Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Alcoholism is characterized by pathological patterns of alcohol use, which are thought to involve unique subsets of neural systems. The identification and characterization of neural targets is important to the understanding and the development of treatments. Mapping of inducible transcription factors (ITFs) encoded by immediate early genes in animals provides a valuable tool for identifying targets during alcohol intoxication (Herdegen and Leah, 1998). ITFs from the Fos family are commonly used in mapping studies due to their low basal expression and increases after drug treatment, which are thought to result from increased neuronal activity (Morgan et al., 1987).

ITF brain mapping after involuntary alcohol administration has identified several brain regions showing c-Fos elevations (Chang et al., 1995; Thiele et al., 1996). Interestingly, we have shown a much different pattern of expression after voluntary alcohol drinking in mice (Bachtell et al., 1999). Specifically, c-Fos induction was not observed in brain regions showing strong reactivity to alcohol injections, including the lateral portion of the central nucleus of amygdala, bed nucleus of the stria terminalis, paraventricular nucleus of the thalamus, and paraventricular nucleus of the hypothalamus. Importantly, the Edinger-Westphal nucleus (EW) showed comparable levels after both voluntary and involuntary administrations. These results agree with other reports demonstrating EW's involvement in alcohol self-administration (Topple et al., 1998; Weitemier et al., 2001).

The EW has received limited attention in alcohol research because of its association with oculomotor functions. However, mounting evidence suggests that the EW is involved in other functions, including temperature regulation, nociception, and anxiety (Loewy et al., 1978; Innis and Aghajanian, 1986; Smith et al., 1998; Weninger et al., 1999). Moreover, the EW is the main brain source of the neuropeptide urocortin, the alternative and strongest ligand of corticotropin-releasing factor receptors (Vaughan et al., 1995). Our recent studies using inbred strains suggest that urocortin-containing neurons of EW are involved in regulation of alcohol-induced hypothermia and alcohol consumption in mice (Bachtell et al., 2002). Therefore, it is essential to characterize the mechanisms of alcohol-induced c-Fos expression in EW neurons. To this aim, we sought to characterize the pharmacological and signal transduction nature of alcohol-induced c-Fos expression in the EW.

The presence of postsynaptic ₂-adrenoceptors in EW seems to play a large role in the pupillary dilation response (i.e., mydriasis) (Heal et al., 1995). In addition, pharmacological studies using benzodiazepines suggest the involvement of GABA_A receptors in activating EW neurons (Cutrera et al., 1993; Skelton et al., 2000). Other evidence shows that microinjections of opioid agonists (e.g., morphine and fentanyl) into the EW produce behavioral effects without significantly modulating mydriasis (Kamenetsky et al., 1997). Thus, the opioid system may play a role in nonmydriatic behaviors stemming from EW neurons. In addition to these receptor systems, tyrosine hydroxylase-containing projections stemming from the highly dopaminergic ventral tegmental area suggest that the dopaminergic system may also play a role. Therefore, these four transmitter systems were tested for their involvement in alcohol-induced c-Fos expression in EW neurons.

The second part of this study analyzing the signal transduction mechanisms in the EW was based on previous reports showing that alcohol administration elevates expression of c-Fos, but does not elevate expression of ITF Egr1 (also known as NGFI-A, Zif268, and Krox-24) in EW (Bachtell et al., 1999; Bachtell and Ryabinin, 2001). The promoter regions of c-fos and egr1 genes contain several common regulatory elements, including the serum response element (SRE) and the calcium/cAMP response element (Ca/CRE). On the other hand, the c-fos promoter contains the c-Sis inducible element (SIE), whereas the egr1 promoter does not (Herdegen and Leah, 1998). We hypothesized, therefore, that alcohol induces c-Fos mainly via an element that is unique to the c-Fos promoter. To address this hypothesis we used immunohistochemistry to analyze the signal transduction machinery not only acting through the SRE and Ca/CRE pathway (the CREB and Elk1 transcription factors), but also acting through the SIE mechanism (the STAT proteins).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Male C57BL/6J mice purchased from Jackson Laboratories (Bar Harbor, ME) were placed four per cage for a minimum of 1 week. All animals were 7 to 9 weeks of age at the beginning of the experimental procedure. Although other strains were considered, C57BL/6J mice were selected for this analysis because of the well characterized ITF expression in our previous experiments in the EW nucleus (Bachtell et al., 2002). Animals were maintained on a 12-h light/dark cycle with lights on beginning at 6:00 AM. Water and food were available ad libitum throughout the experiment in the home cage. All animal procedures were in accordance with National Institutes of Health guidelines.

