Address correspondence and reprint requests to Fulton T. Crews, Director, Bowles Center for Alcohol Studies, University of North Carolina, Thurston-Bowles Bldg. CB #7178, Chapel Hill, NC 27599, USA. E-mail: FTCrews@med.unc.edu
Alcoholism is associated with cognitive deficits and loss of brain mass. Recent studies have indicated that neural progenitor cells proliferate throughout life forming neurons, astrocytes, and oligodendrocytes. The dentate gyrus is one neurogenic region of the adult brain containing neural progenitor cells. To determine if binge ethanol (EtOH) exposure alters neural progenitor cell proliferation and survival, bromodeoxyuridine was administered to adult male rats following an acute or chronic binge exposure paradigm. For an acute binge, rats were gavaged with a 5 g/kg dose of EtOH or vehicle, administered bromodeoxyuridine, and killed either 5 h or 28 days after EtOH treatment. In a 4-day, chronic-binge paradigm, rats were infused with EtOH three times per day (mean dose 9.3 g/kg/day) or isocaloric control diet. Rats were given bromodeoxyuridine once a day for the 4 days of chronic binge treatment, then perfused either immediately following the last dose of EtOH or 28 days later. In both EtOH treatment groups, binge EtOH decreased neural progenitor cell proliferation. Following the chronic four-day binge, neural progenitor cell survival was decreased. These studies are the first to show EtOH inhibition of neural progenitor cell proliferation and survival in the adult, a possible new mechanism underlying alcoholic cognitive dysfunction.
EtOH is known to disrupt neurogenesis during development. The effects of EtOH on fetal neurogenesis depend upon the stage of development, cell type, dose, and duration of EtOH exposure (Miller 1986; Bonthius and West 1990; Marcussen et al. 1994; Goodlett and Johnson 1999; Dunty et al. 2001). EtOH has two major effects on fetal development: altered proliferation and enhanced apoptotic death (Dunty et al. 2001; Jacobs and Miller 2001), although different neurogenic cell groups have different EtOH responses. EtOH treatment during postnatal days 4–12 increases DG neurons at moderate blood EtOH concentrations but at high blood EtOH levels inhibits cell proliferation (Miller 1995). The developmental effects of EtOH appear to be a balance between promotion of cell death and inhibition of cell proliferation (Jacobs and Miller 2001). Binge drinking during temporal windows of vulnerability is known to have prominent effects on fetal neurogenesis (Goodlett and Johnson 1999; Goodlett and Horn 2001). This manuscript investigates the effects of acute and chronic models of binge drinking on adult NPC proliferation and survival.
Adult neurogenesis in the hippocampus occurs in the subgranular cell layer of the DG. New cells migrate into the DG granule cell layer, differentiate, send out axons and dendrites similar to dentate granule cells, express neuron-specific markers, and connect electrically to appropriate afferent and efferent neuronal populations (Gage and Song 2002; Gould and Gross 2002; van Praag et al. 2002). Recent studies indicate that in adult rats, neurogenesis in the DG generates approximately 6% of the total granule cell population every month (Cameron and McKay 2001). This number of new cells could contribute in significant and important ways to hippocampal function (van Praag et al. 2002). We report here that models of binge drinking reduce NPC formation in hippocampus. EtOH-induced disruption of adult neurogenesis could contribute to the neuropathology and cognitive deficits associated with alcoholism.
Materials and methods
Fifty-three, adult male rats (Sprague–Dawley: 275–330 g, Charles River, Raleigh, NC, USA) were used. Rats were maintained on a 12-h light–dark cycle and had ad libitum access to rat chow and water except where noted. All protocols followed NIH guidelines and were approved by the University of North Carolina Institutional Animal Care and Use Committee.
Acute EtOH binge
Rats were food deprived for 12 h and gavaged with EtOH (5 g/kg, 30% w/v in saline) or vehicle. Bromodeoxyuridine (BrdU, Sigma, St Louis, MO, USA), a marker of cell division, was dissolved in saline and administered (100 mg/kg, intraperitoneally) 45 min following EtOH administration. A second dose was administered at 2 h and 45 min following EtOH administration. Rats were killed either 5 h following the EtOH dose (cell proliferation) or 28 days later (cell survival). Blood EtOH concentrations (BEC) were determined from tail blood samples drawn at four time points following EtOH administration (30, 60, 90 and 120 min) using a separate group of rats (n = 8). Serum was extracted and BECs were measured by a GM7 Analyser (Analox, London, UK). Three rats were dropped for tissue-related problems.
