Journal of Neural Transmission

, Volume 116, Issue 5, pp 615–622

N-methyl-d-aspartate 2b receptor subtype (NR2B) promoter methylation in patients during alcohol withdrawal


    • Department of Psychiatry and PsychotherapyUniversity Hospital of Erlangen
  • Udo Reulbach
    • Department of Psychiatry and PsychotherapyUniversity Hospital of Erlangen
    • Department of Epidemiology and Public HealthUniversity College Cork
  • Bernd Lenz
    • Department of Psychiatry and PsychotherapyUniversity Hospital of Erlangen
  • Helge Frieling
    • Department of Psychiatry and PsychotherapyUniversity Hospital of Erlangen
  • Marc Muschler
    • Department of Psychiatry, Social Psychiatry and PsychotherapyMedical School of Hanover
  • Thomas Hillemacher
    • Department of Psychiatry, Social Psychiatry and PsychotherapyMedical School of Hanover
  • Johannes Kornhuber
    • Department of Psychiatry and PsychotherapyUniversity Hospital of Erlangen
  • Stefan Bleich
    • Department of Psychiatry, Social Psychiatry and PsychotherapyMedical School of Hanover
Basic Neurosciences, Genetics and Immunology - Original Article

DOI: 10.1007/s00702-009-0212-2

Cite this article as:
Biermann, T., Reulbach, U., Lenz, B. et al. J Neural Transm (2009) 116: 615. doi:10.1007/s00702-009-0212-2


NMDA receptors and especially the NR2B receptor subtype play a crucial role during chronic ethanol consumption and alcohol withdrawal. Therefore, the NR2B receptor subtype expression in peripheral blood cells of 32 male patients suffering from alcohol dependency were assessed through quantitative RT-PCR and to explore regulating epigenetic mechanisms, a methylation analysis was conducted using bisulfite sequencing of a fragment of the NR2B promoter region. The expression of the NR2B receptor increased significantly during the first 24 h of withdrawal treatment (day 1; t = 4.1, P = 0.001), and also on and day 3 (t = 2.4; P = 0.029). The severity of alcohol drinking pattern, measured by lifetime drinking and daily ethanol intake, was negatively correlated with the methylation of a defined cluster of five CPG-sites within the NR2B promoter (lifetime drinking: Spearman’s rho = −0.55; P = 0.013; daily ethanol intake: rho = −0.46; P = 0.043). These findings might explain the observation of an impact of alcohol consumption patterns on the gravity of withdrawal symptoms.


AlcoholismNR2B receptorDNA methylationmRNAAlcohol withdrawal


The N-methyl-d-aspartate (NMDA) receptor is an ionotropic subtype of glutamate receptors that is involved in a variety of physiological as well as pathophysiological transactions (Darstein et al. 2000), i.e., synaptic plasticity, ischemia, neurodegenerative disorders and epilepsy (Lipton and Rosenberg 1994; Bleich and Kornhuber 2005). Among others, NMDA receptors play a crucial role during chronic ethanol consumption and withdrawal, central reward systems and relapse (Self and Nestler 1995). During chronic ethanol intoxication, the number of NMDA receptors and their functions have been found to be up-regulated, possibly due to the depressant and antagonistic effects of ethanol on the NMDA receptor. These mechanisms may provoke neuroadaptive processes, causing the excitatory syndrome upon alcohol withdrawal (Tsai et al. 1998; Maler et al. 2005). Alcohol is known to cause brain damage manifested by cognitive, physiological and structural changes in several cortical structures (Bleich et al. 2003). NMDA receptors seem to be major cell-membrane bound targets for ethanol and may be responsible for mediating its long-term damaging effects (Dodd et al. 2004).

The NMDA receptors have different subunit compositions, which determine their properties (Nagy et al. 2005). Distinct subunit compositions lead to various ligand binding sites, affinities towards agonists and antagonists and thereby different functional properties (Lynch et al. 1994). All of the NMDA receptor subunit combinations are directly or indirectly affected by ethanol, with the NR1-NR2A or NR1-NR2B channels particularly sensitive to ethanol. (Popp et al. 1998; Little 1999). The various subunit compositions are extremely important in experience-related neuronal plasticity, and thus mediate dependency. Their ability to change their synaptic localization is thought to be a mechanism for ethanol-induced changes in the brain similar to learning and memory processes (Carpenter-Hyland and Chandler 2007).

