- Email: [email protected]
Damage and plasticity in adult rat hippocampal trisynaptic circuit neurons after neonatal exposure to glutamate excitotoxicity
Int. J. Devl Neuroscience 27 (2009) 741–745
Contents lists available at ScienceDirect
International Journal of Developmental Neuroscience journal homepage: www.elsevier.com/locate/ijdevneu
Damage and plasticity in adult rat hippocampal trisynaptic circuit neurons after neonatal exposure to glutamate excitotoxicity I. Gonza´lez-Burgos a,c, D.A. Vela´zquez-Zamora a, C. Beas-Za´rate b,c,* a
Laboratorio de Psicobiologı´a, Divisio´n de Neurociencias, Centro de Investigacio´n Biome´dica de Occidente, IMSS., Guadalajara, Mexico Laboratorio de Neurobiologı´a Celular y Molecular, Divisio´n de Neurociencias, Centro de Investigacio´n Biome´dica de Occidente, IMSS., Guadalajara, Mexico c Depto. de Biologı´a Celular y Molecular, CUCBA, Universidad de Guadalajara, Guadalajara, Jal., Mexico b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 5 May 2009 Received in revised form 26 August 2009 Accepted 27 August 2009
Hippocampal vulnerability to excitotoxicity has been widely studied along with its implication to learning and memory. Neonatal glutamate excitotoxicity induces loss of CA1 pyramidal neurons in adult rats concomitantly with some plastic changes in the dendritic spines of surviving neurons. At least in part, these may underlie the place learning impairments seen in previous studies based on a similar excitotoxicity-inducing model. In the present study, cytoarchitecture of dentate gyrus, CA3 and CA1 fields were evaluated in 120-day-old rats, after they had been neonatally treated with glutamate as monosodium salt. Dentate granule cells and CA1 pyramidal neurons were less than those counted in NaCl-treated control animals. In addition, dentate granule cells had more dendrites as well as more branched spines. Spine density in CA1 pyramidal neurons was greater than in the controls. Additionally, thin and mushroom spines were proportionally more abundant in monosodium glutamate-treated animals. No effects were seen in the hippocampal CA3 field. Our results strongly suggest a long-term induction of plastic changes in the cytoarchitecture of the hippocampal trisynaptic circuit neurons after cell death provoked by the monosodium glutamate-induced excitotoxicity. These plastic events as well as the aberrant expression of the glutamate NMDA receptors resulting from monosodium glutamate neonatal treatment could be strongly associated with the place learning impairments previously reported. ß 2009 ISDN. Published by Elsevier Ltd. All rights reserved.
Keywords: Glutamate Hippocampus Excitotoxicity Trisynaptic circuit Dendritic spines
The hippocampus is one of the most vulnerable regions of the brain to excitotoxic agents (Pulsinelli, 1985). The mechanisms underlying such differential and specific vulnerability are unclear but the presence and high density of N-Methyl-D-Aspartate (NMDA) glutamate receptors seems to play a crucial role (Schmidt-Kastner et al., 1990); in fact, overstimulation of NMDA receptors has been shown to trigger cell death (Rivera-Cervantes et al., 2004; Gasco´n et al., 2008; Martel et al., 2009). In the hippocampus, NMDA receptors are located mostly on the dendritic spines of those neurons constituting the trisynaptic circuit, i.e., dentate gyrus granule cells, CA3 pyramidal neurons, and CA1 pyramidal cells (Siegel et al., 1995). Briefly, the hippocampal trisynaptic circuit is integrated by the terminal axons of the enthorrinal cortex-proceeding perforant pathway establishing synaptic contacts with the dendritic spines located in the two-thirds distal dendritic segment of the dentate gyrus granule cells. Then, axons from granule cells (mossy fibers) reach
* Corresponding author at: Sierra Mojada # 800, Col. Independencia, Guadalajara, Jalisco 44340, Mexico. E-mail address: [email protected] (C. Beas-Za´rate). 0736-5748/$36.00 ß 2009 ISDN. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdevneu.2009.08.016
the hippocampal CA3 subfield and synapse with the thorny excrescences of pyramidal neurons and finally, those axonal bifurcations constituting the Schaffer collaterals emerging from the main axon of the CA3 pyramidal cells establish synaptic contacts with the spines of the dendritic arbor of the CA1 pyramidal neurons (Amaral and Witter, 1995). Previous studies from our group have shown that neonatal exposure to glutamate, as monosodium salt, induces cell death and some cytoarchitectural impairments both in prefrontal (Gonza´lezBurgos et al., 2001) and hippocampal (Beas-Za´rate et al., 2002b) pyramidal neurons in young adult rats, as well as place learning impairments (Olvera-Corte´s et al., 2005). Spatial orientation abilities are sustained by the coordinated activity of the hippocampal trisynaptic circuit neurons (Goodrich-Hunsaker et al., 2008), and some plastic changes in hippocampal neurons’ dendritic spines have been shown to be closely associated with place learning (Gonza´lezBurgos et al., 2005). However, it is unknown whether glutamate receptor overstimulation induced by early exposure to high glutamate concentrations could modify the cytoarchitecture of neurons comprising the trisynaptic circuit in the long term. Therefore, the aim of this work was to evaluate the cytoarchitectonic parameters of the hippocampal neurons critically involved in the
742
I. Gonza´lez-Burgos et al. / Int. J. Devl Neuroscience 27 (2009) 741–745
flow of information through the trisynaptic circuit in adult rats after neonatal treatment with MSG. 1. Experimental procedures Twenty-four neonatal Sprague–Dawley male rats were used, which were obtained from the litters of twelve normal female rats. Before birth and during lactation, dams were maintained under standard conditions: regular 12-h light– dark cycles (07:00–19:00 h), environmental humidity 45–50%, temperature 22 2 8C, and free access to food and water. The dams kept one rat to a cage and, after weaning, the pups used for this study were kept in the same conditions as their mothers, until the day of sacrifice. Experimental procedures were conducted to minimize pain and discomfort for the animals, and performed in accordance with the NIH guide for Care and Use of Laboratory Animals (NIH Publications No. 80-23) 1996 revision, and approved by the Research Ethics Committee of the Instituto Mexicano del Seguro Social, Me´xico. Each pup was randomly assigned to one of three groups of study. Eight experimental animals (MSG group) were subcutaneously injected with a 50% aqueous solution of monosodium L-glutamic acid (MSG) at a dose of 4 mg/g of body weight, on postnatal days one, three, five and seven. As a control group (NaCl group), eight pups were given an equimolar aqueous solution of NaCl (17.27%), at a dose of 1.37 mg/g of body weight. The remaining eight rats were untreated (Intact group). During and after treatment, animals from the three groups remained with their original litter until weaning at 21 days and then were caged until they reached four months of age. At 120 days of age, animals were anesthetized with ethyl ether and thereafter perfused with 100 ml of a washing phosphate-buffered solution (pH 7.4; 0.01 M) containing 1000 IU/l of sodium heparin as anticoagulant, and 1 g/l of procaine hydrochloride as vasodilator (Feria-Velasco and Karnovsky, 1970). Then, 200 ml of a fixing phosphate-buffered 4% formaldehyde solution was perfused. Both solutions flowed at a rate of 11.5 ml/min. Each brain remained for at least 48 h in 100 ml of a fresh fixing solution. The bilateral whole hippocampi were dissected out, and four right and four left hippocampi per group were used for either cresyl violet staining or Golgi impregnation, correspondingly. As