General Procedures. All animals were habituated to handling and injections on the 4 days before the drug treatment day. Our preliminary experiments have shown this procedure to be satisfactory in reducing basal c-Fos expression to minimal levels. During the habituation procedures, animals were exposed to two injections. All injections in all experiments were intraperitoneal. The second injection was made contralateral to the first injection. On the test day, animals were injected with the pretreatment drug and placed back into the home cage. Animals were injected again at precise intervals (depending on the drug properties), on the contralateral side, with the second injection (see specific experiments). Animals were then placed back into the home cage. Animals were sacrificed 100 min after the second injection by CO₂ inhalation. Animals were not perfused, however, brains were postfixed in 2% paraformaldehyde in phosphate-buffered saline (PBS) after removal. Brains were then immersed in 20% sucrose in PBS containing 0.1% NaN₃ for 24 h followed by two 24-h incubations in 30% sucrose in PBS containing 0.1% NaN₃.

Immunohistochemistry. Immunohistochemistry was used because of the small size of EW and the difficulty in isolating it for Western blot analyses. All immunohistochemistry was performed according to previously published protocols (Bachtell et al., 1999), unless otherwise noted. Briefly, frozen brains were sectioned into 40-µm slices on a CM1850 cryostat (Leica Microsystems, Inc., Deerfield, IL). Blocking was performed with 4% goat serum. All primary antibodies were raised in rabbit, whereas the secondary antibodies were raised in goat (see specific experiments for antibody details). The immunoreaction was detected with Vectastain ABC kit (Vector Laboratories, Burlingame, CA), and enzymatic development was performed with the Metal Enhanced DAB kit (Pierce Chemical, Rockford, IL).

Data Analysis. Cell counting was performed using a quantitative image analysis system as reported previously (Bachtell et al., 1999). Two slices containing EW were selected from each animal at approximately 3.00 to 3.40 from bregma as described by Franklin and Paxinos (1997). Our preliminary experiments have determined that sections between 3.00 and 3.40 from bregma produce the most consistent results with c-Fos immunohistochemistry. To control for variability in 3,3-diaminobenzidine tetrachloride intensity, all statistical analyses were performed on sections processed simultaneously. Data from the two sections were averaged and used as a single data point for statistical analysis using an analysis of variance. Significant main and interactive effects were further analyzed with appropriate post hoc tests.

Experiment 1: Tests of GABA-ergic Involvement. The involvement of the GABA-ergic system in alcohol-induced c-Fos expression was first tested by pretreating animals (n = 4-6/group) with saline, the GABA_A antagonist ()-bicuculline (Sigma-Aldrich, St. Louis, MO), or the negative modulator of GABA_A pentylenetetrazole (PTZ; Sigma-Aldrich) dissolved in saline. Ten minutes after the pretreatment, animals were injected with either saline or 2.4 g/kg alcohol (20% v/v in 0.9% saline). This dose of alcohol was previously shown to produce maximal c-Fos expression in EW neurons (Bachtell et al., 2002). Two positive modulators at the GABA_A site, 3-hydroxy-5-pregnan-20-one (35-P, gift from Dr. Deborah Finn, Oregon Health and Science University, Portland, OR) and chlordiazepoxide (CDP, gift from Dr. Robert Hitzemann, Oregon Health and Science University) were dissolved in 2-hydroxypropyl--cyclodextrin and saline, respectively, and were tested for their ability to induce c-Fos expression in the EW. A pretreatment of saline was administered 10 min before a vehicle, 35-P, or CDP injection (n = 4-6/group). A group of animals (n = 4) injected with 2.4 g/kg alcohol was also treated simultaneously to allow comparison of the GABA_A effects with those produced by alcohol. In all cases, animals were sacrificed 100 min after the second injection and were analyzed for c-Fos (1:10,000 dilution; Santa Cruz Biotechnology, Santa Cruz, CA) using the protocol described above.