In particular, the activity of the NMDA receptor subunit 2B (NR2B) is increased during ethanol treatment (Biermann et al. 2007). In animal studies, the expression of NR2B mRNA was elevated due to ethanol treatment in rat brain (Follesa and Ticku 1995) as well as in cultured cortical neurons (Hu et al. 1996). Similar differences have been shown on the protein level with an up-regulation of NR1, NR2A and NR2B in rat hippocampus and cortex after ethanol ingestion (Kalluri et al. 1998). However, knowledge about the regulation of NR2B protein and mRNA expression is limited.

Variances in gene regulation can be caused by epigenetic programming of their expression in relation to environmental exposure (Szyf et al. 2008). The long-term regulation of gene expression is altered by epigenetic mechanisms that constitute a molecular memory of gene and environmental interactions, namely, DNA methylation, histone modifications and chromatin restructuring (Rodenhiser and Mann 2006). DNA methylation is a stable epigenetic marker of gene expression, which is linked to long-lasting gene silencing (Razin and Riggs 1980), mainly explored within the context of tumorigenesis (Kim et al. 2006). To date, two mechanisms of regulation of gene expression via DNA methylation are known: On the one hand, there is site-specific methylation that directly interferes with binding of transcription factors (Comb and Goodman 1990); on the other hand, site-independent promoter-wide methylation attracts methylated DNA binding proteins, leading to an inactive chromatin structure and thereby gene silencing, which is determined by the density of methylated CPG-sites (Nan et al. 1997).

Disruption of epigenetic control mechanisms seems to be of importance in the pathophysiology of a variety of psychiatric disease entities such as schizophrenia (Abdolmaleky et al. 2006), depression (Hillemacher et al. 2007), suicide (McGowan et al. 2008), eating disorders (Frieling et al. 2008) and also alcohol dependence (Bonsch et al. 2005; Bleich et al. 2006; Hillemacher et al. 2008a).

However, little is known about the epigenetic regulation of genes involved in glutamatergic neurotransmission in patients with alcohol dependence. As suggested by previous studies, a marked increase in NR2B receptor subunit expression has been observed in peripheral blood lymphocytes during alcohol withdrawal (Biermann et al. 2007).

These findings inspired us to further investigate the epigenetic regulation of the NR2B receptor subunit by applying sodium bisulfite sequencing, in order to determine whether the promoter methylation pattern in the NR2B receptor is changed in patients suffering from alcohol dependency at different time points during withdrawal at single nucleotide resolution.

Materials and methods

Patients and design

The present study was approved by the Ethics Committee of the University of Erlangen-Nuremberg. Patients, who had an established diagnosis of alcohol dependence according to DSM-IV, were recruited from FARS (Franconian Alcoholism Research Studies, Bleich and Kornhuber 2005). They signed an informed consent form before participating in any part of the study. Female patients were excluded to avoid bias caused by sex specific metabolism aspects. All of those patients were seeking alcohol detoxification treatment of their own free will. Blood samples were drawn at the same time each morning.

To analyze NR2B receptor expression in peripheral blood lymphocytes, 32 male patients were extracted from the FARS data base.

Expression analysis worked only in 17 patients at different time points during ethanol withdrawal (day 0, 1, 3 and days 7–10; mean age 46.2 years, SD 8.6; range: 31–61 years) due to difficulties with RNA quality after longer storage. Each of these 17 patients had abstained from alcohol for between 0–72 h and had a mean history of alcohol consumption of 21.1 years (SD 12.5). Results from a subset of this population concerning NR2B mRNA expression have already been published (Biermann et al. 2007). However, the study in question included only 14 patients and failed to produce statistically significant results. In the present study, the main focus lies on the methylation analysis of the NR2B receptor promoter.

Bisulfite sequencing of the NR2B promoter in peripheral blood worked in only 22 out of the 32 patients from the FARS data base (mean age 45.5 years, SD 7.7; range 31–58 years) with a mean duration of alcohol consumption of 19.0 years (SD 10.2 years).

