- Email: [email protected]
Short-term ethanol exposure alters calbindin D28k and glial fibrillary acidic protein immunoreactivity in hippocampus of mice
Brain Research 879 (2000) 55–64 www.elsevier.com / locate / bres
Research report
Short-term ethanol exposure alters calbindin D28k and glial fibrillary acidic protein immunoreactivity in hippocampus of mice a, a a b b Irawan Satriotomo *, Takanori Miki , Masahiro Itoh , Kiyoshi Ameno , Iwao Ijiri , a Yoshiki Takeuchi b
a Department of Anatomy, Faculty of Medicine, Kagawa Medical University, 1750 -1 Miki-cho, Kita-gun, Ikenobe, Kagawa 761 -0793, Japan Department of Forensic Medicine, Faculty of Medicine, Kagawa Medical University, 1750 -1, Miki-cho, Kita-gun, Ikenobe, Kagawa 761 -0793 Japan
Accepted 18 July 2000
Abstract The effects of a short-term ethanol treatment on hippocampus have been studied in mice exhibiting intoxication signs. The alterations of neurons and astrocytes as well as quantitative changes of calbindin D28k-immunoreactivity and glial fibrillary acidic proteinimmunoreactivity (GFAP-IR) in selected regions of the dorsal hippocampus were examined using anti-calbindin and anti-GFAP monoclonal anti-body (mAb), respectively. The administration of 6% (v / v) ethanol during first week led to the neuronal death and decrease of the total number of calbindin-IR neurons in the examined brain regions. Moreover, the calbindin positive neurons were shown to have diminished processes following short-term ethanol exposure. These neuronal changes were associated with the increase of the GFAP-IR astrocytes. Hypertrophy of cell bodies and cytoplasmic processes of reactive astrocytes were also seen. In addition, dense, thick and highly-stained GFAP-IR cells with long processes in granular cell layer appeared entering into molecular layer of dentate gyrus. In agreement with the discrepancy percentage of neuronal cell loss and increase of reactive astrocytes detected by calbindin and GFAP-IR using image quantitative analysis, the regional differences in the vulnerability to the neurotoxic effects following short-term ethanol exposure were found: CA3.CA2.CA1.DG. These findings also illustrate the importance of correlation between calbindin and GFAP-IR when determining the morphological alteration of neuron and astroglial following short-term ethanol treatment. The disruption of calbindin and GFAP could affect neuronal-astroglial interaction, resulting in disturbance of behaviors dependent on hippocampus. 2000 Elsevier Science B.V. All rights reserved. Theme: Neural basis of behaviour Topic: Drugs of abuse: alcohol, barbiturates, and benzodiazepines Keywords: Hippocampus; Short-term ethanol exposure; Intoxication signs; Glial fibrillary acidic protein (GFAP); Calbindin D28k
1. Introduction Ethanol exposure has been shown to result in alteration in morphology and function of the central nervous system (CNS). Behaviorial studies on the effects of ethanol treatment suggest that the hippocampus, which has been particularly identified as one of the targets for neurotoxic effects, is more sensitive than other regions and plays a prominent role in memory and learning processes [9]. Damage to this structure by acute ethanol treatment *Corresponding author. Tel.: 181-87-891-2087; fax: 181-87-8912088. E-mail address: [email protected] (I. Satriotomo).
resulted in a wide variety of behavioral tasks and processes [7]. Recently it has been recognized that the method used for ethanol exposure may be important, both in the induction of dependence and degree of damage inflicted on the CNS. In this context, it has been demonstrated that high peak blood ethanol concentrations maintained for relatively short periods of time may be critical in disturbing brain functions not only in developing brain [28,29] but also mature brain studies [5,35]. Previous studies suggested that glutamate excitotoxicitylike processes and related changes in calcium levels were implied in the toxic effects of alcohol [25]. Such a mechanism is also possible in hippocampus, since glutamate is the main excitatory neurotransmitter in the hip-
0006-8993 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 00 )02729-3
56
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
pocampus and the importance of Ca 21 for the functioning and survival of hippocampal cells has been well established. Calcium binding proteins have been implicated as important regulators of neuronal degeneration in pathological process [1]. Calbindin-D28k, which is a member of the superfamily of calcium binding proteins is implicated in the regulation of intracellular calcium and as a marker of a neuronal subpopulation [30]. In the adult mammalian brain, calbindin was thought to be present only in neurons, which is believed to serve a neuroprotective role by its calcium-buffering abilities [23]. Astrocytes were originally thought to play only supportive roles in the brain. Recently, there have been many reports concerning the functions of astrocytes including neurotransmitter uptake [15], synthesis and secretion of neurotrophic factors [11]. Analysis targets of astrocytic reaction led the hypothesis that astrocytes are primary targets of chemical injury and mediates the phenotypic expression of chemical injury in the CNS [2]. Glial fibrillary acidic protein (GFAP), which is the major protein of the intermediate filaments of astroglial cells, is the most commonly used method to examine the distribution of astrocytes and the hypertrophy of astrocytes in response to neural degeneration or injury [21]. Our previous report about the effects of short-term ethanol exposure on the suprachiasmatic nucleus (SCN) of the hypothalamus of mice demonstrated that there was no change on the calbindin D28k-IR with the increase on the GFAP-IR [33]. The question arise that whether similar immunoreactivity changes are seen or not in the hippocampus. Therefore, the aim of the present study was to investigate the alteration calbindin D28k-IR neurons and GFAP-IR astrocytes in the hippocampus of mice exhibiting intoxication signs in response to short-term ethanol treatment. To understand which are particularly vulnerable or invulnerable to short-term ethanol exposure, the possible interaction between short-term sensitivity to ethanol and differences in the subfields of hippocampus were also examined.
2. Materials and methods
2.1. Immunohistochemistry procedures Thirty male adult mice (7–8 weeks) of BALB / C strain from SLC (Shizuoka, Japan), weighing 22–25 g, were housed in separate cages under controlled conditions with a constant temperature (23618C) and kept in 12:12 light / dark cycle. In this study the mice were divided into two groups, ethanol-fed and pair-fed controls. The experimental mice (n516) received an unrestricted access liquid diet (Oriental Yeast Co. Ltd., Tokyo, Japan) containing 6% (v / v) ethanol (99.5% ethyl-alcohol, Wako Pure Chemical Ind. Ltd., Osaka, Japan) as the sole fluid source. The pair-fed control mice (n514) were fed an identical liquid
diet except that sucrose was substituted isocalorically for ethanol. After 4–5 days, the mice which exhibited stage 2 or 3 of intoxication signs according to Freund’s classification [10] were selected for morphological observation. The intoxication signs consisted of four stages, such as hyperreactivity and tremor (stage 1); episode of rapid beating motion of the tail, slow, broad-based gait and rapid backward movements (retropulsion); stereotype or repetitive movement (stage 2); generalized tonic–clonic convulsion (stage 3); and death during convulsion (stage 4). The approval of the Kagawa Medical University Animal Committee was obtained for this study. The animals were deeply anaesthetized with sodium pentobarbital (40 mg / kg i.p.). To measure the blood alcohol concentration (BAC) when the animals showed the intoxication signs, blood was taken from the axilla-vessels of the mice and analyzed by the gas chromatograph (Shimadzu GC-8A, Tokyo, Japan). Four mice of the control group and 6 mice of the experimental group were then perfused transcardially with a fixative of 10% formalin neutral buffer solution (pH 7.4, Wako Co. Ltd., Osaka, Japan) for preparation of paraffin embedded sections. Serial 7 mm-thick sections were cut coronally and stained with cresyl-violet. The remaining animals of each group were perfused with 0.02 M phosphate buffered saline (PBS, pH 7.4) followed by a fixative of 4% paraformaldehyde in PBS. The brains were removed and cryoprotected by immersion in 30% (w / v) sucrose at 48C until the brain sank. Serial 40 mm-thick frozen sections were cut coronally and processed for immunohistochemistry. Free floating sections were immunostained for calbindin D28k or GFAP using monoclonal antibody. To stain for either GFAP or calbindin D28k, the sections were incubated in the following solutions and between each steps the sections were rinse with PBS: 10% normal goat serum in PBS with 0.2% Triton X-100 for 30 min; antiGFAP rabbit antibody (1:2000 dilution, DAKO, Glostrup, Denmark) or anti-calbindin D28k mouse monoclonal antibody (1:500 dilution, Sigma Chemical Company, St. Louis, MO) overnight; biotinylated anti rabbit IgG or biotinylated anti mouse IgG for 1 h; avidin–biotin complex (Vectastain, Elite ABC Kit, Vector Lab. Burlingame USA) for 1 h and 0.05% diaminobenzidine–0.03% hydrogen peroxidase for 5 min. Stained sections were mounted on gelatin-coated glass slides and air dried. They were then dehydrated through graded (70, 80, 95, and 100%) alcohol to xylene and coverslipped using Entellan (Merck, Germany). Control sections omitting the primary antibody routinely developed to ensure that any observed staining was due to calbindin-D28k or GFAP. The morphology and expression of calbindin or GFAP were studied with the aid of a light microscope (Nikon, YTHM, Japan).