Experiment 7: Tests of MEK 1/2 Inhibition. Inhibition of MEK 1/2 activity was accomplished with the MEK 1/2 inhibitor SL327 (gift from Dr. James Trzaskos; DuPont Pharmaceuticals, Wilmington, DE), which has the primary advantage of being active via systemic injections (Selcher et al., 1999). SL327 was dissolved in DMSO and pretreated to animals 40 min before a saline or 2.4 g/kg alcohol injection (n = 5/group). Animals were sacrificed 100 min after the second injection, and immunohistochemistry was performed for c-Fos as described above.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Experiment 1: GABA-ergic Involvement in Alcohol-Induced c-Fos Expression in EW Neurons. Administration of GABA-ergic antagonists produced significant changes in alcohol-induced c-Fos expression in the EW (Fig. 1, C and D). Specifically, pretreatment with 4 mg/kg bicuculline significantly attenuated alcohol-induced c-Fos expression [F(1,19) = 100.31, p < 0.001]. Like bicuculline, pretreatment with 45 mg/kg PTZ significantly attenuated alcohol-induced c-Fos expression [F(1,20) = 28.76, p < 0.001]. These doses of bicuculline and PTZ approach doses known to produce seizures (Johansson et al., 1996); however, we observed no convulsive effects in our study. Administration of the positive modulator of the GABA_A receptor also produced significant effects. The neurosteroid 35-P produced a dose-dependent increase in EW c-Fos expression (Fig. 2A). In particular, 10 mg/kg 35-P produced a modest, statistically insignificant increase in c-Fos expression, whereas the 20 mg/kg 35-P showed a significant increase in EW c-Fos expression. This dose of 35-P in mice is expected to produce a similar locomotor response to 2.4 g/kg ethanol (Palmer et al., 2002). Therefore, higher doses of 35-P were not included. The benzodiazepine CDP showed a uniform, but significant, increase in c-Fos expression across all administered doses (Fig. 2B). These elevations in c-Fos expression after both GABA-ergic agonists are significantly below that shown after alcohol administration (Fig. 2).

View larger version (63K): [in this window] [in a new window]   Fig. 1.   Expression of c-Fos after saline/saline (A), saline/2.4 g/kg alcohol (B), 4 mg/kg bicuculline/2.4 g/kg alcohol (C), 45 mg/kg pentylenetetrazole/2.4 g/kg alcohol (D), 10 mg/kg ARC239/2.4 g/kg alcohol (E), 10 mg/kg RX 821002/2.4 g/kg alcohol (F), and c-Fos expression data from GABAA antagonist pretreatment of bicuculline and PTZ (G). Quantification of E and F are shown in Fig. 4. Note the robust elevation in c-Fos expression after alcohol administration alone (B) and significant alterations produced by pretreatment of 4 mg/kg bicuculline (C), 45 mg/kg pentylenetetrazole (D), and 10 mg/kg RX 821002 (F). No changes were seen in animals pretreated with 10 mg/kg ARC 239 (E). , significant post hoc difference from saline (p < 0.05); , significant post hoc difference from saline (p < 0.01); , significant post hoc difference from all groups (p < 0.001).

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Expression of c-Fos after the administration of positive modulators of the GABAA receptor. A, 35-P produced significant increases in c-Fos expression after the 20-mg/kg dose but not the 10-mg/kg dose [F(3,16) = 12.764, p < 0.001]. This increase, however, was significantly below elevations produced by 2.4 g/kg alcohol. B, similarly, chlordiazepoxide produced significant increases in EW c-Fos expression [F(4,20) = 32.66, p < 0.001] that were significantly less than that produced by 2.4 g/kg alcohol. , significant post hoc difference from saline (p < 0.01); , significant post hoc difference from all groups (p < 0.001).

Experiment 2: Opioid Involvement in Alcohol-Induced c-Fos Expression in EW Neurons. Pretreatment with 30 mg/kg naltrexone before 2.4 g/kg alcohol had no effect on alcohol-induced c-Fos expression (Fig. 3). When morphine was administered at 10 and 100 mg/kg, however, a significant increase in EW c-Fos was detected (Fig. 3). As was evident with the positive modulators of the GABA_A complex, morphine-induced c-Fos expression was significantly below c-Fos levels observed after 2.4 g/kg alcohol. Although 30 mg/kg naltrexone did not affect alcohol-induced c-Fos expression in the EW, 30 mg/kg naltrexone significantly blocked both the 10- and 100-mg/kg doses of morphine (Fig. 3).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Expression of c-Fos after administration of morphine and alcohol after the pretreatment of saline or 30 mg/kg naltrexone. Morphine administered alone produced significant elevations of EW c-Fos, whereas this was effectively blocked with the pretreatment of naltrexone [F(2,18) = 3.84, p < 0.05]. Morphine-induced c-Fos expression in EW was significantly below c-Fos expression levels after 2.4 g/kg alcohol and was not affected by pretreatment of naltrexone. , significant post hoc difference from saline and naltrexone pretreated (p < 0.01); , significant post hoc difference from all morphine and saline groups (p < 0.001).

