Elevated levels of endocannabinoids and CB1 receptor-mediated G-protein signaling in the prefrontal cortex of alcoholic suicide victims

Elevated Levels of Endocannabinoids and CB1 Receptor-Mediated G-Protein Signaling in the Prefrontal Cortex of Alcoholic Suicide Victims K. Yaragudri Vinod, Victoria Arango, Shan Xie, Suham A. Kassir, J. John Mann, Thomas B. Cooper, and Basalingappa L. Hungund Background: Alcoholism is often comorbid with mood disorders and suicide. We recently reported an upregulation of CB1 receptor-mediated signaling in the dorsolateral prefrontal cortex (DLPFC) of subjects with major depression who died by suicide. In the present study, we sought to determine whether the changes in depressed suicides would also be present in alcoholic suicides and whether the endocannabinoid (EC) system plays a role in suicide in alcoholism. Methods: The density of CB1 receptor and its mediated [35S]GTP␥S signaling were measured in the DLPFC of alcoholic suicides (AS) (n ⫽ 11) and chronic alcoholics (CA) (n ⫽ 11). The levels of ECs were measured by a liquid chromatograph/mass spectrometry. Results: The CB1 receptor density was higher in AS compared with the CA group in the DLPFC. Western blot analysis confirmed a greater immunoreactivity of the CB1 receptor in AS. The CB1 receptor-mediated [35S]GTP␥S binding indicated a greater signaling in AS. Higher levels of N-arachidonyl ethanolamide and 2-arachidonylglycerol were observed in the DLPFC of AS. Conclusions: The elevated levels of ECs, CB1 receptors, and CB1 receptor-mediated [35S]GTP␥S binding strongly suggest a hyperactivity of endocannabinoidergic signaling in AS. EC system may be a novel therapeutic target for the treatment of suicidal behavior. Key Words: Alcoholism, suicide, human prefrontal cortex, endocannabinoids, CB1 receptor, G-protein

S

everal natural lipids have emerged as candidate modulators of nervous system functions. For example, two putative endocannabinoids (ECs), N-arachidonyl ethanolamide (AEA) and 2-arachidonylglycerol (2-AG), were shown to act as agonists for cannabinoid (CB) receptors (Devane et al 1992; Sugiura et al 1995), giving rise to the concept of an endocannabinoid system. The CB receptor family consists of two G-protein coupled receptors (GPCRs), CB1 and CB2. CB1 receptors are localized primarily in neural tissue, whereas CB2 receptors are located peripherally and are associated with the immune system. CB1 receptors are known to inhibit adenylyl cyclase (AC) (Howlett 1995; Childers and Breivogel 1998a) via Gi-protein and they are the most abundant neuromodulatory receptors in the brain, expressed in the cortex, hippocampus, cerebellum, and basal ganglia (Herkenham et al 1991; Glass et al 1997; Childers and Breivogel 1998b). High levels of ECs have also been found in the cortex, basal ganglia, and limbic structures, and this may suggest involvement of the EC system in cognitive, emotional, and motivational functions (Bisogno et al 1999). Alterations in the monoaminergic neurotransmitter systems have long been implicated in the pathophysiology of depression and more recently in suicide (Arango et al 1990, 1995, 2002; Callado et al 1998; Gonzalez-Maeso et al 2002). Lower serotoFrom the Department of Analytical Psychopharmacology (KYV, SX, TBC, BLH), Nathan Kline Institute for Psychiatric Research, Orangeburg, New York; and Departments of Analytical Psychopharmacology (KYV, TBC, BLH) and Neuroscience (VA, SAK, JJM), New York State Psychiatric Institute, and Department of Psychiatry (VA, JJM, BLH), College of Physicians and Surgeons, Columbia University, New York, New York. Address reprint requests to Basalingappa L. Hungund, Ph.D., Nathan Kline Institute for Psychiatric Research, 140 Old Orangeburg Road, Orangeburg, NY 10962; E-mail: [email protected]. Received July 8, 2004; revised November 17, 2004; accepted November 19, 2004.

0006-3223/05/$30.00 doi:10.1016/j.biopsych.2004.11.033

nergic activity may, to some degree, mediate genetic and developmental changes in suicide, aggression, mood, and alcoholism (for reviews, see Mann 1998, 2003). In recent years, the EC system has been slowly emerging as a target of psychiatric research, and alterations in CB1 receptor levels in neurodegenerative diseases (Glass et al 1993; Westlake et al 1994) and schizophrenia (Dean et al 2001) have been reported. Recently, we reported elevated CB1 receptor-mediated signaling in the dorsolateral prefrontal cortex (DLPFC) of depressed suicide victims (Hungund et al 2004). Our earlier animal studies have also suggested the participation of the EC system in alcoholism (for a review, see Hungund et al 2002). Clinical evidence suggests that alcoholism and affective disorders are independently associated with suicide behavior (Roy 2000; Preuss et al 2002). Alcoholism and other substance use disorders are related to more aggression and impulsivity, also risk factors for suicide (Rossow and Amundson 1995; Rich et al 1998; Koller et al 2002). It has been suggested that chronic alcoholics are at 60 to 120 times higher risk for suicide than the general population, and alcoholism may contribute to about 25% of all suicides (Murphy and Wetzel 1990). Alcoholism and attempted suicide are familial, suggesting a role for genes and rearing (Potash et al 2000), although the responsible genes remain to be identified. An association between the CB1 receptor gene and attention-deficit/hyperactivity disorder (ADHD) in alcoholics has also been recently reported (Ponce et al 2003). However, cellular mechanisms by which alcohol and the EC system produce behavioral changes in humans remain to be elucidated. The modulation of the EC system by alcohol and the frequency of suicidal behavior in alcoholism raises the question whether the EC system plays a role in alcoholism or suicidal behavior per se. In this study, we have analyzed the density of CB1 receptor, CB1 receptor-mediated G-protein signaling and EC levels in the DLPFC, and CB1 receptormediated G-protein signaling in the occipital cortex (OC) of alcoholic suicide victims and chronic alcoholic control subjects dying from other causes. BIOL PSYCHIATRY 2005;57:480 – 486 © 2005 Society of Biological Psychiatry

