- Email: [email protected]
Neonatal alcohol exposure reduces NMDA induced Ca2+ signaling in developing cerebellar granule neurons
Brain Research 793 Ž1998. 12–20
Research report
Neonatal alcohol exposure reduces NMDA induced Ca2q signaling in developing cerebellar granule neurons D.L. Gruol ) , A.E. Ryabinin 1, K.L. Parsons, M. Cole, M.C. Wilson 2 , Z. Qiu Department of Neuropharmacology and Alcohol Research Center, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA Accepted 30 December 1997
Abstract Glutamatergic neurotransmission through NMDA receptors is critical for both neurogenesis and mature function of the central nervous system ŽCNS., and is thought to be one target for developmentally-induced damage by alcohol to brain function. In the current study we examined Ca2q signaling linked to NMDA receptor activation as a potential site for alcohol’s detrimental effects on the developing nervous system. We compared Ca2q signals to NMDA in granule neurons cultured from cerebella of rat neonates exposed to alcohol Žethanol. during development with responses to NMDA recorded in separated control groups. Alcohol exposure was by the vapor chamber method on postnatal days 4–7. An intermittent exposure paradigm was used where the pups were exposed to alcohol vapor for 2.5 hrday to produce peak BALs of ; 320 mg%. Control pups were placed in an alcohol-free chamber for a similar time period or remained with their mother. After culture under alcohol-free conditions for up to 9 days, Ca2q signaling in response to NMDA was measured using fura-2 Ca2q imaging. Results show that the peak amplitude of the Ca2q signal to NMDA was significantly smaller in cultured granule neurons obtained from alcohol-treated pups compared to granule neurons from control pups. In contrast, the Ca2q signal to Kq depolarization was not depressed by the alcohol treatment. Resting Ca2q levels were also altered by the alcohol treatment. These results show that intermittent alcohol exposure during development in vivo can induce long-term changes in CNS neurons that affect the Ca2q signaling pathway linked to NMDA receptors and resting Ca2q levels. Such changes could play an important role in the CNS dysfunction associated with alcohol exposure during CNS development. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Ethanol; CNS neuron; Development; Glutamate receptor; Intracellular calcium
1. Introduction Maternal alcohol consumption during pregnancy can significantly influence the developing infant w1,2x. The central nervous system ŽCNS. is particularly vulnerable to prenatal alcohol exposure, with some of the most serious effects being microcephaly, reduced mental capabilities, developmental delays and behavior problems w7x. These
) Corresponding author. Department of Neuropharmacology, CVN 11, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037. Fax: q1-619-784-7393; E-mail: [email protected] 1 Current address: Dept. Behavioral Neuroscience, L470, Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, Portland, OR 97201. 2 Current address: Department of Neurosciences, BMSB-145, 915 Camino del Salud, University of New Mexico, School of Medicine, Albuquerque, NM 87131-5221.
0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 0 0 1 4 - 6
effects of alcohol are permanent in nature and appear to involve direct actions of alcohol on CNS neurons. Studies in animal models of fetal alcohol exposure and in vitro model systems have shown that many properties of developing neurons are affected by alcohol exposure w17x. Neuronal functions related to synaptic transmission mediated by glutamate are particularly vulnerable. For example, in animal models prenatal alcohol exposure alters excitatory synaptic transmission and synaptic plasticity mediated by glutamate w13,37,38x, developmental expression of glutamate receptors w8,18,29,30,34x, and the response of CNS neurons to glutamate w24,26x. Altered responses to glutamate were also observed in CNS neurons chronically exposed to alcohol during development in vitro w43x. Glutamate is the main excitatory transmitter in the brain and an important trophic factor in the developing nervous system. Thus, the alcohol-induced alterations in glutamate-mediated functions are likely to contribute significantly to the abnormalities observed in animal models
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
of fetal alcohol exposure. Analogous disruptions may underlie the behavioral and cognitive problems observed in children exposed to alcohol during prenatal development. In the current study we examined the effect of intermittent alcohol exposure during development on an aspect of glutamatergic transmission known to play a critical role in CNS function, Ca2q signaling linked to NMDA receptor ŽNMDAR. activation. The NMDA subtype of glutamate receptor is permeable to Ca2q and Ca2q signaling via this receptor plays a pivotal role in synaptic plasticity w5x, neuronal developmental w14,19x, gene expression w4x, and neurotoxicity w6x. Thus, alcohol-induced alterations in Ca2q signaling linked to NMDAR activation could disrupt many aspects of CNS neuronal function, and thereby play a pivotal role in the CNS damage caused in infants by maternal alcohol abuse. Studies in animal models have shown that dose, timing and duration of alcohol exposure are important determining factors in alcohol’s detrimental effects on the developing CNS. For example, in the rat cerebellum where development occurs postnatally, the early postnatal period shows the greatest sensitivity to the toxic effects of alcohol, and a relatively short exposure to high dose of alcohol is more toxic than a prolonged exposure to a low dose of alcohol w42x. In the current studies we focused on this early critical period of cerebellar development and a dose of alcohol that produced significant alterations in cerebellar structure and function in previous in vivo studies w11,21,22,32x. We examined the effect of alcohol on rat pups exposed to alcohol on postnatal ŽPN. days 4–7 and utilized an alcohol dose that produced BALs of ; 320 mg% Ž70 mM. in the pups. Alcohol’s effects were assessed in cerebellar granule neurons, a CNS neuronal type known to express NMDARs during development in vivo and to be highly sensitive to alcohol exposure during the postnatal developmental period w23,27,35x. This developmental period in the rat corresponds to the third trimester of fetal development in humans. Recently we reported that chronic exposure of cultured granule neurons to alcohol Ž33 mM. during development in culture significantly reduced the Ca2q signals to NMDA w12x. While these studies support the proposal that altered Ca2q signals to NMDAR activation are an important component of alcohol’s effects on the developing nervous system, results from cultured neurons may not reflect fully the changes induced when alcohol exposure occurs in vivo, as the cultured neurons are isolated from the rest of the nervous system and from hormonal influences that could be contributing factors to alcohol’s actions. Moreover, in the experiments with cultured neurons alcohol exposure was continuous, an exposure paradigm that does not reflect the intermittent drinking pattern typical of human alcohol consumption. Thus, in the current study we have extended our analysis of alcohol’s effects on Ca2q signaling in developing granule neurons to an in vivo alcohol exposure model. In this model, alcohol is administered to postnatal
13
rat pups via an inhalation chamber for a fixed time period daily w32x. During the remainder of the day the pups remain with their mother. With this exposure paradigm, the blood alcohol level of the pups fluctuates during the day in a manner more reflective of that occurring with human consumption. To assess alcohol’s effects, it was necessary first to remove the neurons from the animal and dissociate them so that Ca2q imaging methodology could be applied. Because the dissociation procedure could alter neuronal properties, we examined NMDAR function after a period of recovery in culture. An added advantage of this approach was that alcohol effects identified after several days in culture were likely to reflect more permanent changes in neuronal function, since transient changes would presumably dissipate during the culture period. Results from these studies show that intermittent alcohol exposure during postnatal development in vivo reduces Ca2q signals linked to NMDAR activation in granule neurons. Small changes in resting Ca2q levels were observed as well. The effects of alcohol on Ca2q signaling to NMDA were selective in that Ca2q signaling to Kq depolarization was not affected similarly. These effects of alcohol appear to reflect relatively long-term changes in neuronal function since the neurons were studied after several days of alcohol free-conditions.
2. Materials and methods 2.1. Animals Timed-pregnant female rats from the Harlan Sprague Dawley laboratories were housed in Plexiglas cages Ž25 = 25 = 20 cm. with sawdust bedding and maintained at 228C " 0.48C on a 12:12 h lightrdark schedule Žlights on at 6:00 a.m.. with food and water continuously available. On postnatal day 2 ŽPN 2; where the day of birth is PN 0. pups were weighed, sexed and randomly assigned to foster mothers keeping the ratio of genders equal when possible. Alcohol exposure started on PN 4. To ensure that alcohol-exposed and control pups reached comparable body weights, litter sizes of control groups were kept at 9–11rlitter, whereas litter sizes of alcoholexposed groups were kept at 6–8 pupsrlitter starting the first day of alcohol-exposure. This reduction in litter size in the alcohol-treated group did not affect brain development w32x. The smaller litter size decreased competition for food in the alcohol-exposed litters and minimized possible nutritional differences between experimental and control animals. Two control groups were used in these studies: Ža. Separated—pups separated from their mother and placed in an alcohol-free chamber for the same time as those exposed to alcohol, and Žb. NaiÕe—normally raised control pups.
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
14
2.2. Alcohol Õapor inhalation Alcohol vapor inhalation was performed as described previously w32x. Ethanol was used for all studies. Briefly, pups were separated from their foster mothers and placed into the alcohol chamber nestled together as a litter in a 30 = 18 = 12 cm cage lined with sawdust bedding. The calculated alcohol concentration in the vapors of the chamber was 25 mgrl, and the exposure lasted for 2.5 h. The method for calculating the chamber alcohol level is described in detail in Rogers et al. w31x. After return to the housing cage, all of the alcohol-exposed pups were retrieved by their mothers and appeared to be suckling within 1r2 h. Alcohol exposure was performed during PN 4–7. Three to four pups closely matched in size and gender were sacrificed on PN 8 for tissue culture. In one experiment pups were sacrificed on PN 10 to access the extend of alcohol-induced microcephaly. Blood alcohol levels ŽBALs. were determined using the NAD-ADH kit ŽSigma.. BALs in trunk blood were measured in PN 4 pups sacrificed immediately after the alcohol exposure. Tail bleeding for BALs was not used on experimental animals in order to avoid the confounding effects of stress on pup development. 2.3. Cell culture Cerebellar granule cells were isolated from control and alcohol-treated pups at PN 8 by a standard enzyme treatment protocol w28x. Briefly, cerebella were dissociated in Ca2q- and Mg 2q-free saline with trypsin and DNAase and plated on Matrigel ŽCollaborative Biomedical Products. coated coverglasses Ž; 10 6 cellrml. in tissue culture dishes containing DMEMrF12 plus 10% horse serum and supplemented with 30 mM glucose, 2 mM glutamine, 20 mM KCl, and 25 m M penicillin–streptomycin. Average cell yield was similar for control and alcohol-treated groups Ž10 = 10 6 " 0.5 = 10 6 cellsrml Žmean " S.E.M.. for control and 9.4 = 10 6 " 0.7 = 10 6 cellsrml for alcoholtreated; 7 dissections for each condition; 3 pupsrdissection; each dissection represents a different litter.. The cultured neurons were maintained under alcohol-free conditions in a 5% CO 2 incubator at 378C for up to 12 days.