Experiment 3: Noradrenergic Involvement in Alcohol-Induced c-Fos Expression in EW Neurons. The nonselective ₂-adrenergic antagonist yohimbine produced a dose-dependent blockade of alcohol-induced c-Fos expression (Fig. 4). Alcohol-induced c-Fos remained elevated when 20 mg/kg propranolol, a nonselective -adrenergic antagonist, was administered before alcohol (Fig. 4). Although only one dose of propranolol was administered, sedative effects were observed, indicating behavioral activity of the drug. To further characterize the ₂-adrenergic antagonist effects, two compounds were used to elucidate ₂-adrenoceptor subtype-specific mechanisms of alcohol-induced c-Fos expression in EW (Fig. 1, E and F). First, 10 mg/kg ARC 239, a _2B/C-selective antagonist, showed no statistically significant changes in alcohol-induced c-Fos expression (Fig. 4). Behaviorally, this dose of ARC 239 was extremely sedative. Second, we used 5 and 10 mg/kg RX 821002, which has been shown to be a _2A/D-selective antagonist (Trendelenburg et al., 1996). When administered before alcohol, RX 821002 dose dependently diminished alcohol-induced c-Fos expression in EW neurons (Fig. 4).

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Alcohol-induced expression of c-Fos in EW after the pretreatment of adrenoceptors antagonists. A, yohimbine significantly blocked alcohol-induced c-Fos expression in EW neurons [F(2,26) = 11.6, p < 0.001]. B, propranolol had no significant effect on alcohol-induced c-Fos expression (F < 1.0). C, ARC 239 had no significant effect on alcohol-induced c-Fos expression (F < 1.0). D, RX 821002 significantly blocked alcohol-induced c-Fos expression in EW neurons [F(2,24) = 20.99, p < 0.001]. , significant post hoc difference from all groups (p < 0.001).

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Expression of c-Fos in EW after administration with chlordiazepoxide and morphine is modulated by 2-adrenoceptors. A, chlordiazepoxide-induced c-Fos expression is significantly blocked by pretreatment with 5 mg/kg yohimbine [F(1,12) = 7.11, p < 0.05]. B, morphine-induced c-Fos expression is significantly blocked by pretreatment with 10 mg/kg RX 821002 [F(1,12) = 96.542, p < 0.001]. , significant post hoc difference from all groups (p < 0.01).

Experiment 4: Morphological Characterization of Tyrosine Hydroxylase-Containing Neurons in the Edinger-Westphal Nucleus. Using tyrosine hydroxylase, the biosynthetic precursor of both dopamine and noradrenaline, it was found that TH-containing neurons project to EW (Fig. 6). Although these projections most likely represent the ventral noradrenergic bundle, which sends ascending noradrenergic fibers from the A7 brainstem nuclei to midbrain and forebrain structures, it is also possible for antibodies to TH to detect dopaminergic fibers. Moreover, the observed projections were seemingly stemming from the highly dopaminergic region of the ventral tegmental area (VTA). Therefore, we sought to investigate the influence of dopaminergic compounds on c-Fos expression in EW.

View larger version (70K): [in this window] [in a new window]   Fig. 6.   Immunohistochemical staining for tyrosine hydroxylase in the Edinger-Westphal nucleus. Lower magnification (left) shows TH-positive fibers extending to EW from the ventral tegmental area. Higher magnification (right) shows presence of few TH-positive cells within EW.

Experiment 5: Dopaminergic Involvement in Alcohol-Induced c-Fos Expression in EW Neurons. Pretreatment with 0.5 mg/kg of the dopamine antagonist haloperidol before alcohol administration resulted in a complete blockade of alcohol-induced c-Fos expression (Fig. 7). Administration of both 1 and 10 mg/kg apomorphine resulted in no significant elevation in c-Fos expression (data not shown).

View larger version (24K): [in this window] [in a new window]   Fig. 7.   Haloperidol significantly blocks alcohol-induced c-Fos expression in the Edinger-Westphal nucleus [F(1,16) = 64.4, p < 0.001]. , significant post hoc difference from all groups (p < 0.001).

