Ethanol transiently suppresses choline-acetyltransferase in basal nucleus of Meynert slices.

The cholinergic system plays a major role in learning and cognition and cholinergic neurons appear to be particularly vulnerable to ethanol (EtOH) exposure. There are conflicting results if EtOH directly damages cholinergic neurons. Thus, the aims of the present study were (1) to investigate the effect of different EtOH concentrations on cholinergic neurons in organotypic brain slices of the nucleus basalis of Meynert (nbM) and (2) to study if the most potent cholinotrophic substance nerve growth factor (NGF) or inhibitors of mitogen activated kinase (MAPK) p38- and nitric-oxide synthase (NOS)-pathways may counteract any EtOH effect. Two-week old organotypic rat brain slices of the nbM were exposed to 1-100 mM EtOH for 7 days with or without drugs and the number of choline-acetyltransferase (ChAT)-positive neurons was counted. Our data show that EtOH significantly reduced the number of ChAT-positive neurons with the most potent effect at a concentration of 50 mM EtOH (54±5 neurons per slice, p<0.001), compared to control slices (120±13 neurons per slice). Inhibition of MAPK p38 (SB 203580, 10 μM) and NOS (L-thiocitrulline, 10 μM) counteracted the EtOH-induced decline of cholinergic neurons and NGF protected cholinergic neurons against the EtOH-induced effect. Withdrawal of EtOH resulted in a reversal of cholinergic neurons to nearly controls. In conclusion, EtOH caused a transient decline of cholinergic neurons, possibly involving MAPK p38- and NOS-pathways suggesting that EtOH does not induce direct cell death, but causes a transient downregulation of the cholinergic key enzyme, possibly reflecting a form of EtOH-associated plasticity.

Introduction

The cholinergic system plays an important role in cognition and memory and is severely impaired in neurodegenerative diseases such as Alzheimer's disease (AD), vascular dementia (vaD) or alcohol-related dementia. A lack of the neurotransmitter acetylcholine directly correlates with cognitive decline. It is well known that chronic ethanol (EtOH) exposure results in decreased levels of acetylcholine, choline-acetyltransferase (ChAT) and acetylcholine-esterase in the basal forebrain (Arendt, 1994; Arendt et al., 1988, 1995; Costa and Guizzetti, 1999; Floyd et al., 1997; Jamal et al., 2009; Kentroti and Vernadakis, 1996; McKinney, 2005; Olton, 1983). There is strong controversy if alcohol consumption has positive or negative influence on development of dementia. Heavy drinking is a risk factor for most stroke subtypes favoring vascular damage in the brain which may be of importance in the development of vaD and possibly AD (Humpel, 2011; Sundell et al., 2008). Moderate alcohol consumption has been reported to lower the risk for AD, as well as other types of dementia (Huang et al., 2002; Ruitenberg et al., 2002). In fact several studies indicate that moderate chronic EtOH does not induce AD development, but rather suggest a protective effect (Anstey et al., 2009; Graves et al., 1990; Neafsey and Collins, 2011; Rosen et al., 1993; Tanaka et al., 2002). Alcohol-related dementia is completely different to AD etiology and pathogenesis, but has some similar clinical symptoms, such as e.g. cognitive decline (Aho et al., 2009). Some of the EtOH-induced toxic effects, especially on cholinergic neurons, are similar to those observed in AD and vaD possibly pointing to a common pathogenesis.