BIOL PSYCHIATRY 2005;57:480 – 486 481

K.Y. Vinod et al Table 1. Characteristics of Alcoholic Suicide and Chronic Alcoholic Subjects Group 1. CA 2. CA 3. CA 4. CA 5. CA 6. CA 7. CA 8. CA 9. CA 10. CA 11. CA Mean ⫾ SD 1. AS 2. AS 3. AS 4. AS 5. AS 6. AS 7. AS 8. AS 9. AS 10. AS 11. AS Mean ⫾ SD

Sex (M/F)

Age (Years)

PMI (hours)

Cause of Death

Toxicology

M M M M F F F F M M M

16 17 18 24 30 35 38 58 64 67 70 39.7 ⫾ 21.2 20 20 29 31 28 39 32 41 42 58 61 36.5 ⫾ 13.5

09 12 13 09 13 13 20 09 12 09 08 11.6 ⫾ 3.4 12 24 22 09 04 13 11 13 12 12 17 13.4 ⫾ 5.6

MVA, passenger MVA, passenger Homicide, stabbing Accident, drowning Natural, cardiovascular Natural, cancer Homicide, GSW MVA, passenger Natural, cardiovascular Natural, hemorrhage Accident, fall from height

None None None CO None None None None None None AE, BZ

Suicide, GSW Suicide, GSW Suicide, GSW Suicide, GSW Suicide, hanging Suicide, hanging Suicide, poisoning Suicide, hanging Suicide, GSW Suicide, hanging Suicide, GSW

None None None CO AE, BZ None None CO None None None

M M M M F F F F M M M

The data are presented as mean and SD. No statistical significance was observed between age (t ⫽ .43, df ⫽ 20, p ⬍ .67) and PMI (t ⫽ .95, df ⫽ 20, p ⬍ .35) across the groups. M, male; F, female; PMI, postmortem interval (rounded to nearest hours); CA, chronic alcoholic; MVA, motor vehicle accident; GSW, gunshot wound; CO, carbon monoxide; AE, antiepileptic; BZ, benzodiazapine; AS, alcoholic suicide.

Methods and Materials Subjects, Tissue Collection, and Dissection Brain samples of prefrontal cortex and occipital cortex were obtained from autopsy material in the Brain Collection of the Department of Neuroscience at the New York State Psychiatric Institute (NYSPI) and Columbia University, New York. All tissue used in this study was provided by the Allegheny County Coroner in accordance with protocols approved by the Institutional Review Board of the University of Pittsburgh, the New York State Psychiatric Institute, and Columbia University. Cortical tissues from Brodmann area 9 and visual association cortex (1.5 g each) were dissected frozen. The meninges and white matter were removed as much as possible, and tissues were returned immediately to ⫺80°C until membrane preparation. Brain samples from 11 chronic alcoholics (CA) (age range: 16 –70 years), who died from causes other than suicide, were studied together with a matched group of 11 subjects of alcoholic suicides (AS) who had a lifetime diagnosis of alcohol dependence and died by suicide (age range: 20 – 61 years). The coroner determined the cause of death and made the verdict of suicide. There were nine pairs of Caucasians and two pairs of African Americans. There were no significant differences in age, sex, and postmortem interval (PMI) distribution between AS and CA subjects. The demographic variables, such as sex, age, PMI, and cause of death, as well as toxicology, are summarized in Table 1. Toxicological analyses were performed on all cases, ruling out recent consumption of substances of abuse (except alcohol [EtOH]) or psychoactive drugs (an antiepileptic and a benzodiazapine were detected in two cases [CA ⫽ 1 and AS ⫽ 1]). All cases were free of neuropathology. Both AS and CA subjects were characterized psychiatrically by interviews with family

