Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Neurosci. Author manuscript; available in PMC 2012 July 4.
Published in final edited form as:
PMCID: PMC3292407

Dihydromyricetin As A Novel Anti-Alcohol Intoxication Medication


Alcohol use disorders (AUD) constitute the most common form of substance abuse. The development of AUD involves repeated alcohol use leading to tolerance, alcohol withdrawal syndrome (AWS), physical and psychological dependence, with loss of ability to control excessive drinking. Currently there is no effective therapeutic agent for AUD without major side-effects. Dihydromyricetin (DHM, 1 mg/kg, i.p. injection), a flavonoid component of herbal medicines, counteracted acute alcohol (EtOH) intoxication, and also withdrawal signs in rats including tolerance, increased anxiety and seizure susceptibility; DHM greatly reduced EtOH consumption in an intermittent voluntary EtOH intake paradigm in rats. GABAA receptors (GABAARs) are major targets of acute and chronic EtOH actions on the brain. At the cellular levels, DHM (1 μM) antagonized both acute EtOH-induced potentiation of GABAARs and EtOH exposure/withdrawal-induced GABAAR plasticity, including alterations in responsiveness of extra- and post-synaptic GABAARs to acute EtOH, and most importantly, increases in GABAAR α4 subunit expression in hippocampus and cultured neurons. DHM anti-alcohol effects on both behavior and CNS neurons were antagonized by flumazenil (10 mg/kg in vivo, 10 μM in vitro), the benzodiazepine (BZ) antagonist. DHM competitively inhibited BZ-site [3H]flunitrazepam binding (IC50, 4.36 μM), suggesting DHM interaction with EtOH involves the BZ-sites on GABAARs. In summary, we determined DHM anti-alcoholic effects on animal models, and determined a major molecular target and cellular mechanism of DHM for counteracting alcohol intoxication and dependence. We demonstrated pharmacological properties of DHM consistent with those expected to underlie successful medical treatment of AUD; therefore DHM is a therapeutic candidate.


Alcohol use disorders (AUD) represent a substantial public health problem worldwide. Over 76 million people present with AUD; 2.5 million deaths were attributed to alcohol (WHO, 2010). The development of AUD involves repeated alcohol use leading to tolerance, withdrawal syndrome (AWS) including hyperexcitability, distress, anxiety, insomnia, agitation, occasional seizures, a n d dependence (Trevisan et al., 1998; Heilig et al., 2010). Only an estimated 13% of people with identified AUD have ever received specialty treatment due to the lack of effective medications that ameliorate AWS and cure alcohol dependence. There is an urgent need for the development of new, more effective medications (

Although alcohol (EtOH) ingestion impacts most organ systems, its effects on the brain are of vital importance to AUD, given EtOH’s many neuropharmacological actions, including its intoxicating, sedative, anxiolytic, reinforcing, and addictive properties (Hobbs et al., 1996; Paul, 2006). EtOH alters the balance between γ-aminobutyric acid (GABA), the primary inhibitory neurotransmitter, and glutamate, the major excitatory neurotransmitter in CNS underlying the development of AUD (Mukherjee et al., 2008). GABAA receptors (GABAARs)-mediated inhibition is enhanced by acute EtOH (Aguayo et al., 2002; Koob, 2004; Lovinger & Homanics, 2007), with effects reported for both presynaptic increased GABA release (Siggins et al., 2005; Weiner and Valenzuela, 2006) and direct enhancement of GABAARs in the postsynaptic cell (Morrow et al., 1990; Boehm et al., 2004; Wallner and Olsen, 2008). GABAAR formed from a family of subunits exhibit different sensitivities to EtOH and are located either in post- or extra-synaptic membranes; this is a dynamic state with regulated trafficking between pools (Olsen and Homanics, 2000; Wallner et al., 2003; Lovinger and Homanics, 2007). Extrasynaptic GABAARs are sensitive to blood EtOH concentrations reached during social drinking (Devaud et al., 1997; Wei et al., 2004; Hanchar et al., 2005; Liang et al., 2006). Chronic EtOH administration to rats produces profound GABAAR plasticity which contributes critically to EtOH withdrawal and dependence (Liang et al., 2004, 2006, 2009; Morrow et al., 2008; Olsen and Spigelman, 2011). After acute EtOH exposure, some GABAARs are down-regulated rapidly, which, with compensatory GABAAR plasticity produces altered physiology and pharmacology consistent with signs of tolerance and withdrawal exhibited in vivo (Liang et al., 2007; Shen et al., 2011). EtOH-induced behavioral changes and GABAAR plasticity can be utilized as a screen for drugs that ameliorate AUD.

Hovenia dulcis is listed among the premier anti-hangover herbal medicines in China’s first pharmacopoeia, Tang Materia Medica. Hovenia extracts ameliorate alcohol-induced liver injuries (Du et al., 2010), and relieve hangover, partly by promoting EtOH elimination via enhancement of alcohol dehydrogenase (ADH) and acetaldehyde dehydrogenase (ALDH) activity (Kim et al., 2000; Chen et al., 2006). However, its true efficacy, active constituents, and mechanisms of action have not been critically examined.

Here we demonstrate that dihydromyricetin (DHM), a flavonoid component from Hovenia, is highly effective in counteracting acute EtOH intoxication, EtOH exposure/withdrawal-induced GABAAR plasticity and AWS symptoms, as well as reducing excessive EtOH consumption. We provide strong evidence that DHM is a candidate for development as a novel AUD therapy.

Materials and Methods


All animal experiments followed The Institutional Animal Care and Use Committee approved protocols. Male adult Sprague-Dawley (SD) rats (200-250 g, from Harlan Laboratory, USA) were used in the study. Pregnant SD rats (250-300 g, from Charles River Laboratories International, USA) were used for primary neuron cultures. The total animals we used in this study were 315 male rats and 5 pregnant rats. For EtOH administration, rats received i.p. injection of EtOH (3 g/kg, as a 15% w/v solution in 0.9% saline, and 4 g/kg, as a 20% w/v solution in 0.9% saline, both injection volumes were 2 ml/100 g body weight; Pharmco Products, Brookfield, CT), or received a single dose EtOH (5 g/kg, gavage). All of the i.p. injections were performed with a single injection at a final volume of 2 ml/100 g body weight in each rat (with a final dose amount of each drug is reported in Results section).


D H M (Dihydromyricetin, (2R,3R)-3,5,7-trihydroxy-2-(3,4,5-trihydroxyphenyl)-2,3-dihydrochromen-4-one) was purchased from ZR Chemical, Shanghai, China (CAS No. 27200-12-0). It was purified by HPLC (98%) from natural products. Other chemicals including flumazenil, picrotoxin and bicuculline were purchased from Sigma (Sigma-Aldrich Co., St. Louis, MO). Stock solutions of DHM and flumazenil were made with dimethyl sulfoxide (DMSO, except for the two-bottle-choice experiments described below, the solution was made in drinking water), and diluted to final concentration in saline for injection or oral administration. Injection volume was 2 ml/100 g body weight. The final DMSO concentration did not exceed 0.01%.

Clinically, Hovenia dosage range used for hangover is 100-650 mg/kg (Compendium of Materia Medica. Chinese). Total flavonoids purified from Hovenia are 4.53% of the extract, of which DHM accounts for approximately 40%, suggesting a 1-15 mg/kg dose for behavioral assays.

Loss of Righting Reflex (LORR) assay

After drug injection (i.p.), a timer was started and rats were placed in the supine position in a V-shaped support. LORR onset time was taken from the endpoint of injection to the start of LORR. LORR ended when the animal was able to flip over three times in 30 s.

Elevated-plus maze (EPM) assay

Anxiety associated with EtOH withdrawal was measured on an elevated plus-maze (Liang et al., 2004). The maze was elevated 1 m above the floor, and contained four 51 cm-long, 11.5 cm-wide arms arranged at right angles. The closed arms had opaque walls 30 cm high, extending the length of the arm. At the time of the test, each animal was placed in the center of the maze facing an open arm and allowed to explore for a 5-min session. During this 5 min test session, the animal’s behavior (e.g., number of arm entries and time spent in each arm per entry) was recorded on a camcorder.

Pentylenetetrazol (PTZ)-induced seizure test

PTZ dose used in this study (42 mg/kg in 0.9% saline, injection volume was 1 ml/100 g body weight) was determined as the dose which induced seizures in 75% of naïve rats. Briefly, after i.p. injection of PTZ, the time to onset and the duration of tonic-clonic seizures was determined (Liang et al., 2004).

The researchers who did the animal behavior experiments were blind to treatment groups. Animals were used once only for any determination.