a result, eight hippocampi per group were processed histologically and embedded in paraffin, while the remaining eight hippocampi were impregnated for the Golgi study. Once embedded in paraffin, each corresponding tissue block was then sectioned using a rotation microtome to yield serial coronal sections 7 mm thick through each hippocampus, and a systematic random sampling procedure was designed to conduct an unbiased stereological study of the granule cell layer of the dentate gyrus (DG), and the CA3 and CA1 fields (Paxinos and Watson, 1986). The first of these sections, used as anatomical reference, presented the beginning of any of the three regions; it was consequently numbered as slice number one. For each rat, a number between 1 and 100 was randomly selected by lottery and used to determine the first section to be sampled for analysis, and this and the consecutive section was used as the ‘‘pair of sections’’ number one. A new random number was selected for each rat studied. The selected section and the consecutive serial section was placed on the same glass slide and then stained with cresyl violet. The following 98 serial sections were discarded, with the subsequent two being sampled as before. This procedure was repeated until the whole DG, CA3 and CA1 regions were completely sectioned. As a result and based on pilot experiments, 6–8 pairs of consecutive sections were used for the study. The stereological study was conducted under ‘‘blind’’ conditions and each slide was identified with a code not revealed until the end of quantifications. The DG, CA3 and CA1 volume was determined by applying the principle of Cavalieri (Gundersen and Jensen, 1987; Pakkenberg and Gundersen, 1988), and the cell number of the
three regions studied was estimated using the dissector principle (Pakkenberg and Gundersen, 1988). Cell counts were made of the total number of nucleoli profiles appearing in photomicrographs from the ‘‘look-up’’ sections but not appearing in the adjacent ‘‘test’’ sections. Cell counts were conducted using the ‘‘forbidden line’’ rule (Gundersen, 1977). The remaining four left and four right hippocampi per group were processed according to a modification of the Golgi method (Gonza´lez-Burgos et al., 1992). Several coronal slices 100 mm thick were mounted on one slide per animal, and eight granule cells from DG, eight CA3 and eight CA1 pyramidal neurons from the dorsal hippocampus were studied per rat. The cytoarchitectonic parameters studied in the hippocampus were: (1) dendritic arborization of both DG granule cells and CA1 pyramidal neurons, (2) area occupied by the thorny excrescences of the long-shaft CA3 pyramidal neurons (Fitch et al., 1989), and (3) spine density and proportion of the thin, stubby, mushroom, wide, double, and branched spines (Fig. 1; see Harris et al., 1989; ˜ a et al., 2000; Gonza´lez-Burgos et al., 2005) of granule cells in the DG Tarelo-Acun and CA1 pyramidal neurons. The dendritic arborization was evaluated by counting the number of dendritic bifurcations in the molecular layer of the DG or in the stratum oriens and radiatum of the CA1 field. The area of thorny excrescences was measured with an image analyzer (Zeiss Image 3.0). Both the spine density and proportion of each type of spine in the DG granule cells were counted in a segment of 50 mm medial in a half primary dendrite distal to the soma. Likewise, in the CA1 pyramidal neurons both spine density and spine types were counted in a segment of 50 mm medial in an oblique dendrite protruding from its parent apical dendrite and in 50 mm medial of a primary basilar dendrite. Spine counting was made by direct observation at 2000 using a magnification changer coupled to the light microscope. The one-way ANOVA and Tukey post hoc tests were used for statistical comparisons relating to stereological measurements, dendritic arborization, area of thorny excrescences and density of spines. In addition, one-way ANOVA and Bonferroni correction post hoc test was used for the proportional density of the different types of spines.
2. Results 2.1. Stereology Tissue volume of the DG and CA1 hippocampal regions were different between groups (F = 14.672, p < 0.0001, CE = 0.05; and F = 8.337, p < 0.002, CE = 0.05; respectively). DG from MSG rats was less voluminous than those from both Intact (p < 0.0001) and NaCl (p < 0.006) animals. Likewise, CA1 volume in MSG-treated rats was lesser than that from both Intact (p < 0.005) and NaCltreated (p < 0.006) animals. There were no differences between Intact and NaCl groups neither in DG nor in CA1 tissue volume. As to CA3 region, no statistical differences among groups were observed between the three groups studied (Table 1). Cell numbers in both DG and CA1 regions was different between groups (F = 13.119, p < 0.0001, CE = 0.05 and F = 8.592, p < 0.002, CE = 0.05; respectively). The number of DG cells of the MSG group was less than that counted in Intact (p < 0.001) and NaCl (p < 0.0001) groups. The number of cells of Intact and NaCl groups was not different. Statistical analysis of this parameter in the CA3
Fig. 1. Photomicrographs showing the six types of spines studied (arrows), and the corresponding diagrams (upper panel). (A) Thin; (B) stubby; (C) mushroom; (D) wide; (E) double; and (F) branched. Scale bar: 5 mm.
I. Gonza´lez-Burgos et al. / Int. J. Devl Neuroscience 27 (2009) 741–745 Table 1 Tissue volume (mm3) of the three hippocampal regions studied.
Dentate gyrus CA3 CA1
Intact
NaCl
MSG
2.61 0.01 4.16 0.01 1.84 0.01
2.57 0.01 4.16 0.01 1.83 0.01
2.50 0.01a,b 4.13 0.01 1.75 0.01a,b
Mean SEM. p < 0.05. a MSG vs. Intact. b MSG vs. NaCl.
743
Table 2 Spine density (number of spines) and proportional density (%) of the different types of spines counted in a 50 mm-length distal segment of a primary dendrite of granule cells from the hippocampal dentate gyrus. Intact
NaCl
MSG
Density
55.6 1.6
58.8 2.0
61.7 2.7
Spines proportion Thin Mushroom Stubby Wide Branched Double
24.8 0.7 16.4 0.5 11.5 0.1 2.1 0.1 0.2 0.07 0.02 0.02
27.5 1.1 18.6 1.0 9.9 0.5 2.1 0.2 0.2 0.04 0.04 0.02
27.9 1.3 18.8 0.7 11.7 0.6b 2.6 0.1 0.6 0.09a,b 0.1 0.05
Mean SEM. p < 0.05. a MSG vs. Intact. b MSG vs. NaCl.
Table 3 Spine density (number of spines) and proportional density (%) of the different types of spines counted in a segment 50 mm-length of a secondary dendrite of pyramidal neurons from the hippocampal CA1 field.
Fig. 2. Graphic comparison between groups, regarding to the number of cells counted in the three hippocampal regions studied. Mean SEM. (a) MSG vs. Intact; (b) MSG vs. NaCl. p < 0.05.
hippocampal region showed no differences between groups in this parameter. For its part, cells in the CA1 field of MSG-treated rats were less than those counted in CA1 from both Intact (p < 0.003) and NaCl animals (p < 0.008), and the number of CA1 cells from Intact and NaCl groups was not different (Fig. 2). 2.2. Golgi studies Dendritic arborization of DG granule cells was different in the three groups studied (F = 12.077, p < 0.001). MSG granule cells showed more bifurcations than both Intact (p < 0.002) and NaCl (p < 0.002) groups; no difference in this parameter was observed between the Intact and NaCl groups of animals (Fig. 3). Density of spines in granule cells was not different between the groups, but the proportional density of stubby and branched spines was increased in the experimental group (F = 4.634, p < 0.027 and
Intact
NaCl
MSG
Density
80.5 2.5
87.3 1.4
91.9 4.7a
Spines proportion Thin Mushroom Stubby Wide Branched Double
37.4 1.4 25.2 0.9 14.3 0.7 3.1 0.3 0.2 0.06 0.06 0.02
39.2 0.8 27.3 0.6 15.2 0.5 3.4 0.4 0.3 0.1 0.08 0.04
44.7 1.3a,b 31.3 1.3a,b 15.7 0.5 4.2 0.5 0.3 0.06 0.2 0.07
Mean SEM. p < 0.05. a MSG vs. Intact. b MSG vs. NaCl.