members and/or close friends using our previously validated psychological autopsy method and a structured clinical interview for Axis I and Axis II diagnoses according to DSM-III-R criteria (Kelly and Mann 1996). Psychological autopsies confirmed that all the subjects had a diagnosis of alcohol dependence or abuse. One of the suicides had Axis I diagnoses of major depression and schizoaffective disorder but was currently in remission. None of the subjects had comorbid nonalcohol substance use disorders. The samples were coded to mask investigators to group assignment. Membrane Preparation Brain tissue (⬃1 g) was homogenized in 20 volumes of ice-cold TME buffer (50 mM Tris-HCl, 3 mM magnesium chloride [MgCl2], and 1 mM of ethylenediaminetetraacetic acid [EDTA], pH 7.4) containing .32 M sucrose and a freshly added protease inhibitor cocktail. The homogenate was centrifuged at 1000g for 10 minutes at 4oC. The resulting supernatant was then centrifuged at 22,000g for 20 minutes. The pellet was dissolved in TME buffer and recentrifuged at 22,000g for 20 minutes. The final pellet was dissolved in TME buffer. Aliquots of membrane fractions were made and stored at ⫺80oC until assay. The protein content of the membrane fraction was determined by Lowry et al (1951) method using bovine serum albumin (BSA) as the standard. [3H]CP-55,940 Binding Assay An aliquot of membrane (100 ␮g protein) was incubated with TME buffer, [3H]CP-55,940 (.05–5.0 nM), and .1% fatty acid free BSA in silicone-treated test tubes for 1 hour at 37oC. The nonspecific binding of radioligand was defined by CP-55,940 (10 ␮M). The reaction was terminated by the addition of 2 mL ice-cold termination buffer (.1% BSA in 50 mM Tris-HCl, pH 7.4). The reaction mixture was rapidly filtered through polyethyleneiwww.elsevier.com/locate/biopsych

482 BIOL PSYCHIATRY 2005;57:480 – 486

K.Y. Vinod et al [35S]GTP␥S binding was achieved when cortical membranes were incubated with 1 ␮M CP-55,940 and 40 ␮M of GDP. The basal activity was estimated in absence of CP-55,940. To determine the specificity of CB1 receptor-stimulated [35S]GTP␥S binding, a concentration-response effect in the presence of the CB1 receptor antagonist SR141716A was also studied. The nonspecific binding of radioligand was determined in the presence of 10 ␮M GTP␥S. The termination and filtration (without presoaking the filters in polyethyleneimine) of reaction mixture was performed as described for [3H]CP-55,940 binding assay. The radioactivity was measured by liquid scintillation spectroscopy at an efficiency of 95% for 35S.

Figure 1. Scatchard transformation of saturation binding of [3H]CP-55,940 (.5–5.0 nm) to the prefrontal cortical membranes of a AS (alcoholic suicide) and CA (chronic alcoholic) subjects. The inset represents the Scatchard transformation of the same binding data.

mine (.1%) pretreated glass fiber filters using a Brandel 24position cell harvester (Brandel, Gaithersburg, Maryland). Filters were washed three times with the termination buffer and transferred to scintillation vials containing 5 mL of scintillation cocktail (ICN Biochemicals, Costa Mesa, California). The vials were incubated overnight at room temperature. The radioactivity was measured by liquid scintillation spectroscopy (Beckman Instruments, Fullerton, California) at an efficiency of 47% for tritium. The Bmax (maximal binding sites) and Kd (apparent dissociation constant) values were determined from saturation isotherms using nonlinear regression analysis to fit the data to the single binding equation. CB1 Receptor Immunoblot Analysis Immunoblot analysis was used to confirm the presence of higher CB1 receptor protein in the DLPFC of AS victims. An aliquot of membrane protein (30 ␮g) separated by 12% polyacrylamide gel was electrophoretically transferred to nitrocellulose membrane. The membrane was treated with blocking buffer (TTBS [10 mM Tris, .9% sodium chloride [NaCl]; 1% Tween 20 containing 3% milk powder] of pH 7.4) for 1 hour at room temperature. The membrane was incubated with human anti-CB1 receptor antibody (1:500, Biosource International, Camarillo, California) overnight at 4oC. The blot was washed three times with TTBS and then incubated with alkaline phosphatase conjugated anti-immunoglobulin G (IgG) for 1 hour at room temperature. After washing the blot for three to four times with TTBS, the immunoreactive band was visualized by CDP-Star reagent (Tropix, Bedford, Massachusetts). The blot was reprobed with ␣-tubulin antibody to ensure equal protein loading. Agonist-Stimulated [35S]GTP␥S Binding Assay The CB1 receptor-mediated [35S]GTP␥S binding assay was performed as described previously (Gonzalez-Maeso et al 2000) with minor modification. Briefly, an aliquot of membrane (50 ␮g protein) was preincubated in assay buffer (TME buffer and .1% fatty acid free BSA and 100 mM NaCl) containing guanosine 5=-diphosphate (GDP) (40 ␮M) in silicone-treated test tubes for 15 minutes. Later, [35S]GTP␥S (.05 nM) was added to the reaction mixture and incubated for 1 hour at 37oC. The CB1 receptor agonist, CP-55,940 (1 ␮M), was used to study CB1 receptormediated [35S]GTP␥S binding. A maximum stimulation of www.elsevier.com/locate/biopsych