2.2. Quantitative procedures For a morphometric study, the dorsal hippocampus was
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
divided in 4 regions: the fields of the cornu Ammonis (CA1–CA3) and the superior limb of the dentate gyrus (DG). The cornu Ammonis of CA1 until CA3 was divided into sub-groups of stratum (str.) oriens, str. pyramidale, str. radiatum and str. lacunosum moleculare. The dentate gyrus was divided into molecular layers of fascia dentata or str. oriens, str. granulosum and hilus of fasciae dentata. Our observations were confined to the dorsal part of hippocampus, as shown in Fig. 1. The number of labeled cells was measured quantitatively using image digital analysis in sections of the ten mice of both groups. The first of these sections which contained dorsal hippocampus was numbered as the first section, with all subsequent sections being numbered sequentially. For each animal, a number between 1 and 4 was selected randomly in order to determine the first section. Every 4th section was picked up and the number of calbindin-IR neurons and GFAP-IR astrocytes (for example, the 2, 6, 10th sections etc. were selected for calbindin-immunohistochemistry (IH) and the 4, 8, 12th sections were chosen for GFAP-IH) were counted, respectively. About 10 sections from each animal were used in this study. Photographs (3400 for calbindin and 31000 for GFAP) that were taken from selected sections with a light microscope (Nikon eclips E600) were captured by CCD color camera (Inotech, Japan) and transformed into digits with the help of Viltz Universal software (Vers. 1.5.1, Interwere Corp.) in Power Macintosh 4400 / 200 personal computer. Photographs of individual cells were used in order to avoid interference from overlapping images. The scanned images were placed permanently on the computer screen then analyzed by National Institute of Health (NIH) image software. The area fraction of the total number of IR-cells (proportion of calbindin-IR area per square measuring fields of 1.35310 4 mm 2 and GFAP-IR area per square measuring field of 1.4310 3 mm 2 , respectively) were taken. The number of IR-positive cells / mm-square tissue from each region of the hippocampal sections of the alcohol treated animals was calculated and then compared with the respective control group (Tables 1 and 2). The statistical analysis of data from each region were performed using Statview-J-4.5 (Abacus Concepts, CA) with Student’s t-test (P,0.05 was taken as being significant).
3. Results The analysis by the gas chromatography showed the BAC ranged from approximately 3.0 to 4.8 mg / ml with means 4.1260.85 (mean6S.D.) mg / ml. These values coincided with Freund’s result, which showed a similar BAC (3.0–6.5 mg / ml) [10]. In all sections examined, the pyramidal cell layer of CA1 contained small round type neurons (10–15 mm) which were tightly packed. CA2– CA3 layers had various sizes of neurons (20–30 mm) which were packed with clear cytoplasm and dense
57
processes. The granule cell layer of the fascia dentate contained the smallest (10 mm) neuronal cell bodies and was the most densely packed layer in the hippocampus. On the other hand, the hilus of fascia dentate was composed of very large (25–35 mm) and loosely packed polymorphic cells (Fig. 1A). The intake of 6% (v / v) ethanol in the mice exhibiting intoxication signs altered the neurons in the hippocampus. Neuronal death or lysis was observed in fields of the cornu Ammonis (CA1–CA3) and the dentate gyrus (DG) of the experimental group (Fig. 1C and E) compared with the control group (Fig. 1B and D). The calbindin-D28k positive neurons were moderately stained in CA1, CA2 and dentate gyrus subfields of mice hippocampus. CA1 subfields contained a higher proportion of calbindin D28k positive pyramidal cells and were more immunoreactive than the CA2 subfields, whereas pyramidal cells in the CA3 subfield remained unreactive (Fig. 2A). The single neurons lying in the str. oriens, str. radiatum and str. lacunosum-moleculare were intensely labelled in the control groups (Fig. 2C). These single neurons were characterized by the arborizated slender processes (Fig. 2E). The short-term ethanol exposure led to significant differences when compared with the control group (Fig. 2B). The number of calbindin-positive neurons was decreased and less intensity of immunostaining was seen when compared with the control group (Fig. 2D). Moreover the single neurons that located in str. oriens, str. radiatum and lacunosum-moleculare showed a lack of processes in the experimental group (Fig. 2F). Table 1 shows the estimate of the number (mean6S.D.) of calbindin-D28k IR neurons / mm 2 in various subregions in the hippocampus of ethanol treated and control mice. The statistical comparison between these groups revealed significant differences in all hippocampal subfields, except str. pyramidale of CA1 subfield and str. granulosum of dentate area. In the control mice, GFAP-IR perikarya and processes were observed throughout the hippocampal formation including CA1–CA3 and DG cells of the hippocampus (Fig. 3A). Both pyramidal and granular cell layers were clearly distinguished from adjacent layers, because only fine GFAP-IR were seen in these layers (Figs. 3C and G). The GFAP-IR astrocytes were observed to have small bodies with the short processes in the control group (Fig. 3E). The short-term ethanol exposure resulted in distinct alteration of GFAP-IR astrocytes in the dorsal hippocampus (Fig. 3B). In alcohol treated mice, the numbers of GFAP-IR astrocytes were increased, whereas the area and intensity of GFAP-IR were marked in CA1–CA3 as well as DG subfields of hippocampus (Fig. 3D). Hypertropic astrocytes with the longer processes were also observed (Fig. 3F). In addition, dense, thick and highly-stained GFAP-IR cells of processes of granular cell layers appeared entering into the molecular layer of DG (Fig. 3H). The estimates of the number (mean6S.D.) of GFAP-IR astrocytes / mm 2 in various subregions of the dorsal hip-
58
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
Fig. 1. Frontal section through the dorsal hippocampus of mice including the area, in which histophotometric measurements were carried out, are presented in Fig. A. Photomicrograph of cresyl-violet staining through the CA1–CA2 subfields and dentate gyrus (DG) of control mice (B, D) and 6% (v / v) alcohol treated mice (C, E). Note that short-term ethanol exposure led neuronal death in hippocampus of treated mice compare the control group. h, hilus fasciae dentatae; sg, stratum granulosum; sl, stratum lacunosum moleculare; ml, molecular layer of the fasciae dentata; so, stratum oriens; sp, stratum pyramidale; sr, stratum radiatum (A). Scale bars, 200 mm (A) and 50 mm (B–E).
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
59
Table 1 Data on the number of calbindin-IR neurons (3 10 3 cells / mm 2 ) in the various regions of mice hippocampus following short-term ethanol treatment a Subfields
St. Oriens St. Pyramidale St. Radiatum St. Moleculare
St. Granulosum
n
b
E Cc Eb Cc Eb Cc Eb Cc
b
E Cc
Ammon’s horn CA1
CA2
CA3
10 10 10 10 10 10 10 10
0.1260.08*** 0.1760.08 3.7260.72 3.9160.79 0.0560.04* 0.0660.04 0.1760.07*** 0.2660.08
0.0760.05*** 0.116 0.06 2.2260.47*** 2.6060.36 0.0560.04** 0.0760.05 0.1060.05** 0.1360.06
0.1760.07*** 0.2460.08 NRd NRd 0.0860.05*** 0.1060.06 0.1460.07*** 0.2560.10
n
Dentate gyrus
10 10
31.4465.58 32.1567.28
a
The values are presented as mean6S.D. E: Experiment groups. c C: Control groups. d NR: no reaction, n, number of animals examined. *P,0.05, **P,0.01, ***P,0.001 Student’s t-test. b
pocampus are presented. The statistical comparison of both groups revealed a significant increase of GFAP-IR in all regions of the hippocampus of alcohol treated mice. The quantitative determination of decrease in the total number of calbindin positive neurons and increase in the total number of reactive astrocytes in the dorsal hippocampal formation showed regional differences in the vulnerability to the neurotoxic effect of short-term ethanol treatment: CA3.CA2.CA1.DG (Fig. 4).