Medium without serum was added every 7 days. Treatment with FUDR Ž20 m grml. on the first and fourth days after plating minimized the number of non-neuronal cells in the cultures. 2.4. Intracellular Ca 2 q measurement Relative levels of intracellular Ca2q were determined for individual granule neurons using standard microscopic fura-2 digital imaging and previously published methods w25x. Granule neurons were loaded with 1.5 m M fura2rAM ŽMolecular Probes. and 0.02% pluronic F-127 ŽMolecular Probes. in physiological saline for 30 min. After removal of the fura-2 solution, the cultures were incubated in saline for an additional 45 min to allow for de-esterification of the fura-2 AM. Live video images of the microscopic fields illuminated at 340 and 380 nm wavelengths were recorded with a SIT-66 video camera ŽDAGE-MTI. and digitized by computer. Ratio images were formed by a pixel-by-pixel division of the 340 nm imagerthe 380 nm image. Real time digitized display, image acquisition, and Ca2q measurements were made with MCID imaging software ŽImaging Research.. Intracellular Ca2q levels were estimated by converting fluorescent images at 340 nm and 380 nm wavelengths to ratios and calculating intracellular Ca2q concentration using the following formula: wCa2q x i s K d Ž R y R min .rŽ R max y R . = ForFs , where R s the ratio value, R min s the ratio for a Ca2q free solution, R max s the ratio for a saturated Ca2q solution, K d s 135 Žthe dissociate constant for fura-2., Fo s the intensity of a Ca2q free solution at 380 nm and Fs s the intensity of a saturated Ca2q solution at 380 nm. Background fluorescence was very low making it unnecessary to use background subtraction. Calibration was done using fura salt Ž100 m M. in solutions of known Ca2q concentration. In situ calibration produced inconsistent results and thus was not used. Typical R max , R min , and ForFs values were 0.61, 2.85 and 2.5, respectively. Dye loading and all experiments were performed at room temperature Ž238C.. Several of the fields studied during Ca2q imaging experiments were also used to measured neuronal cluster size, which reflects neuronal survival in culture. Cluster
Table 1 Body and organ weights Žin grams. of pups exposed to alcohol by inhalation on postnatal days 4–7 Group
Body weight on PN 2; F Ž2,16. s 0.15, p ) 0.1
Body weight on PN 10; F Ž2,16. s 1.3, p ) 0.1
Brain weight on PN 10; F Ž2,16. s 7.83, p - 0.01
Liver weight on PN 10; F Ž2,16. s 1.39, p ) 0.1
Kidney weight on PN 10; F Ž2,16. s 0.33, p ) 0.1
Alcohol Ž n s 5. Separation Ž n s 5. Naive Ž n s 9.
11.8 " 0.48 11.8 " 0.32 11.6 " 0.26
25.6 " 0.80 26.8 " 0.75 26.72 " 0.33
0.89 " 0.017) 0.95 " 0.021 0.97 " 0.009
0.87 " 0.017 0.91 " 0.024 0.91 " 0.014
0.31 " 0.012 0.31 " 0.007 0.30 " 0.007
)Statistically significant from other groups. No statistical difference was found between male and females in any of the tests. Therefore, data on animals of both sexes were pulled together as one uniform group. n s Number of pups studied.
D.L. Gruol et al.r Brain Research 793 (1998) 12–20 Table 2 Cluster size in granule neuron cultures from control and alcohol-treated pups Age ŽDIV.
Control cultures; area Ž m m2 .
Alcohol cultures; area Ž m m2 .
5 6 7 9
1208"97 Ž47. 927"49 Ž21. 1560"104 Ž48. 1481"125 Ž27.
1302"87 Ž39. 1072"81 Ž11. 1419"102 Ž35. 1646"191 Ž17.
Numbers in parenthesis are the number of neuronal clusters measured; areas are mean"S.E.M.
size was defined as the two dimensional area encompassed by the cluster. Measurements were made with MCID imaging software ŽImaging Research.. 2.5. Drug application During experiments granule neurons were stimulated with NMDA Ž200 m M; dissolved in bath saline. a selective agonist at the NMDAR or Kq Ž150 mM; Kq substituted for Naq in physiological saline. a depolarizing agent. These stimulants were applied by a brief microperfusion
15
pulse Ž1 s in duration. from drug pipettes Ž; 3 m m tip. placed near a neuronal cluster. The bath was exchanged after each drug application. For experiments with Kq the bath contained physiological saline of the following composition ŽmM.: NaCl, 140; KCl, 3.5; KH 2 PO4 , 0.4; Na 2 HPO4, 0.33; CaCl 2 , 2.2; MgSO4 , 2; glucose, 10; HEPES–NaOH, 10 ŽpH 7.3.. A similar saline but with the omission of MgSO4 and the addition of 5 m M glycine was used for experiments with NMDA. 2.6. Data analysis Ca2q signals to NMDA and to Kq were quantified by measurement of the peak amplitude Žminus baseline resting Ca2q levels. in the somatic region. Results presented represent data collected from 5 litters; both control and alcohol-treated animals from each litter were used to prepare the respective cultures. Both naive Ž3 culture sets. and separated Ž2 culture sets. control animals were used to prepare cultures. Results were similar for the two control groups. Thus, no distinction was made between these two groups when combining data. On each experimental day, 2 or more cultures were measured for each experimental
Fig. 1. Phase contrast micrographs of cultured granule neurons from control and alcohol-treated pups. ŽA1, A2., granule neurons from control pups at 1 ŽA1. and 6 ŽA2. DIV. ŽB1, B2., granule neurons from alcohol-treated pups at 1 ŽB1. and 6 ŽB2. DIV. The same number of neurons were plated for each condition. Arrows in A1 and B1 point to granule neuron somata Žsmaller arrow. and processes Žlarger arrow.. Arrows in A1 and B1 point to granule neuron somata Žsmaller arrow. and processes Žlarger arrow.. Arrows in A2 and B2 point to a granule neuron cluster Žlarge arrow. and fiber tract Žsmall arrow.. Calibration bar is 20 m m.
16
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
isons. p - 0.05 was considered indicative of a significant difference. In all studies, mean values are expressed as mean " S.E.M. For Ca2q imaging studies, n is the number of neurons studied.
3. Results 3.1. Effects of alcohol treatment on the rat pups
Fig. 2. Effect of in vivo alcohol treatment on resting Ca2q levels in cultured granule neurons. Graph show resting Ca2q levels Žmean" S.E.M.; S.E.M.’s are smaller than data points., normalized to control levels, at each culture age tested. Resting Ca2q levels were significantly elevated in the alcohol-treated neurons compared to controls at 5, 6 and 7 DIV, but were reduced at 9 DIV. At 12 DIV there was no difference in resting Ca2q levels between control and alcohol-treated neurons. Resting Ca2q levels ranged from 15 nM–25 nM in control and alcohol-treated neurons at all ages. ) sSignificant difference Ž p- 0.5, ANOVA.. Numbers in parenthesis are the number of neurons measured Žalcohol culturesrcontrol cultures.. Cultures were obtained from 3 different litters, with measurements made on 2–3 different DIVs for each litter.
condition, with 4–6 microscopic fields containing approximately 10–20 neurons measured in each culture. Data from cultures of a similar age and treatment were pooled for statistical analyses. Statistical significance was determined by one-way Analysis of Variance ŽANOVA. followed by the Fisher post-hoc test for multiple compar-
BALs measured in a set of PN 4 old animals averaged 318 " 17 mg% Ž; 70 mM; 29 pups measured. immediately after alcohol exposure. Our previous studies indicated that these BALs return to baseline control levels 10–12 h after the alcohol exposure, and are sufficient to produce microcephaly and Purkinje cell loss on PN 9 and PN 10 if applied during PN 4–9 w32x. In the present experiments, for the purpose of successful generation of primary cell cultures from cerebellar neurons, animals were sacrificed on PN 8, and thus were exposed to alcohol only during PN 4–7. In order to evaluate whether this shorter alcohol exposure is sufficient to produce microcephaly, in one set of experiments the remaining pups were grown to PN 10, and then sacrificed to evaluate the organ and body weight. A comparison of brain weights revealed that alcohol exposure during PN 4–7 produced significantly smaller brains in alcohol-exposed pups in comparison to separated or naive controls ŽTable 1.. In contrast, body, liver and kidney weights between experimental and control animals were not statistically significant. No effect of pup gender on brain, liver or kidney weight was found in these experiments.
Fig. 3. Effect of in vivo alcohol treatment on the Ca2q signal to NMDA in cultured granule neurons. ŽA. Representative Ca2q signal to NMDA Žapplied at the arrow. in a control and alcohol-treated granule neuron Ž6 DIV.. The Ca2q signals were similar in general form in control and alcohol-treated neurons and consisted of a relatively rapid initial increase in intracellular Ca2q and slower recovery phase. The Ca2q signals were smaller in amplitude in the alcohol-treated neurons compared to control neurons. ŽB. Mean values Ž"S.E.M.. for the peak amplitude Žresting levels subtracted. of the Ca2q signal to NMDA at the three culture ages studied. The Ca2q signal to NMDA was smaller in the alcohol-treated neurons compared to controls at all three culture ages. ) s Significant difference Ž p - 0.5, ANOVA.. Numbers in parenthesis are the number of neurons measured. Cultures were obtained from 3 different litters, with measurements made on 2–3 different DIVs for each litter.