Voluntary EtOH consumption measurement (two-bottle choice intermittent access to EtOH paradigm)

All fluids were presented in 100 ml graduated glass cylinders with stainless-steel drinking spouts inserted 15 min after the lights went off in the reversed light/dark cycle room. Bottles were weighed 30 min and 24 hrs after the fluids were presented. Every rat was weighed daily to monitor health and calculate the grams of EtOH intake per kilogram of body weight.

At the beginning of two-bottle choice paradigm, naïve rats were trained to have free two-bottle choice intermittent access to 20% (w/v) EtOH/water (E/water), EtOH plus DHM (0.05 mg/ml, dissolved in drinking water, E+D)/water, DHM/water (D/water) or water/water for two weeks with three 24-hr-sessions per week (Mondays, Wednesdays and Fridays) (Li et al., 2010). All drinking liquids contained 2 pk sweetener/L (1 pk EQUAL 0 CALORIE SWEETENER contains 1 g dextrose with maltodextrin and aspartame) at the first week, and then decreased to 1 pk sweetener/L at the second week. After two-week training (started at the 3rd week), rats were given two-bottle choice access to E/water, E+D/water, or D/water without any sweetener in all drinking liquids for two-bottle choice. Rats had unlimited access to two bottles of water between the EtOH-access periods. The placement of the EtOH bottle was alternated each EtOH drinking session to control for side preferences. Rats were maintained on 20% EtOH intermittent access two-bottle choice paradigm 6 weeks (21 EtOH-access sessions). Half of E/water-group had DHM added to the EtOH bottle beginning at the 7th week. The rest of the E/water-group continued EtOH-access sessions. The experiment period was 10 weeks total. EtOH consumption was expressed as grams of EtOH consumed per kilogram of body weight. Animals in the control-group had ad libitum access to two bottles of water.

Plasma EtOH concentrations (plasma [EtOH]) assay of LORR

Blood samples from the tail vein of rats at different time points (0, 5, 30, 60, 90, 180 min) after EtOH or E+D i.p. injections were collected for plasma [EtOH] assays. Briefly, the rat was put into a restraint tube and its tail was warm in ~40 °C. The tail vein at the tip of the rail was punched with a sharp blade. Approximate 0.2 ml venous blood was dropped to a capillary blood collection tube containing lithium heparin (Ram Scientific Inc. Yonkers, NY). Blood samples were centrifuged at 2500 rpm for 20 min. The supernatant was collected and stored at −80°C until assay. The EtOH content of each blood sample will be measured in duplicate along with EtOH standards using the alcohol oxidase reaction procedures (GM7 Micro-Stat; Analox Instruments, Lunenberg, MA) (Liang et al., 2007).

For plasma [EtOH] assay of two-bottle choice, blood samples were collected from the rats at 30, 45, 60 and 100 min after the voluntary EtOH two-bottle choice procedure (E/water- and E+D/water-group). The sample preparations were the same as described above.

Primary hippocampal neuron culture

Hippocampal neurons from embryonic day 18 rats were prepared by papain dissociation (Worthington Biochemical, Lakewood, NJ) and cultured in Neurobasal medium (Invitrogen) and B27 supplement as previously reported (Shen et al., 2011). Briefly, embryos were removed from maternal rats anesthetized with isoflurane and euthanized by decapitation. Hippocampi were dissected and placed in Ca2+- and Mg2+-free HEPES-buffered Hank’s buffered salt solution (pH 7.45). Tissues were dissociated by papain digestion followed by trituration through a Pasteur pipette and papain inhibitor treatment. Cells were pelleted and resuspended in Neurobasal medium containing 2% B27 serum-free supplement, 100 U/ml penicillin, 100 μg/ml streptomycin, 0.5 mM glutamine (all from Invitrogen), and 10 μM glutamate (Sigma).

Dissociated cells were then plated at a density of 0.3 × 105 cells/cm2 onto 12 mm round coverslips in 24-well plates (for patch-clamp recording) and/or at a density of 0.5 × 105 cells/cm2 in 6-well plates (for Western blot and biotinylation assays) coated with poly-D-lysine (Sigma, 50 μg/ml). Cultures were kept at 37°C in a 5% CO2 humidified incubator. Thereafter, one third to half of the medium was replaced twice a week with Neurobasal culture medium containing 2% B27 supplement, and 0.5 mM glutamine.

EtOH or vehicle exposure

At 24 hrs before whole-cell patch-clamp recording and/or biochemical experiments, half of the medium of cultured neurons (DIV13-14, DIV: days in vitro) was replaced with Neurobasal culture medium containing 120 mM EtOH (final EtOH concentration was 60 mM) for 30 min and then the entire medium replaced with half fresh Neurobasal culture medium plus half original medium (kept in the dishes, at 37°C). Same procedure was used for EtOH+DHM (final 0.1 μM) and DHM alone treatments. Control neurons were treated with corresponding vehicle. Neurons were maintained in the incubator until use.

Whole-cell patch-clamp recordings from slices and cultured neurons

Transverse slices (400 μm) of dorsal hippocampus were obtained from adult rats using a Vibratome (VT 100, Technical Products International, St. Louis, MO). Slices were continuously perfused with artificial cerebrospinal fluid (ACSF) composed of (in mM) 125 NaCl, 2.5 KCl, 2 CaCl2, 2 MgCl2, 26 NaHCO3, and 10 D-glucose. The ACSF was continuously bubbled with 95% O2-5% CO2 to ensure adequate oxygenation of slices and a pH of 7.4 and kept at 34 ± 0.5°C for perfusion. Patch electrodes were pulled from thin-wall borosilicate glass pipettes with resistances of 5.5-7.5 MΩ and were filled with internal solution containing (in mM): 137 CsCl, 2 MgCl2, 1 CaCl2, 11 EGTA, 10 HEPES and 3 ATP, pH adjusted to 7.30 with CsOH. Recordings targeted dentate gyrus granule cells (DGCs) of hippocampal slices (Liang et al., 2007).

Cultured neurons were transferred to a chamber perfused with an extracellular solution containing (in mM): 137 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 20 glucose and 10 HEPES (pH 7.40, room temperature), and visualized with an inverted microscope (TE200, Nikon). Recordings targeted pyramidal cells (Shen et al., 2011).

Neurons were whole-cell voltage-clamped at a holding potential of −70 mV with an Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA). Access resistances were <25 MΩ before electrical compensation by about 70%. Intracellular signals were low-pass filtered at 3 kHz and digitized at a sampling frequency of 20 kHz. Pharmacologically-isolated GABAAR-mediated miniature inhibitory postsynaptic currents (mIPSCs) were recorded by adding tetrodotoxin (TTX, 0.5 μM), D(-)-2-amino-5-phosphonopentanoate (APV, 40 μM), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX,10 μM), and CGP54626 (1 μM, GABABR antagonist) to the ACSF (or extracellular solution for cultured neurons) from stock solutions. Stock solutions of CGP54626 were made with DMSO. Final concentrations of DMSO did not exceed 0.01% in the recording chamber. For the cultured neurons, control and drug-containing solutions were delivered through a removable tip that was positioned close to the soma of target neurons with a Valvelink 8.02 fast-exchange perfusion system (AutoMate Scientific, USA). The flow rate of the perfusion system was around 0.3-0.5 ml/min, and the total volume of the recording chamber was 3 ml. For GABA concentration–response curves, evoked GABAAR-currents were recorded during acute applications of GABA or DHM onto neurons through a removable pipette tip using a Valvelink 8.02 (Liang et al., 2007; Shen et al., 2011).

Detection and analysis of mIPSCs and GABAAR-mediated tonic currents (Itonic)

Briefly, the recordings were low-pass filtered off-line (Clampfit software 10.3) at 2 kHz. The mIPSCs were detected (Mini Analysis Program, version 6.0.7) with threshold criteria of 8 pA amplitude and 20 pA*ms area. The frequency of mIPSCs was determined from all automatically detected events in a given 100 s recording period. For kinetic analysis, only single event mIPSCs with a stable baseline, sharp rising phase (10 to 90% rise time), and exponential decay were chosen during visual inspection of the recording trace. Double and multiple peak mIPSCs were excluded. At least 100 individual mIPSC events were recorded under each experimental condition. The mIPSC kinetics were obtained from analysis of the averaged chosen single events aligned with half rise time in each cell. Decay time constants were obtained by fitting a double exponential to the falling phase of the averaged mIPSCs. The Itonic magnitudes were obtained from the averaged baseline current of a given recording period. Two methods were used to measure Itonic: shifts in the baseline current, and changes in baseline noise. The baseline noise arises from the recording apparatus (machine noise) and opening and closing of chloride channels and receptors at any holding potential. When bicuculline (10 μM) or picrotoxin (100 μM), both selective GABAAR inhibitors, is applied, both the baseline current and baseline noise are diminished. The baseline noise was quantified by measuring the root mean square (RMS) noise using the RMS analysis routines packaged with the Mini Analysis program (Synaptosoft, Decatur, GA). RMS noise was measured only in the absence of phasic synaptic currents after bicuculline or picrotoxin blocked phasic synaptic currents (Liang et al., 2004; Mangan et al., 2005; Shen et al., 2011). The investigator performing the recordings and mIPSC analysis was blind to the treatments the rats received.