4. Discussion In this study we provide evidence that short-term ethanol exposure alters the morphology and the numbers of calbindin-D28k-IR neurons and GFAP-IR astrocytes in the hippocampus of the mice exhibiting intoxication. It is interesting that our present result differs from our previous study on the SCN. Our previous study showed that there was no change on the calbindin D28k-IR with the increase
Table 2 Data on the number of GFAP-IR astrocytes (310 3 cells / mm 2 ) in the various regions of mice hippocampus following short-term ethanol treatment a Subfields
St. Oriens St. Pyramidale St. Radiatum St. Moleculare
Molecular ly. St. Granulosum Hilus FD a
n
b
E Cc Eb Cc Eb Cc Eb Cc
b
E Cc Eb Cc Eb Cc
CA1
CA2
CA3
10 10 10 10 10 10 10 10
5.0660.86*** 3.6460.71 3.6661.02*** 2.4160.75 5.6061.18*** 3.4360.69 5.6561.15*** 3.5160.81
6.2161.22*** 3.5160.81 3.9661.19*** 2.3660.89 5.7961.16*** 3.7460.80 5.3760.94*** 3.4460.66
5.1960.96*** 2.9960.65 3.7360.99*** 2.0660.77 4.6860.90*** 3.0260.71 4.9560.95*** 3.0460.66
n
Dentate gyrus
10 10 10 10 10 10
5.0960.83*** 3.316 0.90 5.106 1.18*** 3.2960.86 4.3160.98*** 3.3760.86
The values are presented as mean6S.D. E: Experiment groups. c C: Control groups, n, number of animals examined. ***P,0.001; Student’s t-test. b
Ammon’s horn
60
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
Fig. 2. Photomicrographs of calbindin D28k-IR in the hippocampus of control and ethanol treated mice. Calbindin-IR neurons revealed pyramidal cells of CA1–CA2 and granular cells of dentate gyrus (DG) have intensive staining however pyramidal cells of CA3 have no staining (A, C). Weak staining and the decrease of calbindin-positive neurons of the hippocampus were observed following short-term ethanol exposure (B, D). The subfields of the hippocampus (framed area) of the control mice in (C) and experimental mice (D) are magnified in (E) and (F), respectively. Note that the short ethanol exposure produced the poor dendritic arborization of calbindin D28k-IR neurons (black arrowheads) compared to the control group. Scale bars, 400 mm (A and B), 200 mm (C and D), and 50 mm (E and F).
on the GFAP-IR [33]. We suggested that the neuroadaptive response of astrocytes could occur before the neurotoxic effects emerge on the neurons in the SCN. It was also reported that the chronic alcohol consumption and withdrawal do not induce cell death in the SCN [20]. Given that neuronal numbers in the SCN do not fall as a consequence of the intoxication, we considered that the
possibility changes in the GFAP-IR astrocytes in the dorsomedial part of SCN might provide an explanation for the alcohol-induced disruption of circadian rhythms. The analysis data of BAC presented here show that ethanol administrated shortly, such that period with high BAC is toxic to hippocampal subfields of CA1–CA2–CA3 and dentate gyrus after exposure for only 4–5 days.
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
61
Fig. 3. Photomicrographs of GFAP-IR in the dorsal hippocampus of control animals with normal stained astrocytes (A, C, G) and experimental groups (B, D, H). The regions of the hippocampus (framed area) of the control rat (C) and experimental mice (D) are magnified in (E) and (F), respectively. Note that hypertrophic astrocytes were seen after treatment with 6% (v / v) ethanol during the first week compared with the control group. Dense, thick and highly-stained GFAP-IR cells with long processes in granular layers appeared entering the molecular layer of dendate gyrus (H). Scale bars, 400 mm (A and B), 200 mm (C, D, G and H) and 50 mm (E and F).
62
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
Fig. 4. Mean value (6S.D.) of the total number of calbindin-D28k IR neurons and GFAP-IR astrocytes in subfields of hippocampus from ethanol-treated mice, expressed as a percentage of controls. The values for each ethanol group represent a mean of count from ten mice. The counts for each subfield of hippocampus within each group were taken from the same set of ten mice. Note that the decrease of calbindin positive neuron is associated with the increase of GFAP-IR astrocytes in the dorsal hippocampus of ethanol treated mice.
Although the neuronal damage can be correlated with the total amount of ethanol administrated, these results indicate a smaller dose can actually be more harmful than the larger one, provided the smaller dose is consumed in a pattern that produces correspondingly higher blood ethanol concentrations. In fact, there is evidence that the binge exposure (i.e., consuming a given amount of ethanol over a shorter period of time) reduces the amount of ethanol needed to produce brain damage in developing [4,12] or mature rats [5,35]. It appears that ethanol intoxication can trigger directly neuronal degeneration. The neurotoxicity of ethanol has been widely discussed physiologically [13] and morphologically [8,31], as well as in in vivo studies [6]. Ethanol intoxication is associated with changes in the activity of neurons in the central nervous system. However, the cellular and molecular mechanisms underlying these changes are still poorly understood. The disturbance of Ca 21 homeostasis has been proposed as a common step in the development of cytotoxicity and / or cell dysfunction in the CNS. A sustained increase in cytosolic Ca 21 concentration, different from the rapid and transient changes occurring in physiological condition, is invariably associated with neuronal damage [26,27]. The increase in Ca 21 appears to mediate the toxicity of several known neurotoxic agents (cyanide, chlordecone, various heavy metals), which induce either alterations in the physical integrity of the plasma membrane or mitochondrial impairment and consequent ATP depletion [16]. These results suggest that ethanol induces one or more of the mechanisms involved in the cytosistolic buffering of intracellular Ca 21 . Thus intoxication elevated intracellular Ca 21 levels may participate in adaptation of the central nervous system to ethanol and / or to occurrence of ethanol induced neuronal damage. The present study demonstrated the loss of neurons from hippocampal regions following short-term ethanol expo-
sure. These results are similar to earlier reports on animals chronically treated with ethanol [18,19] and on human alcoholics [3]. The calbindin-IR staining has confirmed these findings, significant alterations of calbindin-IR neurons were also observed in experimental mice. Our study also showed the hippocampal neurons appear to be differentially susceptible to toxic effects of ethanol both within and between cell types. The resistance or selective vulnerability of different hippocampal neuron populations seemed to be inversely correlated with the calbindin-D28K contents of cells. It has been assumed that neurons containing certain intracellular calcium binding proteins may have a greater capacity to buffer Ca 21 and therefore would be more resistant to degeneration [14]. A previous study demonstrated that cultured hippocampal neurons expresses calbindin D28k-IR resistant to neurotoxicity induced either by glutamate or calcium ionophore [23]. It is of interest that in our results, the CA3 is more vulnerable than the other subfields of hippocampus, since the immunostaining results showed that the CA3 fails to show the evidence of granular cells containing calbindin-IR neuron. The other fact, CA1 cells are more resistant to the neurotoxic effect of ethanol than CA2 cells because the CA1 subfield contains a higher proportion of calbindin D28k positive pyramidal cells than the CA2 subfield. Rami et al. [32] also showed CA1 cells are more immunoreactive than the CA2 cells and the pyramidal cells of the CA2 subfield are more vulnerable than CA1 to ischemia. Compared to rodents, a number of clinical and experimental reports on epileptic damage indicates that calcium-binding proteins are concentrated in the CA2 area of the monkey or human hippocampus, so called the ‘resistant zone’ [17,34]. Quantitative estimations of changes in calbindin D28kIR and GFAP-IR provide a valuable index for monitoring the neuronal-degeneration and reactive astrocytes. The
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
decrease in the calbindin-D28K contents of these neurons was accompanied by cell damage. According to our results, we suggest that the decrease of calbindin-positive neurons reflected the loss of neurons. Our present data demonstrated that the observed decrease in the total number of calbindin-positive neurons is also directly related to the increased amount of reactive astrocytes among the subfields. The increase in the number of GFAPIR astrocytes in hippocampal subfields might reflect increases in GFAP synthesis. New recent evidence has suggested that astrocytes have important roles in the CNS function including synthesis and / or secretion neurotrophic factors [11], and aid repair of injured neuronal tissue [22]. The data quoted above demonstrated that short-term ethanol exposure produced distinct alteration in calbindin D28k-IR neurons and neighboring GFAP-IR astrocytes. The diminished processes of calbindin positive neurons might reflect in a retraction of the neuronal dendritic arbor which play crucial role in cell-to-cell communication. A previous study of targeted deletion in astrocyte intermediate filament suggested that GFAP is important for astrocytes-neuronal interactions and astrocytes processes play a vital role in modulating synaptic efficacy in the CNS [24]. Thus, it should be further emphasized in the present study that short-term administered ethanol disruption of GFAP and calbindin could affect neuronal-astroglial interaction, resulting in disturbance of behaviors dependent on hippocampus. The link established here between calcium binding protein such as calbindin-IR and GFAPIR provide valuable evidence for discoveries concerning neuronal-astrocytes interactions.