EtOH easily passes the blood–brain barrier (BBB) and interacts with various signal transduction cascades (Aroor and Shukla, 2004; Ku et al., 2007), ion channels (Allgaier, 2002), second messengers (Deng and Deitrich, 2007), neurotransmitters (Foddai et al., 2004; Jamal et al., 2007) and their receptors (Diamond and Messing, 1994). EtOH causes brain damage (Harper and Matsumoto, 2005), induces inflammatory processes (Blanco and Guerri, 2007; Crews and Nixon, 2009; Vallés et al., 2004), increases NF kappaB‐DNA binding (Crews et al., 2006; Zou and Crews, 2006), enhances cytokine-mediated inducible nitric oxide synthase (iNOS) production in astrocytoma cells (Davis et al., 2002) as well as in adolescent brain slice cultures (Zou and Crews, 2010). EtOH alters amyloid-precursor protein (APP) and APP processing enzymes (Kim et al., 2011; Lahiri et al., 2002), enhances the accumulation of hyperphosphorylated tau protein (Sun et al., 2005), and may lead to neuritic plaques in rats (Paula-Barbosa and Tavares, 1984), all pathological hallmarks seen in AD. In order to investigate a direct effect of EtOH on cholinergic neurons we aim to explore the consequence of direct EtOH-exposure on ChAT-positive neurons in organotypic brain slices of the nucleus basalis of Meynert (nbM). Furthermore, we will study if NGF or inhibition of NO- and mitogen activated kinase (MAPK) p38-pathways may mediate any EtOH-induced effects on cholinergic nbM neurons. In fact, we will show that EtOH does not induce cell death of cholinergic neurons, but rather causes a transient decline of the enzyme ChAT, possibly reflecting a form of EtOH-associated plasticity.

Results

Effects of EtOH on cholinergic neurons

Cholinergic neurons in nbM organotypic brain slices were visualized by ChAT-like immunohistochemistry (Fig. 1). In 2-week old control slices approximately 120 neurons were detectable (Figs. 1A, B) and this number did not change when slices were incubated for further 2 weeks without NGF. When 2-week old slices were incubated with different concentrations of EtOH for 7 days, a marked decline of cholinergic neurons was seen at concentrations > 10 mM EtOH (Figs. 1C, D). The EtOH effect was most pronounced at 50 mM but the number of ChAT+ neurons did not further decrease at higher concentrations (Fig. 2A). No effect was visible at 1 mM EtOH (Fig. 1A). When 2 week old slices where incubated with 50 mM EtOH for further 14 days, the number of ChAT+ neurons markedly declined to < 20 neurons per slice (Fig. 2). When EtOH was withdrawn from the culture medium, the number of cholinergic neurons returned to control levels (Fig. 3).

Fig. 1

Immunohistochemical staining of cholinergic neurons in organotypic rat brain slices of the basal nucleus of Meynert. Slices were cultured for 2 weeks with nerve growth factor (NGF) and for 3 days without NGF. Thereafter slices were cultured without (A, B) or with 50 mM ethanol (EtOH; C, D) for 7 days. Slices were then fixed and immunohistochemistry for choline-acetyltransferase (ChAT) was performed. Several ChAT+ neurons are shown in control slices (A, B) whereas EtOH markedly reduced the number of ChAT+ neurons in the slices (C, D). Scale bar in A = 280 μm (A, C) and = 45 μm (B, D).

Fig. 2

Effects of ethanol (EtOH) and inhibitors on cholinergic neurons. Brain slices were cultured for 2 weeks with nerve growth factor (NGF), then 3 days without NGF and incubated further for 7 days without or with 1–100 mM EtOH with or without 10 ng/ml NGF (A). Slices were treated without (Co) or with 50 mM EtOH (−) and with or without 10 μM MAPK p38 inhibitor (INH) or with or without 10 μM L-thiocitrulline (L-Thio) (B). Slices were fixed and stained for choline-acetyltransferase (ChAT). Values are given as mean ± SEM number of ChAT+ neurons per slice (numbers in parenthesis give the number of analyzed slices). Statistical analysis was performed by one way ANOVA followed by Fisher LSD post hoc test; (*p < 0.05;**p < 0.01, ***p < 0.001; ns, not significant; § p < 0.05 versus EtOH alone).

Fig. 3

Effects of ethanol (EtOH) withdrawal on cholinergic nbM neurons. Brain slices were cultured for 2 weeks with nerve growth factor (NGF) and then exposed to medium without NGF or with 50 mM EtOH for 7 days. Some slices were further treated for 7 days with or without NGF and with or without 50 mM EtOH or with NGF and 50 mM EtOH. Slices were fixed and immunohistochemically stained for the enzyme choline-acetyltransferase (ChAT). The number of cholinergic neurons was not different when incubated for 2 weeks with NGF or for 2 weeks without NGF. EtOH treatment resulted in a significant decrease in ChAT+ neurons after 7 days and more pronounced after 14 days. When EtOH was withdrawn, the number of cholinergic neurons nearly recovered to control levels, but NGF did not counteract the EtOH effect, when given together with EtOH. Values are given as mean ± SEM number of ChAT-positive neurons per slice. Values in parenthesis give the number of analyzed slices. Statistical analysis was performed by one way ANOVA followed by Fisher LSD posthoc test (***p < 0.001).