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
17
Fig. 4. Effect of in vivo alcohol treatment on the Ca2q signal to Kq in cultured granule neurons. ŽA. Representative Ca2q signal to Kq Žapplied at the arrow. in a control and alcohol-treated granule neuron Ž12 DIV.. The Ca2q signals were similar in general form under both conditions. The Ca2q signal to Kq was typically larger in amplitude and faster in time course than the Ca2q signal to NMDA Žcompare to Fig. 3.. ŽB. Mean values Ž"S.E.M.. for the peak amplitude Žresting levels subtracted. of the Ca2q signal to Kq at the two culture ages studied. At 6 DIV there was no difference in the amplitude of the Ca2q signal to Kq in control and alcohol-treated neurons. At 12 DIV the Ca2q signal to Kq was slightly larger in the alcohol-treated neurons. ) s Significant difference Ž p - 0.5, ANOVA.. Numbers in parentheses are the number of neurons studied. Cultures were obtained from one litter, with measurements made on both DIVs.
3.2. Morphological features of the granule neuron deÕelopment in culture Granule neurons obtained from PN 8 rat cerebella and plated on Matrigel coated glass coverslips show a distinct developmental pattern of morphological changes in culture w25x. After dissociation, the neurons are round in shape. During the first day in culture they attach to the substrate, extend processes and start migrating. Gradually clusters of neurons form and by 6 days in vitro ŽDIV. prominent fiber tracks are observed. The fiber tracks and neuronal cluster increase in size with culture age before 7 DIV, after which time they appear relatively stable ŽTable 2.. These developmental changes were similar in granule neuron cultures from control and alcohol-treated pups ŽFig. 1.. Neuronal cluster size was also similar in cultures from control and alcohol-treated animals ŽTable 2., suggesting that alcohol treatment prior to culturing did not cause delayed toxicity or enhanced sensitivity to the culture procedure. 3.3. Intracellular Ca 2 q measurements Alcohol treatment of postnatal rat pups produced small, age-dependent changes in resting Ca2q level of the cultured granule neurons obtained from the pups. At 5, 6 and 7 DIV, resting Ca2q levels were significantly higher in granule neurons obtained from the alcohol-treated pups compared to granule neurons obtained from control pups, whereas at 9 DIV resting Ca2q levels were significantly lower in granule neurons from the alcohol-treated pups ŽFig. 2.. By 12 DIV there was no significant difference in the resting Ca2q levels of the cultured granule neurons from control and alcohol-treated pups, suggesting that this neuronal parameter recovered with time under the culture conditions used.
The Ca2q signal to NMDA was also altered by alcohol treatment of the pups. Measurements were made in the granule neuron cultures at 5, 7 and 9 DIV. The peak amplitude Žresting levels subtracted. of the Ca2q signal to NMDA was significantly smaller in cultured granule neurons from alcohol-treated pups compared to cultured granule neurons from control pups at all three culture ages studied Ž5, 7 and 9 DIV.. Sample recordings and mean values for the population of neurons studied are shown in Fig. 3. The Ca2q signal to Kq depolarization was also examined in cultured granule neurons from control and alcoholtreated pups, to determine if alcohol’s effects on the Ca2q signal to NMDA extended to non-receptor-mediated Ca2q signaling. Measurements were made at two culture ages, 6 DIV and 12 DIV. Ca2q signals elicited by Kq in the cultured granule neurons were similar in general form to Ca2q signals elicited by NMDA, and were characterized by an initial peak and slower recovery phase in both control and alcohol-treated neurons. The peak amplitude of the Ca2q signal to Kq was not depressed in the alcoholtreated pups compared to controls, as was observed for the Ca2q signal to NMDA. A representative series of data are shown in Fig. 4. At 6 DIV, the peak amplitude of the Ca2q signal to Kq was similar in granule neurons from control and alcohol-treated pups, whereas at 12 DIV the Ca2q signal to Kq was slightly Žbut significantly. larger in granule neurons from alcohol-treated pups.
4. Discussion Results from the current study show that alcohol exposure of rat pups during postnatal development alters resting Ca2q levels and the Ca2q signal to NMDA in cultured
18
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
cerebellar granule neurons derived from these animals. These results complement and extend previous studies by others demonstrating an alcohol sensitivity of NMDA receptor-mediated Ca2q signaling in CNS neurons exposed to alcohol during the prenatal period. For example, studies by Lee et al. w20x, Weaver et al. w40x, and Spuhler-Phillips et al. w36x showed that alcohol exposure throughout the prenatal period via maternal alcohol consumption Žmaternal BALs ranged from 26–31 mM. resulted in a reduction in the Ca2q signal to NMDA in brain neurons isolated from 1 day postnatal rat pups. Other studies also indicate a disruptive influence of alcohol on NMDAR functions in the immature nervous system. Prenatal alcohol exposure via maternal alcohol consumption in rats was shown to reduce w 3 Hx-MK-801 binding to the NMDAR channel Žmaternal BALss 23–37 mM. w39x and NMDA-sensitive w 3 Hx-glutamate binding Žmaternal BALs s 8 mM. w33,34x in the hippocampus of offspring, effects consistent with an alcohol-induced reduction in the number of functional NMDARs. Similar results were observed in fetal guinea pigs exposed to alcohol during the prenatal period Žmaternal BALss 59 mM. w3x. In contrast, exposure of rat pups to low levels of alcohol Žneonatal BAL s 12 mM. on PN 4–10 did not alter NMDA-sensitive w 3 Hx-glutamate binding in the cerebellum Žmeasured at 45 days of age. w33x, perhaps due to the low BAL achieved in this study. The effect of alcohol exposure on Ca2q signals to NMDA has also been investigated in cultured neurons derived from embryonic or postnatal animals. In these studies, alcohol applied chronically during the culture period altered the Ca2q signals to NMDA. For example, Zou et al. w44x found that chronic alcohol reduced the Ca2q signals to NMDA in cultured embryonic cerebellar neurons exposed to alcohol Ž75 mM. for 96 h in vitro. We reported similar findings in older cultured cerebellar neurons Žequivalent to postnatal neurons. exposed to chronic alcohol Ž50 mM. for ; 1 week during the culture period w12x. Thus, a variety of alcohol exposure paradigms appear to reduce the Ca2q signal to NMDA in developing CNS neurons. However, in other studies of cultured CNS neurons from embryonic or postnatal animals chronic exposure to alcohol Ž50–75 mM. during the culture period enhanced the Ca2q signal to NMDA w15,16x. Consistent with this observation, increased w 3 Hx-MK801 binding w15x and up-regulated NMDAR polypeptide levels w9x were observed in the alcohol-treated cultured neurons. Thus, chronic alcohol treatment can increase or decrease the Ca2q signals to NMDA in cultured CNS neurons, effects that may depend on the developmental stage and concentration of alcohol administered. In contrast to alcohol’s effect on Ca2q signals to NMDA, the Ca2q signal to Kq depolarization was unaltered or slightly increased in the cultured granule neurons from alcohol-treated pups compared to granule neurons from control pups. Kq depolarization produces an increase in intracellular Ca2q by activating voltage-sensitive cal-
cium channels ŽVSCCs. resulting in Ca2q influx that in turn induces Ca2q release from intracellular stores. These same pathways can contribute to Ca2q signals elicited by NMDA: Ca2q influx through the NMDAR can induce Ca2q release from intracellular stores, where as the membrane depolarization elicited by NMDAR activation can activate VSCCs. Thus, the lack of effect of alcohol treatment on the Ca2q signal to Kq depolarization would appear to argue against alterations in Ca2q influx through VSCCs or Ca2q release from stores as the main contributors to the reduced Ca2q signal to NMDA in the granule neurons from alcohol-treated pups. However, the possibility that the VSCCs and intracellular Ca2q stores involved in the Ca2q signal to NMDA differ from those involved in the Ca2q signal Kq depolarization can not be eliminated at this time. The lack of effect of chronic alcohol on the Ca2q signal to Kq depolarization observed in the current study contrasts with results from studies of Lee et al. w20x and Zou et al. w44x where exposure to alcohol in vivo w20x or in vitro w44x produced a depression of the Ca2q signal to Kq depolarization. However, in the studies by Lee et al. w20x and Zou et al. w44x, measurements were made at the time of alcohol withdrawal, whereas in our studies alcohol had been removed for several days. Thus, the differences in results may reflect recovery of function in our neurons. The intermittent exposure of postnatal rat pups to high levels of alcohol Žreaching BALs of 320 mg% or 70 mM. administered either by vapor inhalation w32x as in our experiments or artificial rearing procedures w41x models exposure of the developing human brain during the third trimester to ‘binge’ drinking. The effects of alcohol exposure in animal exposed to these paradigms have been correlated to teratogenic effects of alcohol and brain retardation seen in children with fetal alcohol syndrome ŽFAS. w32,41x. In the current study intermittent exposure to a relatively high dose Žaveraging ; 320 mg%. of alcohol during postnatal days 4–7 was shown to produce microencephaly in the alcohol-treated pups, reflecting a condition commonly observed in humans with FAS. However, alcohol treatment of the pups did not appear to influence the developmental pattern or the size of granule neuron clusters formed in culture in the absence of alcohol. This lack of effect was surprising since Ca2q influx through NMDA receptors is important for granule neuron migration and development w19x. However, the standard growth medium for granule neurons contains a relatively high concentration of Kq Ž20 mM.. The Kq depolarizes the neurons resulting in activation of VSCCs, Ca2q influx and increased intracellular Ca2q levels. The higher intracellular Ca2q level is thought to be a substitute for trophic factors needed for growth and development of the granule neurons w10x, and may have compensated for alterations in function produced by the chronic alcohol treatment. Moreover, the neurons treated with chronic alcohol exhibited a slightly increased resting Ca2q level at early culture ages when the neurons are most dependent on Ca2q w19x. This higher
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
resting Ca2q level could have compensated for the reduced Ca2q signal to NMDA. However, under physiological conditions in vivo, the developing brain may not be able to counteract the alcohol-mediated deficiencies in NMDA receptor function, ultimately leading to cell loss and CNS dysfunction characteristic of FAS.
w16x
w17x
w18x
Acknowledgements This work was supported in part by NIAAA grants AAO6665, ARC6420 and training grant AA07456. We thank Jodi Caguioa for technical help and Floriska Chizer for secretarial assistance.