Measurement of EC Levels by Gas Chromatography/Mass Spectrometry Endocannaboid levels were measured by a liquid chromatograph/mass spectrometry (LC-MS) method using the isotopic dilution procedure as described (Di Marzo et al 2000) with minor modifications. Briefly, brain tissue was homogenized in 4 mL of chloroform-methanol-Tris buffer (2:1:1, pH 7.4) containing .25 mM phenylmethylsulfonyl fluoride (PMSF), 20 ␮L of 1% butylated hydroxytoluene (BHT), .05 mL of AEA-d8 (50 ng), and .050 mL of 2-AG-d8 (500 ng). The homogenate was centrifuged and the organic layer removed and taken to dryness under nitrogen. The residue was dissolved in .3 mL of ethyl acetate and recentrifuged. The supernatant was transferred to another tube and dried down under nitrogen. The residue was redissolved in ethyl alcohol (40 ␮L) and transferred to a vial for the measurement of AEA and 2-AG by LC-MS (1100 series mass LC-MSD, Agilent Technologies, Wilmington, Delaware). The separation was achieved on a Supelcosil LC-8 column (Supelco, Bellefonte, Pennsylvania) (25 cm x 4.6 mm, 5 ␮m) using methanol-ammonium acetate-acetic acid (85:15:.05) as a mobile phase. The standard curves were fitted with a quadratic equation with the curve encompassing a range of .5–50 and 50 –1500 ng for AEA and 2-AG, respectively, and were processed similarly with quality controls with each batch of samples.

Figure 2. The density of CB1 receptor was estimated using [3H]CP-55,940 in the prefrontal cortical membranes of AS (alcoholic suicide, n ⫽ 11) and CA (chronic alcoholic, n ⫽ 11) subjects. **Significantly higher than that of CA subjects (t ⫽ 3.59, df ⫽ 18, p ⬍ .001).

K.Y. Vinod et al

BIOL PSYCHIATRY 2005;57:480 – 486 483

Figure 3. The Western blot analysis for CB1 receptor. A representative immunoblot of the CB1 receptor (A) and levels of CB1 receptor immunoreactivity (B) in the prefrontal cortical membranes of AS (alcoholic suicide, n ⫽ 11) and CA (chronic alcoholic; n ⫽ 11) subjects expressed in arbitrary densitometric units. *Significantly higher than that of CA subjects (t ⫽ 2.66, df ⫽ 20, p ⬍ .01).

Statistical Analyses Statistical analysis was performed by using GraphPad Prism (GraphPad software, San Diego, California). Experiment in each sample was run in at least duplicate unless otherwise indicated. All values presented are mean and standard deviation. Immunoblots were analyzed using the National Institutes of Health (NIH) image software program. The demographic variables (age and PMI) were compared between CA and AS subjects with independent t tests. Differences in binding parameters, protein, and EC levels between CA and AS groups as a whole were analyzed using independent-sample unpaired t test with two-tailed p values considered significant at p ⬍ .05.

Results The Density of CB1 Receptor: Radioligand Binding and Immunoblot Analyses The saturation binding analysis suggests that [3H]CP-55,940 binding sites were saturable at 5.0 nM concentration (Figure 1). The nonspecific binding was 10% to 18% of total [3H]CP-55,940 binding. A Scatchard analysis demonstrates high affinity (apparent dissociation constant [Kd] was 1.3 ⫾ .3 nm), saturable, monophasic binding, and a Hill coefficient of near unity indicates a single class of receptor at the concentrations used. Significantly higher (39%) density (Bmax) of the CB1 receptor was observed in the DLPFC of AS (820.1 ⫾ 139.2 fmol/mg protein) compared with CA (590.2 ⫾ 91.7 fmol/mg protein) (t ⫽ 3.59, df ⫽ 18, p ⬍ .0021) (Figure 2). There was no significant difference in the affinity of receptor between the groups (t ⫽ 1.59, df ⫽ 18, p ⫽ .12), indicating the upregulation of receptor density may not be due to its altered affinity. Western blot analysis also revealed a significantly higher (67%, t ⫽ 2.66, df ⫽ 20, p ⫽ .014) CB1 receptor immunoreactivity in the DLPFC of AS than CA subjects (Figure 3B). Furthermore, significant positive correlations between the Bmax (radioligand binding) and protein expression (immunoreactivity) of the CB1 receptor were observed in both CA (r ⫽ .94) and AS (r ⫽ .85) groups. A representative CB1 receptor immunoblot of a CA and matched AS is provided in Figure 3A. The Western blot analysis clearly demonstrated single immunoreactivity band, which corresponds to the relative molecular weight of the CB1 receptor. CB1 Receptor-Mediated [35S]GTP␥S Binding CB1 receptor agonist, CP-55,940-stimulated [35S]GTP␥S binding assay was performed in membranes isolated from the DLPFC and the OC of CA and AS subjects to asses the functional coupling between CB1 receptor and G-protein. Greater stimulation (34%; t ⫽ 5.38, df ⫽ 20, p ⫽ .0001) of CB1 receptor-mediated