Acknowledgements This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sport and Culture of Japan (09307018). We would like to thank Dr. Gail S. Tucker for her constructive comments and suggestions, and Mr. Kensaku Miyamoto and Mrs. Mizue Fukutomi for their technical assistance.
References [1] M.S. Airaksinen, J. Eilers, O. Garaschuk, H. Thoenen, A. Konnerth, M. Meyer, Ataxia and altered dendritic calciums signaling in mice carryng a targeted null mutations of calbindin D28k gene, Proc. Natl Acad. Sci. USA. 94 (1997) 1488–1493. [2] M. Aschner, R.M. LoPachin Jr., Astrocytes: Targets and mediators of chemical-induced CNS injury, J. Toxicol. Environ. Health 38 (1993) 329–342. [3] O. Bengochea, L.M. Gonzalo, Effect of chronic alcoholism on human hippocampus, Histol. Histopathol. 5 (1990) 349–357. [4] D.J. Bonthius, J.R. West, Ethanol induced neuronal lost in the developing rats: increased brain damage with binge exposure, Alcohol. Clin. Exp. Res. 14 (1990) 107–118. [5] M.A. Collins, T.D. Corso, E.J. Neafsey, Neuronal degeneration in
[6]
[7]
[8]
[9]
[10] [11]
[12]
[13]
[14] [15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23]
[24]
[25]
63
rat cerebrocortical and olfactory regions during subchronic ‘binge’ intoxication with ethanol: possible explanation for olfactory deficits in alcoholics, Alcohol. Clin. Exp. Res. 20 (1996) 284–292. D.L. Davies, W.E. Cox Dela, Delayed growth and maturation of astrocytic cultures following exposure to ethanol: electron microscopic observations, Brain Res. 547 (1991) 53–61. L.D. Devenport, R.L. Hale, Contributions of hippocampus and neocortex to the expression of ethanol effects, Psychopharmacology (Berl.) 99 (1989) 337–344. D. Durand, J.A. Saint-Cyr, N. Gurevich, P.L. Carlen, Ethanolinduced dendritic alterations in hippocampal granule cells, Brain Res. 477 (1989) 373–377. H. Franke, H. Kittner, P. Berger, K. Wirkner, J. Schramek, The reaction of astrocytes in the hippocampus of adult rats during chronic ethanol treatment and correlations to behaviour impairments, Alcohol 14 (1997) 445–454. G. Freund, Alcohol withdrawal syndrome in mice, Arch. Neurol. 21 (1969) 315–320. S. Furukawa, Y. Furukawa, E. Satoyoshi, K. Hayashi, Synthesis and secretion of nerve growth factor by mouse astroglia cells in culture, Biochem. Biophys. Res. Commun. 136 (1986) 57–63. C.R. Goodlett, B.L. Marcussen, J.R. West, A single day of alcohol exposure during the brain growth spurt induces brain weight restriction and cerebellar Purkinje cell loss, Alcohol 7 (1990) 107– 114. D.L. Gruol, J.G. Curry, Calcium signals elicited by quisqualate in cultured Purkinje neurons show developmental changes in sensitivity to acute alcohol, Brain Res. 673 (1995) 1–12. C.W. Heizmann, Calcium signaling in the brain, Acta Neurobiol. Exp. (Warsz) 51 (1993) 15–23. H.K. Kimelberg, D.M. Katz, High-affinity uptake of serotonin into immunocytochemically identified astrocytes, Science 228 (1985) 889–891. H. Komulainen, S.C. Bondy, Increased free intracellular Ca21 by toxic agents: an index of potential neurotoxicity?, Trends Pharmacol. Sci. 9 (1988) 154–156. C. Leranth, C.E. Ribak, Calcium binding protein are concentrated in the CA2 field of the monkey hippocampus: a possible key to this region’s resistance to epileptic damage, Exp. Brain Res. 85 (1991) 129–136. N.V. Lukoyanov, M.D. Madeira, M.M. Paula-Barbosa, Behavioral and neuroanatomical consequence of chronic ethanol intake and withdrawal, Physiol. Behav. 66 (1999) 337–346. C. Lundqvist, C. Alling, R. Knoth, B. Volk, Intermittent ethanol exposure of adults rats: hippocampal cell loss after one month of treatment, Alcohol Alcohol. 30 (1995) 737–748. M.D. Madiera, J.P. Andrade, A.R. Lieberman, N. Sousa, O.F.X. Almeida, M.M. Paula-Barbosa, Chronic alcohol consumption and withdrawal do not induce cell death in the suprachiasmatic nucleus, but lead to irreversible depression of peptide immunoreactivity and mRNA levels, J. Neurosci. 17 (1997) 1302–1319. P.M. Martin, J.P. O’Callaghan, A direct comparison of GFAP immunocytochemistry and GFAP concentration in various regions of ethanol-fixed rat and mouse brain, J. Neurosci. Methods 58 (1995) 181–192. A.J. Mathewson, M. Berry, Observation on astrocytes response to the cerebral wound in adult rats, Brain Res. 327 (1985) 61–69. M.P. Mattson, B. Rychlik, C. Chu, S. Christakos, Evidence for calcium-reducing and excitoprotective roles for the calcium-binding protein-D28k in cultured hippocampal neurons, Neuron 6 (1991) 41–51. M.A. McCall, R.G. Gregg, R.R. Behringer, M. Brenner, C.L. Delaney, E.J. Galbreath, C. Zhang L, R.A. Pearce, S.Y. Chiu, A. Messing, Targeted deletion in astrocytes intermediate filament (GFAP) alters neuronal physiology, Proc. Natl. Acad. Sci. USA 93 (1996) 6361–6366. E.K. Michaelis, Fetal alcohol exposure: Cellular toxicity and
64
[26]
[27] [28]
[29] [30]
[31]
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64 molecular events involved intoxicity, Alcohol. Clin. Exp. Res. 14 (1990) 819–826. P. Nicotera, G. Bellomo, S. Orrenius, Calcium mediated mechanisms in chemically induced cell death, Ann. Rev. Pharmacol. Toxicol. 32 (1992) 449–470. S. Orrenius, P. Nicotera, The calcium ion and cell death, J. Neural Transm. 43 (1994) 1–11. D.R. Pierce, J.R. West, Alcohol-induced microcephaly during the third trimester equivalent: relationship to dose and blood alcohol concentration, Alcohol 3 (1986) 185–191. D.R. Pierce, J.R. West, Blood alcohol concentration: a critical factor for producing fetal alcohol effects, Alcohol 3 (1986) 269–272. B. Pfeiffer, A.W. Norman, B. Hamprecht, Immunocytochemical characterization of neuron-rich rat brain primary cultures: calbindin D28k as marker of neuronal subpopulation, Brain Res. 476 (1989) 120–128. S.C. Philips, B.G. Cragg, Chronic consumption of alcohol by adult
[32]
[33]
[34]
[35]
mice: Effect on hippocampal cells and synapses, Exp. Neurol. 80 (1983) 218–226. A. Rami, A. Rabie, M. Thomasset, J. Krieglstein, Calbindin-D28k and ischemic damage of pyramidal cells in rat hippocampus, J. Neurosci. Res. 31 (1992) 89–95. I. Satriotomo, T. Miki, M. Itoh, Q. Xie, K. Ameno, Y. Takeuchi, Effect of short-term ethanol exposure on the suprachiasmatic nucleus of hypothalamus: immunohystochemical study in mice, Brain Res. 847 (1999) 124–129. R.S. Sloviter, A.L. Sollas, N.M. Barbaro, K.D. Laxer, calcium binding protein (calbindin-D28k) and parvalbumin immunocytochemistry in the normal and epileptic human hippocampus, J. Comp. Neurol. 308 (1991) 381–396. J.Y. Zou, D.B. Martinez, E.J. Neafsey, M.A. Collins, Binge ethanol– induced brain damage in rats: effect of inhibitors of nitric oxide synthase, Alcohol. Clin. Exp. Res. 20 (1996) 1406–1411.