F = 6.101, p < 0.012; respectively). Stubby spines in MSG granule cells were more abundant than in the NaCl group (p < 0.04), and branched spines in MSG group were more abundant than in the Intact (p < 0.02) and NaCl (p < 0.02) groups. No significant differences in the proportion of thin, mushroom, wide, and double spines in the DG granule cells were observed between the three groups studied (Table 2). The area of thorny excrescences of Intact, NaCl, and MSG groups was not different (data not shown). The dendritic arborization of CA1 pyramidal neurons was not different between the three groups studied (data not shown). As to spine density, a significant difference between groups was seen (F = 8.729, p < 0.003). Rats from the MSG group had more dendritic spines than animals from the Intact (p < 0.002) and NaCl (p < 0.04) groups; the density of spines in Intact and NaCl control groups was not different. In addition, proportional density of thin (F = 9.106, p < 0.003) and mushroom (F = 9.124, p < 0.003) spines were different between groups. Thin spines of CA1 pyramidal cells from MSG were more abundant than those from both Intact (p < 0.003) and NaCl (p < 0.022). Likewise, mushroom spines were more abundant in the MSG group than in the Intact (p < 0.002) and NaCl (p < 0.049) groups. There was no statistical difference in the proportional density of thin and mushroom spines between the Intact and NaCl control groups. Proportional density of stubby, wide, branched, and double spines was not different (Table 3). 3. Discussion
Fig. 3. Graph showing the comparison of dendritic bifurcations counted in dentate gyrus granule cells, between the three groups studied. Mean SEM. (a) MSG vs. Intact; (b) MSG vs. NaCl. p < 0.05.
The hippocampus is a limbic structure strongly related with spatial learning and memory (O’Keefe and Nadel, 1978) by virtue of the concerted activity of the hippocampal trisynaptic circuit
744
I. Gonza´lez-Burgos et al. / Int. J. Devl Neuroscience 27 (2009) 741–745
neurons (Goodrich-Hunsaker et al., 2008) from the dorsal hippocampus (Goodrich-Hunsaker et al., 2005). Glutamate-mediated excitotoxicity in the hippocampus underlies, at least in part, in an overstimulation of the glutamate NMDA receptors (Choi, 1988; Rivera-Cervantes et al., 2004) located on the external membrane of dendritic spines from those neurons constituting the hippocampal trisynaptic circuit. In fact, cerebral ischemia-inducing excitotoxicity leads to cell death in the hippocampus, along with place learning and memory impairments (Letechipı´a-Vallejo et al., 2007) similar to those seen in adult rats after neonatal MSG treatment (Olvera-Corte´s et al., 2005). In the present study, dentate gyrus and CA1 field of the MSGtreated animals were less voluminous than those from controls. This corresponded with the reduced cell population seen in the dentate gyrus (8%) and the CA1 field (15%). This data is in agreement with previous studies showing that hippocampal CA1 pyramidal neurons and dentate granule cells are among the most vulnerable nervous system cells to excitotoxicity (Pulsinelli, 1985). Tissue volume and cell numbers in the hippocampal CA3 field were not affected after MSG treatment, which could be sustained on the low density of glutamate NMDA receptors existing on the thorny excrescences of the CA3 pyramidal neurons (Jaffe and Brown, 1997; Reid et al., 2001) where mossy fibers synapse. Dentate granule cells of MSG-treated rats had more dendrites than the controls. Expressions of the different NMDA receptor subunits have been shown to affect dendritic growth of hippocampal neurons during development (Scheetz and Constantine-Paton, 1994). In this sense, changes in the NMDA receptor subunits were previously reported in adult rats as a result of the neonatal exposure to MSG (Beas-Za´rate et al., 2002a; RiveraCervantes et al., 2009). Despite spine density in dendrites of dentate granule cells was not modified, the fact that these neurons showed more dendritic ramifications allows us to assume that they had more spines per neuron. We also found that branched spines were more abundant than those seen in granule cells from control animals. Branched spines have been strongly related both with spatial learning and induction of long-term potentiation (LTP) (Moser et al., 1994; Muller et al., 2000), which has been associated with the acquisition and consolidation of learning and memory (Bliss and Lømo, 1973; Morris et al., 1986). In this sense, after LTP induction perforation of macular synapses takes place (Geinisman et al., 1993; Toni et al., 2001), and perforated synapses ultimately divide into two small independent spines through the transient formation of a branched spine (Moser et al., 1994; Muller et al., 2000). Thus, along with the putative increase of total spines, the proportional increase of branched spines seen in the MSG-treated animals could be interpreted as compensatory events after the excitotoxic effects exerted by MSG on granule cells during development, such as cell death. As a result, these putative plastic events could influence the organization of the metric spatial information in which dentate granule cells are closely involved (Goodrich-Hunsaker et al., 2008; Kesner et al., 2004). Glutamate-inducing neurotoxicity is strongly related with the entry of excessive calcium to postsynaptic cells (Gasco´n et al., 2008) after glutamate NMDA receptor stimulation (Grishin et al., 2004). Based on these, the lack of significant damage or plastic changes to CA3 pyramidal neurons after MSG treatment could be due to the fact that thorny excrescences of CA3 pyramidal neurons have a low density of glutamate NMDA receptors (Jaffe and Brown, 1997; Reid et al., 2001). CA3 pyramidal neurons are involved in the metric representations of spatial information processed by granule cells (Goodrich-Hunsaker et al., 2008) by encoding them within short-term memory (Kesner, 2007). Because these two components of the trisynaptic circuit act in concert through the mossy fibers to process metric representations of space (GoodrichHunsaker et al., 2008; Kesner, 2007), these findings would mean
that the putative alterations of spatial information proceeding from the dentate granule cells would in turn be abnormally processed by the CA3 pyramidal neurons. CA1 pyramidal neurons receive CA3-proceeding Schaffer collaterals after which, CA1 efferent fibers leave the hippocampus to innervate several brain regions (Amaral and Witter, 1995). Lesion of the dorsal hippocampus has been shown to impair the topological representations of space (Goodrich-Hunsaker et al., 2005; Rogers and Kesner, 2006), a cognitive ability closely related with the functional activity of the hippocampal CA1 field (Goodrich-Hunsaker et al., 2008). Topology of space consists in those inclusion, vicinity, adjacency, and order-related properties of objects in the surrounding space. Since the CA1 field constitutes the last ‘‘relay station’’ before the space-related information flowing through the trisynaptic circuit leaves the hippocampus, its functional activity would be related with the integration of both the dentate gyrus-and-CA3-dependent metric properties of space, as well as the topological information necessary to achieve spatial information to resolve a given