Chronic EtOH binge
Chronic EtOH binge treatment was administered via intragastric catheter as described elsewhere (Majchrowicz 1975; Collins et al. 1996; Knapp and Crews 1999). Briefly, rats were administered either a 25% (w/v) EtOH solution in a nutritionally complete diet (Vanilla Ensure®) or diet plus dextrose (isocaloric to EtOH diet) every 8 h for 4 days. The first EtOH dose was 5 g/kg with subsequent doses determined using a six-point behavior scale, modified from Majchrowicz (1975) to maintain consistent, intoxicating blood EtOH levels while preventing mortality (Collins et al. 1996; Knapp and Crews 1999). Briefly, the rats were scored according to the following behaviors: normal rat, 0; hypoactive, 1; ataxia, 2; ataxia with dragging abdomen, delayed righting reflex 3; loss of righting reflex, 4; and loss of righting reflex and loss of eye-blink reflex, 5. Each score was then associated with a particular dose of EtOH (0–5 g/kg). One EtOH rat died and one control was dropped due to catheter problems. BEC was determined from tail blood extracted on days 3 and 4, 90 min after the 3.00 pm infusion of EtOH. An additional group of three rats remained untreated except for BrdU administration (‘unhandled controls’). BrdU was administered once each day (50 mg/kg, intraperitoneally) during the 4-day binge EtOH treatment. Rats were killed either immediately following the last dose of EtOH (cell proliferation) or 28 days later (cell survival).
All rats were killed by an overdose of sodium pentobarbital (Nembutal®, 100 mg/kg) and transcardially perfused with 0.1 m phosphate-buffered saline (pH 7.4) followed by 4% paraformaldehyde in 0.1 m phosphate buffer (pH 7.4). Extracted whole brains were postfixed in paraformaldehyde then transferred to Tris-buffered saline (TBS; pH 7.4) until sectioning. Coronal sections were cut 40-µm thick using a vibratome (Leica) and stored in cryoprotectant at −20°C. Every sixth section was kept for BrdU processing such that each section was 240 µm apart. BrdU immunohistochemistry followed the methods of Kuhn et al. (1997) with an additional denaturing step (Dolbeare 1995). In brief, free-floating sections were treated with H2O2 to block endogenous peroxidases. DNA was denatured and sections blocked in TBS + (TBS, 3% normal horse serum, 0.1% Triton X) for 30 min followed by washes in MgCl2 and incubation in DNase at room temperature for 1 h. After a 5-min rinse in TBS+, sections were incubated overnight in anti-mouse BrdU (Chemicon MAB3424) diluted 1 : 400 in TBS+. Sections were rinsed well in TBS+, incubated for 1 h in biotinylated horse anti-mouse, rat adsorbed, secondary antibody, and washed three times in TBS. Avidin–biotin–peroxidase complex (ABC Elite Kit, Vector Laboratories, Burlingame, CA, USA) was applied for 1 h and was detected with nickel-enhanced diaminobenzidine as a chromagen. Sections were mounted, dried overnight, counterstained with cresyl violet, dehydrated in a series of alcohols, cleared in xylene, and coverslipped with Cytoseal® (Stephens Scientific, Kalamazoo, MI, USA). Both positive and negative controls were run to verify specificity of BrdU staining. Negative controls consisted of (i) brain tissue from animals with no BrdU treatment and (ii) BrdU-treated brain sections with the omission of primary antibody. In both cases, no BrdU staining was visible. Positive control brain tissue (kainate-induced seizures followed by BrdU injections, which produces vast labeling) was run in the same well as several EtOH-treated sections to confirm staining in each assay.
Quantification and analysis
Brains were coded so that the experimenter was blind to the experimental condition during counting. The number of BrdU-positive (BrdU+) cells in the dorsal DG and subgranular zone (Bregma −3.14 to − 4.52) was estimated using unbiased stereological methods (Sterio 1984; West 1991) with the exception of a smaller x,y step and larger counting frame. These parameters were optimized for BrdU label counting in pilot work with control sections. These changes improved the coefficients of error (mean of 0.12 ± 0.003), as BrdU-stained cells are not a homogeneously distributed population. Systematic random sampling was used at all levels of the design: (i) a one in six series of sections was collected with a random starting point at the level of the hippocampal commissure; (ii) application of the x,y grid was placed randomly on the section by software (CAST® stereology system, Olympus, Denmark), and (iii) a 10-µm thickness was used for counting in the z-axis. Thus, the section sampling fraction was 1/6 and the area sampling fraction was 50%. The thickness sampling fraction was the ratio of the height of the disector, 10 µm, to the mean thickness of the sections, which averaged 15.5 µm. The mean thickness of the section was determined at three locations within the region of interest by measuring the distance between the upper and lower focal planes with the microcator. The counting frame and 100 µm grid were superimposed over live video images by the software. After tracing the DG and subgranular zone under a 10 × objective lens, cells were counted under a 60 × oil immersion objective (Olympus Plan Apo oil, numerical aperture = 1.4).