[35S]GTP␥S binding was observed in cortical membranes of AS (102.1 ⫾ 10.5 fmol/mg protein) compared with CA (76.4 ⫾ 11.8 fmol/mg protein) subjects (Figure 4). However, this parameter was found to be unaltered (t ⫽ 1.30, df ⫽ 20, p ⫽ .20) in the OC of AS victims (58.9 ⫾ 8.8) when compared with CA group (54.2 ⫾ 8.1) (Figure 5). The pharmacological specificity of the CB1 receptor-mediated [35S]GTP␥S binding was tested by CB1 receptor antagonist, SR SR141716A. One-site competition analysis indicated a concentration-dependent inhibition of CP-55,940 stimulated [35S]GTP␥S binding by SR141716A with a median effective concentration (EC50) of 11.9 nM (Figure 6). Levels of Endocannabinoids Levels of both ECs, AEA and 2-AG, were measured in the DLPFC of AS and CA subjects. The levels of AEA (98.4 ⫾ 26.4 pmol/gm tissue) were significantly lower than 2-AG (2.20 ⫾ .80 nmol/gm tissue) in the DLPFC of CA subjects. Levels of both AEA (79%, 176.8 ⫾ 42.5 pmol/gm tissue, t ⫽ 5.18, df ⫽ 20, p ⫽ .0001) and 2-AG (47%, 3.24 ⫾ .94 nmol/gm tissue, t ⫽ 2.75, df ⫽ 20, p

Figure 4. The CB1 receptor-mediated [35S]GTP␥S binding was greater in the prefrontal cortical membranes of AS (alcoholic suicide, n ⫽ 11) compared with CA (chronic alcoholic, n ⫽ 11) subjects. Data are presented as percentage of stimulation over the basal. ***Significantly higher than that of CA subjects (t ⫽ 5.38, df ⫽ 20, p ⬍ .0001).

www.elsevier.com/locate/biopsych

484 BIOL PSYCHIATRY 2005;57:480 – 486

Figure 5. The CB1 receptor-mediated [35S]GTP␥S binding was unaltered (t ⫽ 1.30, df ⫽ 20, p ⬍ .20) in membranes isolated from the occipital cortex of AS (alcoholic suicide, n ⫽ 11) compared with CA (chronic alcoholic, n ⫽ 11) subjects. Data are presented as percentage of stimulation over the basal.

⫽.012) were significantly higher in the DLPFC of AS when compared with CA subjects (Figure 7).

Discussion We previously reported higher levels of CB1 receptors and its mediated G-protein signaling in the DLPFC of depressed suicide victims compared with nondepressed nonsuicide normal subjects (Hungund et al 2004). Postmortem studies of brains of depressed suicide victims, however, cannot address whether the observed alterations are due to the presence of major depression or suicidal behavior. Therefore, to further address this question, the present study was designed to investigate the prefrontal cortical EC system in alcoholics, comparing alcoholics who committed suicide to alcoholics who died by other causes. The results of this study indicating enhanced EC function in relation to suicide and considering our recent finding in depressed suicides (Hungund et al 2004) provides further evidence linking a hyperactive EC system in the DLPFC to suicide in both mood disorder and in alcoholism. The study reveals elevated levels of CB1 receptors in prefrontal cortical membranes of AS as compared with CA. In addition, immunoblot analysis demonstrated a higher immunoreactivity for CB1 receptor in the DLPFC of AS, corroborating the results of radioligand binding assay. The observed upregulation of CB1 receptors may reflect either increased rate of synthesis or decreased degradation of receptors but is not due to altered affinity of the receptors, as Kd values are not different between groups. G-protein coupled receptors play an important role in various cellular functions. Several studies have shown an abnormal receptor-G-protein coupling in the brain of suicide victims (Gurguis et al 1997; Gonzalez-Maeso et al 2000; Dwivedi et al 2002b). In the present study, [35S]GTP␥S was employed to asses the coupling efficiency of CB1 receptors and G-proteins in cortical membranes. We observed enhanced CB1 receptor-mediated [35S]GTP␥S binding in the DLPFC of AS compared with CA subjects. The pharmacological specificity of the CB1 receptormediated [35S]GTP␥S binding was confirmed by the CB1 receptor antagonist, SR SR141716A. Enhanced agonist-stimulated G-prowww.elsevier.com/locate/biopsych

K.Y. Vinod et al tein signaling does not appear to be due to altered affinity between the receptor and G-protein, as basal [35S]GTP␥S binding was similar in both the groups. It could be due to the observed high density of CB1 receptors in the DLPFC of AS. The consequence of elevated levels of CB1 receptors in the DLPFC of AS observed in this study is not known. Upregulation of CB1 receptors due to the feedback or a homeostatic response to low levels of ECs is not the explanation, because we observed elevated levels of both AEA and 2-AG in the DLPFC of AS. Elevated levels of both ECs (Leweke et al 1999; De Marchi et al 2003) and CB1 receptors have also been found in schizophrenic patients (Dean et al 2001), indicating that this pattern of abnormality may not be unique to suicide. However, it is not clear whether the alterations reported in these studies were associated with suicidal behavior or with schizophrenia. Delta-9tetrahydrocannabinol, a primary constituent of marijuana, which is a CB1 receptor agonist and mimics the action of ECs, appears to increase certain forms of impulsive behaviors in humans (McDonald et al 2003). The high levels of ECs/CB1 receptor may facilitate impulsive behavior, which is characteristic of chronic alcoholics and depressed subjects with suicidal behavior. Impulsivity is thought to be part of a diathesis for suicidal and aggressive behaviors, increasing the probability of acting on such feelings (for a review, see Mann et al 1999). The regulatory mechanisms governing the EC system remain largely unknown. Neuroanatomical and electrophysiological studies of the mammalian nervous system have revealed that the CB1 receptors are located in presynaptic terminals of neurons (Kreitzer and Regehr 2001). Fatty acid amide hydrolase (FAAH), a membrane bound serine hydrolase that inactivates ECs, appears to be preferentially located in the somatodendritic compartments of neurons. Endocannabinoids are released on demand by stimulated neurons and activate presynaptic CB1 receptors. Based on these findings, it has been suggested that ECs are synthesized by postsynaptic cells that act as retrograde messengers to modulate neurotransmitter release (Kreitzer and Regehr 2001; Wilson and Nicoll 2002). Therefore, there may be elevated retrograde signaling mediated by the EC system in the DLPFC of suicide victims. The relevance of impaired cannabinoidergic signaling to the pathophysiology of suicide remains to be elucidated. The