space-related behavioral task. Spine density and proportional density of thin and mushroom spines were greater in CA1 pyramidal neurons. The greater the amount of spines, the nearer they are, and then, the greater their capacity is to associate with neighbor synaptic stimuli (Harris and Kater, 1994; De Roo et al., 2008). On the other hand, it is well known that both thin and mushroom spines are related with cognitive efficiency (Kasai et al., 2003). Accordingly, it would be expected that hippocampal-dependent cognitive abilities would be highly efficient in those MSG-treated animals. However, in spite of the putative plastic events that took place in the hippocampal trisynaptic circuit neurons after cell death, behavioral performance in the Morris maze was severely impaired according to a previous study from our work group using the same experimental paradigm (Olvera-Corte´s et al., 2005) than that used in the present study. Yet, it has been reported that place learning is performed normally even after up to a 60% loss of CA1 neurons in some ischemiainducing models (Block, 1999), a much higher value than the 15% observed in this study. In addition, plastic events observed, i.e., more profuse dendritic arborization, greater density of spines, and a higher proportion of those spines related with the highly efficient processing of mnemonic information such as branched, thin, and stubby spines (Bourne and Harris, 2007; Kasai et al., 2003; Matsuzaki et al., 2004; Muller et al., 2000; Yang et al., 2008); were not enough to ‘‘protect’’ the functional integrity of the hippocampal trisynaptic circuit as to perform efficiently in the Morris maze. Based on these results, some other neuropathological and/or plastic events could be involved in the place learning impairments occurred after neonatal exposure to glutamate neurotoxicity previously reported. NMDA receptors anchored to postsynaptic density scaffolding proteins mediate plastic remodeling of excitatory synapses and dendritic spines (Calabrese et al., 2006; Schubert and Dotti, 2007). Therefore, since NMDA receptor composition disrupts due to MSG-mediated excitotoxicity (BeasZa´rate et al., 2002a; Rivera-Cervantes et al., 2009), the pathological/plastic modifications in dendritic spines seen in this study could be related to changes in NMDA receptor. This hypothesis must be examined in further experiments. References Amaral, D.G., Witter, M.P., 1995. Hippocampal formation. In: Paxinos, G. (Ed.), The Rat Nervous System. Academy Press, London, pp. 443–493. Beas-Za´rate, C., Flores-Soto, M.E., Armendariz-Borunda, J., 2002a. NMDAR-2C and 2D subunits gene expresio´n is induced in brain by neonatal exposure of monosodium L-glutamate to adult rats. Neurosci. Lett. 321, 9–12. Beas-Za´rate, C., Pe´rez-Vega, M.I., Gonza´lez-Burgos, I., 2002b. Neonatal exposure to monosodium L-glutamate induces loss of neurons and cytoarchitectural alterations in hippocampal pyramidal neurons of adult rats. Brain Res. 952, 275– 281.
I. Gonza´lez-Burgos et al. / Int. J. Devl Neuroscience 27 (2009) 741–745 Bliss, T.V., Lømo, T., 1973. Long-lasting potentiation of synaptic transmission in the dentate area of the anesthetized rabbit following stimulation of the perforant path. J. Physiol. 232, 331–356. Block, F., 1999. Global ischemia and behavioral deficits. Prog. Neurobiol. 58, 279–295. Bourne, J., Harris, K.M., 2007. Do thin spines learn to be mushroom spines that remember? Curr. Opin. Neurobiol. 17, 1–6. Calabrese, B., Wilson, M.S., Halpain, S., 2006. Development and regulation of dendritic spine synapses. Physiology 21, 38–47. Choi, D.W., 1988. Calcium-mediated neurotoxicity: relationship to specific channel types and role in ischemic damage. Trends Neurosci. 11, 465–469. De Roo, M., Klauser, P., Muller, D., 2008. LTP promotes a selective long-term stabilization and clustering of dendritic spines. PLoS Biol. 6, 1850–1860. Feria-Velasco, A., Karnovsky, M., 1970. Preservacio´n o´ptima del sistema nervioso central por perfusio´n con glutaraldehı´do para estudio ultraestructural. Arch. Invest. Med. 1, 201–220. Fitch, J.M., Juraska, J.M., Washington, L.M., 1989. The dendritic morphology of pyramidal neurons in the rat hippocampal CA3 area. l. Cell types. Brain Res. 479, 105–114. ˜ a, A., Dı´az-Guerra, M., 2008. Gasco´n, S., Sobrado, M., Roda, J.M., Rodrı´guez-Pen Excitotoxicity and focal cerebral ischemia induce truncation of the NR2A and NR2B subunits of the NMDA receptor and cleavage of the scaffolding protein PSD-95. Mol. Psychiatry 13, 99–114. Geinisman, Y., de Toledo-Morrell, L., Morrell, F., Heller, R.E., Rossi, M., Parshall, R.F., 1993. Structural synaptic correlate of long-term potentiation: formation of axospinous synapses with multiple, completely partitioned transmission zones. Hippocampus 3, 435–446. Gonza´lez-Burgos, I., Alejandre-Go´mez, M., Cervantes, M., 2005. Spine-type densities of hippocampal CA1 neurons vary in proestrus and estrus rats. Neurosci. Lett. 379, 52–54. Gonza´lez-Burgos, I., Pe´rez-Vega, M.I., Beas-Za´rate, C., 2001. Neonatal exposure to monosodium glutamate induces cell death and dendritic hypotrophy in rat prefrontal pyramidal neurons. Neurosci. Lett. 297, 69–72. Gonza´lez-Burgos, I., Tapia-Arizmendi, G., Feria-Velasco, A., 1992. Golgi method without osmium tetroxide for the study of the central nervous system. Biotech. Histochem. 67, 288–296. Goodrich-Hunsaker, N.J., Hunsaker, M.R., Kesner, R.P., 2005. Dissociating the role of the parietal cortex and dorsal hippocampus for spatial information processing. Behav. Neurosci. 119, 1307–1315. Goodrich-Hunsaker, N.J., Hunsaker, M.R., Kesner, R.P., 2008. The interactions and dissociations of the dorsal hippocampus subregions: how the dentate gyrus, CA3, and CA1 process spatial information. Behav. Neurosci. 122, 16–26. Grishin, A.A., Gee, C.E., Gerber, U., Benquet, P., 2004. Differential calcium-dependent modulation of NMDA currents in CA1 and CA3 hippocampal pyramidal cells. J. Neurosci. 24, 350–355. Gundersen, H.J.G., 1977. Notes on the estimation of the numerical density of arbitrary profiles: the edge effect. J. Microsc. 111, 219–223. Gundersen, H.J.G., Jensen, E.B., 1987. The efficiency of systematic sampling in stereology and its prediction. J. Microsc. 147, 229–263. Harris, K.M., Jensen, F.E., Tsao, B.H., 1989. Ultrastructure, development and plasticity of dendritic spine synapses in area CA1 of the rat hippocampus: extending our vision with serial electron microscopy and three-dimensional analyses. In: Chan-Palay, V., Koˆhler, Ch. (Eds.), The Hippocampus—New Vistas. Alan R. Liss, pp. 33–52. Harris, K.M., Kater, S.B., 1994. Dendritic spines: cellular specializations imparting both stability and flexibility to synaptic function. Annu. Rev. Neurosci. 17, 341– 371. Jaffe, D.V., Brown, T.H., 1997. Calcium dynamics in thorny excrescences of CA3 pyramidal neurons. J. Neurophysiol. 78, 10–18. Kasai, H., Matsuzaki, M., Noguchi, J., Yasumatsu, N., Nakahara, H., 2003. Structurestability-function relationships of dendritic spines. Trends Neurosci. 26, 360– 368. Kesner, R.P., 2007. A behavioural analysis of dentate gyrus function. Prog. Brain Res. 163, 567–576. Kesner, R.P., Lee, I., Gilbert, P., 2004. A behavioral assessment of hippocampal function based on a subregional analysis. Rev. Neurosci. 15, 333–351.