Western blots

Hippocampal tissues from rats were lysed in RIPA-buffer containing 1% Triton X-100, 0.1% sodium dodecyl sulfate (SDS), and 50 (in mM) Na3PO4, 150 NaCl, 2 EDTA, 50 NaF, 10 sodium pyrophosphate, 1.0 sodium orthovanadate, 1.0 phenylmethylsulfonyl fluoride (PMSF) and Complete protease inhibitor cocktail (Roche). The lysate was centrifuged for 15 min (14,000 × g, 4°C) and the supernatant collected for Western blot analysis. Western blots were performed using rabbit anti-GABAAR α4 (aa 379-421, gift of W. Sieghart) and mouse anti-β-actin (Sigma) followed by HRP-conjugated secondary antibodies. Bands were analyzed by densitometric measurements using ImageQuant 5.2 (Molecular Dynamics).

Biotinylation assay for cell-surface receptors

The biotinylation procedure was performed as previously described (Shen et al., 2011). Briefly, neurons were placed on ice, washed and then incubated for 30 min with PBS containing 1 mg/ml sulfo-NHS-LC-biotin (ProteoChem). After quenching the biotin-reaction with Tris-buffered saline (TBS) neurons were lysed in modified RIPA-buffer. After centrifugation 10% of the supernatant was removed to measure β-actin. The remaining supernatant was incubated with 60 μl of Neutravidin agarose (Pierce Chemical Company). Agarose-bound proteins were taken up in SDS sample buffer and boiled. Western blots were performed as mentioned above.

Tissue preparation and [3H]flunitrazepam binding assays

Naïve rat cortex was homogenized in 0.32 M sucrose and 10 mM HEPES buffer (pH 7.4), and centrifuged at 650 × g, 4°C. The subsequent supernatant was centrifuged at 150,000 × g to collect the desired membrane-containing pellet. The pellet was washed and centrifuged two more times, first using ice-cold water and second using membrane buffer containing (in mM) 50 KH2PO4, 1 EDTA, 2 benzamidine HCl, 0.5 DTT, 0.1 benzethonium HCl, 0.01% bacitracin, 0.2 PMSF (pH 7.4), and the resulting pellet was frozen. On the day of binding assay, the pellet was homogenized in assay buffer containing (in mM) 50 KH2PO4, 1 EDTA, 200 KCl (pH 7.4) and centrifuged, and resuspended in fresh assay buffer to a final protein concentration of 1 mg/ml. [3H]flunitrazepam (85.2 Ci/mmol, PerkinElmer, Boston, MA), brain homogenate, and DHM were placed into glass culture tubes for a final assay volume of 0.5 ml. The solutions were vortexed and allowed to incubate on ice for 30 min, and filtered onto Whatman GF/B filter paper using a Brandel cell harvester M-24R (Brandel Co, Gaithersburg MD). Filters were placed into minivials (Fisher Scientific) and 2.5 ml scintillation fluid (MP Biomedicals) and allowed to shake for 1 hr. Samples were counted in a Beckman LS-3801 liquid scintillation counter. Specific binding was defined as the total amount bound (zero unlabeled ligand) minus the binding in the presence of 10 μM final concentration flurazepam (Sigma). Data were analyzed with GraphPad Prism 4.0 Software (San Diego, CA) to determine IC50 (One-site competition equation) and Hill slope (Sigmoidal Dose-Response equation). Experiments were conducted in triplicate (Li et al., 2010).

Statistical analysis

SigmaPlot, SigmaStat (Systat Software Inc) and GraphPad Prism 4.0 (Graph-Pad Software Inc) were used for data display and statistical analysis. Data were expressed as mean ± SEM. t-test, one-way or one-way repeated measurement (RM) ANOVA followed by multiple comparison analyses based on Dunnett’s Method, two-way or two-way RM followed by post hoc multiple comparison analyses based on Holm-Sidak were used to determine significant levels between treatments and controls.


DHM counteracts EtOH intoxication, and this effect is antagonized by flumazenil

We examined DHM effects on EtOH intoxication (LORR) in rats. Studies on the neuroprotective effects have shown that flavonoids, including DHM purified from Hovenia (Fig. 1 A) are able to pass the blood-brain barrier (Youdim et al., 2003).

Figure 1
DHM counteracts EtOH intoxication and the effect of DHM is antagonized by flumazenil. A, Chemical structure of DHM. B, Upper panel: Plasma [EtOH] assay associated with EtOH-induced Loss of Righting Reflex (LORR). X axis shows time points after i.p. injection ...

Rats were divided into 7 groups and intraperitoneally (i.p.) injected with saline, EtOH (3 g/kg, E3), EtOH combined with DHM (0.3, 0.5, 1, and 10 mg/kg, E3+D0.3, E3+D0.5, E3+D1 and E3+D10), or DHM (D1) alone (single injection at a volume of 2 ml/100 g body weight). The onset of LORR induced by E3 was 1.8 ± 0.2 min, while DHM dose-dependently increased LORR onset (n = 4-12 rats/group, one-way ANOVA, F(3,25) = 10.64, p < 0.001, * represents statistical significance vs. E3-group in Fig. 1 B). E3 induced 73.6 ± 8.0 min LORR; DHM dose-dependently reduced LORR (n = 4-12 rats/group, one-way ANOVA, F(3,22) = 26.58, p < 0.001, * represents statistical significance vs. E3-group in Fig. 1 B). 1 mg/kg DHM, as saline, did not induce LORR. These results suggest that DHM counteracts acute EtOH intoxication.

During the LORR assay, we took venous blood samples at the various points from 0-180 min to measure plasma [EtOH] from EtOH- and E+D-groups. E3 induced onset of LORR within 5 min. Plasma [EtOH] rapidly increased for 5 min followed by a slower increase to around 60 min, then [EtOH] declined gradually. The time course and magnitude of plasma [EtOH] was not affected in E3+D0.3 and E3+D0.5 groups; In E3+D1 and E3+D10 (DHM 1, 10 mg/kg) groups, the rise time of plasma [EtOH] was slowed at early time (Fig. 1 B). From 30 to 60 min, when the rats in E3-group were sleeping while E3+D1- and E3+D10-groups were awake, the differences in plasma [EtOH] were not statistically significant between E3- and E3+D1-group (30 min: E3 vs. E3+D1 = 334.9 ± 37.8 vs. 287.7 ± 21.5 mg/dl, and 60 min: E3 vs. E3+D1 = 353.7 ± 35.4 vs. 296.3 ± 17.8 mg/dl); Higher dose of DHM (D10) slightly decreased [EtOH] (n = 3-4 rats/group, two-way RM ANOVA, F(4,92) = 4.28, p = 0.025, with DHM concentrations by sampling time interaction F(20,92) = 2.70, p = 0.002; Post hoc analysis revealed a statistical significance at 30 min but not at 60 min, * represents significance vs. E3-group in Fig. 1 B).

We also tested the effects of DHM on higher dose EtOH (4 g/kg) as shown in Table 1. 4 g/kg of EtOH induced 112.8 ± 8.0 min LORR. DHM at dose of 1 mg/kg greatly reduced LORR to 28.2 ± 4.0 min. 1 mg/kg DHM did not have significant effect on plasma [EtOH] (n = 4 rats/group, two-way RM ANOVA, p > 0.05; Table 1). These results suggest that DHM affects EtOH pharmacokinetics, but this effect is not sufficient to account for the DHM antagonism of EtOH-induced LORR. DHM counteracts EtOH-induced LORR even at low concentrations that do not affect the kinetics (See Discussion).

Table 1
The effects of 1 mg/kg DHM on 4 g/kg EtOH-induced blood [EtOH] kinetics and LORR duration. High doses of DHM alone induced short-time LORR, while high dose flumazenil alone did not induced LORR. Data (mean ± SEM) are obtained from naïve ...