Figure 6. To determine the specificity of CB1 receptor-stimulated [35S]GTP␥S binding, a concentration-response effect in the presence of the CB1 receptor antagonist SR141716A was studied in the cortical membranes of CA (chronic alcoholic) subjects. Inhibition of CP-55,940-stimulated [35S]GTP␥S binding by SR141716A was concentration-dependent with an EC50 of 11.9 nM. EC50, median effective concentration.

K.Y. Vinod et al

BIOL PSYCHIATRY 2005;57:480 – 486 485

Figure 7. Levels of AEA (A) and 2-AG (B) were estimated in the DLPFC of AS (alcoholic suicide, n ⫽ 11) and CA (chronic alcoholic, n ⫽ 11) subjects. aSignificantly higher in the levels of AEA in AS than that of CA subjects (t ⫽ 5.18, df ⫽ 20, p ⬍ .0001). bSignificantly higher in the levels of 2-AG in AS than that of CA subjects (t ⫽ 2.75, df ⫽ 20, p ⬍ .01). AEA, Narachidonyl ethanolamide; 2-AG, 2-arachidonylglycerol; DLPFC, dorsolateral prefrontal cortex.

sensitization of cannabinoidergic signaling could lead to decreased cyclic adenosine monophosphate (cAMP) content of the cell, as these receptors are negatively coupled to AC, resulting in the reduced cAMP dependent protein kinase A (PKA) and cAMP response element binding protein (CREB) activities. A reduced activity of cAMP/PKA and CREB signaling in the cortex of suicide brain has been reported (Dowlatshahi et al 1999; Reiach et al 1999; Dwivedi et al 2002a, 2003). Many biological responses are regulated by the state of phosphorylation of specific proteins, which in turn affects the regulation of cellular functions. Thus, a hyperactivity of cannabinoidergic signaling may result in impairments in various physiological functions that manifest in suicidal behavior. Several studies have highlighted the effect of age, sex, and PMI on the levels and function of receptors and G-proteins (Dowlatshahi et al 1999) and also the effect of PMI on the levels ECs (Schmid et al 1995). However, the brain samples analyzed in this study were closely matched with regard to sex, age, ethnic background, and PMI, and no significant group differences in these demographic variables were observed. Suicide is associated with several psychiatric disorders and also drugs of abuse (Suominen et al 1996; Rich et al 1998; Kessler et al 1999; Garlow 2002). However, all the subjects included in this study had a diagnosis of alcoholism and no drug of abuse other than EtOH was detected postmortem, and yet the elevation in CB1 receptors occurred only in the suicide group. Whether the observed alterations in the DLPFC of AS is due to the pathophysiology of suicide or psychotropic medication is of particular relevance. However, our toxicological analyses revealed that only two cases (one from each group) had medications (antiepileptic and benzodiazapine) at the time of death and no other psychoactive drugs were detected in remaining subjects. Therefore, the present findings in brains of AS are unlikely to be the result of short-term drug effects and more likely to be related to the pathophysiology of suicide. In the present study, we included CA as a psychiatric control group for AS instead of normal control groups. This enabled us to identify the biochemical alterations in the brain of alcoholic suicide victims with greater specificity. Alteration of CB1 receptor-mediated G-protein signaling seems to be region-specific because the CB1 receptor-mediated [35S]GTP␥S signaling was found to be unaltered in the OC of AS victims. This may be due to unaltered density of CB1 receptor in the OC of AS victims. Future studies are needed to examine the components of EC system in the human postmortem brains of alcoholics, alcoholic suicides, and normal control subjects in

