatment, was associated with a full recovery in AN patients (% change from baseline to follow-up in full recovery patients: +35.13% (SD: 47.56); mixed linear model: <italic>P</italic> &lt; 0.0001). To test for potential predictive properties of mean <italic>LEP</italic> DNA methylation a <italic>LEP</italic> DNA methylation cut-off (31.25% DNA methylation) was calculated, which significantly discriminated full recovery vs. full syndrome AN patients. This cut-off was then tested in a group of previously unclassified patients (missing follow-up data of the Structured Interview for Anorexic and Bulimic disorders; <italic>n</italic> = 33). Patients below the cut-off (31.25% <italic>LEP</italic> DNA methylation) showed an increase in BMI over time, while those above the cut-off had a decrease in BMI (ANOVA at the 12-months follow-up: <italic>P</italic> = 0.0142). To our knowledge, this is the first study investigating epigenetic alterations in AN over time. Our findings indicate that <italic>LEP</italic> DNA methylation might be involved in the disease course of AN.</p>