Research report
Short-term ethanol exposure alters calbindin D28k and glial fibrillary acidic protein immunoreactivity in hippocampus of mice a, a a b b Irawan Satriotomo *, Takanori Miki , Masahiro Itoh , Kiyoshi Ameno , Iwao Ijiri , a Yoshiki Takeuchi b
a Department of Anatomy, Faculty of Medicine, Kagawa Medical University, 1750 -1 Miki-cho, Kita-gun, Ikenobe, Kagawa 761 -0793, Japan Department of Forensic Medicine, Faculty of Medicine, Kagawa Medical University, 1750 -1, Miki-cho, Kita-gun, Ikenobe, Kagawa 761 -0793 Japan
Accepted 18 July 2000
Abstract The effects of a short-term ethanol treatment on hippocampus have been studied in mice exhibiting intoxication signs. The alterations of neurons and astrocytes as well as quantitative changes of calbindin D28k-immunoreactivity and glial fibrillary acidic proteinimmunoreactivity (GFAP-IR) in selected regions of the dorsal hippocampus were examined using anti-calbindin and anti-GFAP monoclonal anti-body (mAb), respectively. The administration of 6% (v / v) ethanol during first week led to the neuronal death and decrease of the total number of calbindin-IR neurons in the examined brain regions. Moreover, the calbindin positive neurons were shown to have diminished processes following short-term ethanol exposure. These neuronal changes were associated with the increase of the GFAP-IR astrocytes. Hypertrophy of cell bodies and cytoplasmic processes of reactive astrocytes were also seen. In addition, dense, thick and highly-stained GFAP-IR cells with long processes in granular cell layer appeared entering into molecular layer of dentate gyrus. In agreement with the discrepancy percentage of neuronal cell loss and increase of reactive astrocytes detected by calbindin and GFAP-IR using image quantitative analysis, the regional differences in the vulnerability to the neurotoxic effects following short-term ethanol exposure were found: CA3.CA2.CA1.DG. These findings also illustrate the importance of correlation between calbindin and GFAP-IR when determining the morphological alteration of neuron and astroglial following short-term ethanol treatment. The disruption of calbindin and GFAP could affect neuronal-astroglial interaction, resulting in disturbance of behaviors dependent on hippocampus. 2000 Elsevier Science B.V. All rights reserved. Theme: Neural basis of behaviour Topic: Drugs of abuse: alcohol, barbiturates, and benzodiazepines Keywords: Hippocampus; Short-term ethanol exposure; Intoxication signs; Glial fibrillary acidic protein (GFAP); Calbindin D28k
1. Introduction Ethanol exposure has been shown to result in alteration in morphology and function of the central nervous system (CNS). Behaviorial studies on the effects of ethanol treatment suggest that the hippocampus, which has been particularly identified as one of the targets for neurotoxic effects, is more sensitive than other regions and plays a prominent role in memory and learning processes [9]. Damage to this structure by acute ethanol treatment *Corresponding author. Tel.: 181-87-891-2087; fax: 181-87-8912088. E-mail address: [email protected] (I. Satriotomo).
resulted in a wide variety of behavioral tasks and processes [7]. Recently it has been recognized that the method used for ethanol exposure may be important, both in the induction of dependence and degree of damage inflicted on the CNS. In this context, it has been demonstrated that high peak blood ethanol concentrations maintained for relatively short periods of time may be critical in disturbing brain functions not only in developing brain [28,29] but also mature brain studies [5,35]. Previous studies suggested that glutamate excitotoxicitylike processes and related changes in calcium levels were implied in the toxic effects of alcohol [25]. Such a mechanism is also possible in hippocampus, since glutamate is the main excitatory neurotransmitter in the hip-
0006-8993 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 00 )02729-3
56
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
pocampus and the importance of Ca 21 for the functioning and survival of hippocampal cells has been well established. Calcium binding proteins have been implicated as important regulators of neuronal degeneration in pathological process [1]. Calbindin-D28k, which is a member of the superfamily of calcium binding proteins is implicated in the regulation of intracellular calcium and as a marker of a neuronal subpopulation [30]. In the adult mammalian brain, calbindin was thought to be present only in neurons, which is believed to serve a neuroprotective role by its calcium-buffering abilities [23]. Astrocytes were originally thought to play only supportive roles in the brain. Recently, there have been many reports concerning the functions of astrocytes including neurotransmitter uptake [15], synthesis and secretion of neurotrophic factors [11]. Analysis targets of astrocytic reaction led the hypothesis that astrocytes are primary targets of chemical injury and mediates the phenotypic expression of chemical injury in the CNS [2]. Glial fibrillary acidic protein (GFAP), which is the major protein of the intermediate filaments of astroglial cells, is the most commonly used method to examine the distribution of astrocytes and the hypertrophy of astrocytes in response to neural degeneration or injury [21]. Our previous report about the effects of short-term ethanol exposure on the suprachiasmatic nucleus (SCN) of the hypothalamus of mice demonstrated that there was no change on the calbindin D28k-IR with the increase on the GFAP-IR [33]. The question arise that whether similar immunoreactivity changes are seen or not in the hippocampus. Therefore, the aim of the present study was to investigate the alteration calbindin D28k-IR neurons and GFAP-IR astrocytes in the hippocampus of mice exhibiting intoxication signs in response to short-term ethanol treatment. To understand which are particularly vulnerable or invulnerable to short-term ethanol exposure, the possible interaction between short-term sensitivity to ethanol and differences in the subfields of hippocampus were also examined.
2. Materials and methods
2.1. Immunohistochemistry procedures Thirty male adult mice (7–8 weeks) of BALB / C strain from SLC (Shizuoka, Japan), weighing 22–25 g, were housed in separate cages under controlled conditions with a constant temperature (23618C) and kept in 12:12 light / dark cycle. In this study the mice were divided into two groups, ethanol-fed and pair-fed controls. The experimental mice (n516) received an unrestricted access liquid diet (Oriental Yeast Co. Ltd., Tokyo, Japan) containing 6% (v / v) ethanol (99.5% ethyl-alcohol, Wako Pure Chemical Ind. Ltd., Osaka, Japan) as the sole fluid source. The pair-fed control mice (n514) were fed an identical liquid
diet except that sucrose was substituted isocalorically for ethanol. After 4–5 days, the mice which exhibited stage 2 or 3 of intoxication signs according to Freund’s classification [10] were selected for morphological observation. The intoxication signs consisted of four stages, such as hyperreactivity and tremor (stage 1); episode of rapid beating motion of the tail, slow, broad-based gait and rapid backward movements (retropulsion); stereotype or repetitive movement (stage 2); generalized tonic–clonic convulsion (stage 3); and death during convulsion (stage 4). The approval of the Kagawa Medical University Animal Committee was obtained for this study. The animals were deeply anaesthetized with sodium pentobarbital (40 mg / kg i.p.). To measure the blood alcohol concentration (BAC) when the animals showed the intoxication signs, blood was taken from the axilla-vessels of the mice and analyzed by the gas chromatograph (Shimadzu GC-8A, Tokyo, Japan). Four mice of the control group and 6 mice of the experimental group were then perfused transcardially with a fixative of 10% formalin neutral buffer solution (pH 7.4, Wako Co. Ltd., Osaka, Japan) for preparation of paraffin embedded sections. Serial 7 mm-thick sections were cut coronally and stained with cresyl-violet. The remaining animals of each group were perfused with 0.02 M phosphate buffered saline (PBS, pH 7.4) followed by a fixative of 4% paraformaldehyde in PBS. The brains were removed and cryoprotected by immersion in 30% (w / v) sucrose at 48C until the brain sank. Serial 40 mm-thick frozen sections were cut coronally and processed for immunohistochemistry. Free floating sections were immunostained for calbindin D28k or GFAP using monoclonal antibody. To stain for either GFAP or calbindin D28k, the sections were incubated in the following solutions and between each steps the sections were rinse with PBS: 10% normal goat serum in PBS with 0.2% Triton X-100 for 30 min; antiGFAP rabbit antibody (1:2000 dilution, DAKO, Glostrup, Denmark) or anti-calbindin D28k mouse monoclonal antibody (1:500 dilution, Sigma Chemical Company, St. Louis, MO) overnight; biotinylated anti rabbit IgG or biotinylated anti mouse IgG for 1 h; avidin–biotin complex (Vectastain, Elite ABC Kit, Vector Lab. Burlingame USA) for 1 h and 0.05% diaminobenzidine–0.03% hydrogen peroxidase for 5 min. Stained sections were mounted on gelatin-coated glass slides and air dried. They were then dehydrated through graded (70, 80, 95, and 100%) alcohol to xylene and coverslipped using Entellan (Merck, Germany). Control sections omitting the primary antibody routinely developed to ensure that any observed staining was due to calbindin-D28k or GFAP. The morphology and expression of calbindin or GFAP were studied with the aid of a light microscope (Nikon, YTHM, Japan).