Mean blood alcohol levels at admission of non-abstinent patients was 153 mg/dl (SD 112.7). Mean daily ethanol intake was 259.8 g (SD 118.5). Lifetime drinking (LD, daily alcohol intake in kg × 365 × years of drinking) was 1,930 kg (SD 1300.4). The patients were all seen in a closed detoxification unit and took no vitamin supplements or other drugs (including withdrawal medication) before being enrolled in the study. The patients showed no nutritional abnormalities, liver cirrhosis, renal or pancreatic dysfunction or other psychiatric comorbidities. Study characteristics of the patients such as sociodemographic data, time of alcohol consumption, last alcohol intake, daily alcohol intake, and estimation of life-time drinking and medical history were taken using a semi-structured interview (Bleich and Kornhuber 2005).

NMDA 2b receptor expression analyses in peripheral blood lymphocytes

Fasting blood samples for RNA extraction were drawn on admission, on day 1, day 3 and days 7–10 in ethylenediaminetetraacetic (EDTA) acid-containing tubes and were stored at −80°C immediately after collection. Total RNA was extracted from whole frozen EDTA–blood using a modified Qiagen-protocol: A phenol-extraction in Qiazol® (QIAGEN GmbH, Hilden, Germany) was followed by column-purification with Rneasy® Mini Kit (QIAGEN GmbH, Hilden, Germany), including DNase digestion. Reverse transcription was done by using poly-dT primers and AMV reverse transcriptase from the Roche® cDNA Synthesis System (Roche Applied Science, Indianapolis, IN, USA). Quantitative PCR was performed using SYBR Green I® Master Mix buffer (Applied Biosystems), and reactions were run on an iCycler (BioRad™) using a three-step standard protocol. The annealing temperature was optimized for all primer pairs and was 61.5°C (Biermann et al. 2007). PCR products were visualized on standard 2.0% agarose gels with ethidium. β-Actin was used as an internal standard, and ∆CT values were calculated from differences between β-Actin and NR2B. All experiments were repeated at least three times and the mean value was used for further analysis.

The following primer pairs were used:




Bisulfite sequencing for methylation analysis of the promoter of NR2B receptor

Fasting blood samples for DNA extraction were drawn on admission and on days 7–10 in ethylenediaminetetraacetic (EDTA) acid-containing tubes and were stored at −80°C immediately after collection. Genomic DNA was extracted from whole frozen EDTA–blood with the QIAamp® DNA Blood Mini Kit (QIAGEN Inc., Valencia, CA, USA) according to the manufacturer’s instructions. Afterwards, 500 ng of genomic DNA were modified by sodium bisulfite using the EpiTect® Bisulfite Kit (QIAGEN Inc., Valencia, CA, USA) to convert cytosines to uracils, while leaving methylated cytosines unaltered.

The NR2B promoter region comprises 806 base pairs containing two CPG-islands. Primers were designed to amplify a region covering a fragment of 348 base pairs (from −956 to −608 bp prior to the starting base pair of the exon of NR2b) containing 42 CPG-sites within the first CPG-island of the NR2B promoter region.

To amplify a specific product, a semi-nested PCR was performed as follows: the first PCR was carried out with a final volume of 25 μl, containing 0.4 μM of each primer, 2 μl of template DNA and 12.5 μl HotStarTaq® Polymerase Mastermix (QIAGEN Inc., Valencia, CA, USA). The first amplification was carried out with initial denaturing at 97°C for 10 min, followed by 25 cycles of denaturing at 95°C for 2 min, annealing for 90 s at 53°C and extension for 1 min at 72°C. Afterwards, a second PCR was performed with a final volume of 25 μl, containing 0.4 μM of each primer for the semi-nested PCR, 2 μl of template DNA of the first PCR and 12.5 μl HotStarTaq® Polymerase Mastermix (QIAGEN Inc., Valencia, CA, USA). The amplification of the first PCR product was amplified over 25 cycles accordingly, the annealing temperature was 58°C.