w19x w20x
w21x
References w1x Fetal Alcohol Syndrome: Diagnosis, Epidemiology, Prevention, and Treatment, National Academy Press, Washington, DC, 1996, pp. 1–213. w2x Fetal Alcohol Syndrome: From Mechanism to Prevention, CRC Press, Boca Raton, 1996, pp. 1–336. w3x S. Abdollah, J.F. Brien, Effect of chronic maternal ethanol administration on glutamate and N-methyl-D-aspartate binding sites in the hippocampus of near-term fetal guinea pig, Alcohol 12 Ž1995. 377–382. w4x H. Bading, M.M. Segal, N.J. Sucher, H. Dudek, S.A. Lipton, M.E. Greenberg, N-methyl-D-aspartate receptors are critical for mediating the effects of glutamate on intracellular calcium concentration and immediate early gene expression in cultured hippocampal neurons, Neuroscience 64 Ž1995. 653–664. w5x T.V.P. Bliss, G.L. Collingridge, A synaptic model of memory: long-term potentiation in the hippocampus, Nature 361 Ž1993. 31–39. w6x D.W. Choi, Excitotoxic cell death, J. Neurobiol. 23 Ž1992. 1261– 1276. w7x W. Colangelo, D.G. Jones, The fetal alcohol syndrome: a review and assessment of the syndrome and its neurological sequelae, Prog. Neurobiol. 19 Ž1982. 271–314. w8x K.L. Farr, C.Y. Montano, L.L. Paxton, D.D. Savage, Prenatal ethanol exposure decreases hippocampal 3 H-glutamate binding in 45-day old rats, Alcohol 5 Ž1988. 125–133. w9x P. Follesa, M.K. Ticku, Chronic ethanol mediated up regulation of the N-methyl-D-asparate receptor polypeptide subunits in mouse cortical neurons in culture, J. Biol. Chem. 271 Ž1996. 13297–13299. w10x V. Gallo, A. Kingsbury, R. Balazs, O.S. Jorgensen, The role of depolarization in the survival and differentiation of cerebellar granule neurons in culture, J. Neurosci. 7 Ž1987. 2203–3313. w11x C.R. Goodlett, J.D. Thomas, J.R. West, Long term deficits in cerebellar growth and rotarod performance of rats following ‘bingelike’ alcohol exposure during the neonatal brain growth spurt, Neurotoxicol. Teratol. 13 Ž1991. 69–74. w12x D.L. Gruol, K.L. Parsons, Chronic alcohol reduces calcium signaling elicited by glutamate receptor stimulation in developing cerebellar neurons, Brain Res. 728 Ž1996. 166–174. w13x J.J. Hablitz, Prenatal exposure to alcohol alters short-term plasticity in hippocampus, Exp. Neurol. 93 Ž1986. 423–427. w14x J. Holliday, N.C. Spitzer, Calcium channels in the regulation of cell development and cellular interactions, in: L. Hurwitz, L.D. Partidge, J.K. Leach ŽEds.., Calcium Channels: Their Properties, Functions, Regulation, and Clinical Relevance, CRC Press, Boca Raton, FL, 1991, pp. 137–155. w15x X.-J. Hu, M. Ticku, Chronic ethanol treatment upregulates the
w22x
w23x
w24x
w25x
w26x
w27x
w28x
w29x
w30x
w31x
w32x
w33x
w34x
19
NMDA receptor function and binding in mammalian cortical neurons, Mol. Brain Res. 30 Ž1995. 347–356. K.R. Iorio, L. Reinlib, B. Tabakoff, P.L. Hoffman, Chronic exposure of cerebellar granule cells to ethanol results in increased N-methylD-aspartate receptor function, Mol. Pharmacol. 41 Ž1992. 1142–1148. D.G. Jones, Influence of ethanol on neuronal and synaptic maturation in the central nervous system—morphological investigations, Prog. Neurobiol. 31 Ž1988. 171–197. G.M. Kelly, M.J. Druse, D.A. Tonetti, B.G. Oden, Maternal ethanol consumption: binding of L-glutamate to synaptic membranes from whole brain, cortices, and cerebella of offspring, Exp. Neurol. 91 Ž1986. 219–228. H. Komuro, P. Rakic, Modulation of neuronal migration by NMDA receptors, Science 260 Ž1993. 95–97. Y.-H. Lee, K. Spuhler-Phillips, P.K. Randall, S.W. Leslie, Effects of prenatal ethanol exposure on voltage dependent calcium entry into neonatal whole brain dissociated neurons, Alcoholism: Clin. Exper. Res. 20 Ž1996. 921–928. K.E. Light, D.C. Serbus, M. Santiago, Exposure of rats to ethanol from postnatal days 4 to 8: alterations of cholinergic neurochemistry in the cerebral cortex and corpus striatum at day 20, Alcoholism: Clin. Exper. Res. 13 Ž1989. 29–35. L.S. Meyer, L.E. Kotch, E.P. Riley, Alterations in gait following ethanol exposure during the brain growth spurt in rats, Alcoholism: Clin. Exper. Res. 14 Ž1990. 23–27. C. Moffett-Bauer, J. Altman, The effect of ethanol chronically administered to pre-weaning rats on cerebellar development: a morphological study, Brain Res. 119 Ž1977. 249–268. R.A. Morrisett, D. Martin, W.A. Wilson, D.D. Savage, H.S. Swartzwelder, Prenatal exposure to ethanol decreases the sensitivity of the adult rat hippocampus to N-methyl-D-aspartate, Alcohol 6 Ž1989. 415–420. I. Nevo, M. Hamon, Neurotransmitter and neuromodulatory mechanisms involved in alcohol abuse and alcoholism, Neurochem. Int. 26 Ž1995. 305–336. E.P. Nobel, T. Ritchie, Prenatal ethanol exposure reduces the effects of excitatory amino acids in the rat hippocampus, Life Sci. 45 Ž1989. 803–810. D.R. Pierce, C.R. Goodlett, J.R. West, Differential neuronal loss following early postnatal alcohol exposure, Teratology 40 Ž1989. 113–126. Z. Qiu, K.L. Parsons, D.L. Gruol, Interleukin-6 selectively enhances the intracellular calcium response to NMDA in developing CNS neurons, J. Neurosci. 15 Ž1995. 6688–6699. S.A. Queen, C.F. Sanchez, S.R. Lopez, L.L. Paxton, D.D. Savage, Dose and age dependent effects of prenatal ethanol exposure on hippocampal metabotropic-glutamate receptor stimulated phosphoinositide hydrolysis, Alcoholism: Clin. Exper. Res. 17 Ž1993. 887– 893. J.D. Reynolds, J.F. Brien, Ethanol neurobehavioural teratogenesis and the role of L-glutamate in the fetal hippocampus, Can. J. Physiol. Pharmacol. 73 Ž1995. 1209–1223. J. Rogers, S.G. Wiener, F.E. Bloom, Long-term administration methods for rats: advantage of inhalation over intubation or liquid diets, Behav. Neural. Biol. 27 Ž1979. 466–486. A.E. Ryabinin, M. Cole, F.E. Bloom, M.C. Wilson, Exposure of neonatal rats to alcohol by vapor inhalation demonstrates specificity of microcephaly and Purkinje cell loss but not astrogliosis, Alcoholism: Clin. Exper. Res. 19 Ž1995. 784–791. D.D. Savage, C.Y. Montano, M.A. Otero, L.L. Paxton, Prenatal ethanol exposure decreases hippocampal NMDA-sensitive w 3 Hxglutamate binding site density in 45-day old rats, Alcohol 8 Ž1991. 193–201. D.D. Savage, S.A. Queen, C.F. Sanchez, L.L. Paxton, J.C. Mahoney, C.R. Goodlett, J.R. West, Prenatal ethanol exposure during the last third of gestation in rat reduces hippocampal NMDA agonist
20
w35x
w36x
w37x
w38x
w39x
D.L. Gruol et al.r Brain Research 793 (1998) 12–20 binding site density in 45-day old offspring, Alcohol 9 Ž1991. 37–41. D.E. Smith, A. Foundas, J. Canale, Effect of perinatally administered ethanol on the development of the cerebellar granule cell, Exp. Neurol. 92 Ž1986. 491–501. K. Spuhler-Phillips, Y.-H. Lee, P. Hughes, L. Randoll, S. Leslie, Effects of Prenatal ethanol exposure on brain region NMDA-mediated increase in intracellular calcium and the NMDAR1 subunit in forebrain, Alcoholism: Clin. Exper. Res. 21 Ž1997. 68–75. H.S. Swartzwelder, K.L. Farr, W.A. Wilson, D.D. Savage, Prenatal exposure to ethanol decreases physiological plasticity in the hippocampus of the adult rat, Alcohol 5 Ž1988. 121–124. S.E. Tan, R.F. Berman, E.L. Abel, C.S. Zajac, Prenatal alcohol exposure alters hippocampal slice electrophysiology, Alcohol 7 Ž1990. 507–511. S. Valles, V. Felipo, C. Montoliu, C. Guerri, Alcohol exposure during brain development reduces 3 H-MK-801 binding and enhances metabotropic-glutamate receptor-stimulated phosphinositide hydrolysis in rat hippocampus, Life Sci. 56 Ž1995. 1373–1383.
w40x M.S. Weaver, Y.H. Lee, J.L. Morris, P.K. Randall, T. Schallert, S.W. Leslie, Effects of in vitro ethanol and fetal ethanol exposure on glutathione stimulation of N-methyl-D-aspartate receptor function, Alcoholism: Clin. Exper. Res. 17 Ž1993. 643–650. w41x J.R. West, C.R. Goodlett, Teratogenic effects of alcohol on brain development, Ann. Med. 22 Ž1991. 319–325. w42x J.R. West, C.R. Goodlett, D.J. Bonthius, D.R. Pierce, Manipulating peak blood alcohol concentrations in neonatal rats: review of an animal model for alcohol related developmental effects, Neurotoxicity 10 Ž1989. 347–366. w43x A.J. Yool, D.L. Gruol, Development of spontaneous and glutamate evoked activity is altered by chronic ethanol in cultured cerebellar Purkinje neurons, Brain Res. 420 Ž1987. 205–219. w44x J.-Y. Zou, C. Cohan, R.A. Rabin, R.J. Pentney, Continuous exposure of cultured rat cerebellar macroneurons to ethanol-depressed NMDA and KCI-stimulated elevations of intracellular calcium, Alcoholism: Clin. Exper. Res. 19 Ž1995. 840–845.