2.2. Quantitative procedures For a morphometric study, the dorsal hippocampus was
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
divided in 4 regions: the fields of the cornu Ammonis (CA1–CA3) and the superior limb of the dentate gyrus (DG). The cornu Ammonis of CA1 until CA3 was divided into sub-groups of stratum (str.) oriens, str. pyramidale, str. radiatum and str. lacunosum moleculare. The dentate gyrus was divided into molecular layers of fascia dentata or str. oriens, str. granulosum and hilus of fasciae dentata. Our observations were confined to the dorsal part of hippocampus, as shown in Fig. 1. The number of labeled cells was measured quantitatively using image digital analysis in sections of the ten mice of both groups. The first of these sections which contained dorsal hippocampus was numbered as the first section, with all subsequent sections being numbered sequentially. For each animal, a number between 1 and 4 was selected randomly in order to determine the first section. Every 4th section was picked up and the number of calbindin-IR neurons and GFAP-IR astrocytes (for example, the 2, 6, 10th sections etc. were selected for calbindin-immunohistochemistry (IH) and the 4, 8, 12th sections were chosen for GFAP-IH) were counted, respectively. About 10 sections from each animal were used in this study. Photographs (3400 for calbindin and 31000 for GFAP) that were taken from selected sections with a light microscope (Nikon eclips E600) were captured by CCD color camera (Inotech, Japan) and transformed into digits with the help of Viltz Universal software (Vers. 1.5.1, Interwere Corp.) in Power Macintosh 4400 / 200 personal computer. Photographs of individual cells were used in order to avoid interference from overlapping images. The scanned images were placed permanently on the computer screen then analyzed by National Institute of Health (NIH) image software. The area fraction of the total number of IR-cells (proportion of calbindin-IR area per square measuring fields of 1.35310 4 mm 2 and GFAP-IR area per square measuring field of 1.4310 3 mm 2 , respectively) were taken. The number of IR-positive cells / mm-square tissue from each region of the hippocampal sections of the alcohol treated animals was calculated and then compared with the respective control group (Tables 1 and 2). The statistical analysis of data from each region were performed using Statview-J-4.5 (Abacus Concepts, CA) with Student’s t-test (P,0.05 was taken as being significant).
3. Results The analysis by the gas chromatography showed the BAC ranged from approximately 3.0 to 4.8 mg / ml with means 4.1260.85 (mean6S.D.) mg / ml. These values coincided with Freund’s result, which showed a similar BAC (3.0–6.5 mg / ml) [10]. In all sections examined, the pyramidal cell layer of CA1 contained small round type neurons (10–15 mm) which were tightly packed. CA2– CA3 layers had various sizes of neurons (20–30 mm) which were packed with clear cytoplasm and dense
57
processes. The granule cell layer of the fascia dentate contained the smallest (10 mm) neuronal cell bodies and was the most densely packed layer in the hippocampus. On the other hand, the hilus of fascia dentate was composed of very large (25–35 mm) and loosely packed polymorphic cells (Fig. 1A). The intake of 6% (v / v) ethanol in the mice exhibiting intoxication signs altered the neurons in the hippocampus. Neuronal death or lysis was observed in fields of the cornu Ammonis (CA1–CA3) and the dentate gyrus (DG) of the experimental group (Fig. 1C and E) compared with the control group (Fig. 1B and D). The calbindin-D28k positive neurons were moderately stained in CA1, CA2 and dentate gyrus subfields of mice hippocampus. CA1 subfields contained a higher proportion of calbindin D28k positive pyramidal cells and were more immunoreactive than the CA2 subfields, whereas pyramidal cells in the CA3 subfield remained unreactive (Fig. 2A). The single neurons lying in the str. oriens, str. radiatum and str. lacunosum-moleculare were intensely labelled in the control groups (Fig. 2C). These single neurons were characterized by the arborizated slender processes (Fig. 2E). The short-term ethanol exposure led to significant differences when compared with the control group (Fig. 2B). The number of calbindin-positive neurons was decreased and less intensity of immunostaining was seen when compared with the control group (Fig. 2D). Moreover the single neurons that located in str. oriens, str. radiatum and lacunosum-moleculare showed a lack of processes in the experimental group (Fig. 2F). Table 1 shows the estimate of the number (mean6S.D.) of calbindin-D28k IR neurons / mm 2 in various subregions in the hippocampus of ethanol treated and control mice. The statistical comparison between these groups revealed significant differences in all hippocampal subfields, except str. pyramidale of CA1 subfield and str. granulosum of dentate area. In the control mice, GFAP-IR perikarya and processes were observed throughout the hippocampal formation including CA1–CA3 and DG cells of the hippocampus (Fig. 3A). Both pyramidal and granular cell layers were clearly distinguished from adjacent layers, because only fine GFAP-IR were seen in these layers (Figs. 3C and G). The GFAP-IR astrocytes were observed to have small bodies with the short processes in the control group (Fig. 3E). The short-term ethanol exposure resulted in distinct alteration of GFAP-IR astrocytes in the dorsal hippocampus (Fig. 3B). In alcohol treated mice, the numbers of GFAP-IR astrocytes were increased, whereas the area and intensity of GFAP-IR were marked in CA1–CA3 as well as DG subfields of hippocampus (Fig. 3D). Hypertropic astrocytes with the longer processes were also observed (Fig. 3F). In addition, dense, thick and highly-stained GFAP-IR cells of processes of granular cell layers appeared entering into the molecular layer of DG (Fig. 3H). The estimates of the number (mean6S.D.) of GFAP-IR astrocytes / mm 2 in various subregions of the dorsal hip-
58
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
Fig. 1. Frontal section through the dorsal hippocampus of mice including the area, in which histophotometric measurements were carried out, are presented in Fig. A. Photomicrograph of cresyl-violet staining through the CA1–CA2 subfields and dentate gyrus (DG) of control mice (B, D) and 6% (v / v) alcohol treated mice (C, E). Note that short-term ethanol exposure led neuronal death in hippocampus of treated mice compare the control group. h, hilus fasciae dentatae; sg, stratum granulosum; sl, stratum lacunosum moleculare; ml, molecular layer of the fasciae dentata; so, stratum oriens; sp, stratum pyramidale; sr, stratum radiatum (A). Scale bars, 200 mm (A) and 50 mm (B–E).
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
59
Table 1 Data on the number of calbindin-IR neurons (3 10 3 cells / mm 2 ) in the various regions of mice hippocampus following short-term ethanol treatment a Subfields
St. Oriens St. Pyramidale St. Radiatum St. Moleculare
St. Granulosum
n
b
E Cc Eb Cc Eb Cc Eb Cc
b
E Cc
Ammon’s horn CA1
CA2
CA3
10 10 10 10 10 10 10 10
0.1260.08*** 0.1760.08 3.7260.72 3.9160.79 0.0560.04* 0.0660.04 0.1760.07*** 0.2660.08
0.0760.05*** 0.116 0.06 2.2260.47*** 2.6060.36 0.0560.04** 0.0760.05 0.1060.05** 0.1360.06
0.1760.07*** 0.2460.08 NRd NRd 0.0860.05*** 0.1060.06 0.1460.07*** 0.2560.10
n
Dentate gyrus
10 10
31.4465.58 32.1567.28
a
The values are presented as mean6S.D. E: Experiment groups. c C: Control groups. d NR: no reaction, n, number of animals examined. *P,0.05, **P,0.01, ***P,0.001 Student’s t-test. b
pocampus are presented. The statistical comparison of both groups revealed a significant increase of GFAP-IR in all regions of the hippocampus of alcohol treated mice. The quantitative determination of decrease in the total number of calbindin positive neurons and increase in the total number of reactive astrocytes in the dorsal hippocampal formation showed regional differences in the vulnerability to the neurotoxic effect of short-term ethanol treatment: CA3.CA2.CA1.DG (Fig. 4).