745
Letechipı´a-Vallejo, G., Lo´pez-Loeza, E., Espinoza-Gonza´lez, V., Gonza´lez-Burgos, I., Olvera-Corte´s, M.E., Moralı´, G., Cervantes, M., 2007. Long-term morphological and functional evaluation of the neuroprotective effects of post-ischemic treatment with melatonin in rats. J. Pineal Res. 42, 139–146. Martel, M.A., Sorianom, F.X., Baxter, P., Rickman, C., Duncan, R., Wyllie, D.J., Hardingham, G.E., 2009. Inhibiting pro-death NMDA receptor signaling dependent on the NR2 PDZ ligand may not affect synaptic function or synaptic NMDA receptor signaling to gene expression. Channels (Austin) 3, 12–15. Matsuzaki, M., Honkura, N., Ellis-Davies, G.C.R., Kasai, H., 2004. Structural basis of long-term potentiation in single dendritic spines. Nature 429, 761–766. Morris, R.G., Anderson, E., Lynch, G.S., Baurdy, M., 1986. Selective impairment of learning and blockade of long-term potentiation by an N-methyl D-aspartate receptor antagonist, AP5. Nature 319, 774–776. Moser, M.B., Trommald, M., Andersen, P., 1994. An increase in dendritic spine density on hippocampal CA1 pyramidal cells following spatial learning in adult rats suggests the formation of new synapses. Proc. Natl. Acad. Sci. U.S.A. 91, 12673–12675. Muller, D., Toni, N., Buchs, P.A., 2000. Spine changes associated with long-term potentiation. Hippocampus 10, 596–604. O’Keefe, J., Nadel, L., 1978. The Hippocampus as a Cognitive Map. Oxford University Press, Oxford. Olvera-Corte´s, E., Lo´pez-Va´zquez, M.A., Beas-Za´rate, C., Gonza´lez-Burgos, I., 2005. Neonatal exposure to monosodium glutamate disrupts place learning ability in adult rats. Pharmacol. Biochem. Behav. 82, 247–251. Pakkenberg, B., Gundersen, J.G., 1988. Total number of neuron and glial cells in human brain nuclei estimated by the dissector and the fractionator. J. Microsc. 150, 1–20. Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates, 2nd ed. Oxford University Press, USA. W.A. Pulsinelli, 1985. Selective neuronal vulnerability: morphological and molecular characteristics. In: K. Kogure, K.A. Hossman, B.K. Siesjo¨, F.A. Welsch (Eds.), Molecular Mechanisms of Ischemic Brain Damage. Prog. Brain Res., vol. 63. Elsevier, New York, pp. 29–37. Reid, C.A., Fabian-Fine, R., Fabian, A., 2001. Postsynaptic calcium transients evoked by activation of individual hippocampal mossy fiber synapses. J. Neurosci. 21, 2206–2214. Rivera-Cervantes, M.C., Torres, J.S., Feria-Velasco, A., Armendariz-Borunda, J., BeasZarate, C., 2004. NMDA and AMPA receptor expression and cortical neuronal death are associated with p38 in glutamate-induced excitotoxicity in vivo. J. Neurosci. Res. 76, 678–687. Rivera-Cervantes, M.C., Flores-Soto, M.E., Chaparro-Huerta, V., Reyes-Go´mez, J., Feria-Velasco, A., Schliebs, R., Beas-Za´rate, C., 2009. Changes in hippocampal NMDA-R subunit composition induced by exposure of neonatal rats to Lglutamate. Int. J. Dev. Neurosci. 27, 197–204. Rogers, J.L., Kesner, R.P., 2006. Hippocampal-parietal cortex interactions: evidence from a disconnection study in the rat. Behav. Brain Res. 179, 19–27. Scheetz, A.J., Constantine-Paton, M., 1994. Modulation of NMDA receptor function: implications for vertebrate neuronal development. FASEB J. 8, 745–752. Schmidt-Kastner, R., Ophoff, B.G., Hossmann, K.A., 1990. Pattern of neuronal vulnerability in the cat hippocampus after one hour of global cerebral ischemia. Acta Neuropathol. 19, 444–455. Schubert, V., Dotti, C.G., 2007. Transmitting on actin: synaptic control of dendritic architecture. J. Cell Sci. 120 (Pt 2), 205–212. Siegel, S.J., Janssen, W.G., Tullai, J.W., Rogers, S.W., Moran, T., Heinesmann, S.F., Morrison, J.H., 1995. Distribution of the excitatory amino acid receptor subunits GluR2(4) in monkey hippocampus and colocalization with subunits GluR5-7 and NMDAR1. J. Neurosci. 15, 2707–2719. ˜ a, L., Olvera-Corte´s, E., Gonza´lez-Burgos, I., 2000. Prenatal and postnatal Tarelo-Acun exposure to ethanol induces changes in the shape of the dendritic spines from hippocampal CA1 pyramidal neurons of the rat. Neurosci. Lett. 286, 13–16. Toni, N., Buchs, P.A., Nikonenko, I., Povilaitite, P., Parisi, L., Muller, D., 2001. Remodeling of synaptic membranes after induction of long-term potentiation. J. Neurosci. 21, 6245–6251. Yang, Y., Wang, X., Frerking, M., Zhou, Q., 2008. Spine expansion and stabilization associated with long-term potentiation. J. Neurosci. 28, 5740–5751.
Contents lists available at ScienceDirect
International Journal of Developmental Neuroscience journal homepage: www.elsevier.com/locate/ijdevneu
Damage and plasticity in adult rat hippocampal trisynaptic circuit neurons after neonatal exposure to glutamate excitotoxicity I. Gonza´lez-Burgos a,c, D.A. Vela´zquez-Zamora a, C. Beas-Za´rate b,c,* a
Laboratorio de Psicobiologı´a, Divisio´n de Neurociencias, Centro de Investigacio´n Biome´dica de Occidente, IMSS., Guadalajara, Mexico Laboratorio de Neurobiologı´a Celular y Molecular, Divisio´n de Neurociencias, Centro de Investigacio´n Biome´dica de Occidente, IMSS., Guadalajara, Mexico c Depto. de Biologı´a Celular y Molecular, CUCBA, Universidad de Guadalajara, Guadalajara, Jal., Mexico b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 5 May 2009 Received in revised form 26 August 2009 Accepted 27 August 2009
Hippocampal vulnerability to excitotoxicity has been widely studied along with its implication to learning and memory. Neonatal glutamate excitotoxicity induces loss of CA1 pyramidal neurons in adult rats concomitantly with some plastic changes in the dendritic spines of surviving neurons. At least in part, these may underlie the place learning impairments seen in previous studies based on a similar excitotoxicity-inducing model. In the present study, cytoarchitecture of dentate gyrus, CA3 and CA1 fields were evaluated in 120-day-old rats, after they had been neonatally treated with glutamate as monosodium salt. Dentate granule cells and CA1 pyramidal neurons were less than those counted in NaCl-treated control animals. In addition, dentate granule cells had more dendrites as well as more branched spines. Spine density in CA1 pyramidal neurons was greater than in the controls. Additionally, thin and mushroom spines were proportionally more abundant in monosodium glutamate-treated animals. No effects were seen in the hippocampal CA3 field. Our results strongly suggest a long-term induction of plastic changes in the cytoarchitecture of the hippocampal trisynaptic circuit neurons after cell death provoked by the monosodium glutamate-induced excitotoxicity. These plastic events as well as the aberrant expression of the glutamate NMDA receptors resulting from monosodium glutamate neonatal treatment could be strongly associated with the place learning impairments previously reported. ß 2009 ISDN. Published by Elsevier Ltd. All rights reserved.