Effects of NGF on EtOH-induced decline of cholinergic neurons

When 2-week old slices were incubated with EtOH and NGF, NGF counteracted the EtOH-induced effect at 100 mM (Fig. 2A), but not at 50 mM EtOH (Fig. 2A). When 2-week old slices were incubated with 50 mM EtOH for 1 week and then for further 1 week with or without NGF, the number of ChAT+ neurons returned to nearly control levels (Fig. 3). However, when 2-week old slices were incubated with 50 mM EtOH for further 7 to 14 days with NGF and EtOH, the number of cholinergic neurons did not change (Fig. 3).

Effects of inhibitors of MAPK p38 and NOS on cholinergic neurons

In order to test underlying pathway mechanisms of EtOH-induced effects, slices were treated with NOS and MAPK p38 inhibitors. Treatment of slices with 50 mM EtOH together with SB203580 (MAPK p38 inhibitor) for 7 days counteracted the EtOH-induced decline (Fig. 2B). Incubation of slices with SB203580 alone did not have any effects on ChAT+ neurons (Fig. 2B). Treatment of slices with 50 mM EtOH together with L-thiocitrulline (NOS inhibitor) for 7 days counteracted the EtOH-induced decline (Fig. 2B). Incubation of slices with L-thiocitrulline alone did not show any effects (Fig. 2B).

Effects of EtOH on inflammatory markers

Inflammatory markers were measured in slices by multiplex ELISAs and the levels of control slices were 45 ± 18 pg/mg (macrophage-inflammatory protein-2, MIP-2), 18 ± 4 pg/mg (tumor necrosis factor-alpha, TNF-α), 171 ± 34 pg/mg (macrophage-chemotactic protein-1, MCP-1), 42 ± 15 pg/mg (interleukin-1beta, IL-1β) and 1421 ± 593 pg/mg (matrix-metalloproteinase-2, MMP-2) (all, n = 6). EtOH (50 mM, 7 days) did not affect the levels of all measured markers.

Discussion

The present study shows that EtOH induces a decline of cholinergic neurons, most likely by involving MAPK p38 and NO pathways. The decrease of cholinergic neurons in nbM slices may be caused by a transient downregulation of the enzyme ChAT pointing rather to a form of adaptive plasticity in response to the EtOH treatment, than to neurodegeneration.

The organotypic brain slice model is well-established in our working group and serves as a validated tool to study toxic, degenerative and developmental changes as well as synaptic recovery, survival and cell death of neurons (Gähwiler et al., 1997; Humpel and Weis, 2002; Moser et al., 2003, 2006; Stoppini et al., 1991; Weis et al., 2001; Zassler et al., 2003). In this model cholinergic neurons are axotomized, however, the normal cytoarchitecture is retained similar to the in vivo situation and functional connections including transport and diffusion probabilities are maintained. The brain tissue is derived from postnatal day 10 brains and therefore it is not completely comparable to adult brains, which is a limitation of the present study. In further studies it would be of particular interest to investigate the effects of EtOH in adult nbM slices and thus to compare with the neuropathological changes in adult brains. In fact, culturing of adult brain slices has been reported (Bickler et al., 2010; Hassen et al., 2004; Xiang et al., 2000), although this technique is not trivial and such slices are not easy to culture for long time.