4. Discussion In this study we provide evidence that short-term ethanol exposure alters the morphology and the numbers of calbindin-D28k-IR neurons and GFAP-IR astrocytes in the hippocampus of the mice exhibiting intoxication. It is interesting that our present result differs from our previous study on the SCN. Our previous study showed that there was no change on the calbindin D28k-IR with the increase
Table 2 Data on the number of GFAP-IR astrocytes (310 3 cells / mm 2 ) in the various regions of mice hippocampus following short-term ethanol treatment a Subfields
St. Oriens St. Pyramidale St. Radiatum St. Moleculare
Molecular ly. St. Granulosum Hilus FD a
n
b
E Cc Eb Cc Eb Cc Eb Cc
b
E Cc Eb Cc Eb Cc
CA1
CA2
CA3
10 10 10 10 10 10 10 10
5.0660.86*** 3.6460.71 3.6661.02*** 2.4160.75 5.6061.18*** 3.4360.69 5.6561.15*** 3.5160.81
6.2161.22*** 3.5160.81 3.9661.19*** 2.3660.89 5.7961.16*** 3.7460.80 5.3760.94*** 3.4460.66
5.1960.96*** 2.9960.65 3.7360.99*** 2.0660.77 4.6860.90*** 3.0260.71 4.9560.95*** 3.0460.66
n
Dentate gyrus
10 10 10 10 10 10
5.0960.83*** 3.316 0.90 5.106 1.18*** 3.2960.86 4.3160.98*** 3.3760.86
The values are presented as mean6S.D. E: Experiment groups. c C: Control groups, n, number of animals examined. ***P,0.001; Student’s t-test. b
Ammon’s horn
60
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
Fig. 2. Photomicrographs of calbindin D28k-IR in the hippocampus of control and ethanol treated mice. Calbindin-IR neurons revealed pyramidal cells of CA1–CA2 and granular cells of dentate gyrus (DG) have intensive staining however pyramidal cells of CA3 have no staining (A, C). Weak staining and the decrease of calbindin-positive neurons of the hippocampus were observed following short-term ethanol exposure (B, D). The subfields of the hippocampus (framed area) of the control mice in (C) and experimental mice (D) are magnified in (E) and (F), respectively. Note that the short ethanol exposure produced the poor dendritic arborization of calbindin D28k-IR neurons (black arrowheads) compared to the control group. Scale bars, 400 mm (A and B), 200 mm (C and D), and 50 mm (E and F).
on the GFAP-IR [33]. We suggested that the neuroadaptive response of astrocytes could occur before the neurotoxic effects emerge on the neurons in the SCN. It was also reported that the chronic alcohol consumption and withdrawal do not induce cell death in the SCN [20]. Given that neuronal numbers in the SCN do not fall as a consequence of the intoxication, we considered that the
possibility changes in the GFAP-IR astrocytes in the dorsomedial part of SCN might provide an explanation for the alcohol-induced disruption of circadian rhythms. The analysis data of BAC presented here show that ethanol administrated shortly, such that period with high BAC is toxic to hippocampal subfields of CA1–CA2–CA3 and dentate gyrus after exposure for only 4–5 days.
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
61
Fig. 3. Photomicrographs of GFAP-IR in the dorsal hippocampus of control animals with normal stained astrocytes (A, C, G) and experimental groups (B, D, H). The regions of the hippocampus (framed area) of the control rat (C) and experimental mice (D) are magnified in (E) and (F), respectively. Note that hypertrophic astrocytes were seen after treatment with 6% (v / v) ethanol during the first week compared with the control group. Dense, thick and highly-stained GFAP-IR cells with long processes in granular layers appeared entering the molecular layer of dendate gyrus (H). Scale bars, 400 mm (A and B), 200 mm (C, D, G and H) and 50 mm (E and F).
62
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
Fig. 4. Mean value (6S.D.) of the total number of calbindin-D28k IR neurons and GFAP-IR astrocytes in subfields of hippocampus from ethanol-treated mice, expressed as a percentage of controls. The values for each ethanol group represent a mean of count from ten mice. The counts for each subfield of hippocampus within each group were taken from the same set of ten mice. Note that the decrease of calbindin positive neuron is associated with the increase of GFAP-IR astrocytes in the dorsal hippocampus of ethanol treated mice.
Although the neuronal damage can be correlated with the total amount of ethanol administrated, these results indicate a smaller dose can actually be more harmful than the larger one, provided the smaller dose is consumed in a pattern that produces correspondingly higher blood ethanol concentrations. In fact, there is evidence that the binge exposure (i.e., consuming a given amount of ethanol over a shorter period of time) reduces the amount of ethanol needed to produce brain damage in developing [4,12] or mature rats [5,35]. It appears that ethanol intoxication can trigger directly neuronal degeneration. The neurotoxicity of ethanol has been widely discussed physiologically [13] and morphologically [8,31], as well as in in vivo studies [6]. Ethanol intoxication is associated with changes in the activity of neurons in the central nervous system. However, the cellular and molecular mechanisms underlying these changes are still poorly understood. The disturbance of Ca 21 homeostasis has been proposed as a common step in the development of cytotoxicity and / or cell dysfunction in the CNS. A sustained increase in cytosolic Ca 21 concentration, different from the rapid and transient changes occurring in physiological condition, is invariably associated with neuronal damage [26,27]. The increase in Ca 21 appears to mediate the toxicity of several known neurotoxic agents (cyanide, chlordecone, various heavy metals), which induce either alterations in the physical integrity of the plasma membrane or mitochondrial impairment and consequent ATP depletion [16]. These results suggest that ethanol induces one or more of the mechanisms involved in the cytosistolic buffering of intracellular Ca 21 . Thus intoxication elevated intracellular Ca 21 levels may participate in adaptation of the central nervous system to ethanol and / or to occurrence of ethanol induced neuronal damage. The present study demonstrated the loss of neurons from hippocampal regions following short-term ethanol expo-
sure. These results are similar to earlier reports on animals chronically treated with ethanol [18,19] and on human alcoholics [3]. The calbindin-IR staining has confirmed these findings, significant alterations of calbindin-IR neurons were also observed in experimental mice. Our study also showed the hippocampal neurons appear to be differentially susceptible to toxic effects of ethanol both within and between cell types. The resistance or selective vulnerability of different hippocampal neuron populations seemed to be inversely correlated with the calbindin-D28K contents of cells. It has been assumed that neurons containing certain intracellular calcium binding proteins may have a greater capacity to buffer Ca 21 and therefore would be more resistant to degeneration [14]. A previous study demonstrated that cultured hippocampal neurons expresses calbindin D28k-IR resistant to neurotoxicity induced either by glutamate or calcium ionophore [23]. It is of interest that in our results, the CA3 is more vulnerable than the other subfields of hippocampus, since the immunostaining results showed that the CA3 fails to show the evidence of granular cells containing calbindin-IR neuron. The other fact, CA1 cells are more resistant to the neurotoxic effect of ethanol than CA2 cells because the CA1 subfield contains a higher proportion of calbindin D28k positive pyramidal cells than the CA2 subfield. Rami et al. [32] also showed CA1 cells are more immunoreactive than the CA2 cells and the pyramidal cells of the CA2 subfield are more vulnerable than CA1 to ischemia. Compared to rodents, a number of clinical and experimental reports on epileptic damage indicates that calcium-binding proteins are concentrated in the CA2 area of the monkey or human hippocampus, so called the ‘resistant zone’ [17,34]. Quantitative estimations of changes in calbindin D28kIR and GFAP-IR provide a valuable index for monitoring the neuronal-degeneration and reactive astrocytes. The
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64
decrease in the calbindin-D28K contents of these neurons was accompanied by cell damage. According to our results, we suggest that the decrease of calbindin-positive neurons reflected the loss of neurons. Our present data demonstrated that the observed decrease in the total number of calbindin-positive neurons is also directly related to the increased amount of reactive astrocytes among the subfields. The increase in the number of GFAPIR astrocytes in hippocampal subfields might reflect increases in GFAP synthesis. New recent evidence has suggested that astrocytes have important roles in the CNS function including synthesis and / or secretion neurotrophic factors [11], and aid repair of injured neuronal tissue [22]. The data quoted above demonstrated that short-term ethanol exposure produced distinct alteration in calbindin D28k-IR neurons and neighboring GFAP-IR astrocytes. The diminished processes of calbindin positive neurons might reflect in a retraction of the neuronal dendritic arbor which play crucial role in cell-to-cell communication. A previous study of targeted deletion in astrocyte intermediate filament suggested that GFAP is important for astrocytes-neuronal interactions and astrocytes processes play a vital role in modulating synaptic efficacy in the CNS [24]. Thus, it should be further emphasized in the present study that short-term administered ethanol disruption of GFAP and calbindin could affect neuronal-astroglial interaction, resulting in disturbance of behaviors dependent on hippocampus. The link established here between calcium binding protein such as calbindin-IR and GFAPIR provide valuable evidence for discoveries concerning neuronal-astrocytes interactions.