The respective primers for the first PCR were:


And for the semi-nested second PCR:


Primers were designed using specific and freely available software (; the results are shown in Fig. 1. Furthermore, Fig. 2 shows the analyzed sequence of the promoter comparing the original sequence to the analyzed sample using the BiQ Analyzer software (Bock et al. 2005).
Fig. 1

The promotor region of the NMDA-2B receptor The promoter region of the NMDA-2B receptor contains 2 CPG-island. The first was analyzed in the present study as it contains more CPG sites as indicated by the vertical marks below. The grey region shows the two CPG-islands of the promoter region in general. The figure was prepared using methylation primer design software by Urogene: (
Fig. 2

The genomic sequence in comparison to a methylated sequence of a patient’s sample. The CPG sites are underlined and thereby highlighted. This was done using the BiQ Analyzer Software (Bock et al. 2005)

The PCR products were visualized on standard 2.0% agarose gels and purified with the Qiaquick® PCR Purification Kit (QIAGEN Inc., Valencia, CA, USA). Subsequently, an extension reaction was performed with the reverse primer of the second PCR to prepare for sequencing, which was performed using a BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) and Applied Biosystems 3,730 × l DNA Analyzer (Applied Biosystems) according to the manufacturer’s instructions. The resulting sequences were analyzed using special software called ESME (Epigenomics AG, Berlin, Germany) to determine the DNA methylation levels from the sequence trace files. Briefly, ESME performs quality control, normalizes signals, corrects for incomplete bisulfite conversion, and maps positions in the trace file to CpGs in the reference sequence. The program calculates methylation levels by comparing the C to T peaks at CpG sites (Lewin et al. 2004).

Statistical analyses

Distributions of all utilized variables were tested using the method of Kolgomorov–Smirnov. A general linear model for repeated measurements was performed. The alteration of the expression of the NR2B receptor subunits between the four time points was tested for differences with paired t tests. All statistical tests were two-tailed; a significance level of α = 0.05 or less was considered significant.

The methylation of each CPG-site in the CPG-island of the NR2B promoter was assessed using the ESME computer program, which compares each methylation site to the original sequence of the promoter. Afterwards, a cluster selection algorithm was performed to discover regions of interest in the CPG-island, including 42 that we were able to sequence. The definition of the cluster was as follows:
  • Criterion 1: median methylation in % >0%; mean methylation ≥10% for each single CPG-site.

  • Criterion 2: a cluster should comprise at least three CPG-sites.

  • Criterion 3: all of the CPG-sites have to be positively and significantly correlated.

  • Criterion 4: there are no statistically significant differences between each of the CPG-sites within one cluster. Of the 42 CPG-sites included, 18 fragments containing CPGs met criterion 1, whereas only three areas met criterion 2. At least two areas within the fragment of 348 bp conformed to criterion 3, though only the last fragment containing 5 methylated CPG-sites from −778 to −796 bp prior to the start base pair of the exon of NR2B met criterion 4. Therefore, these five CPG-sites were defined as a relevant cluster and implemented in the statistical analysis.

For methylation analyses, general linear models as well as regression analyses were performed. Data were analyzed using SPSS™ for Windows 16.0.2 (SPSS Inc., Chicago, IL).


The general linear model for repeated measurements showed a statistically significant change in the expression of the NR2B receptor in peripheral blood lymphocytes (GLM; F = 5.1; P = 0.013) between day 0, 1, 3 and days 7–10. In relation to the baseline values at day 0, the expression of the NR2B receptor was significantly increased compared with day 1 (t = 4.1; P = 0.001) and day 3 (t = 2.4; P = 0.029), whereas there was no statistically significant change between day 0 and 7. The results are highlighted in Fig. 3.
Fig. 3

Changes in the expression (ΔCT β-Actin) of the NR2B receptors on day 0, day 1, day 3 and day 7 of alcohol withdrawal. “○” represent cases with more than 1.5 box lengths from the upper or lower edge of the box. The box length is the interquartile range

Altogether, the methylation of each of the 42 analyzed CPG-sites within the first CPG-island of the NR2B promoter did not differ significantly between day 0 and 7. The mean methylation of all analyzed CPGs also did not differ.

To discover regions of interest within the analyzed fragment of the NR2B promoter, a cluster analysis was performed as described above. Accordingly, the fragment containing 5 methylated CPG-sites from −778 to −796 bp prior to the start base pair of the exon of NR2b was defined as a region of interest. In a correlation matrix, all five CPG-sites were significantly correlated with each other (0.522 ≤ rho ≤ 0.816; 0.0005 ≤ P ≤ 0.018). Therefore, the mean values of methylation were used for further analysis. The methylation of this cluster on day 0 was dependent on the severity of alcohol addiction. The amount of consumed alcohol [measured by life-time drinking (LD) and daily intake (DI) as described above] was negatively correlated with the methylation of this cluster, so that the higher the quantity of ethanol consumption, the lower the methylation of this cluster of CPGs within the promoter (LD: Spearman’s rho = −0.55; P = 0.013; DI: rho = −0.46; P = 0.043).