Research report
Neonatal alcohol exposure reduces NMDA induced Ca2q signaling in developing cerebellar granule neurons D.L. Gruol ) , A.E. Ryabinin 1, K.L. Parsons, M. Cole, M.C. Wilson 2 , Z. Qiu Department of Neuropharmacology and Alcohol Research Center, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA Accepted 30 December 1997
Abstract Glutamatergic neurotransmission through NMDA receptors is critical for both neurogenesis and mature function of the central nervous system ŽCNS., and is thought to be one target for developmentally-induced damage by alcohol to brain function. In the current study we examined Ca2q signaling linked to NMDA receptor activation as a potential site for alcohol’s detrimental effects on the developing nervous system. We compared Ca2q signals to NMDA in granule neurons cultured from cerebella of rat neonates exposed to alcohol Žethanol. during development with responses to NMDA recorded in separated control groups. Alcohol exposure was by the vapor chamber method on postnatal days 4–7. An intermittent exposure paradigm was used where the pups were exposed to alcohol vapor for 2.5 hrday to produce peak BALs of ; 320 mg%. Control pups were placed in an alcohol-free chamber for a similar time period or remained with their mother. After culture under alcohol-free conditions for up to 9 days, Ca2q signaling in response to NMDA was measured using fura-2 Ca2q imaging. Results show that the peak amplitude of the Ca2q signal to NMDA was significantly smaller in cultured granule neurons obtained from alcohol-treated pups compared to granule neurons from control pups. In contrast, the Ca2q signal to Kq depolarization was not depressed by the alcohol treatment. Resting Ca2q levels were also altered by the alcohol treatment. These results show that intermittent alcohol exposure during development in vivo can induce long-term changes in CNS neurons that affect the Ca2q signaling pathway linked to NMDA receptors and resting Ca2q levels. Such changes could play an important role in the CNS dysfunction associated with alcohol exposure during CNS development. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Ethanol; CNS neuron; Development; Glutamate receptor; Intracellular calcium
1. Introduction Maternal alcohol consumption during pregnancy can significantly influence the developing infant w1,2x. The central nervous system ŽCNS. is particularly vulnerable to prenatal alcohol exposure, with some of the most serious effects being microcephaly, reduced mental capabilities, developmental delays and behavior problems w7x. These
) Corresponding author. Department of Neuropharmacology, CVN 11, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037. Fax: q1-619-784-7393; E-mail: [email protected] 1 Current address: Dept. Behavioral Neuroscience, L470, Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, Portland, OR 97201. 2 Current address: Department of Neurosciences, BMSB-145, 915 Camino del Salud, University of New Mexico, School of Medicine, Albuquerque, NM 87131-5221.
0006-8993r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 8 . 0 0 0 1 4 - 6
effects of alcohol are permanent in nature and appear to involve direct actions of alcohol on CNS neurons. Studies in animal models of fetal alcohol exposure and in vitro model systems have shown that many properties of developing neurons are affected by alcohol exposure w17x. Neuronal functions related to synaptic transmission mediated by glutamate are particularly vulnerable. For example, in animal models prenatal alcohol exposure alters excitatory synaptic transmission and synaptic plasticity mediated by glutamate w13,37,38x, developmental expression of glutamate receptors w8,18,29,30,34x, and the response of CNS neurons to glutamate w24,26x. Altered responses to glutamate were also observed in CNS neurons chronically exposed to alcohol during development in vitro w43x. Glutamate is the main excitatory transmitter in the brain and an important trophic factor in the developing nervous system. Thus, the alcohol-induced alterations in glutamate-mediated functions are likely to contribute significantly to the abnormalities observed in animal models
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
of fetal alcohol exposure. Analogous disruptions may underlie the behavioral and cognitive problems observed in children exposed to alcohol during prenatal development. In the current study we examined the effect of intermittent alcohol exposure during development on an aspect of glutamatergic transmission known to play a critical role in CNS function, Ca2q signaling linked to NMDA receptor ŽNMDAR. activation. The NMDA subtype of glutamate receptor is permeable to Ca2q and Ca2q signaling via this receptor plays a pivotal role in synaptic plasticity w5x, neuronal developmental w14,19x, gene expression w4x, and neurotoxicity w6x. Thus, alcohol-induced alterations in Ca2q signaling linked to NMDAR activation could disrupt many aspects of CNS neuronal function, and thereby play a pivotal role in the CNS damage caused in infants by maternal alcohol abuse. Studies in animal models have shown that dose, timing and duration of alcohol exposure are important determining factors in alcohol’s detrimental effects on the developing CNS. For example, in the rat cerebellum where development occurs postnatally, the early postnatal period shows the greatest sensitivity to the toxic effects of alcohol, and a relatively short exposure to high dose of alcohol is more toxic than a prolonged exposure to a low dose of alcohol w42x. In the current studies we focused on this early critical period of cerebellar development and a dose of alcohol that produced significant alterations in cerebellar structure and function in previous in vivo studies w11,21,22,32x. We examined the effect of alcohol on rat pups exposed to alcohol on postnatal ŽPN. days 4–7 and utilized an alcohol dose that produced BALs of ; 320 mg% Ž70 mM. in the pups. Alcohol’s effects were assessed in cerebellar granule neurons, a CNS neuronal type known to express NMDARs during development in vivo and to be highly sensitive to alcohol exposure during the postnatal developmental period w23,27,35x. This developmental period in the rat corresponds to the third trimester of fetal development in humans. Recently we reported that chronic exposure of cultured granule neurons to alcohol Ž33 mM. during development in culture significantly reduced the Ca2q signals to NMDA w12x. While these studies support the proposal that altered Ca2q signals to NMDAR activation are an important component of alcohol’s effects on the developing nervous system, results from cultured neurons may not reflect fully the changes induced when alcohol exposure occurs in vivo, as the cultured neurons are isolated from the rest of the nervous system and from hormonal influences that could be contributing factors to alcohol’s actions. Moreover, in the experiments with cultured neurons alcohol exposure was continuous, an exposure paradigm that does not reflect the intermittent drinking pattern typical of human alcohol consumption. Thus, in the current study we have extended our analysis of alcohol’s effects on Ca2q signaling in developing granule neurons to an in vivo alcohol exposure model. In this model, alcohol is administered to postnatal
13
rat pups via an inhalation chamber for a fixed time period daily w32x. During the remainder of the day the pups remain with their mother. With this exposure paradigm, the blood alcohol level of the pups fluctuates during the day in a manner more reflective of that occurring with human consumption. To assess alcohol’s effects, it was necessary first to remove the neurons from the animal and dissociate them so that Ca2q imaging methodology could be applied. Because the dissociation procedure could alter neuronal properties, we examined NMDAR function after a period of recovery in culture. An added advantage of this approach was that alcohol effects identified after several days in culture were likely to reflect more permanent changes in neuronal function, since transient changes would presumably dissipate during the culture period. Results from these studies show that intermittent alcohol exposure during postnatal development in vivo reduces Ca2q signals linked to NMDAR activation in granule neurons. Small changes in resting Ca2q levels were observed as well. The effects of alcohol on Ca2q signaling to NMDA were selective in that Ca2q signaling to Kq depolarization was not affected similarly. These effects of alcohol appear to reflect relatively long-term changes in neuronal function since the neurons were studied after several days of alcohol free-conditions.
2. Materials and methods 2.1. Animals Timed-pregnant female rats from the Harlan Sprague Dawley laboratories were housed in Plexiglas cages Ž25 = 25 = 20 cm. with sawdust bedding and maintained at 228C " 0.48C on a 12:12 h lightrdark schedule Žlights on at 6:00 a.m.. with food and water continuously available. On postnatal day 2 ŽPN 2; where the day of birth is PN 0. pups were weighed, sexed and randomly assigned to foster mothers keeping the ratio of genders equal when possible. Alcohol exposure started on PN 4. To ensure that alcohol-exposed and control pups reached comparable body weights, litter sizes of control groups were kept at 9–11rlitter, whereas litter sizes of alcoholexposed groups were kept at 6–8 pupsrlitter starting the first day of alcohol-exposure. This reduction in litter size in the alcohol-treated group did not affect brain development w32x. The smaller litter size decreased competition for food in the alcohol-exposed litters and minimized possible nutritional differences between experimental and control animals. Two control groups were used in these studies: Ža. Separated—pups separated from their mother and placed in an alcohol-free chamber for the same time as those exposed to alcohol, and Žb. NaiÕe—normally raised control pups.
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
14
2.2. Alcohol Õapor inhalation Alcohol vapor inhalation was performed as described previously w32x. Ethanol was used for all studies. Briefly, pups were separated from their foster mothers and placed into the alcohol chamber nestled together as a litter in a 30 = 18 = 12 cm cage lined with sawdust bedding. The calculated alcohol concentration in the vapors of the chamber was 25 mgrl, and the exposure lasted for 2.5 h. The method for calculating the chamber alcohol level is described in detail in Rogers et al. w31x. After return to the housing cage, all of the alcohol-exposed pups were retrieved by their mothers and appeared to be suckling within 1r2 h. Alcohol exposure was performed during PN 4–7. Three to four pups closely matched in size and gender were sacrificed on PN 8 for tissue culture. In one experiment pups were sacrificed on PN 10 to access the extend of alcohol-induced microcephaly. Blood alcohol levels ŽBALs. were determined using the NAD-ADH kit ŽSigma.. BALs in trunk blood were measured in PN 4 pups sacrificed immediately after the alcohol exposure. Tail bleeding for BALs was not used on experimental animals in order to avoid the confounding effects of stress on pup development. 2.3. Cell culture Cerebellar granule cells were isolated from control and alcohol-treated pups at PN 8 by a standard enzyme treatment protocol w28x. Briefly, cerebella were dissociated in Ca2q- and Mg 2q-free saline with trypsin and DNAase and plated on Matrigel ŽCollaborative Biomedical Products. coated coverglasses Ž; 10 6 cellrml. in tissue culture dishes containing DMEMrF12 plus 10% horse serum and supplemented with 30 mM glucose, 2 mM glutamine, 20 mM KCl, and 25 m M penicillin–streptomycin. Average cell yield was similar for control and alcohol-treated groups Ž10 = 10 6 " 0.5 = 10 6 cellsrml Žmean " S.E.M.. for control and 9.4 = 10 6 " 0.7 = 10 6 cellsrml for alcoholtreated; 7 dissections for each condition; 3 pupsrdissection; each dissection represents a different litter.. The cultured neurons were maintained under alcohol-free conditions in a 5% CO 2 incubator at 378C for up to 12 days.