In the day 28 groups, cell survival was calculated by dividing the total number of BrdU+ cells by the day 0 values (Kempermann et al. 1998a). This calculation allows for the analysis of survival independent of effects on proliferation. Analyses on cell counts and survival were performed using single factor anova. For simplicity, behavioral scores were blocked into groups of days then analyzed using Friedman's two-way anova for ranks with Dunn's post-hoc test.
To investigate the effects of an acute EtOH binge on proliferation and survival of NPCs, rats were administered a 5 g/kg dose of EtOH and killed 5 h after EtOH treatment or 28 days later. This dose of EtOH produced BECs as shown in Table 1. BrdU+ cells were seen in all sections counted. Data are reported as estimated total number of BrdU+ cells for the dorsal DG as described in the methods. Control values at both time points investigated are similar to those reported by others (Kempermann et al. 1998b; Gould et al. 1999; Cameron and McKay 2001). Five hours following EtOH administration, BrdU+ cells were clustered primarily in the DG subgranular zone (Fig. 1). At this time, the number of BrdU+ cells was decreased by 40% (F1,10 = 13.630, p < 0.01) in EtOH-treated rats (n = 6) compared with saline controls (n = 5; Fig. 2). Twenty-eight days following acute binge exposure, the majority of BrdU+ cells are present in the granule cell layer and appear as individual cells. Other studies have shown that the number of BrdU+ cells decreases during this migration and differentiation period though neither apoptotic nor necrotic cells are found in controls (Gage 2000; Kempermann 2002). Twenty-eight days following acute binge exposure, BrdU+ cells were markedly reduced for both control and acute EtOH groups (n = 5 both groups; Fig. 3). The EtOH group was approximately half that of the control group, but was not statistically different. Thus, acute EtOH reduced the number of BrdU+ cells during intoxication.
Table 1. A time course of blood EtOH concentrations shown as mean ± SEM following an acute binge (5 g/kg)
Blood EtOH concentration (mg/dL)
BrdU was administered once the animals were intoxicated at 45 min and 165 min following EtOH exposure.
156.2 ± 12.9
203.2 ± 28.9
233.2 ± 28.3
195.4 ± 25.3
To investigate the effects of EtOH tolerance and dependence on NPC proliferation and survival, rats were administered EtOH using a 4-day chronic binge model. The average dose of EtOH was 9.3 g/kg/day, a dose similar to what rodents will self-administer (Lukoyanov et al. 2000). Rats reached high levels of intoxication and became tolerant as indicated by decreasing behavioral scores (Fig. 4). For example, the average behavioral score on day 2 was 2.8 ± 0.2 compared with 1.8 ± 0.2 [Friedman's χ2 (11) = 37.39, p < 0.0001; Dunn's post-hoc test, p < 0.05] on day 4 though blood levels were high (Fig. 4).
Rats were killed immediately after the last EtOH dose of the 4-day binge or 28 days later. Following 4 days of chronic binge EtOH treatment, BrdU+ cells were observed in the subgranular zone, though clusters were rare (Fig. 1). The ‘unhandled’ controls were similar to diet controls and were combined in the analysis, indicating that the stress of the intragastric administration procedure was not a factor in this study. After the 4-day chronic binge, there were 47% fewer BrdU+ cells in EtOH-treated animals (n = 5) than controls (n = 5; F1,8 = 19.290, p < 0.01; Fig. 3). One month following the chronic binge treatment, shown in Fig. 3, the EtOH group (n = 4) had a decreased BrdU+ cell survival of 3.9 ± 2% compared with 34.6 ± 9% in controls (n = 5; F1,8 = 8.730, p < 0.05). In summary, chronic binge EtOH treatment decreased both NPC proliferation and survival.