The influence of lifetime drinking and daily ethanol intake on methylation status of this region of interest on day 0 (admission) was modelled by multiple linear regressions (dependent variable: methylation of the CPG-cluster adjusted for age and blood alcohol concentration on admission). In this context, LD and DI were taken as measurable values for the extent of severity of alcohol addiction.

A significant influence could be proven for both models. Both lifetime drinking (quality of model R = 0.58; T = −2.7; P = 0.015) and daily intake (R = 0.61, T = −2.9; P = 0.010) were significantly associated with the CPG-cluster methylation.

Mean values for the variables in the model are displayed in Table 1. For clarity, the metric methylation values were categorized into three groups.
Table 1

CPG-cluster methylation and daily intake and lifetime drinking of alcohol

CPG-cluster methylation

Age in years (mean, SD)

BAC day 0 in mg/dl (mean, SD)

Daily intake in g (mean, SD)

Lifetime drinking in kg (mean, SD)

NR2B-expression Delta CT β-actin (mean, SD)

Lowest (0.0–0.03%)

45.0 (6.3)

150.5 (140.8)

344 (83)

2,597 (989)

5.9 (5.6)

Low (0.031–0.158%)

47.1 (9.3)

154.6 (118.8)

240 (118)

2,087 (1,510)

7.6 (4.8)

Medium-high (0.159–0.42%)

44.4 (8.5)

153.7 (84.9)

192 (108)

1,130 (1,089)

9.7 (4.0)

The CPG methylation groups (lowest, low and medium-high) are defined by a split of the methylation values on day 0 into three size-equal groups with 8, rsp. 7 participants. The values are given as means and standard deviations (SD). The higher the Delta-CT values of the NR2B receptor, the lower the expression of this receptor. The assumed relation between the CPG-cluster methylation status and the peripheral expression of the NR2B receptor could not be proven by statistical means

This result could be confirmed when the methylation of the region of interest was normalized for the expression of the NR2B receptor in peripheral blood lymphocytes (Model LD: R = 0.74; T = −2.8; P = 0.015; DI: R = 0.74 T = −2.7; P = 0.017). In these statistical models, the blood alcohol concentration on admission was also a significant predictor of quantitative methylation status for peripheral expression (3.38 < T ≤ 3.52; 0.004 ≤ P ≤ 0.005).

Concerning days 7–10 in alcohol withdrawal, the region of interest from −778 to −796 bp prior to the start base pair of the exon of NR2B was less methylated than on the day of admission (14.2%; SD: 16.0 vs. 10.9%; SD: 14.5). However, this decrease failed to reach a significant level. The “severity” of alcohol dependency measured through the amount of consumed ethanol (LD and DI) was likewise not correlated with the methylation status of the fragment of interest on day 7.


The main purpose of the present study was to explore whether the methylation status of the NR2B promoter is changed during ethanol withdrawal in a clinical sample.

Concerning peripheral NR2B expression, the majority of distinctions could be observed during early withdrawal from day 0 to day 3, in accordance with previous findings (Biermann et al. 2007). These results confirm the role of NMDA receptors as contributors to the expression of hyperexcitability during alcohol withdrawal (Wong et al. 1998). Ethanol consumption is known to inhibit NMDA receptors (Li et al. 1994), leading to an increase in their number and functioning (Tsai et al. 1998). According to the results of the present study, the NR2B receptor subtype expression changed mostly during early withdrawal and nearly reached baseline levels after cessation of clinical withdrawal symptoms 7–10 days after admission, a finding also observed in animal studies (Kalluri et al. 1998).