more brain regions of interest to further understanding the role of EC system in the neurobiology of alcoholism and suicide. In conclusion, convergence of three different approaches pointing to the upregulation of CB1 receptors taken together with higher levels of ECs strongly suggests that hyperactivity of the EC signaling in the DLPFC may be a potential factor in suicide in alcoholism. The present study corroborates our previous observations and further elucidates the association between hyperactivity of EC system and suicide. Pharmacological agents that modulate EC function would be of interest in the potential development of treatment for suicidal behavior. This study was supported by Grants AA13003 and National Alliance for Research on Schizophrenia and Depression, Independent Investigator Award (BLH); AA09004 and MH40210 (VA); MH62185 (JJM); and American Foundation for Suicide Prevention Young Investigator Grant (KYV). We thank Dr. Veeranna, Center for Dementia Research, Nathan Kline Institute, for his technical advice. Arango V, Ernsberger P, Marzuk PM, Chen JS, Tierney H, Stanley M, et al (1990): Autoradiographic demonstration of increased serotonin 5-HT2 and beta-adrenergic receptor binding sites in the brain of suicide victims. Arch Gen Psychiatry 47:1038 –1047. Arango V, Underwood MD, Gubbi AV, Mann JJ (1995): Localized alterations in pre- and postsynaptic serotonin binding sites in the ventrolateral prefrontal cortex of suicide victims. Brain Res 688:121–133. Arango V, Underwood MD, Mann JJ (2002): Serotonin brain circuits involved in major depression and suicide. Prog Brain Res 136:443– 453. Bisogno T, Berrendero F, Ambrosino G, Cebeira M, Ramos JA, FernandezRuiz JJ, et al (1999): Brain regional distribution of endocannabinoids: Implications for their biosynthesis and biological function. Biochem Biophys Res Commun 256:377–380. Callado LF, Meana JJ, Grijalba B, Pazos A, Sastre M, Garcia-Sevilla JA (1998): Selective increase of alpha2A-adrenoceptor agonist binding sites in brains of depressed suicide victims. J Neurochem 70:1114 –1123. Childers SR, Breivogel CS (1998a): Cannabis and endogenous cannabinoid systems. Drug Alcohol Depend 51:173–187. Childers SR, Breivogel CS (1998b): The functional neuroanatomy of brain cannabinoid receptors. Neurobiol Dis 5:417– 431. De Marchi N, De Petrocellis L, Orlando P, Daniele F, Fezza F, Di Marzo V (2003): Endocannabinoid signalling in the blood of patients with schizophrenia. Lipids Health Dis 19:2–5. Dean B, Sundram S, Bradbury R, Scarr E, Copolov D (2001): Studies on [3H]CP-55940 binding in the human central nervous system: Regional specific changes in density of cannabinoid-1 receptors associated with schizophrenia and cannabis use. Neuroscience 103:9 –15.

www.elsevier.com/locate/biopsych

486 BIOL PSYCHIATRY 2005;57:480 – 486 Devane WA, Hanus L, Breuer A, Pertwee RG, Stevenson LA, Griffin G, et al (1992): Isolation and structure of a brain constituent that binds to the cannabinoid receptor. Science 258:1946 –1949. Di Marzo V, Breivogel CS, Tao Q, Bridgen DT, Razdan RK, Zimmer AM, et al (2000): Levels, metabolism, and pharmacological activity of anandamide in CB1 cannabinoid receptor knockout mice: Evidence for non-CB1, nonCB2 receptor-mediated actions of anandamide in mouse brain. J Neurochem 75:2434 –2444. Dowlatshahi D, MacQueen GM, Wang JF, Reiach JS, Young LT (1999): G Protein-coupled cyclic AMP signaling in postmortem brain of subjects with mood disorders: Effects of diagnosis, suicide, and treatment at the time of death. J Neurochem 73:1121–1126. Dwivedi Y, Conley RR, Roberts RC, Tamminga CA, Pandey GN (2002a): [3H]cAMP binding sites and protein kinase A activity in the prefrontal cortex of suicide victims. Am J Psychiatry 159:66 –73. Dwivedi Y, Rao JS, Rizavi HS, Kotowski J, Conley RR, Roberts RC, et al (2003): Abnormal expression and functional characteristics of cyclic adenosine monophosphate response element binding protein in postmortem brain of suicide subject. Arch Gen Psychiatry 60:273–282. Dwivedi Y, Rizavi HS, Conley RR., Roberts RC, Tamminga CA, Pandey GN (2002b): mRNA and protein expression of selective alpha subunits of G proteins are abnormal in prefrontal cortex of suicide victims. Neuropsychopharmacology 27:499 –517. Garlow SJ (2002): Age, gender, and ethnicity differences in patterns of cocaine and ethanol use preceding suicide. Am J Psychiatry 159:615– 619. Glass M, Dragunow M, Faull RL (1997): Cannabinoid receptors in the human brain: A detailed anatomical and quantitative autoradiographic study in the fetal, neonatal and adult human brain. Neuroscience 77:299 –318. Glass M, Faull RL, Dragunow M (1993): A significant loss of cannabinoid receptors in the substantia nigra in Huntington’s disease. Neuroscience 56:523–527. Gonzalez-Maeso J, Rodriguez-Puertas R, Gabilondo J, Meana JJ (2000): Characterization of receptor mediated [35S]GTP␥S binding to cortical membrane from post-mortem human brain. Eur J Pharmacol 390:25–36. Gonzalez-Maeso J, Rodriguez-Puertas R, Meana JJ, Garcia-Sevilla JA, Guimon J (2002): Neurotransmitter receptor-mediated activation of G-proteins in the brains of suicide victims with mood disorders: Selective supersensitivity of ␣2A-adrenoceptors. Mol Psychiatry 7:755–767. Gurguis GN, Turkka J, George DT, Linnoila M (1997): Beta-adrenoreceptor coupling to GS protein in alcohol dependence, panic disorder, and patients with both conditions. Neuropsychopharmacology 16:69 –76. Herkenham M, Lynn AB, Johnson MR, Melvin LS, deCost BR, Rice KC (1991): Characterization and localization of cannabinoid receptors in rat brain; a quantitative in vitro autoradiographic study. J Neurosci 16:8057– 8066. Howlett AC (1995): Pharmacology of cannabinoid receptors. Annu Rev Pharmacol Toxicol 35:607– 634. Hungund BL, Basavarajappa BS, Vadasz C, Kunos G, Rodriguez de Fonseca F, Colombo G, et al (2002): Ethanol, endocannabinoids, and the cannabinoidergic signaling system. Alcohol Clin Exp Res 26:565–574. Hungund BL, Vinod KY, Kassir SA, Basavarajappa BS, Yalamanchili R, Cooper TB, et al (2004): Upregulation of CB1 receptors and agonist-stimulated [35S]GTP␥S binding in the prefrontal cortex of depressed suicide victims. Mol Psychiatry 9:184 –190. Kelly TM, Mann JJ (1996): Validity of DSM-III-R diagnosis by psychological autopsy: A comparison with antemortem diagnosis. Acta Psychiatr Scand 94:337–343. Kessler RC, Borges G, Walters EE (1999): Prevalence of risk factors for lifetime suicide attempts in the National Comorbidity Survey. Arch Gen Psychiatry 56:617– 626.