Adolescent brains distinctively response to EtOH exposure compared to adult brains (Smith, 2003) and the context of ongoing plasticity in the adolescence faces the continuous production of new neurons during adult neurogenesis (Nixon et al., 2010). Indeed, adolescents are more prone to the neurotoxic effects of EtOH than adults (Crews et al., 2007). In the present model brain slices are normally cultured for 2 weeks before staining in experiments correlating to adolescent age. In fact the basal forebrain cholinergic neurogenesis is already completed before birth (E17) (Semba and Fibiger, 1988). In the present study detection of cholinergic neurons was performed using the immunohistochemical marker for the enzyme ChAT, which is expressed in cell bodies and nerve fibers of cholinergic neurons. In our experiments control slices displayed around 120 ChAT-positive neurons, which is in line with previous work (Weis et al., 2001). ChAT serves as a marker for the functional activity of cholinergic neurons (Oda, 1999) and a decreased number directly correlates with cognitive impairment (Counts and Mufson, 2005). Indeed, a dysfunction of the cholinergic system and the loss of cholinergic neurons is in concert with low levels of acetylcholine in the cortex and resulted in cognitive impairment (Mesulam, 2010). Interestingly, an impairment of the cholinergic system (Floyd et al., 1997) and cognitive decline has also been reported after long-term EtOH treatment in vivo (Arendt et al., 1988; Ehrlich et al., 2012). Accordingly, the activity of cholinergic neurons after EtOH exposure possibly represents a depression of the enzyme ChAT and not cell death. In fact in the present in vitro study we show that cholinergic neurons in adolescent nbM slices reverse to nearly control levels after withdrawal of EtOH suggesting reversible effects of EtOH. This is also in agreement with Arendt et al. (1989) showing that in rats treated for 8 weeks with EtOH a full recovery of declined ChAT activity in the basal forebrain is seen after 4 weeks of EtOH withdrawal. The effects of EtOH on the cholinergic system may reflect a form of adaptive plasticity, rather than neurodegeneration after EtOH exposure in our brain slice model.

In the present study we used EtOH concentrations from 1 mM (5 mg/dl) up to 100 mM (500 mg/dl). EtOH concentrations between 50 mM (250 mg/dl) to 70 mM (350 mg/dl) are of particular interest, because these levels has been reported in alcohol dependent adults as well as adolescent humans (Deas et al., 2000; Jones and Holmgren, 2009) and were also used in several in vitro studies (Cheema et al., 2000; Mooney and Miller, 2003; Zou and Crews, 2010). The sensitivity of the cholinergic system to EtOH has been reported in previous in vivo studies (Arendt et al., 1988, 1995; Floyd et al., 1997). In rats prolonged intake of EtOH resulted in a neurotoxic effect on the basal forebrain cholinergic projection system (Floyd et al., 1997) and leads to a partial cholinergic denervation of the cortex, hippocampus and amygdala (Arendt et al., 1988). Beside the cholinergic system, EtOH affects also other brain areas and EtOH-induced apoptotic cell death in the developing cortex has been observed in organotypic cultures (Mooney and Miller, 2003). In the present study the most prominent decrease of cholinergic neurons of approximately 60% of total neurons was found after treatment with 50 mM (250 mg/dl) EtOH, but not at higher concentrations. This surprising finding is consistent with data reported by Cheema et al. (2000), who showed that cell death was enhanced in cultures after treatment with 64 mM (320 mg/dl) EtOH but not at 190 mM (950 mg/dl). Others reported that 80 mM (400 mg/dl) EtOH increased cell death in the cortical plate in cultures of rat cerebral cortex, whereas the highest amounts of 160 mM (800 mg/dl) had no effect (Mooney and Miller, 2003). Interestingly, also in human astroglia cells EtOH displayed a biphasic effect: 50 mM EtOH stimulated while 200 mM EtOH inhibited a cytokine-induced iNOS activity (Davis et al., 2002). In our present study we suggest that a saturating effect of < 50 mM EtOH may stimulate a specific single pathway, while at EtOH levels > 50 mM a second, independent and protective pathway may become activated.