Experiment 6: Signal Transduction Cascades after Acute Alcohol Administration. Overall, relatively few changes were observed in the targeted upstream modulators of the c-Fos promotor at 5, 15, or 30 min after alcohol injection (Fig. 8). In particular, no changes were observed in phospho-CREB in EW neurons, suggesting that the pathway targeting the Ca/CREB element is not involved in alcohol-induced c-Fos expression in the EW. Changes in phospho-CREB were observed in the paraventricular nucleus of hypothalamus. EW neurons were also absent of changes in phospho-Elk1 levels, which is one of the primary regulators of the SRE site, thus lending support to the hypothesis that the SRE is not involved in alcohol-induced c-Fos expression. Interestingly, however, alcohol produced significant elevations in serine 727-phosphorylated Stat3 in the nuclei of EW neurons at the 15-min time point (Fig. 9). Tyrosine 705-phosphorylated Stat3, however, was not elevated at any time points after alcohol or saline treatment. EW neurons did not show any changes in other tested members of the Stat family, suggesting specificity for Stat3, with preferential phosphorylation activity at the serine 727. In addition to the changes in serine 727-phosphorylated Stat3, we observed changes in the upstream regulator of many substrates, phospho-ERK1/2. Specifically, significantly higher levels of phospho-ERK 1/2 were observed in EW neurons in animals receiving alcohol (Fig. 9). As with the elevations noted with phospho-Stat3, these changes were apparent at 15-min postinjection. No changes in other upstream regulators, including p90RSK, phospho-Stat1, and phospho-MAPK 38 were observed (data not shown).

View larger version (145K): [in this window] [in a new window]   Fig. 8.   Administration of 2.4 g/kg alcohol produced robust expression of phosphorylated Elk1 in the cingulate cortex (A) at 15 min postinjection. Relatively low levels of expression, however, were observed in Edinger-Westphal neurons at this same time point after 2.4 g/kg alcohol (B). Likewise, robust levels of phosphorylated CREB were detected in the paraventricular nucleus of the hypothalamus (C) 15 min after 2.4 g/kg alcohol, whereas no changes were detected in Edinger-Westphal neurons (D).

View larger version (68K): [in this window] [in a new window]   Fig. 9.   Significant changes were observed in the expression of serine 727-phosphorylated Stat3 at 15 min postinjection [F(1,27) = 4.6, p < 0.05, A] in Edinger-Westphal neurons. Expression changes between saline (C) and 2.4 g/kg alcohol (D) are shown to the right for the 15-min time point. Importantly, no changes were observed in the tyrosine 705 phosphorylated form of Stat3. Small punctate staining (arrows) was observed in the medial amygdaloid nucleus (E) and other regions; however, the Edinger-Westphal nucleus was absent of staining (F). Significant changes were also observed in the expression of phosphorylated ERK 1/2 15 min postinjection [F(1,20) = 4.6, p < 0.05, B]. These expression changes between saline (G) and 2.4 g/kg alcohol (H) are shown to the right for the 15-min time point. , significant post hoc difference from all groups (p < 0.01); , significant post hoc difference from all groups (p < 0.001).

Experiment 7: Expression of c-Fos in EW after Pretreatment of the MEK 1/2 Inhibitor SL327. In an attempt to evaluate the role of the MEK/ERK pathway in alcohol-induced c-Fos in EW neurons, we blocked MEK 1/2 activity pharmacologically with SL327. Administration of SL327 40 min before alcohol or saline treatment showed a dose-dependent blockade of c-Fos expression in EW neurons (Fig. 10). Specifically, 50 mg/kg SL327 produced relatively no effect on alcohol-induced c-Fos expression. Pretreatment with 100 mg/kg SL327, however, completely blocked alcohol-induced c-Fos expression in EW neurons, suggesting an MEK/ERK-dependent pathway.