Keywords: Glutamate Hippocampus Excitotoxicity Trisynaptic circuit Dendritic spines
The hippocampus is one of the most vulnerable regions of the brain to excitotoxic agents (Pulsinelli, 1985). The mechanisms underlying such differential and specific vulnerability are unclear but the presence and high density of N-Methyl-D-Aspartate (NMDA) glutamate receptors seems to play a crucial role (Schmidt-Kastner et al., 1990); in fact, overstimulation of NMDA receptors has been shown to trigger cell death (Rivera-Cervantes et al., 2004; Gasco´n et al., 2008; Martel et al., 2009). In the hippocampus, NMDA receptors are located mostly on the dendritic spines of those neurons constituting the trisynaptic circuit, i.e., dentate gyrus granule cells, CA3 pyramidal neurons, and CA1 pyramidal cells (Siegel et al., 1995). Briefly, the hippocampal trisynaptic circuit is integrated by the terminal axons of the enthorrinal cortex-proceeding perforant pathway establishing synaptic contacts with the dendritic spines located in the two-thirds distal dendritic segment of the dentate gyrus granule cells. Then, axons from granule cells (mossy fibers) reach
* Corresponding author at: Sierra Mojada # 800, Col. Independencia, Guadalajara, Jalisco 44340, Mexico. E-mail address: [email protected] (C. Beas-Za´rate). 0736-5748/$36.00 ß 2009 ISDN. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijdevneu.2009.08.016
the hippocampal CA3 subfield and synapse with the thorny excrescences of pyramidal neurons and finally, those axonal bifurcations constituting the Schaffer collaterals emerging from the main axon of the CA3 pyramidal cells establish synaptic contacts with the spines of the dendritic arbor of the CA1 pyramidal neurons (Amaral and Witter, 1995). Previous studies from our group have shown that neonatal exposure to glutamate, as monosodium salt, induces cell death and some cytoarchitectural impairments both in prefrontal (Gonza´lezBurgos et al., 2001) and hippocampal (Beas-Za´rate et al., 2002b) pyramidal neurons in young adult rats, as well as place learning impairments (Olvera-Corte´s et al., 2005). Spatial orientation abilities are sustained by the coordinated activity of the hippocampal trisynaptic circuit neurons (Goodrich-Hunsaker et al., 2008), and some plastic changes in hippocampal neurons’ dendritic spines have been shown to be closely associated with place learning (Gonza´lezBurgos et al., 2005). However, it is unknown whether glutamate receptor overstimulation induced by early exposure to high glutamate concentrations could modify the cytoarchitecture of neurons comprising the trisynaptic circuit in the long term. Therefore, the aim of this work was to evaluate the cytoarchitectonic parameters of the hippocampal neurons critically involved in the
742
I. Gonza´lez-Burgos et al. / Int. J. Devl Neuroscience 27 (2009) 741–745
flow of information through the trisynaptic circuit in adult rats after neonatal treatment with MSG. 1. Experimental procedures Twenty-four neonatal Sprague–Dawley male rats were used, which were obtained from the litters of twelve normal female rats. Before birth and during lactation, dams were maintained under standard conditions: regular 12-h light– dark cycles (07:00–19:00 h), environmental humidity 45–50%, temperature 22 2 8C, and free access to food and water. The dams kept one rat to a cage and, after weaning, the pups used for this study were kept in the same conditions as their mothers, until the day of sacrifice. Experimental procedures were conducted to minimize pain and discomfort for the animals, and performed in accordance with the NIH guide for Care and Use of Laboratory Animals (NIH Publications No. 80-23) 1996 revision, and approved by the Research Ethics Committee of the Instituto Mexicano del Seguro Social, Me´xico. Each pup was randomly assigned to one of three groups of study. Eight experimental animals (MSG group) were subcutaneously injected with a 50% aqueous solution of monosodium L-glutamic acid (MSG) at a dose of 4 mg/g of body weight, on postnatal days one, three, five and seven. As a control group (NaCl group), eight pups were given an equimolar aqueous solution of NaCl (17.27%), at a dose of 1.37 mg/g of body weight. The remaining eight rats were untreated (Intact group). During and after treatment, animals from the three groups remained with their original litter until weaning at 21 days and then were caged until they reached four months of age. At 120 days of age, animals were anesthetized with ethyl ether and thereafter perfused with 100 ml of a washing phosphate-buffered solution (pH 7.4; 0.01 M) containing 1000 IU/l of sodium heparin as anticoagulant, and 1 g/l of procaine hydrochloride as vasodilator (Feria-Velasco and Karnovsky, 1970). Then, 200 ml of a fixing phosphate-buffered 4% formaldehyde solution was perfused. Both solutions flowed at a rate of 11.5 ml/min. Each brain remained for at least 48 h in 100 ml of a fresh fixing solution. The bilateral whole hippocampi were dissected out, and four right and four left hippocampi per group were used for either cresyl violet staining or Golgi impregnation, correspondingly. As a result, eight hippocampi per group were processed histologically and embedded in paraffin, while the remaining eight hippocampi were impregnated for the Golgi study. Once embedded in paraffin, each corresponding tissue block was then sectioned using a rotation microtome to yield serial coronal sections 7 mm thick through each hippocampus, and a systematic random sampling procedure was designed to conduct an unbiased stereological study of the granule cell layer of the dentate gyrus (DG), and the CA3 and CA1 fields (Paxinos and Watson, 1986). The first of these sections, used as anatomical reference, presented the beginning of any of the three regions; it was consequently numbered as slice number one. For each rat, a number between 1 and 100 was randomly selected by lottery and used to determine the first section to be sampled for analysis, and this and the consecutive section was used as the ‘‘pair of sections’’ number one. A new random number was selected for each rat studied. The selected section and the consecutive serial section was placed on the same glass slide and then stained with cresyl violet. The following 98 serial sections were discarded, with the subsequent two being sampled as before. This procedure was repeated until the whole DG, CA3 and CA1 regions were completely sectioned. As a result and based on pilot experiments, 6–8 pairs of consecutive sections were used for the study. The stereological study was conducted under ‘‘blind’’ conditions and each slide was identified with a code not revealed until the end of quantifications. The DG, CA3 and CA1 volume was determined by applying the principle of Cavalieri (Gundersen and Jensen, 1987; Pakkenberg and Gundersen, 1988), and the cell number of the
three regions studied was estimated using the dissector principle (Pakkenberg and Gundersen, 1988). Cell counts were made of the total number of nucleoli profiles appearing in photomicrographs from the ‘‘look-up’’ sections but not appearing in the adjacent ‘‘test’’ sections. Cell counts were conducted using the ‘‘forbidden line’’ rule (Gundersen, 1977). The remaining four left and four right hippocampi per group were processed according to a modification of the Golgi method (Gonza´lez-Burgos et al., 1992). Several coronal slices 100 mm thick were mounted on one slide per animal, and eight granule cells from DG, eight CA3 and eight CA1 pyramidal neurons from the dorsal hippocampus were studied per rat. The cytoarchitectonic parameters studied in the hippocampus were: (1) dendritic arborization of both DG granule cells and CA1 pyramidal neurons, (2) area occupied by the thorny excrescences of the long-shaft CA3 pyramidal neurons (Fitch et al., 1989), and (3) spine density and proportion of the thin, stubby, mushroom, wide, double, and branched spines (Fig. 1; see Harris et al., 1989; ˜ a et al., 2000; Gonza´lez-Burgos et al., 2005) of granule cells in the DG Tarelo-Acun and CA1 pyramidal neurons. The dendritic arborization was evaluated by counting the number of dendritic bifurcations in the molecular layer of the DG or in the stratum oriens and radiatum of the CA1 field. The area of thorny excrescences was measured with an image analyzer (Zeiss Image 3.0). Both the spine density and proportion of each type of spine in the DG granule cells were counted in a segment of 50 mm medial in a half primary dendrite distal to the soma. Likewise, in the CA1 pyramidal neurons both spine density and spine types were counted in a segment of 50 mm medial in an oblique dendrite protruding from its parent apical dendrite and in 50 mm medial of a primary basilar dendrite. Spine counting was made by direct observation at 2000 using a magnification changer coupled to the light microscope. The one-way ANOVA and Tukey post hoc tests were used for statistical comparisons relating to stereological measurements, dendritic arborization, area of thorny excrescences and density of spines. In addition, one-way ANOVA and Bonferroni correction post hoc test was used for the proportional density of the different types of spines.