In the brain, nerve growth factor (NGF) serves as the most potent trophic substance to support survival of cholinergic neurons (Humpel and Weis, 2002; Levi-Montalcini et al., 1996). In vivo studies showed an increase of NGF mRNA levels in a number of brain areas, including the basal forebrain and their cortical target areas, after chronic EtOH treatment in rats (Arendt et al., 1995). In addition, it has been reported that NGF is able to counteract EtOH-induced deficits in the cholinergic system (Aloe and Tirassa, 1992; Lukoyanov et al., 2003; Paula-Barbosa et al., 2001), as well as in cultured cortical (Mooney and Miller, 2007), hippocampal (Webb et al., 1997) and cerebellar neurons (Luo et al., 1997). In our slice model cholinergic neurons were cultivated for two weeks with NGF from beginning resulting in around 120 detectable ChAT+ neurons. This number did not change, when slices were cultured for further 2 weeks without NGF. We have well established that cholinergic neurons survive at least for 2 weeks without NGF, but not longer (Weis et al., 2001). In the present study only the EtOH-induced effect was counteracted by NGF at 100 mM, but not at 50 mM effect. This again may point to a second independent (possibly neuroprotective) intracellular pathway, which is only activated at higher EtOH concentrations.

In order to investigate intracellular pathways of EtOH-induced effects on cholinergic neurons, we investigated two well established pathways. (1) The MAPK pathway may play an important role in EtOH-induced neurotoxicity. EtOH induces oxidative stress, which further has been shown to activate all three MAPK cascades, p42/44, JNK/SAPK and the MAPK p38 (Owuor and Kong, 2002). The role of MAPK p38 is divergent, because MAPK p38 pathways may be involved in anti-apoptotic processes (Roberts et al., 2000), but may also increase the vulnerability to cell death (Aroor and Shukla, 2004). It has been shown that MAPK p38 cascades may be responsible for EtOH-induced cell cycle arrest and inhibition (Koteish et al., 2002). Interestingly, in the present study the treatment with a MAPK p38 inhibitor counteracted the EtOH-induced decline of cholinergic neurons. (2) EtOH is able to activate free radical generating enzymes, such as NAPDH oxidase and iNOS, may induce reactive oxygen species (Alikunju et al., 2011) and modulates NO activity by inducing oxidative stress. EtOH directly alters NOS expression and activity in the brain (Davis and Syapin, 2005; Syapin, 1998) causing blood pressure elevation and regional blood flow reduction (Toda and Ayajiki, 2010). Inhibition of NO has been suggested as a possible treatment against EtOH-induced excitotoxicity and addiction (Lancaster, 1995). However, there is strong indication that NO is not involved in EtOH-associated brain damage (Vassiljev et al., 1998; Zou et al., 1996). In the present study the EtOH-induced decline of cholinergic neurons in the nbM was counteracted by inhibition of NOS activity suggesting that the NO cascade is involved in EtOH-mediated in vitro effects. However, in vivo NO may induce some additional protective pathways. Unfortunately, a shortcoming in our slice model is the lack of functional vascularization to study aspects of NO-mediated vasodilatation. Further investigations are necessary to explore if NO has a direct effect against the decline of neurons in the organotypic slice model and if this is comparable to the in vivo situation.

In our slices treatment with EtOH did not result in enhanced cytokine production. It seems likely that this treatment was not strong enough to induce an inflammatory cascade in the nbM. Indeed, EtOH-induced inflammation in humans has been shown after chronic alcoholism and is not a short time effect. In addition, cytokines found in the brains of individuals after heavy EtOH consume are originally produced by the liver cells (Crews and Nixon, 2009). Thus, any lack of direct EtOH on inflammation is in line with such a peripheral inflammatory process. In hippocampal–entorhinal brain slice cultures EtOH induced inflammatory gene expression (Zou and Crews, 2010), suggesting that this region may be more sensitive to EtOH-induced cytokine upregulation than the nbM. Further studies are necessary to investigate if the lack of inflammation in our slice model is area-related or a methodological limitation.

Taken together, our data show that EtOH-induced a decline of cholinergic neurons in vitro, which was partly counteracted by NGF. Inhibition of MAPK p38 and NOS ameliorated the EtOH effect suggesting a role in the underlying mechanism of EtOH-mediated effects in vitro. In conclusion, the data may suggest that direct EtOH exposure to cholinergic nbM neurons may transiently suppress the enzyme ChAT and may not induce cell death of cholinergic neurons, but rather may reflect a form of neuronal plasticity in response to EtOH.