Nevertheless, the peripheral NR2B receptor expression was not correlated with methylation status, at least of the analyzed fragment. There are several explanations for the missing connection between expression and methylation status: firstly, it may be due to the relatively small number of participants. Secondly, the regulation of peripheral NR2B expression may be controlled via the methylation status of other CPG-sites of the relatively large NR2B promoter region of 906 bp not included in the analysis, or even through mechanisms other than DNA methylation (McGowan et al. 2008). The latter include trafficking of NMDA receptors that have been described to play a key function underlying channel function (Qiang et al. 2007) and may therefore have an impact on peripherally measurable receptor expression. However, an important conclusion that can be derived from in vitro studies of neurons originating from different brain regions of various animals is that chronic ethanol exposure does not always result in compensatory changes in ionotropic glutamate receptor subunit expression (Ferreira et al. 2001).

Concerning methylation status, a cluster of 5 CPG sites could be identified that was significantly influenced by the severity of alcoholism, with lesser methylation associated with higher amounts of ethanol intake in the patients’ histories.

These findings are in line with the results of animal studies, which have described that acute ethanol exposure inhibits NMDA receptor functioning and that long-term ethanol exposure results in maladaptive increases in the expression of some of these receptors in neurons (Ferreira et al. 2001). This might explain the presented result of an inverse correlation between the amount of alcohol intake and methylation. The more alcohol consumed by alcohol dependent patients, the lesser their NR2B methylation status, leading to a higher expression of NR2B receptors. The NR2B receptor subunit particularly contributes to hyperexcitability during alcohol withdrawal, although at the same time it is known that persistence of ethanol consumption prevents the genesis of withdrawal symptoms generally associated with down-regulation of NR2B receptor number and functioning (Nagy et al. 2005). Consequently, the methylation status of the described cluster might lead to an up-regulation of NR2B receptor subunits that are inhibited directly or indirectly by ethanol itself (Li et al. 1994; Grant and Lovinger 1995; Calton et al. 1999). When ethanol intake is abruptly stopped, the increased number of glutamate receptors might explain the aggravated withdrawal symptoms or ethanol tolerance in relation to the length and severity of alcohol addiction and the quantity of ethanol intake during the time before withdrawal (Littleton 2001; De Witte et al. 2003).

According to the results of the present study, relevant correlations between severity of dependence and blood alcohol concentration on admission were no longer found after completion of clinically relevant withdrawal symptoms 7–10 days after admission. The methylation status was lower than on the day of admission, but failed to reach the level of significance. Therefore, it should be emphasized that the cluster of interest of the NR2B promoter in the present study is relevant concerning early detoxification, but is not altered measurably during long-term withdrawal with a predominance of affective and cognitive disturbances or craving (Fadda and Rossetti 1998).

However, little is known about the role of each individual CPG-site within the NR2B promoter region and their clinical relevance. This study demonstrates a method of defining a potentially relevant cluster of CPG-sites within the NR2B promoter whose methylation was dependent on the “severity” of alcoholism.

Limitations of the study

Methylation levels in general are highly tissue and region specific. One limitation of the present study may be that we measured epigenetic changes in leukocyte DNA instead of brain cells. However, it is difficult to obtain brain cells to study central nervous transmission pathways. It has been hypothesized that methylation and RNA-expression in peripheral blood cells may reflect changes in the central nervous system (Czermak et al. 2004; Hillemacher et al. 2008b). Nevertheless, it is possible that peripheral and central NMDA2B expression are not correlated at all. However, the pathophysiological explanation of a lower methylation status of the NMDA2B promoter region in connection with severity of alcohol dependence as measured by life time drinking and higher daily ethanol intake seems to be plausible so that the presented results in peripheral blood lymphocytes at least might reflect brain status.

Furthermore, the present study fails to answer the question whether ethanol or withdrawal medications were causative in changing peripheral methylation levels. As the number of patients in this preliminary study is relatively small, the given results should be treated cautiously. Furthermore, only the samples with the most CPG sites were analyzed in the present study. Further research is under way on other fragments of the NMDA-promoter and the pathophysiological function of NMDA receptors in alcoholism in greater detail.

The method of direct bisulfite sequencing has mostly been applied in cancer research (Bock and Lengauer 2008). So far, little is known about the applications of this method in psychiatric research.

Thus, the results of this study are of particular importance despite the small sample size, demonstrating a significant effect of the severity of alcohol dependence on the NR2B methylation.


We thank Claire Coffey who is a Research Officer in the National Suicide Research Foundation in Cork, Ireland, for having done a wonderful job by giving advice for the improvement of the clarity and readability of the manuscript.

Copyright information

© Springer-Verlag 2009