Medium without serum was added every 7 days. Treatment with FUDR Ž20 m grml. on the first and fourth days after plating minimized the number of non-neuronal cells in the cultures. 2.4. Intracellular Ca 2 q measurement Relative levels of intracellular Ca2q were determined for individual granule neurons using standard microscopic fura-2 digital imaging and previously published methods w25x. Granule neurons were loaded with 1.5 m M fura2rAM ŽMolecular Probes. and 0.02% pluronic F-127 ŽMolecular Probes. in physiological saline for 30 min. After removal of the fura-2 solution, the cultures were incubated in saline for an additional 45 min to allow for de-esterification of the fura-2 AM. Live video images of the microscopic fields illuminated at 340 and 380 nm wavelengths were recorded with a SIT-66 video camera ŽDAGE-MTI. and digitized by computer. Ratio images were formed by a pixel-by-pixel division of the 340 nm imagerthe 380 nm image. Real time digitized display, image acquisition, and Ca2q measurements were made with MCID imaging software ŽImaging Research.. Intracellular Ca2q levels were estimated by converting fluorescent images at 340 nm and 380 nm wavelengths to ratios and calculating intracellular Ca2q concentration using the following formula: wCa2q x i s K d Ž R y R min .rŽ R max y R . = ForFs , where R s the ratio value, R min s the ratio for a Ca2q free solution, R max s the ratio for a saturated Ca2q solution, K d s 135 Žthe dissociate constant for fura-2., Fo s the intensity of a Ca2q free solution at 380 nm and Fs s the intensity of a saturated Ca2q solution at 380 nm. Background fluorescence was very low making it unnecessary to use background subtraction. Calibration was done using fura salt Ž100 m M. in solutions of known Ca2q concentration. In situ calibration produced inconsistent results and thus was not used. Typical R max , R min , and ForFs values were 0.61, 2.85 and 2.5, respectively. Dye loading and all experiments were performed at room temperature Ž238C.. Several of the fields studied during Ca2q imaging experiments were also used to measured neuronal cluster size, which reflects neuronal survival in culture. Cluster
Table 1 Body and organ weights Žin grams. of pups exposed to alcohol by inhalation on postnatal days 4–7 Group
Body weight on PN 2; F Ž2,16. s 0.15, p ) 0.1
Body weight on PN 10; F Ž2,16. s 1.3, p ) 0.1
Brain weight on PN 10; F Ž2,16. s 7.83, p - 0.01
Liver weight on PN 10; F Ž2,16. s 1.39, p ) 0.1
Kidney weight on PN 10; F Ž2,16. s 0.33, p ) 0.1
Alcohol Ž n s 5. Separation Ž n s 5. Naive Ž n s 9.
11.8 " 0.48 11.8 " 0.32 11.6 " 0.26
25.6 " 0.80 26.8 " 0.75 26.72 " 0.33
0.89 " 0.017) 0.95 " 0.021 0.97 " 0.009
0.87 " 0.017 0.91 " 0.024 0.91 " 0.014
0.31 " 0.012 0.31 " 0.007 0.30 " 0.007
)Statistically significant from other groups. No statistical difference was found between male and females in any of the tests. Therefore, data on animals of both sexes were pulled together as one uniform group. n s Number of pups studied.
D.L. Gruol et al.r Brain Research 793 (1998) 12–20 Table 2 Cluster size in granule neuron cultures from control and alcohol-treated pups Age ŽDIV.
Control cultures; area Ž m m2 .
Alcohol cultures; area Ž m m2 .
5 6 7 9
1208"97 Ž47. 927"49 Ž21. 1560"104 Ž48. 1481"125 Ž27.
1302"87 Ž39. 1072"81 Ž11. 1419"102 Ž35. 1646"191 Ž17.
Numbers in parenthesis are the number of neuronal clusters measured; areas are mean"S.E.M.
size was defined as the two dimensional area encompassed by the cluster. Measurements were made with MCID imaging software ŽImaging Research.. 2.5. Drug application During experiments granule neurons were stimulated with NMDA Ž200 m M; dissolved in bath saline. a selective agonist at the NMDAR or Kq Ž150 mM; Kq substituted for Naq in physiological saline. a depolarizing agent. These stimulants were applied by a brief microperfusion
15
pulse Ž1 s in duration. from drug pipettes Ž; 3 m m tip. placed near a neuronal cluster. The bath was exchanged after each drug application. For experiments with Kq the bath contained physiological saline of the following composition ŽmM.: NaCl, 140; KCl, 3.5; KH 2 PO4 , 0.4; Na 2 HPO4, 0.33; CaCl 2 , 2.2; MgSO4 , 2; glucose, 10; HEPES–NaOH, 10 ŽpH 7.3.. A similar saline but with the omission of MgSO4 and the addition of 5 m M glycine was used for experiments with NMDA. 2.6. Data analysis Ca2q signals to NMDA and to Kq were quantified by measurement of the peak amplitude Žminus baseline resting Ca2q levels. in the somatic region. Results presented represent data collected from 5 litters; both control and alcohol-treated animals from each litter were used to prepare the respective cultures. Both naive Ž3 culture sets. and separated Ž2 culture sets. control animals were used to prepare cultures. Results were similar for the two control groups. Thus, no distinction was made between these two groups when combining data. On each experimental day, 2 or more cultures were measured for each experimental
Fig. 1. Phase contrast micrographs of cultured granule neurons from control and alcohol-treated pups. ŽA1, A2., granule neurons from control pups at 1 ŽA1. and 6 ŽA2. DIV. ŽB1, B2., granule neurons from alcohol-treated pups at 1 ŽB1. and 6 ŽB2. DIV. The same number of neurons were plated for each condition. Arrows in A1 and B1 point to granule neuron somata Žsmaller arrow. and processes Žlarger arrow.. Arrows in A1 and B1 point to granule neuron somata Žsmaller arrow. and processes Žlarger arrow.. Arrows in A2 and B2 point to a granule neuron cluster Žlarge arrow. and fiber tract Žsmall arrow.. Calibration bar is 20 m m.
16
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
isons. p - 0.05 was considered indicative of a significant difference. In all studies, mean values are expressed as mean " S.E.M. For Ca2q imaging studies, n is the number of neurons studied.
3. Results 3.1. Effects of alcohol treatment on the rat pups
Fig. 2. Effect of in vivo alcohol treatment on resting Ca2q levels in cultured granule neurons. Graph show resting Ca2q levels Žmean" S.E.M.; S.E.M.’s are smaller than data points., normalized to control levels, at each culture age tested. Resting Ca2q levels were significantly elevated in the alcohol-treated neurons compared to controls at 5, 6 and 7 DIV, but were reduced at 9 DIV. At 12 DIV there was no difference in resting Ca2q levels between control and alcohol-treated neurons. Resting Ca2q levels ranged from 15 nM–25 nM in control and alcohol-treated neurons at all ages. ) sSignificant difference Ž p- 0.5, ANOVA.. Numbers in parenthesis are the number of neurons measured Žalcohol culturesrcontrol cultures.. Cultures were obtained from 3 different litters, with measurements made on 2–3 different DIVs for each litter.
condition, with 4–6 microscopic fields containing approximately 10–20 neurons measured in each culture. Data from cultures of a similar age and treatment were pooled for statistical analyses. Statistical significance was determined by one-way Analysis of Variance ŽANOVA. followed by the Fisher post-hoc test for multiple compar-
BALs measured in a set of PN 4 old animals averaged 318 " 17 mg% Ž; 70 mM; 29 pups measured. immediately after alcohol exposure. Our previous studies indicated that these BALs return to baseline control levels 10–12 h after the alcohol exposure, and are sufficient to produce microcephaly and Purkinje cell loss on PN 9 and PN 10 if applied during PN 4–9 w32x. In the present experiments, for the purpose of successful generation of primary cell cultures from cerebellar neurons, animals were sacrificed on PN 8, and thus were exposed to alcohol only during PN 4–7. In order to evaluate whether this shorter alcohol exposure is sufficient to produce microcephaly, in one set of experiments the remaining pups were grown to PN 10, and then sacrificed to evaluate the organ and body weight. A comparison of brain weights revealed that alcohol exposure during PN 4–7 produced significantly smaller brains in alcohol-exposed pups in comparison to separated or naive controls ŽTable 1.. In contrast, body, liver and kidney weights between experimental and control animals were not statistically significant. No effect of pup gender on brain, liver or kidney weight was found in these experiments.
Fig. 3. Effect of in vivo alcohol treatment on the Ca2q signal to NMDA in cultured granule neurons. ŽA. Representative Ca2q signal to NMDA Žapplied at the arrow. in a control and alcohol-treated granule neuron Ž6 DIV.. The Ca2q signals were similar in general form in control and alcohol-treated neurons and consisted of a relatively rapid initial increase in intracellular Ca2q and slower recovery phase. The Ca2q signals were smaller in amplitude in the alcohol-treated neurons compared to control neurons. ŽB. Mean values Ž"S.E.M.. for the peak amplitude Žresting levels subtracted. of the Ca2q signal to NMDA at the three culture ages studied. The Ca2q signal to NMDA was smaller in the alcohol-treated neurons compared to controls at all three culture ages. ) s Significant difference Ž p - 0.5, ANOVA.. Numbers in parenthesis are the number of neurons measured. Cultures were obtained from 3 different litters, with measurements made on 2–3 different DIVs for each litter.