Experimental procedures

Organotypic brain slices

Cholinergic neurons in organotypic brain slices were cultured, as described in detail previously (Humpel and Weis, 2002; Weis et al., 2001). Briefly, the basal nucleus of Meynert (nbM) of postnatal day 10 (P10) rats was dissected under aseptic conditions. Further, 400 μm slices were cut with a tissue chopper (McIlwain, USA) and placed on 30 mm Millicell-CM 0.4 μm pore membrane culture plate inserts (6–8 slices per membrane). It needs to be pointed out that a single experiment included approximately 8–12 pups. In one dissection experiment 4 pups were dissected and all brain slices were randomly distributed on all 6-wells. An experiment was normally repeated 3 times on different groups, so that a single treatment contained at least slices from 9 different rat pups. Slices were cultured in 6-well plates at 37 °C and 5% CO2 with 1.2 ml/well of slice medium (50% MEM/HEPES (Gibco), 25% heat inactivated horse serum (Gibco/Lifetech, Austria), 25% Hanks' solution (Gibco), 2 mM NaHCO3 (Merck, Austria), 6.5 mg/ml glucose (Merck), 2 mM glutamine (Merck), pH 7.2) including 10 ng/ml nerve growth factor (NGF) for 2 weeks. It is well established that the 400 μm brain slices become thinner during the 2 weeks of incubation resulting in a thickness of approx. 100 μm, which is a sign of healthy cultures. Slices, which did not flatten were immediately removed from the experiments. For all experiments the 2-week old slices were cultured for 3 days without NGF and then transferred to different media with 1 to 100 mM EtOH. In some experiments 10 μM L-thiocitrulline (nitric oxide synthase inhibitor) or 10 μM SB203580 (MAPK p38 inhibitor) were added for further 7 or 14 days, respectively.

Immunohistochemistry

Immunohistochemistry using the avidin–biotin technique was performed to detect cholinergic neurons, as described previously (Zassler and Humpel, 2006). All incubations were processed free-floating at 4 °C for 2 days including 0.1% Triton, allowing good penetration of the antibody into the slices from both sides. Fixed slices were rinsed 30 min with 0.1% Triton/PBS (T-PBS) at room temperature and pre-treated for 20 min with 20% methanol/1% H2O2/PBS. Then the slices were washed three times for 10 min with PBS, blocked with 20% horse serum/0.2% BSA/T-PBS and then incubated with the primary antibody against goat anti-choline‐acetyltransferase (1:750, Millipore, USA) in 0.2% BSA/T-PBS for 2 days at 4 °C. Slices were washed and incubated with secondary biotinylated anti-goat antibody (1:200, Vector Laboratories), for 1 h at room temperature. After rinsing three times in PBS, slices were incubated in avidin–biotin complex solution (ABC; Elite Standard PK 6100, Vector Laboratories) for 1 h, then washed three times in 50 mM Tris-buffered saline (TBS) and the signal was detected using 0.5 mg/ml 3,3′diaminobenzidine (DAB) in TBS with 0.003% H2O2 as substrate. Slices were then rinsed in PBS and mounted on glass slides.

ELISA (MIP-2, TNF-α, MCP-1, IL-1β and MMP-2)

Slices were extracted in 100 μl ice-cold sodium-phosphate buffer (PBS) with protease-inhibitor cocktail (Sigma, Germany) using an ultrasonic device (Branson sonifier 250) and centrifuged at 16 000 ×g for 10 min at 4 °C. Inflammatory markers and MMP-2 were analyzed in slice extracts using a rat Multiplex ELISA (SearchLight®; Aushon Biosystems), as described previously (Marksteiner et al., 2011; Pirchl et al., 2010).

Quantitative analysis and statistics

All neuronal counts were based on individual sections and show total number of neurons per slices. The number of microscopically detectable immunoreactive ChAT+ neurons was counted in the whole slice visualized under a 40× objective by an investigator blinded to the treatment code. Quantitative data are presented as mean ± SEM. Multistatistical analysis was obtained by one way ANOVA, followed by a subsequent Fisher PLSD posthoc test by comparing controls against the respective treatments, where p < 0.05 represents statistical significance.