To examine the target of DHM’s anti-EtOH effects, we tested flumazenil, the selective BZ antagonist of GABAARs (Hunkeler et al., 1981). EtOH induced 69.0 ± 11.3 min LORR; co-injection of DHM (3 mg/kg) and EtOH reduced LORR to 2.7 ± 1.7 min (Fig. 1 C). Flumazenil (10 mg/kg) reversed the DHM reduction in EtOH-induced LORR (56.1 ± 4.6 min). Increasing DHM dose to 10 mg/kg decreased the flumazenil effect (29.3 ± 4.8 min), while increasing the flumazenil dose to 30 mg/kg increased its antagonism of the DHM effect (58.2 ± 3.9 min). Flumazenil co-injected with EtOH did not alter LORR compared with the E3-group (n = 5-13 rats, one-way ANOVA, F(5,34) = 16.34, p < 0.001, * represents statistical significance vs. E3-group; Fig. 1 C). These results suggest that GABAARs play a major role in the behavioral effects of EtOH-induced LORR in vivo. Flumazenil competitively antagonizes DHM effects on EtOH-induced LORR. In addition, the results suggest that the interactions of DHM and EtOH involve DHM action at GABAAR-BZ sites that may underlie DHM therapeutic effects on EtOH intoxication.

We further tested whether DHM application prior to, or after EtOH exposure would affect EtOH-induced LORR. 30 min prior to EtOH injection (D1+E), DHM reduced LORR to 8.2 ± 4.1 min (n = 5 rats/group, t-test, t = 15.7, df = 4, p < 0.001; Fig. 1 D). 30 min after injection of EtOH that induced LORR, LORR went on for an additional 42 ± 9.1 min in rats injected with saline; while injection of DHM reduced the remaining LORR to 19 ± 1.0 min (n = 4 rats/group, t-test, df = 3, t = 14.2, p < 0.001; Fig. 1 D). These results suggest DHM effectively ameliorates moderate to high dose EtOH intoxication even when it is administered 30 min prior or 30 min post-EtOH exposure.

We tested high doses of DHM (doses hundreds-fold higher than that for its antagonistic effects on EtOH intoxication). DHM (100, 300 mg/kg, D100, D300) induced only 2.9 ± 1.3 and 9.8 ± 4.3 min LORR, respectively (n = 8-10 rats/group, Table 1), suggesting that DHM is not merely a typical benzodiazepine. High doses of flumazenil (20, 200 mg/kg, F20, F200) did not induce LORR (n = 5 rats/group, Table 1).

DHM ameliorates EtOH withdrawal symptoms in rats

We examined DHM effects on EtOH withdrawal symptoms in rats. Rats were divided into 4 groups and injected (i.p.) with vehicle (V/W), EtOH (3 g/kg, E/W), EtOH combined with DHM (1 mg/kg, E+D/W) or DHM (D/W) respectively; all of the injections in the 4 groups were performed with a single injection at a final volume of 2 ml/100 g body weight in each rat. W represents withdrawal. Then, rats were sub-divided into 3 groups to measure signs of EtOH withdrawal at 48 hrs.

Anxiety was assayed with the EPM (Fig. 2 A). V/W-group spent 57.4 ± 9.8 % of total time (time in open arm, closed arm and center, total 5 min) in open arms and 32.6 ± 8.6% in closed arms. E/W-group spent significantly shorter time in open arms and longer time in closed arms than those of V/W group (* represents statistical significance vs. V/W-group, n = 5-6 rats/group; Fig. 2 A); while E+D/W-group spent similar times. Our two-way ANOVA analysis revealed significant DHM effects by EtOH exposure/withdrawal interaction (F(1, 22) = 5.51, p = 0.03 for close arm stay and F(1, 22) = 5.20, p = 0.034 for open arm). DHM alone did not affect the time rats spent in either arm (Fig. 2 A). These data suggest that EtOH exposure/withdrawal produces anxiety, while DHM co-application with EtOH prevents EtOH-induced anxiety; DHM alone does not affect anxiety levels.

Figure 2
DHM ameliorates EtOH withdrawal symptoms in rats. 4 groups of rats were injected (i.p.) with single-dose vehicle (V/W), EtOH (3 g/kg, E/W), EtOH plus DHM (1 mg/kg, E+D/W), or DHM alone (D/W). W represents subsequent 48 hr withdrawal. A, Anxiety was measured ...

Tolerance to acute EtOH was measured with EtOH-induced LORR (Fig. 2 B). EtOH exposure/withdrawal (E/W-group) reduced LORR to 10.8 ± 3.8 min compared to 63.6 ± 7.0 min LORR in V/W-group (* represents statistical significance vs. V/W-group; n = 5 rats/group; Fig. 2 B). Co-application of DHM and EtOH (E+D/W-group) reversed this effect on LORR to 61.0 ± 3.8 min. Two-way ANOVA analysis revealed a significant DHM effect by EtOH exposure/withdrawal interaction on LORR (F(1, 19) = 15.64, p = 0.001). DHM alone (D/W-group) did not affect LORR (65.6 ± 8.4 min). These results suggest that a single EtOH exposure/removal produces tolerance to EtOH; while DHM prevents EtOH exposure-induced tolerance to EtOH.

Hyperexcitability was assayed with PTZ-induced seizure duration (Fig. 2 C). PTZ-induced 0.9 ± 0.2 min seizures in V/W-group and 6.5 ± 1.1 min seizures in E/W-group (n = 6-13 rats/group, * represents statistical significance vs. V/W-group in Fig. 2 C); while seizure duration was minimized in E+D/W-group (1.7 ± 0.8 min). Two-way ANOVA analysis revealed a significant DHM effect by EtOH exposure/withdrawal interaction on PTZ-induced seizure (F(1,32) = 4.37, p = 0.046). PTZ-induced seizure duration in DHM-group was similar to V/W-group (0.6 ± 0.4 min). These results suggest that EtOH exposure/withdrawal produces hyperexcitability and increases seizure susceptibility; while DHM ameliorates these effects of EtOH.

DHM reduced EtOH consumption in an intermittent voluntary alcohol intake paradigm in rats

We determined DHM effects on EtOH consumption, the most important sign of EtOH dependence. Rats were divided into 4 groups and offered intermittent access to two bottle choice of 20% EtOH/water (E/water), EtOH+DHM/water (DHM 0.05 mg/ml, E+D/water), DHM/water (D/water) or water/water, respectively. After two weeks, in E/water-group, EtOH consumption increased from 3.9 ± 1.4 at the end of the 2nd week to 7.4 ± 0.6 g/kg/day at the beginning of the 4th week. In E+D/water-group, DHM inhibited the increase in EtOH intake (2.6 ± 0.5 g/kg/day at the beginning of the 4th week). After 6 weeks, E/water-group was sub-divided into 2 groups: one continued with E/water, while the other one was offered E+D/water. The E/water sub-group kept up the high level of EtOH intake, while in E+D/water sub-group, EtOH intake was greatly reduced to 1.8 ± 1.0 g/kg/day at the end of the 7th week, and 1.2 ± 0.2 g/kg/day at the end of the 8th week, similar to that of the group started with E+D/water (n = 6-8 rats/group, two-way RM ANOVA, F(2,227) = 169.69, p < 0.001, with significant interaction between treatments and treatment days, F(36,227) = 4.11, *, p < 0.001, E+D/water-group vs. E/water-group; †, p < 0.001, E+D/water from 7th week vs. E/water group in weeks 8, 9 and 10; Fig. 3 A). E/water-group showed significantly increased preference to bottle containing EtOH (n = 6-8 rats/group, two-way RM ANOVA, F(1,31) = 7.06, p = 0.019, with interaction F(1,31) = 16.0, p = 0.001; Fig. 3 B), while E+D/water did not change this preference and remained the same level of drinking from the bottle containing E+D. We also noted that, as a control for taste aversion of DHM, there was no significant difference between fluid intake from the bottle of water and from the bottle of DHM (~50%) in the D/water groups. There were no significant differences in total fluid consumption between the 4 groups. There was no significant difference in body weight between the 4 group rats at the end of the experiments. These results suggest that DHM prevents excessive EtOH consumption and maintain effective for a long time (at least 10 weeks) if taken with EtOH. DHM reduces EtOH consumption when the high voluntary EtOH consumption is already established by exposure to EtOH.

Figure 3
DHM prevents the escalation of EtOH consumption in the voluntary intermittent two-bottle choice paradigm in rats. A, EtOH consumption gradually escalated in the group of intermittent access to two-bottle choice of 20% EtOH/water. Co-administration of ...

At the end of the 6th week, plasma [EtOH] from E+D/water-group was significantly lower than that from E/water-group (n = 2 rats/group, t-test, t = 28.3, df = 1, p = 0.023; Fig. 3 C). Plasma [EtOH] correlated well with the measured amount of EtOH consumed. Plasma [EtOH] (mg/dl) for each animal was measured following 30, 45, 60 and 100 min of voluntary 20% EtOH started at the EtOH day of the end of the 6th week. These data suggest that DHM prevents high voluntary EtOH consumption.