2. Results 2.1. Stereology Tissue volume of the DG and CA1 hippocampal regions were different between groups (F = 14.672, p < 0.0001, CE = 0.05; and F = 8.337, p < 0.002, CE = 0.05; respectively). DG from MSG rats was less voluminous than those from both Intact (p < 0.0001) and NaCl (p < 0.006) animals. Likewise, CA1 volume in MSG-treated rats was lesser than that from both Intact (p < 0.005) and NaCltreated (p < 0.006) animals. There were no differences between Intact and NaCl groups neither in DG nor in CA1 tissue volume. As to CA3 region, no statistical differences among groups were observed between the three groups studied (Table 1). Cell numbers in both DG and CA1 regions was different between groups (F = 13.119, p < 0.0001, CE = 0.05 and F = 8.592, p < 0.002, CE = 0.05; respectively). The number of DG cells of the MSG group was less than that counted in Intact (p < 0.001) and NaCl (p < 0.0001) groups. The number of cells of Intact and NaCl groups was not different. Statistical analysis of this parameter in the CA3
Fig. 1. Photomicrographs showing the six types of spines studied (arrows), and the corresponding diagrams (upper panel). (A) Thin; (B) stubby; (C) mushroom; (D) wide; (E) double; and (F) branched. Scale bar: 5 mm.
I. Gonza´lez-Burgos et al. / Int. J. Devl Neuroscience 27 (2009) 741–745 Table 1 Tissue volume (mm3) of the three hippocampal regions studied.
Dentate gyrus CA3 CA1
Intact
NaCl
MSG
2.61 0.01 4.16 0.01 1.84 0.01
2.57 0.01 4.16 0.01 1.83 0.01
2.50 0.01a,b 4.13 0.01 1.75 0.01a,b
Mean SEM. p < 0.05. a MSG vs. Intact. b MSG vs. NaCl.
743
Table 2 Spine density (number of spines) and proportional density (%) of the different types of spines counted in a 50 mm-length distal segment of a primary dendrite of granule cells from the hippocampal dentate gyrus. Intact
NaCl
MSG
Density
55.6 1.6
58.8 2.0
61.7 2.7
Spines proportion Thin Mushroom Stubby Wide Branched Double
24.8 0.7 16.4 0.5 11.5 0.1 2.1 0.1 0.2 0.07 0.02 0.02
27.5 1.1 18.6 1.0 9.9 0.5 2.1 0.2 0.2 0.04 0.04 0.02
27.9 1.3 18.8 0.7 11.7 0.6b 2.6 0.1 0.6 0.09a,b 0.1 0.05
Mean SEM. p < 0.05. a MSG vs. Intact. b MSG vs. NaCl.
Table 3 Spine density (number of spines) and proportional density (%) of the different types of spines counted in a segment 50 mm-length of a secondary dendrite of pyramidal neurons from the hippocampal CA1 field.
Fig. 2. Graphic comparison between groups, regarding to the number of cells counted in the three hippocampal regions studied. Mean SEM. (a) MSG vs. Intact; (b) MSG vs. NaCl. p < 0.05.
hippocampal region showed no differences between groups in this parameter. For its part, cells in the CA1 field of MSG-treated rats were less than those counted in CA1 from both Intact (p < 0.003) and NaCl animals (p < 0.008), and the number of CA1 cells from Intact and NaCl groups was not different (Fig. 2). 2.2. Golgi studies Dendritic arborization of DG granule cells was different in the three groups studied (F = 12.077, p < 0.001). MSG granule cells showed more bifurcations than both Intact (p < 0.002) and NaCl (p < 0.002) groups; no difference in this parameter was observed between the Intact and NaCl groups of animals (Fig. 3). Density of spines in granule cells was not different between the groups, but the proportional density of stubby and branched spines was increased in the experimental group (F = 4.634, p < 0.027 and
Intact
NaCl
MSG
Density
80.5 2.5
87.3 1.4
91.9 4.7a
Spines proportion Thin Mushroom Stubby Wide Branched Double
37.4 1.4 25.2 0.9 14.3 0.7 3.1 0.3 0.2 0.06 0.06 0.02
39.2 0.8 27.3 0.6 15.2 0.5 3.4 0.4 0.3 0.1 0.08 0.04
44.7 1.3a,b 31.3 1.3a,b 15.7 0.5 4.2 0.5 0.3 0.06 0.2 0.07
Mean SEM. p < 0.05. a MSG vs. Intact. b MSG vs. NaCl.
F = 6.101, p < 0.012; respectively). Stubby spines in MSG granule cells were more abundant than in the NaCl group (p < 0.04), and branched spines in MSG group were more abundant than in the Intact (p < 0.02) and NaCl (p < 0.02) groups. No significant differences in the proportion of thin, mushroom, wide, and double spines in the DG granule cells were observed between the three groups studied (Table 2). The area of thorny excrescences of Intact, NaCl, and MSG groups was not different (data not shown). The dendritic arborization of CA1 pyramidal neurons was not different between the three groups studied (data not shown). As to spine density, a significant difference between groups was seen (F = 8.729, p < 0.003). Rats from the MSG group had more dendritic spines than animals from the Intact (p < 0.002) and NaCl (p < 0.04) groups; the density of spines in Intact and NaCl control groups was not different. In addition, proportional density of thin (F = 9.106, p < 0.003) and mushroom (F = 9.124, p < 0.003) spines were different between groups. Thin spines of CA1 pyramidal cells from MSG were more abundant than those from both Intact (p < 0.003) and NaCl (p < 0.022). Likewise, mushroom spines were more abundant in the MSG group than in the Intact (p < 0.002) and NaCl (p < 0.049) groups. There was no statistical difference in the proportional density of thin and mushroom spines between the Intact and NaCl control groups. Proportional density of stubby, wide, branched, and double spines was not different (Table 3). 3. Discussion
Fig. 3. Graph showing the comparison of dendritic bifurcations counted in dentate gyrus granule cells, between the three groups studied. Mean SEM. (a) MSG vs. Intact; (b) MSG vs. NaCl. p < 0.05.