Acknowledgements This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sport and Culture of Japan (09307018). We would like to thank Dr. Gail S. Tucker for her constructive comments and suggestions, and Mr. Kensaku Miyamoto and Mrs. Mizue Fukutomi for their technical assistance.
References [1] M.S. Airaksinen, J. Eilers, O. Garaschuk, H. Thoenen, A. Konnerth, M. Meyer, Ataxia and altered dendritic calciums signaling in mice carryng a targeted null mutations of calbindin D28k gene, Proc. Natl Acad. Sci. USA. 94 (1997) 1488–1493. [2] M. Aschner, R.M. LoPachin Jr., Astrocytes: Targets and mediators of chemical-induced CNS injury, J. Toxicol. Environ. Health 38 (1993) 329–342. [3] O. Bengochea, L.M. Gonzalo, Effect of chronic alcoholism on human hippocampus, Histol. Histopathol. 5 (1990) 349–357. [4] D.J. Bonthius, J.R. West, Ethanol induced neuronal lost in the developing rats: increased brain damage with binge exposure, Alcohol. Clin. Exp. Res. 14 (1990) 107–118. [5] M.A. Collins, T.D. Corso, E.J. Neafsey, Neuronal degeneration in
[6]
[7]
[8]
[9]
[10] [11]
[12]
[13]
[14] [15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23]
[24]
[25]
63
rat cerebrocortical and olfactory regions during subchronic ‘binge’ intoxication with ethanol: possible explanation for olfactory deficits in alcoholics, Alcohol. Clin. Exp. Res. 20 (1996) 284–292. D.L. Davies, W.E. Cox Dela, Delayed growth and maturation of astrocytic cultures following exposure to ethanol: electron microscopic observations, Brain Res. 547 (1991) 53–61. L.D. Devenport, R.L. Hale, Contributions of hippocampus and neocortex to the expression of ethanol effects, Psychopharmacology (Berl.) 99 (1989) 337–344. D. Durand, J.A. Saint-Cyr, N. Gurevich, P.L. Carlen, Ethanolinduced dendritic alterations in hippocampal granule cells, Brain Res. 477 (1989) 373–377. H. Franke, H. Kittner, P. Berger, K. Wirkner, J. Schramek, The reaction of astrocytes in the hippocampus of adult rats during chronic ethanol treatment and correlations to behaviour impairments, Alcohol 14 (1997) 445–454. G. Freund, Alcohol withdrawal syndrome in mice, Arch. Neurol. 21 (1969) 315–320. S. Furukawa, Y. Furukawa, E. Satoyoshi, K. Hayashi, Synthesis and secretion of nerve growth factor by mouse astroglia cells in culture, Biochem. Biophys. Res. Commun. 136 (1986) 57–63. C.R. Goodlett, B.L. Marcussen, J.R. West, A single day of alcohol exposure during the brain growth spurt induces brain weight restriction and cerebellar Purkinje cell loss, Alcohol 7 (1990) 107– 114. D.L. Gruol, J.G. Curry, Calcium signals elicited by quisqualate in cultured Purkinje neurons show developmental changes in sensitivity to acute alcohol, Brain Res. 673 (1995) 1–12. C.W. Heizmann, Calcium signaling in the brain, Acta Neurobiol. Exp. (Warsz) 51 (1993) 15–23. H.K. Kimelberg, D.M. Katz, High-affinity uptake of serotonin into immunocytochemically identified astrocytes, Science 228 (1985) 889–891. H. Komulainen, S.C. Bondy, Increased free intracellular Ca21 by toxic agents: an index of potential neurotoxicity?, Trends Pharmacol. Sci. 9 (1988) 154–156. C. Leranth, C.E. Ribak, Calcium binding protein are concentrated in the CA2 field of the monkey hippocampus: a possible key to this region’s resistance to epileptic damage, Exp. Brain Res. 85 (1991) 129–136. N.V. Lukoyanov, M.D. Madeira, M.M. Paula-Barbosa, Behavioral and neuroanatomical consequence of chronic ethanol intake and withdrawal, Physiol. Behav. 66 (1999) 337–346. C. Lundqvist, C. Alling, R. Knoth, B. Volk, Intermittent ethanol exposure of adults rats: hippocampal cell loss after one month of treatment, Alcohol Alcohol. 30 (1995) 737–748. M.D. Madiera, J.P. Andrade, A.R. Lieberman, N. Sousa, O.F.X. Almeida, M.M. Paula-Barbosa, Chronic alcohol consumption and withdrawal do not induce cell death in the suprachiasmatic nucleus, but lead to irreversible depression of peptide immunoreactivity and mRNA levels, J. Neurosci. 17 (1997) 1302–1319. P.M. Martin, J.P. O’Callaghan, A direct comparison of GFAP immunocytochemistry and GFAP concentration in various regions of ethanol-fixed rat and mouse brain, J. Neurosci. Methods 58 (1995) 181–192. A.J. Mathewson, M. Berry, Observation on astrocytes response to the cerebral wound in adult rats, Brain Res. 327 (1985) 61–69. M.P. Mattson, B. Rychlik, C. Chu, S. Christakos, Evidence for calcium-reducing and excitoprotective roles for the calcium-binding protein-D28k in cultured hippocampal neurons, Neuron 6 (1991) 41–51. M.A. McCall, R.G. Gregg, R.R. Behringer, M. Brenner, C.L. Delaney, E.J. Galbreath, C. Zhang L, R.A. Pearce, S.Y. Chiu, A. Messing, Targeted deletion in astrocytes intermediate filament (GFAP) alters neuronal physiology, Proc. Natl. Acad. Sci. USA 93 (1996) 6361–6366. E.K. Michaelis, Fetal alcohol exposure: Cellular toxicity and
64
[26]
[27] [28]
[29] [30]
[31]
I. Satriotomo et al. / Brain Research 879 (2000) 55 – 64 molecular events involved intoxicity, Alcohol. Clin. Exp. Res. 14 (1990) 819–826. P. Nicotera, G. Bellomo, S. Orrenius, Calcium mediated mechanisms in chemically induced cell death, Ann. Rev. Pharmacol. Toxicol. 32 (1992) 449–470. S. Orrenius, P. Nicotera, The calcium ion and cell death, J. Neural Transm. 43 (1994) 1–11. D.R. Pierce, J.R. West, Alcohol-induced microcephaly during the third trimester equivalent: relationship to dose and blood alcohol concentration, Alcohol 3 (1986) 185–191. D.R. Pierce, J.R. West, Blood alcohol concentration: a critical factor for producing fetal alcohol effects, Alcohol 3 (1986) 269–272. B. Pfeiffer, A.W. Norman, B. Hamprecht, Immunocytochemical characterization of neuron-rich rat brain primary cultures: calbindin D28k as marker of neuronal subpopulation, Brain Res. 476 (1989) 120–128. S.C. Philips, B.G. Cragg, Chronic consumption of alcohol by adult
[32]
[33]
[34]
[35]
mice: Effect on hippocampal cells and synapses, Exp. Neurol. 80 (1983) 218–226. A. Rami, A. Rabie, M. Thomasset, J. Krieglstein, Calbindin-D28k and ischemic damage of pyramidal cells in rat hippocampus, J. Neurosci. Res. 31 (1992) 89–95. I. Satriotomo, T. Miki, M. Itoh, Q. Xie, K. Ameno, Y. Takeuchi, Effect of short-term ethanol exposure on the suprachiasmatic nucleus of hypothalamus: immunohystochemical study in mice, Brain Res. 847 (1999) 124–129. R.S. Sloviter, A.L. Sollas, N.M. Barbaro, K.D. Laxer, calcium binding protein (calbindin-D28k) and parvalbumin immunocytochemistry in the normal and epileptic human hippocampus, J. Comp. Neurol. 308 (1991) 381–396. J.Y. Zou, D.B. Martinez, E.J. Neafsey, M.A. Collins, Binge ethanol– induced brain damage in rats: effect of inhibitors of nitric oxide synthase, Alcohol. Clin. Exp. Res. 20 (1996) 1406–1411.