DHM antagonizes EtOH-induced GABAAR potentiation, and the effect is blocked by flumazenil

We performed patch-clamp recordings in DGCs from hippocampal slices in vitro. Bath application of EtOH (60 mM) increased Itonic from 22.0 ± 0.7 to 46.9 ± 1.4 pA (one-way RM ANOVA, F(3,31) = 49.92, p < 0.001; Fig. 4 A, B) and enhanced mIPSCs from 0.53 ± 0.02 to 0.64 ± 0.02 nC (n = 8 neurons/3 rats, one-way RM ANOVA, F(3,34) = 45.24, p < 0.001, * represents statistical significance vs. drug 0; Fig. 4 A, C), as previously reported (Liang et al., 2007). EtOH effects were concentration-dependently antagonized by DHM (0.3 and 1.0 μM; Fig. 4 A-C). We then tested flumazenil effects on the anti-EtOH actions of DHM. DHM (3 μM) decreased EtOH-potentiated Itonic from 44.8 ± 2.3 to 21.0 ± 0.9 pA and mIPSCs from 0.78 ± 0.01 to 0.70 ± 0.02 nC. Flumazenil (10 μM) reversed the DHM actions on Itonic (one-way RM ANOVA, F(3,18) = 115.81, p < 0.001) and mIPSCs (one-way RM ANOVA, F(3,19) = 47.48, p < 0.001, * represents statistical significance vs. drug 0, n = 5 neurons/2 rats; Fig. 4 D-F). These data suggest that DHM antagonizes EtOH-induced potentiation of both extrasynaptic and synaptic GABAARs, and the effects are blocked by flumazenil. These data are consistent with the behavioral experiment observations (Fig. 1 C) indicating that interaction of DHM and EtOH on GABAAR-BZ sites is a cellular mechanism underlying the therapeutic effects of DHM on EtOH intoxication.

Figure 4
DHM antagonizes EtOH-induced GABAAR potentiation and the effects are blocked by flumazenil. All recordings were whole-cell voltage-clamped at −70 mV. A, Recording from dentate gyrus cells (DGCs) in hippocampus slices. The gray dashed lines represent ...

We also tested quercetin, another flavonoid purified from Hovenia with similar chemical structure. Acute EtOH potentiated Itonic and mIPSCs, while subsequent application of quercetin (0.3 and 1 μM) did not change EtOH-induced potentiation (n = 5 neurons/2 rats, one-way RM ANOVA, F(3,18) = 43.20, p < 0.001; Fig. 4 H-J). Post hoc multiple comparison shows no significant change in Itonic and mIPSCs after initial potentiation by EtOH. These data indicate that the antagonism of DHM of EtOH effects on GABAARs is, to some extent, is unique.

DHM is a positive modulator of GABAARs at the BZ sites

We further studied DHM (0.1 to 30 μM) effects on GABAAR-mediated Itonic and mIPSCs of DGCs in hippocampal slices. DHM (1 μM) enhanced Itonic (one-way RM ANOVA, F(5,32) = 78.16, p < 0.001; Fig. 5 A, B) and increased mIPSC area (one-way RM ANOVA, F(5,34) = 46.80, p < 0.001, * represents statistical significance vs. drug 0, n = 8 neurons/3 rats; Fig. 5 A,C) in a concentration-dependent manner (0.1 to 30 μM). We also analyzed DHM effects on mIPSC frequency and kinetics as shown in Table 2. The DHM enhancement of mIPSCs was primarily due to decay time prolongation. These results suggest that DHM potentiates both extra- and post-synaptic GABAARs.

Figure 5
DHM is a positive modulator of GABAARs at BZ sites. All recordings were whole-cell voltage-clamped at −70 mV. A, Recording from DGCs (left panel) in hippocampal slices and superimposed averaged mIPSCs (right). B, C, Summary of Itonic and mIPSC ...
Table 2
The effects of DHM on mIPSC kinetics and Itonic magnitudes in DGCs of naïve rats. Data (mean ± SEM) are obtained from hippocampal slices. n = 5-7 neurons/3 rats *, p < 0.05 vs. baseline (one-way ANOVA).

To further examine the site of DHM actions on GABAARs, we tested flumazenil effects on DHM enhancement of GABAAR function in cultured hippocampal neurons at DIV13-14. DHM (1 μM) potentiated Itonic (194.9 ± 13.6% of baseline, Fig. 5 D, E) and mIPSC area (181.8 ± 9.2% of baseline, Fig. 5 D, F). Flumazenil inhibited the DHM enhancement on GABAAR-currents in a concentration-dependent manner (Itonic: one-way RM ANOVA, F(5,57) = 7.55, p < 0.001; and mIPSCs: one-way RM ANOVA, F(5,54) = 10.16, p < 0.001, * represents statistical significance vs. flumazenil 0, n = 8 neurons/3 rats; Fig. 5 D-F). These observations suggest that DHM acts on the same sites on GABAARs as BZs.

We then examined DHM (0.03-100 μM) actions on BZ sites using [3H]flunitrazepam binding in cortical membrane homogenates from naïve adult rats. Significant inhibition of [3H]flunitrazepam binding by DHM was observed, starting at 0.3 μM in a concentration-dependent manner, with an IC50 of 4.36 μM and Hill slope of −0.73 (Fig. 5 G). These data suggest that DHM directly inhibits [3H]flunitrazepam binding to GABAARs, apparently competitively, indicating that DHM likely acts on GABAAR-BZ sites.

We further examined the concentration dependence of DHM effects on GABAAR-mediated currents in cultured hippocampal neurons. DHM concentration-dependently potentiated Itonic (one-way RM ANOVA, F(6,74) = 26.34, p < 0.001; EC50 was ~0.20 μM) and increased mIPSCs (one-way RM ANOVA, F(5,66) = 28.29, p < 0.001, * represents statistical significance vs. drug 0, n = 9-10; EC50 was ~0.20 μM; the responses to higher than 1 μM DHM decreased slightly, Fig. 6 A, B).

Figure 6
DHM potentiates GABAAR-mediated inhibition in a concentration-dependent manner in primary cultured hippocampal neurons (DIV14). Neurons were whole-cell voltage-clamped at −70 mV. A, B, Dose-response curves of DHM on Itonic and mIPSCs (n = 9-10 ...

We also tested DHM effects on GABAAR-currents induced by focal puffs of 10 and 300 μM GABA in the presence of TTX in cultured neurons at DIV14. Co-application of DHM (0.3 and 1 μM) with GABA increased peak GABA-currents and produced a left shift of the GABA concentration-response curve (n = 6-8, Fig. 6 C, D). These results suggest that DHM acts on GABAARs directly and potently potentiates extra- and/or post-synaptic GABAARs.

DHM prevents EtOH exposure/withdrawal-induced GABAAR plasticity in rat hippocampus

We examined the total protein content of GABAAR α4 subunit in rat hippocampus 48 hr after withdrawal from vehicle (V/W), EtOH (3 g/kg, E/W), EtOH combined with DHM (1 mg/kg, E+D/W) or DHM (D/W) treatments (i.p. injection, the same as mentioned in Fig. 2), respectively; W represents withdrawal. Western blots showed that EtOH exposure increased the total α4 protein level to 184.0 ± 26.0% compared to V/W-group; this increase was blocked in E+D/W-group (93.0 ± 21.0% of control; * represents significant difference vs. V/W-group). Two-way ANOVA analysis revealed a significant DHM effect by EtOH exposure/withdrawal interaction on the total α4 protein level (F(1,15) = 8.93, p = 0.011; n = 3/group; Fig. 7 A, B). DHM exposure had no effect on α4 subunit level. These data indicate that DHM prevents EtOH exposure/withdrawal-induced GABAAR plasticity in vivo.

Figure 7
DHM prevents EtOH exposure/withdrawal-induced alteration in GABAAR α4 subunit expression in rat hippocampus. 4 groups of rats were injected (i.p.) with single-dose vehicle (V/W), EtOH (3 g/kg, E/W), EtOH plus DHM (1 mg/kg, E+D/W), or DHM alone ...

We previously demonstrated that EtOH exposure/withdrawal produces profound GABAAR plasticity which manifests as a switch in the EtOH sensitivity from extrasynaptic to synaptic GABAAR-mediated currents (Liang et al., 2007; Shen et al., 2011), which could be a mechanism underlying EtOH withdrawal signs. Here, we tested whether DHM prevents EtOH-induced GABAAR plasticity in CNS neurons.