This is the first report of EtOH inhibition of NPC proliferation in adult rat hippocampus. Many studies have found EtOH to inhibit fetal neurogenesis depending on cell type, timing, dose, and duration of EtOH exposure (Miller 1986; Bonthius and West 1990; Marcussen et al. 1994; Goodlett and Johnson 1999; Dunty et al. 2001). Neurodevelopmental studies of the teratogenic actions of EtOH indicate that EtOH can retard cell proliferation and increase cell death particularly through apoptosis (Miller 1995; Jacobs and Miller 2001). We found that acute EtOH decreased the number of BrdU+ cells consistent with decreased NPC proliferation in adult DG. However, unlike developmental studies, acute and chronic binge EtOH do not increase apoptosis in adult DG (Obernier et al. 2002a). BrdU is incorporated into DNA as the cell passes through the DNA synthesis (S) phase of the cell cycle and this is not inhibited by EtOH (Jacobs and Miller 2001). Although BrdU could be incorporated through DNA repair, this low level of incorporation is not detectable immunohistochemically (Palmer et al. 2000). Thus, a decrease in mitotic BrdU labeling indicates a decrease in NPC proliferation. Decreases in cell proliferation can be due to either (i) a slower cell cycle or (ii) a decrease in the number of cells actively cycling. In fetal neurogenesis, EtOH affects either mechanism depending on the cell population and timing of exposure (Miller and Nowakowski 1991; Miller 1995). We found that acute EtOH decreased adult hippocampal NPCs by 40% at 5 h. Hippocampal NPCs are in S phase for approximately 9.5 h with the entire cell cycle approximately 25 h in length (Cameron and McKay 2001). Our finding of decreased BrdU+ cells in adult hippocampus 5 h after an acute dose of EtOH is consistent with EtOH decreasing the number of cells in S phase. Additional studies will be needed to determine the detailed mechanisms of EtOH-induced inhibition of NPC proliferation.
Although adult neurogenesis has been reported in rodents, monkeys and humans, the number of new cells is relatively small compared with the total number of neurons in the hippocampus. However, NPC may have unique functions in the hippocampus (Kempermann 2002). Recent studies indicate that in adult rats, neurogenesis in the DG generates approximately 6% of the total granule cell population every month, i.e. about 138 000 new cells each month compared with approximately 2.4 million total DG granule cells (West 1991; Cameron and McKay 2001). It is not clear if these new cells are replacing old DG granule cells or adding new neurons. Some reports have indicated that the number of hippocampal granule cells appears to increase with age in the adult (Bayer et al. 1982; Wimer et al. 1988). These numbers of cells are large enough to contribute to the function of the DG. However, studies suggest that new granule cell neurons may act as a distinct functional population. New granule cells show increased long-term potentiation and reduced GABA-ergic inhibition when compared with mature granule cell neurons (Wang et al. 2000). Thus, the decrease in NPC proliferation by EtOH may disrupt hippocampal function.
In rodents, there is considerable evidence that NPCs play a unique role in associative learning tasks. Both genetic factors and environmental factors regulate NPC proliferation correlating with learning performance (Kempermann 2002). In rats, the cytostatic agent, methylazoxymethanol acetate, blocks NPC proliferation similar to what we report here for EtOH. These rats perform worse than controls on a hippocampus-dependent conditioning task, whereas hippocampus-independent task performance was not altered (Shors et al. 2001). In birds, cortical regions that control seasonal behavior and song correlate with neurogenesis suggesting that new neurons are related to changes in behavior (Nottebohm 2002). Taken together, these studies suggest that NPCs play a unique role in hippocampal function [see Kemperman (2002) for further discussions]. We have found that the 4-day chronic binge treatment, identical to that shown in this study to reduce NPC proliferation and survival, produces re-learning deficits on a hippocampus-dependent task (Obernier et al. 2002b). In adult rats, Walker et al. found that five months of chronic EtOH consumption decreases DG granule cell number approximately 20%, reduces dendritic spines and produces residual impairment on a variety of behavioral tasks (Walker et al. 1980). Thus, the effects of EtOH on hippocampal toxicity and inhibition of NPC proliferation in adult hippocampus may contribute to cognitive dysfunction in alcoholics.
Previous studies have found that 28 days after BrdU labeling, these cells have migrated and differentiated, such that they share morphology, electrophysiological properties, and antigens consistent with DG granule cells (van Praag et al. 2002). The number of BrdU+ cells is greatly reduced at 28 days in both groups due to cell loss and/or BrdU dilution. Four days of binge treatment, a model that produces EtOH tolerance and dependence, resulted in few surviving cells. NPCs form neurons, astrocytes, and/or oligodendrocytes and therefore could be related to both gray and white matter loss associated with alcoholic neurodegeneration. Decreased hippocampal volume has been found in both postmortem and in vivo brain imaging studies of alcoholics (Sullivan et al. 1995; Harding et al. 1997). Additionally, alcoholics drink progressively more as the disease advances. Multiple EtOH binges could lead to additional EtOH-induced reductions in NPC proliferation and survival, possibly contributing to hippocampal volume seen in alcoholics and the cognitive deficits indicative of hippocampal dysfunction (Eckardt and Martin 1986). Thus, EtOH inhibition of NPC proliferation could contribute to both the cognitive and neurodegenerative effects associated with chronic alcoholism.
The authors would like to acknowledge Jennifer A. Obernier and Silvia Bison for their outstanding technical assistance. This work was supported by National Institute of Alcohol Abuse and Alcoholism.