View larger version (18K): [in this window] [in a new window]   Fig. 10.   Pretreatment with the MEK 1/2 inhibitor SL327 significantly blocked alcohol-induced c-Fos expression [F(2,18) = 19.21, p < 0.001]. , significant post hoc difference from all groups (p < 0.001).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Alcohol-induced elevations in c-Fos in EW neurons result from both voluntary and involuntary alcohol administration. Recent data indicate that both routes of alcohol administration lead to c-Fos expression in cells containing the neuropeptide urocortin, which may function to regulate alcohol-induced hypothermia and alcohol consumption (Bachtell et al., 2002). The findings presented herein suggest that, upon involuntary alcohol exposure, EW neurons are influenced by several neurotransmitter receptors. Specifically, it seems that GABA_A receptor activation is necessary and sufficient for increased expression of c-Fos in EW neurons. This suggests that alcohol is acting directly upon GABA_A receptors. This is supported by data showing that positive modulators of the GABA_A receptor (e.g., CDP and 3 5-P) lead to elevated c-Fos in EW neurons as well as data showing attenuation of alcohol-induced c-Fos upon GABA_A antagonist pretreatment. In agreement with our findings are reports showing elevated c-Fos expression and increased urocortin mRNA in EW neurons upon treatment with benzodiazepines (Cutrera et al., 1993; Skelton et al., 2000).

Alcohol actions at GABA_A receptors do not, however, seem to be the only transmitter system involved. This is evidenced by the inability of either CDP or 35-P to induce c-Fos expression in EW to levels comparable with that of alcohol alone. Furthermore, complete blockade of alcohol-induced c-Fos expression was observed by pretreatment of _2A/D-adrenoceptor and dopaminergic receptor antagonists. These receptors are therefore necessary for alcohol-induced c-Fos expression. Administration of direct agonists at these sites, however, suggests that these receptors are not sufficient for increased c-Fos expression in EW neurons. Thereby, neither the ₂-adrenoceptor agonist clonidine nor the dopaminergic agonist apomorphine produced significant elevations of EW c-Fos. The latter supports our previous data showing no increased EW c-Fos expression after cocaine administration, which is a potent activator of extracellular dopamine (Ryabinin et al., 2000). The dopaminergic effects demonstrated in this report may represent nonspecific effects in that we only note dopaminergic modification with haloperidol, which is thought to have actions at serotonin receptors in addition to dopaminergic D2 receptors (for review, see Millan, 2000). Regardless, it is believed that alcohol is acting directly at GABA_A receptors, and this activity is modified by actions at _2A/D and dopaminergic receptors. Support for this claim stems from data in which CDP-induced c-Fos expression in EW is blocked by the nonselective ₂-adrenoceptor antagonist yohimbine.

The opioid system is not necessary for alcohol-induced c-Fos as was evidenced by the inability of naltrexone to modulate alcohol-induced c-Fos in the EW. Interestingly, the opioid system does seem sufficient for c-Fos expression in EW neurons in that morphine significantly elevates c-Fos levels. As was shown with CDP- and 35-P-induced c-Fos expression, morphine-induced c-Fos was significantly below that produced by alcohol alone. Furthermore, morphine-induced c-Fos expression may be influenced by _2A/D-receptors in a similar manner to that of alcohol and CDP. This notion is supported by the data presented herein that shows that the _2A/D-antagonist RX 821002 blocks morphine-induced c-Fos expression in EW neurons. This finding is intriguing given data suggesting that morphine and fentanyl are acting directly on EW neurons to produce nociception and catalepsy, but not mydriasis, which is thought to be primarily mediated by postsynaptic _2A/D-receptors in EW (Kamenetsky et al., 1997).

Taken together, it seems that there are several neurotransmitter systems operating to influence the EW upon alcohol intoxication. Furthermore, these data implicate the involvement of other nuclei upstream from the EW nucleus. Although thorough tracing experiments have not been performed to map the upstream targets of the EW upon alcohol exposure, our data suggest that dopaminergic/noradrenergic neurons expressing tyrosine hydroxylase may play an important role in EW neuronal activity. These neurons likely represent noradrenergic projection neurons of the ventral noradrenergic bundle that perhaps stem from the locus coeruleus (Breen et al., 1983).

Support for this notion comes from a report that emphasizes that EW neurons are under the influence of two primary sources of input (Szabadi and Bradshaw, 1996). The first source is a tonic inhibitory input likely stemming from the locus coeruleus (LC). Physiological and pharmacological evidence supports this idea, showing that this pathway releases noradrenaline onto inhibitory postsynaptic ₂-adrenoceptors to influence autonomic functions (Koss, 1986). Supporting the notion that the LC is an upstream target of EW is evidence that alcohol decreases noradrenergic activity in the LC (Pohorecky and Brick, 1977). This decrease is reversed by antagonists acting at ₂-adrenoceptors (Verbanck et al., 1991). Interestingly, morphine also acts on LC neurons to decrease the activity of noradrenergic neurons (Aghajanian, 1978). In any case, the influence of LC input onto EW neurons demands further study given the pharmacological evidence presented herein.