Four groups of rats were gavaged with vehicle (V/W), EtOH (5 g/kg, E/W), EtOH combined with DHM (1 mg/kg, E+D/W), or DHM (D/W). W represents 48 hr withdrawal. Then whole-cell GABAAR-mediated currents were recorded on DGCs in hippocampal slices of rats after 48 hr withdrawal from the 4 treatments, respectively. In V/W-group, bath application of EtOH (60 mM) enhanced Itonic from 28.8 ± 3.1 to 62.1 ± 3.3 pA (n = 4-7 neurons/group, two-way RM ANOVA, F(3,29) = 48.45, p < 0.001; with DHM effects by EtOH exposure/withdrawal interaction F(3,29) = 39.21, p < 0.001; Fig. 8 A, E). It enhanced mIPSC area from 0.67 ± 0.06 to 0.78 ± 0.05 nC (n = 4-7 neurons/group, two-way RM ANOVA, F(3,29) = 6.98, p = 0.007; with DHM effects by EtOH exposure/withdrawal interaction F(3,29) = 12.15, p < 0.001; * represents significant difference between 60 mM vs. 0 EtOH, † represents significant difference vs. V/W group in Fig. 8 A, F). In E/W-group, EtOH did not increase Itonic (Fig. 8 B, E), while greatly enhancing mIPSC area from 0.70 ± 0.03 to 1.4 ± 0.04 nC (Fig. 8 B, F). In E+D/W- group, EtOH increased Itonic from 30.0 ± 2.8 to 60.0 ± 2.2 pA (Fig. 8 C, E), while mIPSC modulation was unchanged (Fig. 8 C, F). In D/W-group, the responses of Itonic and mIPSCs to EtOH were similar to those of V/W-group (Fig. 8 D-E). The baseline mIPSC frequency and kinetics as well as Itonic magnitudes from DGCs of the 4 group rats were shown in Table 3. These results suggest that intragastric EtOH with DHM prevents both the subsequent tolerance to EtOH, and EtOH-induced GABAAR plasticity. Interestingly, DHM acts equally well orally (gavage) or by i.p. injection.

Figure 8
DHM prevents EtOH exposure/withdrawal-induced GABAAR plasticity. Rats were divided into 4 groups and gavaged with vehicle (V/W), EtOH (5 g/kg, E/W), EtOH combined with DHM (1 mg/kg, E+D/W) or DHM (D/W). W represents subsequent 48 hr withdrawal. Then whole-cell ...
Table 3
Baseline mIPSC kinetics and Itonic in DGCs from rats after vehicle, EtOH, EtOH+DHM, or DHM exposure/withdrawal. Data (mean ± SEM) were obtained from hippocampus slice from rats 48 hr after vehicle (V/W), EtOH (E/W), E+D (E+D/W) and DHM (D/W) administration ...

We also tested DHM effects on cultured neurons pre-exposed to EtOH. 24 hrs after withdrawal from EtOH exposure (w represents withdrawal), bath application of DHM enhanced Itonic and mIPSCs concentration-dependently (0.03-30 μM; Itonic: one-way RM ANOVA, F(6,35) = 79.90, p < 0.001; and mIPSCs: one-way RM ANOVA, F(5,38) = 11.75, p < 0.001, * represents significant difference vs. drug 0, n = 8-9 neurons/group; Fig. 9 A, B). The EC50 for enhancing Itonic (~0.20 μM) and mIPSCs (~0.15 μM) were similar to those in control neurons without pre-exposed to EtOH (Fig. 6 A, B). The data suggest that DHM remains effective in potentiating extrasynaptic and/or postsynaptic GABAARs even following EtOH-exposure that leads to tolerance to EtOH.

Figure 9
DHM potentiates GABAAR-mediated inhibition in EtOH pre-exposed cultured neurons; Co-administration of DHM with EtOH prevents EtOH exposure/removal-induced GABAAR plasticity in vitro. In culture hippocampal neurons (DIV13-14) 24 hrs after EtOH-exposure ...

We measured the surface expression of α4 subunit in cultured neurons using cell-surface biotinylation followed by Western blot analysis. EtOH pre-exposed neurons showed increased α4 subunit surface expression (* represents significant difference vs. V/W-group); while in neurons pre-exposed to EtOH + DHM, this increase was blocked (two-way ANOVA analysis revealed a significant DHM effect by EtOH exposure/withdrawal interaction on the total α4 subunit surface expression, F(1,15) = 29.71, p < 0.001; Fig. 9 C, D). DHM did not alter α4 surface expression. These data indicate that co-administration of EtOH with DHM prevents EtOH exposure/removal-induced GABAAR plasticity in vitro.


Dihydromyricetin (DHM) is a flavonoid compound first found in Hovenia dulcis, in the plant Ampelopsis grossedentata, and in teas (Li et al., 2003); few of its pharmacological properties have been examined. In this study, we found that: 1) DHM potently (1 mg/kg) counteracts acute EtOH intoxication; 2) the counteracting effects are antagonized by the general BZ antagonist, flumazenil in vivo; 3) DHM ameliorates EtOH exposure/withdrawal-induced behavioral changes, including a) tolerance to EtOH; b) increase in basal anxiety and c) hypersensitivity to PTZ-induced seizures (hyperexcitability). At the same doses, 4) DHM does not cause intoxication, sedation, anesthesia, nor hyperexcitability; 5) DHM prevents the escalation of alcohol consumption in an intermittent voluntary alcohol intake paradigm in rats. DHM reduces alcohol consumption in rats that drink high levels of alcohol following intermittent alcohol exposure; yet, at the cellular level in CNS neurons (in slices and cultured neurons); 6) DHM antagonizes acute EtOH-induced potentiation of GABAARs; 7) DHM antagonizes EtOH exposure/removal-induced alterations in responsiveness of GABAARs to acute EtOH including loss of Itonic modulation and increased mIPSC sensitivity; 8) DHM potentiates GABAARs in hippocampal slices and cultured neurons, and retains efficacy in potentiating GABAARs even after EtOH exposure/withdrawal which induces tolerance to EtOH; 9) DHM effects on GABAARs are inhibited by flumazenil in vitro; 10) DHM competitively inhibits BZ site [3H]flunitrazepam binding in rat cortical membrane homogenates; 11) DHM blocks EtOH exposure/withdrawal-induced increases in the amount of GABAAR α4 subunits in rat hippocampus; 12) DHM blocks EtOH pre-exposure-induced increase in α4 subunit surface expression in culture hippocampal neurons. In summary, we determined the effects of DHM on EtOH actions in animal models, a major molecular target of DHM for counteracting EtOH intoxication and dependence, as well as its underlying cellular mechanisms.

DHM counteracts acute EtOH intoxication and ameliorates AWS by acting on GABAARs

We and others (Valenzuela, 1997; Cagetti et al., 2003; Liang et al., 2006, 2007; Kumar et al., 2009; Heilig et al., 2010) have shown that acute and chronic EtOH administration produce GABAAR plasticity in rats that contribute to altered behaviors in AWS and alcohol dependence. We identified DHM by screening herbal medicines used for alcohol disorders for the ability to ameliorate EtOH-induced GABAAR plasticity, and found that DHM not only had that action but also was effective in blocking acute EtOH intoxication, and withdrawal, as well as accelerated EtOH voluntary consumption. Our findings are consistent with current theories linking acute and chronic effects of EtOH with GABAARs, and suggest strongly that DHM blocking EtOH’s effect involves GABAARs, especially in light of flumazenil’s block of all DHM effects in vivo and on GABAAR-mediated current in vitro.

Current medications for EtOH intoxication and AUD

BZs are classical medications for reducing symptoms of AWS, and they act on GABAARs. However, BZs (e.g., diazepam) exhibit cross-tolerance in behavioral depression and GABAAR enhancement after EtOH treatment (Devaud et al., 1996). Yet frequent use of BZs can lead to dependence. Therefore, although BZs can treat life-threatening AWS, developing dual addictions makes BZs unsuitable for alcohol abuse therapy.

Besides BZs, only three medications (oral naltrexone, acamprosate, and disulfiram) are currently approved for treating alcohol dependence by the US FDA. All these medications aim to treat relapse to heavy drinking, however, all have limited success. In all of these cases the receptors for alcohol actions are not directly affected, or even targeted, nor is the underlying mechanisms of interaction with alcohol clarified (except for disulfiram, which blocks acetaldehyde metabolism). Considering the heterogeneous nature of alcohol use disorders, many patients have limited or no response to these medications. Other candidate targets are under study but so far lack convincing evidence for success (Hester and Miller, 2003; O’Malley et al., 2003; Heilig and Egli, 2006; Johnson, 2010).