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
17
Fig. 4. Effect of in vivo alcohol treatment on the Ca2q signal to Kq in cultured granule neurons. ŽA. Representative Ca2q signal to Kq Žapplied at the arrow. in a control and alcohol-treated granule neuron Ž12 DIV.. The Ca2q signals were similar in general form under both conditions. The Ca2q signal to Kq was typically larger in amplitude and faster in time course than the Ca2q signal to NMDA Žcompare to Fig. 3.. ŽB. Mean values Ž"S.E.M.. for the peak amplitude Žresting levels subtracted. of the Ca2q signal to Kq at the two culture ages studied. At 6 DIV there was no difference in the amplitude of the Ca2q signal to Kq in control and alcohol-treated neurons. At 12 DIV the Ca2q signal to Kq was slightly larger in the alcohol-treated neurons. ) s Significant difference Ž p - 0.5, ANOVA.. Numbers in parentheses are the number of neurons studied. Cultures were obtained from one litter, with measurements made on both DIVs.
3.2. Morphological features of the granule neuron deÕelopment in culture Granule neurons obtained from PN 8 rat cerebella and plated on Matrigel coated glass coverslips show a distinct developmental pattern of morphological changes in culture w25x. After dissociation, the neurons are round in shape. During the first day in culture they attach to the substrate, extend processes and start migrating. Gradually clusters of neurons form and by 6 days in vitro ŽDIV. prominent fiber tracks are observed. The fiber tracks and neuronal cluster increase in size with culture age before 7 DIV, after which time they appear relatively stable ŽTable 2.. These developmental changes were similar in granule neuron cultures from control and alcohol-treated pups ŽFig. 1.. Neuronal cluster size was also similar in cultures from control and alcohol-treated animals ŽTable 2., suggesting that alcohol treatment prior to culturing did not cause delayed toxicity or enhanced sensitivity to the culture procedure. 3.3. Intracellular Ca 2 q measurements Alcohol treatment of postnatal rat pups produced small, age-dependent changes in resting Ca2q level of the cultured granule neurons obtained from the pups. At 5, 6 and 7 DIV, resting Ca2q levels were significantly higher in granule neurons obtained from the alcohol-treated pups compared to granule neurons obtained from control pups, whereas at 9 DIV resting Ca2q levels were significantly lower in granule neurons from the alcohol-treated pups ŽFig. 2.. By 12 DIV there was no significant difference in the resting Ca2q levels of the cultured granule neurons from control and alcohol-treated pups, suggesting that this neuronal parameter recovered with time under the culture conditions used.
The Ca2q signal to NMDA was also altered by alcohol treatment of the pups. Measurements were made in the granule neuron cultures at 5, 7 and 9 DIV. The peak amplitude Žresting levels subtracted. of the Ca2q signal to NMDA was significantly smaller in cultured granule neurons from alcohol-treated pups compared to cultured granule neurons from control pups at all three culture ages studied Ž5, 7 and 9 DIV.. Sample recordings and mean values for the population of neurons studied are shown in Fig. 3. The Ca2q signal to Kq depolarization was also examined in cultured granule neurons from control and alcoholtreated pups, to determine if alcohol’s effects on the Ca2q signal to NMDA extended to non-receptor-mediated Ca2q signaling. Measurements were made at two culture ages, 6 DIV and 12 DIV. Ca2q signals elicited by Kq in the cultured granule neurons were similar in general form to Ca2q signals elicited by NMDA, and were characterized by an initial peak and slower recovery phase in both control and alcohol-treated neurons. The peak amplitude of the Ca2q signal to Kq was not depressed in the alcoholtreated pups compared to controls, as was observed for the Ca2q signal to NMDA. A representative series of data are shown in Fig. 4. At 6 DIV, the peak amplitude of the Ca2q signal to Kq was similar in granule neurons from control and alcohol-treated pups, whereas at 12 DIV the Ca2q signal to Kq was slightly Žbut significantly. larger in granule neurons from alcohol-treated pups.
4. Discussion Results from the current study show that alcohol exposure of rat pups during postnatal development alters resting Ca2q levels and the Ca2q signal to NMDA in cultured
18
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
cerebellar granule neurons derived from these animals. These results complement and extend previous studies by others demonstrating an alcohol sensitivity of NMDA receptor-mediated Ca2q signaling in CNS neurons exposed to alcohol during the prenatal period. For example, studies by Lee et al. w20x, Weaver et al. w40x, and Spuhler-Phillips et al. w36x showed that alcohol exposure throughout the prenatal period via maternal alcohol consumption Žmaternal BALs ranged from 26–31 mM. resulted in a reduction in the Ca2q signal to NMDA in brain neurons isolated from 1 day postnatal rat pups. Other studies also indicate a disruptive influence of alcohol on NMDAR functions in the immature nervous system. Prenatal alcohol exposure via maternal alcohol consumption in rats was shown to reduce w 3 Hx-MK-801 binding to the NMDAR channel Žmaternal BALss 23–37 mM. w39x and NMDA-sensitive w 3 Hx-glutamate binding Žmaternal BALs s 8 mM. w33,34x in the hippocampus of offspring, effects consistent with an alcohol-induced reduction in the number of functional NMDARs. Similar results were observed in fetal guinea pigs exposed to alcohol during the prenatal period Žmaternal BALss 59 mM. w3x. In contrast, exposure of rat pups to low levels of alcohol Žneonatal BAL s 12 mM. on PN 4–10 did not alter NMDA-sensitive w 3 Hx-glutamate binding in the cerebellum Žmeasured at 45 days of age. w33x, perhaps due to the low BAL achieved in this study. The effect of alcohol exposure on Ca2q signals to NMDA has also been investigated in cultured neurons derived from embryonic or postnatal animals. In these studies, alcohol applied chronically during the culture period altered the Ca2q signals to NMDA. For example, Zou et al. w44x found that chronic alcohol reduced the Ca2q signals to NMDA in cultured embryonic cerebellar neurons exposed to alcohol Ž75 mM. for 96 h in vitro. We reported similar findings in older cultured cerebellar neurons Žequivalent to postnatal neurons. exposed to chronic alcohol Ž50 mM. for ; 1 week during the culture period w12x. Thus, a variety of alcohol exposure paradigms appear to reduce the Ca2q signal to NMDA in developing CNS neurons. However, in other studies of cultured CNS neurons from embryonic or postnatal animals chronic exposure to alcohol Ž50–75 mM. during the culture period enhanced the Ca2q signal to NMDA w15,16x. Consistent with this observation, increased w 3 Hx-MK801 binding w15x and up-regulated NMDAR polypeptide levels w9x were observed in the alcohol-treated cultured neurons. Thus, chronic alcohol treatment can increase or decrease the Ca2q signals to NMDA in cultured CNS neurons, effects that may depend on the developmental stage and concentration of alcohol administered. In contrast to alcohol’s effect on Ca2q signals to NMDA, the Ca2q signal to Kq depolarization was unaltered or slightly increased in the cultured granule neurons from alcohol-treated pups compared to granule neurons from control pups. Kq depolarization produces an increase in intracellular Ca2q by activating voltage-sensitive cal-
cium channels ŽVSCCs. resulting in Ca2q influx that in turn induces Ca2q release from intracellular stores. These same pathways can contribute to Ca2q signals elicited by NMDA: Ca2q influx through the NMDAR can induce Ca2q release from intracellular stores, where as the membrane depolarization elicited by NMDAR activation can activate VSCCs. Thus, the lack of effect of alcohol treatment on the Ca2q signal to Kq depolarization would appear to argue against alterations in Ca2q influx through VSCCs or Ca2q release from stores as the main contributors to the reduced Ca2q signal to NMDA in the granule neurons from alcohol-treated pups. However, the possibility that the VSCCs and intracellular Ca2q stores involved in the Ca2q signal to NMDA differ from those involved in the Ca2q signal Kq depolarization can not be eliminated at this time. The lack of effect of chronic alcohol on the Ca2q signal to Kq depolarization observed in the current study contrasts with results from studies of Lee et al. w20x and Zou et al. w44x where exposure to alcohol in vivo w20x or in vitro w44x produced a depression of the Ca2q signal to Kq depolarization. However, in the studies by Lee et al. w20x and Zou et al. w44x, measurements were made at the time of alcohol withdrawal, whereas in our studies alcohol had been removed for several days. Thus, the differences in results may reflect recovery of function in our neurons. The intermittent exposure of postnatal rat pups to high levels of alcohol Žreaching BALs of 320 mg% or 70 mM. administered either by vapor inhalation w32x as in our experiments or artificial rearing procedures w41x models exposure of the developing human brain during the third trimester to ‘binge’ drinking. The effects of alcohol exposure in animal exposed to these paradigms have been correlated to teratogenic effects of alcohol and brain retardation seen in children with fetal alcohol syndrome ŽFAS. w32,41x. In the current study intermittent exposure to a relatively high dose Žaveraging ; 320 mg%. of alcohol during postnatal days 4–7 was shown to produce microencephaly in the alcohol-treated pups, reflecting a condition commonly observed in humans with FAS. However, alcohol treatment of the pups did not appear to influence the developmental pattern or the size of granule neuron clusters formed in culture in the absence of alcohol. This lack of effect was surprising since Ca2q influx through NMDA receptors is important for granule neuron migration and development w19x. However, the standard growth medium for granule neurons contains a relatively high concentration of Kq Ž20 mM.. The Kq depolarizes the neurons resulting in activation of VSCCs, Ca2q influx and increased intracellular Ca2q levels. The higher intracellular Ca2q level is thought to be a substitute for trophic factors needed for growth and development of the granule neurons w10x, and may have compensated for alterations in function produced by the chronic alcohol treatment. Moreover, the neurons treated with chronic alcohol exhibited a slightly increased resting Ca2q level at early culture ages when the neurons are most dependent on Ca2q w19x. This higher
D.L. Gruol et al.r Brain Research 793 (1998) 12–20
resting Ca2q level could have compensated for the reduced Ca2q signal to NMDA. However, under physiological conditions in vivo, the developing brain may not be able to counteract the alcohol-mediated deficiencies in NMDA receptor function, ultimately leading to cell loss and CNS dysfunction characteristic of FAS.
w16x
w17x
w18x
Acknowledgements This work was supported in part by NIAAA grants AAO6665, ARC6420 and training grant AA07456. We thank Jodi Caguioa for technical help and Floriska Chizer for secretarial assistance.