www.elsevier.com/locate/biopsych

K.Y. Vinod et al Koller G, Preuss UW, Bottlender M, Wenzel K, Soyka M (2002): Impulsivity and aggression as predictors of suicide attempts in alcoholics. Eur Arch Psychiatry Clin Neurosci 252:155–160. Kreitzer AC, Regehr WG (2001): Retrograde inhibition of presynaptic calcium influx by endogenous cannabinoids at excitatory synapses onto Purkinje cells. Neuron 29:717–727. Leweke FM, Giuffrida A, Wurster U, Emrich HM, Piomelli D (1999): Elevated aspects of the metabolism in schizophrenia. Neuroreport 10: 1665–1667. Lowry OH, Rosebrough NJ, Farr AG, Randall RJ (1951): Protein measurement with folin phenol reagent. J Biol Chem 193:265–275. Mann JJ (1998): The neurobiology of suicide. Nat Med 1:25–30. Mann JJ (2003): Neurobiology of suicidal behaviour. Nat Rev Neurosci 4:819 – 828. Mann JJ, Waternaux C, Haas GL, Malone KM (1999): Toward a clinical model of suicidal behavior in psychiatric patients. Am J Psychiatry 156:181–189. McDonald J, Schleifer L, Richards JB, de Wit H (2003): Effects of THC on behavioral measures of impulsivity in humans. Neuropsychopharmacology 28:1356 –1365. Murphy GE, Wetzel RD (1990): The lifetime risk of suicide in alcoholism. Arch Gen Psychiatry 47:383–392. Ponce G, Hoenicka J, Rubio G, Ampuero I, Jimenez-Arriero MA, RodriguezJimenez R, et al (2003): Association between cannabinoid receptor gene (CNR1) and childhood attention deficit/hyperactivity disorder in Spanish male alcoholic patients. Mol Psychiatry 8:466 – 467. Potash JB, Kane HS, Chiu YF, Simpson SG, McKinnon DF, McInnis MG, et al (2000): Attempted suicide and alcoholism in bipolar disorder: Clinical and familial relationships. Am J Psychiatry 157:2048 –2050. Preuss UW, Schuckit MA, Smith TL, Danko GR, Dasher AC, Hesselbrock MN, et al (2002): A comparison of alcohol-induced and independent depression in alcoholics with histories of suicide attempts. J Stud Alcohol 63:498 –502. Reiach JS, Li PP, Warsh JJ, Kish SJ, Young LT (1999): Reduced adenylyl cyclase immunolabeling and activity in postmortem temporal cortex of depressed suicide victims. J Affect Disord 56:141–151. Rich CL, Dhossche DM, Ghani S, Isacsson G (1998): Suicide methods and presence of intoxicating abusable substances: Some clinical and public health implications. Ann Clin Psychiatry 10:169 –175. Rossow I, Amundsen A (1995): Alcohol abuse and suicide: A 40-year prospective study of Norwegian conscripts. Addiction 90:685– 691. Roy A (2000): Relation of family history of suicide to suicide attempts in alcoholics. Am J Psychiatry 157:2050 –2051. Schmid PC, Krebsbach RJ, Perry SR, Dettmer TM, Maasson JL, Schmid HH (1995): Occurrence and postmortem generation of anandamide and other long-chain N-acylethanolamines in mammalian brain. FEBS Lett 375:117–120. Sugiura T, Konda S, Sukagawa A, Nakane S, Shinoda A, Itoh K, et al (1995): 2-Arachidonoylglycerol: A possible endogenous cannabinoid receptor ligand in brain. Biochem Biophys Res Commun 215:89 –97. Suominen K, Henriksson M, Suokas J, Isometsa E, Ostamo A, Lonnqvist J (1996): Mental disorders and comorbidity in attempted suicide. Acta Psychiatr Scand 94:234 –240. Westlake TM, Howlett AC, Bonner TI, Matsuda LA, Herkenham M (1994): Cannabinoid receptor binding and messenger RNA expression in human brain: An in vitro receptor autoradiography and in situ hybridization histochemistry study of normal aged and Alzheimer’s brains. Neuroscience 63:637– 652. Wilson RI, Nicoll RA (2002): Endocannabinoid signaling in the brain. Science 296:678 – 682.