Kudzu is the only Eastern medication mentioned by NIAAA ( Possible treatment of alcohol craving, and reduced EtOH drinking by animals and human heavy drinkers was reported for Kudzu extract (Keung and Vallee, 1993; Lukas et al., 2005), due to some isoflavone ingredients, like daidzein, puerarin, genistein (Johnson, 2010), mimicked by a synthetic analog CVT-10216 (Arolfo et al., 2009). Kudzu’s mechanism of action was suggested to reduce blood EtOH levels involving mitochondrial acetaldehyde dehydrogenase ALDH2 and/or gastric emptying effects. Reduction of EtOH-induced LORR by daidzin was observed for EtOH administered gastrically but not intraperitoneally (Lin and Li, 1998), consistent with delayed stomach emptying.

DHM effects beyond EtOH pharmacokinetics

It has been reported that Hovenia relieves hangover by decreasing blood EtOH concentrations, promoting alcohol clearance and enhancing the alcohol metabolic enzymes, ADH and ALDH (Kim et al., 2000; Chen et al., 2006). In this study, we showed that DHM, like other flavonoids, has a relatively weak ability to reduce the rate of rise of blood alcohol concentration. However, this cannot account for the actions of DHM on blocking EtOH effects in vivo, since when we decreased the doses of DHM to 0.3 and 0.5 mg/kg, DHM apparently did not affect EtOH pharmacokinetics while still strongly counteracted EtOH-induced LORR (Fig. 1 B), suggesting potency of DHM. When we increased the dose of EtOH (4 g/kg), 1 mg/kg DHM did not have significant effect on the magnitude nor the time cause of plasma [EtOH] (Table 1) while greatly reduce the EtOH-induced LORR. In this study we demonstrated a likely more important mechanism. DHM has direct effects on GABAARs, the major targets of EtOH that contribute to AWS and EtOH dependence. More importantly, the DHM effects on EtOH intoxication and GABAAR plasticity can be reversed by flumazenil in vivo (Fig. 1) and in vitro (Figs. 4, ,5).5). These results indicate that DHM has complex interactions with GABAARs which counteract EtOH effects. DHM potentiates GABAARs in CNS neurons and retains efficacy in potentiating GABAARs even after EtOH exposure/withdrawal which induces tolerance to EtOH (Fig. 6, ,9);9); thus, it may be an effective pharmacotherapy for patients who are tolerant to other medications for alcoholism and AWS such as BZs.

DHM compared with other BZ ligands: preliminary observations suggest novel pharmacology

There is some evidence that selected BZ ligands may block some EtOH actions (Suzdak et al., 1986; Koob, 2004; Hanchar et al., 2005; Paul, 2006; Wallner et al., 2006; Lovinger and Homanics, 2007), but this has not resulted in any therapeutic agents. Notably the imidazobenzodiazepine Ro15-4513 was reported (Suzdak et al., 1986; Paul, 2006) to specifically block EtOH behavioral effects in vivo but not other GABAergic sedatives like pentobarbital; Ro15-4513, but not other inverse agonists, was able to inhibit EtOH enhancement of GABAAR function in vitro (Suzdak et al., 1986). More recent data show that Ro15-4513 and selected analogues that are also able to block alcohol behavioral effects can antagonize the EtOH enhancement of extrasynaptic δ subunit-containing subtypes of GABAARs (Wallner et al., 2006, 2008). However, although DHM also is able to block EtOH effects a n d i s a GABAAR ligand, DHM and Ro15-4513 are very different, both chemically and pharmacologically. Ro15-4513 has pro-convulsant and anxiogenic effects and inhibits most GABAAR subtypes (Paul, 2006). On the contrary, DHM potentiates GABAAR-mediated synaptic and especially extrasynaptic tonic currents (Fig. 5 A-C and Fig. 6) and antagonizes EtOH potentiation of GABAARs. DHM is a positive modulator of GABAARs acting on the benzodiazepine site potentiating both extra- and post-synaptic GABAARs (Fig. 5 A-C), but appears to differ also from “traditional” BZ like diazepam. It is not anxiolytic or sedative at the dose (1 mg/kg) that blocks EtOH effects. It induces short time LORR only when doses of hundreds-fold higher than its pharmacological dose are applied (Table 1). It will be interesting to study other possible pharmacological actions of DHM consistent with a positive modulation of GABAARs, as well as the possible subunit selectivity. Such information may be helpful in understanding how DHM interacts with EtOH in vivo. DHM prevents the escalation of EtOH consumption and remains effective for a long period of time in a chronic intermittent voluntary EtOH intake paradigm. The underlying mechanisms might be related to DHM preventing EtOH-induced GABAAR dysfunction / plasticity by multiple withdrawals from EtOH (Figs. 7, ,8).8). In other words, alcohol craving induced by the hyperexcitable state of AWS is ameliorated by DHM co-administered with EtOH which reduces initial responses and thus AWS to EtOH. Yet, DHM greatly reduces EtOH consumption when the high voluntary EtOH consumption is already established by exposure to EtOH (Fig. 3). DHM, even being applied one half hour before EtOH administration counteracts EtOH intoxication as effectively as if DHM is concurrently applied with EtOH (Fig. 1 D). Therefore, DHM is a realistic, unique candidate for development as a medication for AUD.


This work was supported by United States Public Health Service Grants AA017991 to JL, AA07680 to RWO, AA016100 to IS, and Tobacco-Related Disease Research Program (California) grant 18XT-0183 to XMS. We thank Dr. Werner Sieghart for providing antibodies, Delia Tio for plasma [EtOH] assays, and Martin Wallner for helpful discussions.


The authors declare that they have no competing interests.

Author contributions: Y.S., A.K.L., C.G., and J.L. performed the research; X.M.S., I.S., R.W.O., and J.L. designed the research; Y.S., A.K.L., C.G., and J.L. analyzed the research; Y.S., X.M.S., R.W.O., and J.L. wrote the paper.