w19x w20x
w21x
References w1x Fetal Alcohol Syndrome: Diagnosis, Epidemiology, Prevention, and Treatment, National Academy Press, Washington, DC, 1996, pp. 1–213. w2x Fetal Alcohol Syndrome: From Mechanism to Prevention, CRC Press, Boca Raton, 1996, pp. 1–336. w3x S. Abdollah, J.F. Brien, Effect of chronic maternal ethanol administration on glutamate and N-methyl-D-aspartate binding sites in the hippocampus of near-term fetal guinea pig, Alcohol 12 Ž1995. 377–382. w4x H. Bading, M.M. Segal, N.J. Sucher, H. Dudek, S.A. Lipton, M.E. Greenberg, N-methyl-D-aspartate receptors are critical for mediating the effects of glutamate on intracellular calcium concentration and immediate early gene expression in cultured hippocampal neurons, Neuroscience 64 Ž1995. 653–664. w5x T.V.P. Bliss, G.L. Collingridge, A synaptic model of memory: long-term potentiation in the hippocampus, Nature 361 Ž1993. 31–39. w6x D.W. Choi, Excitotoxic cell death, J. Neurobiol. 23 Ž1992. 1261– 1276. w7x W. Colangelo, D.G. Jones, The fetal alcohol syndrome: a review and assessment of the syndrome and its neurological sequelae, Prog. Neurobiol. 19 Ž1982. 271–314. w8x K.L. Farr, C.Y. Montano, L.L. Paxton, D.D. Savage, Prenatal ethanol exposure decreases hippocampal 3 H-glutamate binding in 45-day old rats, Alcohol 5 Ž1988. 125–133. w9x P. Follesa, M.K. Ticku, Chronic ethanol mediated up regulation of the N-methyl-D-asparate receptor polypeptide subunits in mouse cortical neurons in culture, J. Biol. Chem. 271 Ž1996. 13297–13299. w10x V. Gallo, A. Kingsbury, R. Balazs, O.S. Jorgensen, The role of depolarization in the survival and differentiation of cerebellar granule neurons in culture, J. Neurosci. 7 Ž1987. 2203–3313. w11x C.R. Goodlett, J.D. Thomas, J.R. West, Long term deficits in cerebellar growth and rotarod performance of rats following ‘bingelike’ alcohol exposure during the neonatal brain growth spurt, Neurotoxicol. Teratol. 13 Ž1991. 69–74. w12x D.L. Gruol, K.L. Parsons, Chronic alcohol reduces calcium signaling elicited by glutamate receptor stimulation in developing cerebellar neurons, Brain Res. 728 Ž1996. 166–174. w13x J.J. Hablitz, Prenatal exposure to alcohol alters short-term plasticity in hippocampus, Exp. Neurol. 93 Ž1986. 423–427. w14x J. Holliday, N.C. Spitzer, Calcium channels in the regulation of cell development and cellular interactions, in: L. Hurwitz, L.D. Partidge, J.K. Leach ŽEds.., Calcium Channels: Their Properties, Functions, Regulation, and Clinical Relevance, CRC Press, Boca Raton, FL, 1991, pp. 137–155. w15x X.-J. Hu, M. Ticku, Chronic ethanol treatment upregulates the
w22x
w23x
w24x
w25x
w26x
w27x
w28x
w29x
w30x
w31x
w32x
w33x
w34x
19
NMDA receptor function and binding in mammalian cortical neurons, Mol. Brain Res. 30 Ž1995. 347–356. K.R. Iorio, L. Reinlib, B. Tabakoff, P.L. Hoffman, Chronic exposure of cerebellar granule cells to ethanol results in increased N-methylD-aspartate receptor function, Mol. Pharmacol. 41 Ž1992. 1142–1148. D.G. Jones, Influence of ethanol on neuronal and synaptic maturation in the central nervous system—morphological investigations, Prog. Neurobiol. 31 Ž1988. 171–197. G.M. Kelly, M.J. Druse, D.A. Tonetti, B.G. Oden, Maternal ethanol consumption: binding of L-glutamate to synaptic membranes from whole brain, cortices, and cerebella of offspring, Exp. Neurol. 91 Ž1986. 219–228. H. Komuro, P. Rakic, Modulation of neuronal migration by NMDA receptors, Science 260 Ž1993. 95–97. Y.-H. Lee, K. Spuhler-Phillips, P.K. Randall, S.W. Leslie, Effects of prenatal ethanol exposure on voltage dependent calcium entry into neonatal whole brain dissociated neurons, Alcoholism: Clin. Exper. Res. 20 Ž1996. 921–928. K.E. Light, D.C. Serbus, M. Santiago, Exposure of rats to ethanol from postnatal days 4 to 8: alterations of cholinergic neurochemistry in the cerebral cortex and corpus striatum at day 20, Alcoholism: Clin. Exper. Res. 13 Ž1989. 29–35. L.S. Meyer, L.E. Kotch, E.P. Riley, Alterations in gait following ethanol exposure during the brain growth spurt in rats, Alcoholism: Clin. Exper. Res. 14 Ž1990. 23–27. C. Moffett-Bauer, J. Altman, The effect of ethanol chronically administered to pre-weaning rats on cerebellar development: a morphological study, Brain Res. 119 Ž1977. 249–268. R.A. Morrisett, D. Martin, W.A. Wilson, D.D. Savage, H.S. Swartzwelder, Prenatal exposure to ethanol decreases the sensitivity of the adult rat hippocampus to N-methyl-D-aspartate, Alcohol 6 Ž1989. 415–420. I. Nevo, M. Hamon, Neurotransmitter and neuromodulatory mechanisms involved in alcohol abuse and alcoholism, Neurochem. Int. 26 Ž1995. 305–336. E.P. Nobel, T. Ritchie, Prenatal ethanol exposure reduces the effects of excitatory amino acids in the rat hippocampus, Life Sci. 45 Ž1989. 803–810. D.R. Pierce, C.R. Goodlett, J.R. West, Differential neuronal loss following early postnatal alcohol exposure, Teratology 40 Ž1989. 113–126. Z. Qiu, K.L. Parsons, D.L. Gruol, Interleukin-6 selectively enhances the intracellular calcium response to NMDA in developing CNS neurons, J. Neurosci. 15 Ž1995. 6688–6699. S.A. Queen, C.F. Sanchez, S.R. Lopez, L.L. Paxton, D.D. Savage, Dose and age dependent effects of prenatal ethanol exposure on hippocampal metabotropic-glutamate receptor stimulated phosphoinositide hydrolysis, Alcoholism: Clin. Exper. Res. 17 Ž1993. 887– 893. J.D. Reynolds, J.F. Brien, Ethanol neurobehavioural teratogenesis and the role of L-glutamate in the fetal hippocampus, Can. J. Physiol. Pharmacol. 73 Ž1995. 1209–1223. J. Rogers, S.G. Wiener, F.E. Bloom, Long-term administration methods for rats: advantage of inhalation over intubation or liquid diets, Behav. Neural. Biol. 27 Ž1979. 466–486. A.E. Ryabinin, M. Cole, F.E. Bloom, M.C. Wilson, Exposure of neonatal rats to alcohol by vapor inhalation demonstrates specificity of microcephaly and Purkinje cell loss but not astrogliosis, Alcoholism: Clin. Exper. Res. 19 Ž1995. 784–791. D.D. Savage, C.Y. Montano, M.A. Otero, L.L. Paxton, Prenatal ethanol exposure decreases hippocampal NMDA-sensitive w 3 Hxglutamate binding site density in 45-day old rats, Alcohol 8 Ž1991. 193–201. D.D. Savage, S.A. Queen, C.F. Sanchez, L.L. Paxton, J.C. Mahoney, C.R. Goodlett, J.R. West, Prenatal ethanol exposure during the last third of gestation in rat reduces hippocampal NMDA agonist
20
w35x
w36x
w37x
w38x
w39x
D.L. Gruol et al.r Brain Research 793 (1998) 12–20 binding site density in 45-day old offspring, Alcohol 9 Ž1991. 37–41. D.E. Smith, A. Foundas, J. Canale, Effect of perinatally administered ethanol on the development of the cerebellar granule cell, Exp. Neurol. 92 Ž1986. 491–501. K. Spuhler-Phillips, Y.-H. Lee, P. Hughes, L. Randoll, S. Leslie, Effects of Prenatal ethanol exposure on brain region NMDA-mediated increase in intracellular calcium and the NMDAR1 subunit in forebrain, Alcoholism: Clin. Exper. Res. 21 Ž1997. 68–75. H.S. Swartzwelder, K.L. Farr, W.A. Wilson, D.D. Savage, Prenatal exposure to ethanol decreases physiological plasticity in the hippocampus of the adult rat, Alcohol 5 Ž1988. 121–124. S.E. Tan, R.F. Berman, E.L. Abel, C.S. Zajac, Prenatal alcohol exposure alters hippocampal slice electrophysiology, Alcohol 7 Ž1990. 507–511. S. Valles, V. Felipo, C. Montoliu, C. Guerri, Alcohol exposure during brain development reduces 3 H-MK-801 binding and enhances metabotropic-glutamate receptor-stimulated phosphinositide hydrolysis in rat hippocampus, Life Sci. 56 Ž1995. 1373–1383.
w40x M.S. Weaver, Y.H. Lee, J.L. Morris, P.K. Randall, T. Schallert, S.W. Leslie, Effects of in vitro ethanol and fetal ethanol exposure on glutathione stimulation of N-methyl-D-aspartate receptor function, Alcoholism: Clin. Exper. Res. 17 Ž1993. 643–650. w41x J.R. West, C.R. Goodlett, Teratogenic effects of alcohol on brain development, Ann. Med. 22 Ž1991. 319–325. w42x J.R. West, C.R. Goodlett, D.J. Bonthius, D.R. Pierce, Manipulating peak blood alcohol concentrations in neonatal rats: review of an animal model for alcohol related developmental effects, Neurotoxicity 10 Ž1989. 347–366. w43x A.J. Yool, D.L. Gruol, Development of spontaneous and glutamate evoked activity is altered by chronic ethanol in cultured cerebellar Purkinje neurons, Brain Res. 420 Ž1987. 205–219. w44x J.-Y. Zou, C. Cohan, R.A. Rabin, R.J. Pentney, Continuous exposure of cultured rat cerebellar macroneurons to ethanol-depressed NMDA and KCI-stimulated elevations of intracellular calcium, Alcoholism: Clin. Exper. Res. 19 Ž1995. 840–845.