The second source of input to the EW involves an indirect pathway that is mediated through a non-noradrenergic mechanism. Given the evidence presented herein, it is tempting to assign this non-noradrenergic input to dopamine, perhaps originating in the VTA as the TH-positive projections and pharmacological data purport. This may occur via alcohol's actions in the VTA to increase dopaminergic activity (Bunney et al., 2000). Contradictory to this hypothesis is data showing that agonists for dopamine do not produce c-Fos induction in EW. This is supported by our data showing that cocaine does not induce-c-Fos expression in EW neurons. Therefore, it is extremely likely that another source of input is modulating EW activity.

Physiological as well as morphological data suggest that EW neurons are influenced by an inhibitory input from the hypothalamus (Saper et al., 1976; Koss, 1986; Zheng et al., 1995). Although the transmitter systems targeted herein do not address this notion, the existence of Neuropeptide Y Y5 receptors in the EW gives credence to this hypothesis (Grove et al., 2000). Other evidence suggests that substance P projections from the olivary pretectal nucleus mediate EW activity through NK1 (Klooster et al., 1995, 2000). Therefore, it seems apparent that the receptor systems impacting EW neurons are quite complex and most likely involve a number of interactions occurring through both pre- and postsynaptic actions. Additionally, it is apparent from the results presented herein and previous work that input to the EW is primarily inhibitory, and it is proposed herein that alcohol acts to disinhibit the tonic inhibitory influence through one or more of these upstream targets of the EW.

Aside from the transmitters systems, it is important to consider the downstream actions of the receptors in EW neurons and the type of intracellular events that are initiated through their activation. Interestingly, all of the receptors suspected to modulate alcohol-induced c-Fos expression in EW neurons (including Y5 and NK1 receptors) act to down-regulate adenylate cyclase activity. This alone is suggestive of a unique signal transduction cascade culminating in alcohol-induced c-Fos expression. Further suggesting that a unique cascade may be involved is the observation that c-Fos, but not Egr1 is induced after alcohol administration (Bachtell et al., 1999; Bachtell and Ryabinin, 2001). Indeed, a unique sequence of events was observed in our experiments. Thus, we did not find phospho-CREB to be increased at any of the time points analyzed. This indicates that adenylate cyclase activation of protein kinase A, and subsequent phosphorylation of CREB are not involved with alcohol-induced c-Fos expression in EW neurons. Thereby, we do not believe that the Ca/CRE site is responsible for alcohol-induced c-Fos expression in EW neurons. We also detected that the Elk1 phosphorylation was not up-regulated at any of the time points, suggesting that SRE actions are not responsible for alcohol-induced c-Fos expression.

Rather, we observed that serine 727-phosphorylated, but not tyrosine 705-phosphorylated Stat3 was enhanced after alcohol administration. This enhancement occurred at a similar time point to the alcohol-induced elevations in phosphorylation of ERK 1/2. The similar time course in elevations suggests that these are not causally related; however, more precise time intervals need to be analyzed to make definite conclusions on the relationship of these two proteins. It does seem, however, that ERK 1/2 phosphorylation is necessary for alcohol-induced c-Fos induction in the EW. This is evidenced by significant blockade of alcohol-induced c-Fos expression when the MEK 1/2 inhibitor SL327 was administered before alcohol. Taken together, the data implicate the activation of an MEK 1/2-ERK 1/2 pathway that may phosphorylate Stat3 upon alcohol exposure, which culminates in c-Fos expression. Interestingly, a similar pathway has been observed in tissue culture experiments where it was shown that MEK 1/2 phosphorylation of ERK 1/2 occurs simultaneously with serine 727 phosphorylation, but not tyrosine 705 phosphorylation, of Stat3 (Ceresa et al., 1997).

Taken together, these findings provide important information for the development of research tools enabling the functional significance of alcohol-induced c-Fos expression in EW neurons to be revealed. Specifically, the data gathered in this report provide a first glance at the neurochemical mechanisms affected by alcohol in EW neurons and lend suggestions on how to directly manipulate it. Direct manipulation combined with behavioral analyses of alcohol sensitivity measures (e.g., alcohol-induced hypothermia and loss of righting reflex) and addiction models (e.g., alcohol drinking and conditioned place preference) will allow the determination of the role of EW during alcohol intoxication.