  • Aguayo LG, Peoples RW, Yeh HH, Yevenes GE. GABAA receptors as molecular sites of ethanol action. Direct or indirect actions? Curr Top Med Chem. 2002;2:869–885. [PubMed]
  • Arolfo MP, Overstreet DH, Yao L, Fan P, Lawrence AJ, Tao G, Keung WM, Vallee BL, Olive MF, Gass JT, Rubin E, Anni H, Hodge CW, Besheer J, Zablocki J, Leung K, Blackburn BK, Lange LG, Diamond I. Suppression of heavy drinking and alcohol seeking by a selective ALDH-2 inhibitor. Alcohol Clin Exp Res. 2009;33:1935–1944. [PubMed]
  • Boehm SL, 2nd, Ponomarev I, Jennings AW, Whiting PJ, Rosahl TW, Garrett EM, Blednov YA, Harris RA. gamma-Aminobutyric acid A receptor subunit mutant mice: new perspectives on alcohol actions. Biochem Pharmacol. 2004;68:1581–1602. [PubMed]
  • Cagetti E, Liang J, Spigelman I, Olsen RW. Withdrawal from chronic intermittent ethanol treatment changes subunit composition, reduces synaptic function, and decreases behavioral responses to positive allosteric modulators of GABAA receptors. Mol Pharmacol. 2003;63:53–64. [PubMed]
  • Chen SH, Zhong GS, Li AL, Li SH, Wu LK. Influence of Hovenia dulcis on alcohol concentration in blood and activity of alcohol dehydrogenase (ADH) of animals after drinking. Zhongguo Zhong Yao Za Zhi. 2006;31:1094–1096. [PubMed]
  • Devaud LL, Purdy RH, Finn DA, Morrow AL. Sensitization of gamma-aminobutyric acid A receptors to neuroactive steroids in rats during ethanol withdrawal. J Pharmacol Exp Ther. 1996;278:510–517. [PubMed]
  • Devaud LL, Fritschy JM, Sieghart W, Morrow AL. Bidirectional alterations of GABAA receptor subunit peptide levels in rat cortex during chronic ethanol consumption and withdrawal. J Neurochem. 1997;69:126–130. [PubMed]
  • Du J, He D, Sun LN, Han T, Zhang H, Qin LP, Rahman K. Semen Hoveniae extract protects against acute alcohol-induced liver injury in mice. Pharm Biol. 2010;48:953–8. [PubMed]
  • Hanchar HJ, Dodson PD, Olsen RW, Otis TS, Wallner M. Alcohol-induced motor impairment caused by increased extrasynaptic GABAA receptor activity. Nat Neurosci. 2005;8:339–345. [PMC free article] [PubMed]
  • Heilig M, Egli M. Pharmacological treatment of alcohol dependence: target symptoms and target mechanisms. Pharmacol Ther. 2006;111:855–876. [PubMed]
  • Heilig M, Egli M, Crabbe JC, Becker HC. Acute withdrawal, protracted abstinence and negative affect in alcoholism: are they linked? Addict Biol. 2010;15:169–84. [PMC free article] [PubMed]
  • Hester R, Miller W. Treatment approaches: effective alternatives. Allyn and Bacon; Needham Heights: 2003. Needham Heights: Allyn and Bacon.
  • Hobbs WR, Rall TW, Verdoorn TA. Hypnotics and sedatives; ethanol. In: Hardman JG, et al., editors. The pharmacological basis of therapeutics. 9th ed. McGraw-Hill; New York: 1996. pp. 361–396.
  • Hunkeler W, Mohler H, Pieri L, Polc P, Bonetti EP, Cumin R, Schaffner R, Haefely W. Selective antagonists of benzodiazepines. Nature. 1981;290:514–516. [PubMed]
  • Johnson BA. Medication treatment of different types of alcoholism. Am J Psychiatry. 2010;167:630–639. [PMC free article] [PubMed]
  • Keung WM, Vallee BL. Daidzin: a potent, selective inhibitor of human mitochondrial aldehyde dehydrogenase. Proc Natl Acad Sci U S A. 1993;90:1247–1251. [PubMed]
  • Kim KH, Chung YT, Lee JH, Park YS, Shin MK, Kim HS, Kim DH, Lee HY. Hepatic detoxification activity and reduction of serum alcohol concentration of Hovenia dulcis Thunb from Korea and China. Korean J Med Crop Sci. 2000;8:225–233.
  • Koob GF. A role for GABA mechanisms in the motivational effects of alcohol. Biochem Pharmacol. 2004;68:1515–1525. [PubMed]
  • Kumar S, Porcu P, Werner DF, Matthews DB, Diaz-Granados JL, Helfand RS, Morrow AL. The role of GABAA receptors in the acute and chronic effects of ethanol: a decade of progress. Psychopharmacology (Berl) 2009;205:529–564. [PMC free article] [PubMed]
  • Li GD, Chiara DC, Cohen JB, Olsen RW. Numerous classes of general anesthetics inhibit etomidate binding to gamma-aminobutyric acid type A (GABAA) receptors. J Biol Chem. 2010;285:8615–8620. [PMC free article] [PubMed]
  • Li YQ, Yu ZG, Lu WC. Dihydromyricetin and myricetin content in Ampelopsis grossedentata measured with HPLC technique. Chin. Tradit. Herb. Drugs. 2003;34:1098.
  • Liang J, Cagetti E, Olsen RW, Spigelman I. Altered pharmacology of synaptic and extrasynaptic GABAA receptors on CA1 hippocampal neurons is consistent with subunit changes in a model of alcohol withdrawal and dependence. J Pharmacol Exp Ther. 2004;310:1234–1245. [PubMed]
  • Liang J, Zhang N, Cagetti E, Houser CR, Olsen RW, Spigelman I. Chronic intermittent ethanol-induced switch of ethanol actions from extrasynaptic to synaptic hippocampal GABAA receptors. J Neurosci. 2006;26:1749–1758. [PubMed]
  • Liang J, Suryanarayanan A, Abriam A, Snyder B, Olsen RW, Spigelman I. Mechanisms of reversible GABAA receptor plasticity after ethanol intoxication. J Neurosci. 2007;27:12367–12377. [PubMed]
  • Liang J, Spigelman I, Olsen RW. Tolerance to sedative/hypnotic actions of GABAergic drugs correlates with tolerance to potentiation of extrasynaptic tonic currents of alcohol-dependent rats. J Neurophysiol. 2009;102:224–233. [PubMed]
  • Lin RC, Li TK. Effects of isoflavones on alcohol pharmacokinetics and alcohol-drinking behavior in rats. Am J Clin Nutr. 1998;68:1512S–1515S. [PubMed]
  • Lovinger DM, Homanics GE. Tonic for what ails us? high-affinity GABAA receptors and alcohol. Alcohol. 2007;41:139–143. [PMC free article] [PubMed]
  • Lukas SE, Penetar D, Berko J, Vicens L, Palmer C, Mallya G, Macklin EA, Lee DY. An extract of the Chinese herbal root kudzu reduces alcohol drinking by heavy drinkers in a naturalistic setting. Alcohol Clin Exp Res. 2005;29:756–762. [PubMed]
  • Mangan PS, Sun C, Carpenter M, Goodkin HP, Sieghart W, Kapur J. Cultured hippocampal pyramidal neurons express two kinds of GABAA receptors. Mol Pharmacol. 2005;67:775–788. [PubMed]
  • Morrow AL, Biggio G, Serra M, Becker HC, Lopez MF, Porcu P, Alward SE, O’Buckley TK. The role of neuroactive steroids in ethanol/stress interactions: proceedings of symposium VII at the Volterra conference on alcohol and stress. Alcohol. 2008;43:521–30. [PMC free article] [PubMed]
  • Morrow AL, Montpied P, Lingford-Hughes A, Paul SM. Chronic ethanol and pentobarbital administration in the rat: effects on GABAA receptor function and expression in brain. Alcohol. 1990;7:237–244. [PubMed]
  • Mukherjee S, Das SK, Vaidyanathan K, Vasudevan DM. Consequences of alcohol consumption on neurotransmitters-an overview. Curr Neurovasc Res. 2008;5:266–272. [PubMed]
  • Olsen RW, Homanics GE. In: Function of GABAA receptors: insights from mutant and knockout mice, in GABA in the nervous system: the view at fifty years. Martin DL, Olsen RW, editors. Lippincott Williams & Wilkins; Philadelphia: 2000. pp. 81–96.
  • Olsen RW, Spigelman I. GABAAR plasticity in alcohol withdrawal. In: Noebels JL, Avoli M, Rogawski MA, Olsen RW, Delgado-Escueta AV, editors. Jasper’s Basic Mechanisms of the Epilepsies. Fourth Edition. Oxford University Press; New York: 2011. in press.
  • O’Malley SS, Rounsaville BJ, Farren C, Namkoong K, Wu R, Robinson J, O’Connor PG. Initial and maintenance naltrexone treatment for alcohol dependence using primary care vs specialty care: a nested sequence of 3 randomized trials. Arch Intern Med. 2003;163:1695–1704. [PubMed]
  • Paul SM. Alcohol-sensitive GABA receptors and alcohol antagonists. Proc Natl Acad Sci U S A. 2006;103:8307–8308. [PubMed]
  • Siggins GR, Roberto M, Nie Z. The tipsy terminal: presynaptic effects of ethanol. Pharmacol Ther. 2005;107:80–98. [PubMed]
  • Shen Y, Lindemeyer AK, Spigelman I, Sieghart W, Olsen RW, Liang J. Plasticity of GABAA receptors following ethanol pre-exposure in cultured hippocampal neurons. Mol Pharmacol. 2011;79:432–442. [PubMed]
  • Suzdak PD, Glowa JR, Crawley JN, Schwartz RD, Skolnick P, Paul SM. A selective imidazobenzodiazepine antagonist of ethanol in the rat. Science. 1986;234:1243–7. [PubMed]
  • Trevisan LA, Boutros N, Petrakis IL, Krystal JH. Complications of alcohol withdrawal: pathophysiological insights. Alcohol Health Res World. 1998;22:61–66. [PubMed]
  • Valenzuela CF. Alcohol and neurotransmitter interactions. Alcohol Health & Research World. 1997;21:144–148. [PubMed]
  • Wallner M, Hanchar HJ, Olsen RW. Ethanol enhances α4β3δ and α6β3δ gamma-aminobutyric acid type A receptors at low concentrations known to affect humans. Proc Natl Acad Sci U S A. 2003;100:15218–15223. [PubMed]
  • Wallner M, Hanchar HJ, Olsen RW. Low-dose alcohol actions on α4β3δ GABAA receptors are reversed by the behavioral alcohol antagonist Ro15-4513. Proc Natl Acad Sci U S A. 2006;103:8540–8545. [PubMed]
  • Wallner M, Olsen RW. Physiology and pharmacology of alcohol: the imidazobenzodiazepine alcohol antagonist site on subtypes of GABAARs as an opportunity for drug development? Br J Pharmacol. 2008;154:288–298. [PMC free article] [PubMed]
  • Wei W, Faria LC, Mody I. Low ethanol concentrations selectively augment the tonic inhibition mediated by δ subunit-containing GABAA receptors in hippocampal neurons. J Neurosci. 2004;24:8379–8382. [PubMed]
  • Weiner JL, Valenzuela CF. Ethanol modulation of GABAergic transmission: the view from the slice. Pharmacol Ther. 2006;111:533–554. [PubMed]
  • Youdim KA, Dobbie MS, Kuhnle G, Proteggente AR, Abbott NJ, Rice-Evans C. Interaction between flavonoids and the blood-brain barrier: in vitro studies. J Neurochem. 2003;85:180–192. [PubMed]