The hippocampus is a limbic structure strongly related with spatial learning and memory (O’Keefe and Nadel, 1978) by virtue of the concerted activity of the hippocampal trisynaptic circuit
744
I. Gonza´lez-Burgos et al. / Int. J. Devl Neuroscience 27 (2009) 741–745
neurons (Goodrich-Hunsaker et al., 2008) from the dorsal hippocampus (Goodrich-Hunsaker et al., 2005). Glutamate-mediated excitotoxicity in the hippocampus underlies, at least in part, in an overstimulation of the glutamate NMDA receptors (Choi, 1988; Rivera-Cervantes et al., 2004) located on the external membrane of dendritic spines from those neurons constituting the hippocampal trisynaptic circuit. In fact, cerebral ischemia-inducing excitotoxicity leads to cell death in the hippocampus, along with place learning and memory impairments (Letechipı´a-Vallejo et al., 2007) similar to those seen in adult rats after neonatal MSG treatment (Olvera-Corte´s et al., 2005). In the present study, dentate gyrus and CA1 field of the MSGtreated animals were less voluminous than those from controls. This corresponded with the reduced cell population seen in the dentate gyrus (8%) and the CA1 field (15%). This data is in agreement with previous studies showing that hippocampal CA1 pyramidal neurons and dentate granule cells are among the most vulnerable nervous system cells to excitotoxicity (Pulsinelli, 1985). Tissue volume and cell numbers in the hippocampal CA3 field were not affected after MSG treatment, which could be sustained on the low density of glutamate NMDA receptors existing on the thorny excrescences of the CA3 pyramidal neurons (Jaffe and Brown, 1997; Reid et al., 2001) where mossy fibers synapse. Dentate granule cells of MSG-treated rats had more dendrites than the controls. Expressions of the different NMDA receptor subunits have been shown to affect dendritic growth of hippocampal neurons during development (Scheetz and Constantine-Paton, 1994). In this sense, changes in the NMDA receptor subunits were previously reported in adult rats as a result of the neonatal exposure to MSG (Beas-Za´rate et al., 2002a; RiveraCervantes et al., 2009). Despite spine density in dendrites of dentate granule cells was not modified, the fact that these neurons showed more dendritic ramifications allows us to assume that they had more spines per neuron. We also found that branched spines were more abundant than those seen in granule cells from control animals. Branched spines have been strongly related both with spatial learning and induction of long-term potentiation (LTP) (Moser et al., 1994; Muller et al., 2000), which has been associated with the acquisition and consolidation of learning and memory (Bliss and Lømo, 1973; Morris et al., 1986). In this sense, after LTP induction perforation of macular synapses takes place (Geinisman et al., 1993; Toni et al., 2001), and perforated synapses ultimately divide into two small independent spines through the transient formation of a branched spine (Moser et al., 1994; Muller et al., 2000). Thus, along with the putative increase of total spines, the proportional increase of branched spines seen in the MSG-treated animals could be interpreted as compensatory events after the excitotoxic effects exerted by MSG on granule cells during development, such as cell death. As a result, these putative plastic events could influence the organization of the metric spatial information in which dentate granule cells are closely involved (Goodrich-Hunsaker et al., 2008; Kesner et al., 2004). Glutamate-inducing neurotoxicity is strongly related with the entry of excessive calcium to postsynaptic cells (Gasco´n et al., 2008) after glutamate NMDA receptor stimulation (Grishin et al., 2004). Based on these, the lack of significant damage or plastic changes to CA3 pyramidal neurons after MSG treatment could be due to the fact that thorny excrescences of CA3 pyramidal neurons have a low density of glutamate NMDA receptors (Jaffe and Brown, 1997; Reid et al., 2001). CA3 pyramidal neurons are involved in the metric representations of spatial information processed by granule cells (Goodrich-Hunsaker et al., 2008) by encoding them within short-term memory (Kesner, 2007). Because these two components of the trisynaptic circuit act in concert through the mossy fibers to process metric representations of space (GoodrichHunsaker et al., 2008; Kesner, 2007), these findings would mean
that the putative alterations of spatial information proceeding from the dentate granule cells would in turn be abnormally processed by the CA3 pyramidal neurons. CA1 pyramidal neurons receive CA3-proceeding Schaffer collaterals after which, CA1 efferent fibers leave the hippocampus to innervate several brain regions (Amaral and Witter, 1995). Lesion of the dorsal hippocampus has been shown to impair the topological representations of space (Goodrich-Hunsaker et al., 2005; Rogers and Kesner, 2006), a cognitive ability closely related with the functional activity of the hippocampal CA1 field (Goodrich-Hunsaker et al., 2008). Topology of space consists in those inclusion, vicinity, adjacency, and order-related properties of objects in the surrounding space. Since the CA1 field constitutes the last ‘‘relay station’’ before the space-related information flowing through the trisynaptic circuit leaves the hippocampus, its functional activity would be related with the integration of both the dentate gyrus-and-CA3-dependent metric properties of space, as well as the topological information necessary to achieve spatial information to resolve a given space-related behavioral task. Spine density and proportional density of thin and mushroom spines were greater in CA1 pyramidal neurons. The greater the amount of spines, the nearer they are, and then, the greater their capacity is to associate with neighbor synaptic stimuli (Harris and Kater, 1994; De Roo et al., 2008). On the other hand, it is well known that both thin and mushroom spines are related with cognitive efficiency (Kasai et al., 2003). Accordingly, it would be expected that hippocampal-dependent cognitive abilities would be highly efficient in those MSG-treated animals. However, in spite of the putative plastic events that took place in the hippocampal trisynaptic circuit neurons after cell death, behavioral performance in the Morris maze was severely impaired according to a previous study from our work group using the same experimental paradigm (Olvera-Corte´s et al., 2005) than that used in the present study. Yet, it has been reported that place learning is performed normally even after up to a 60% loss of CA1 neurons in some ischemiainducing models (Block, 1999), a much higher value than the 15% observed in this study. In addition, plastic events observed, i.e., more profuse dendritic arborization, greater density of spines, and a higher proportion of those spines related with the highly efficient processing of mnemonic information such as branched, thin, and stubby spines (Bourne and Harris, 2007; Kasai et al., 2003; Matsuzaki et al., 2004; Muller et al., 2000; Yang et al., 2008); were not enough to ‘‘protect’’ the functional integrity of the hippocampal trisynaptic circuit as to perform efficiently in the Morris maze. Based on these results, some other neuropathological and/or plastic events could be involved in the place learning impairments occurred after neonatal exposure to glutamate neurotoxicity previously reported. NMDA receptors anchored to postsynaptic density scaffolding proteins mediate plastic remodeling of excitatory synapses and dendritic spines (Calabrese et al., 2006; Schubert and Dotti, 2007). Therefore, since NMDA receptor composition disrupts due to MSG-mediated excitotoxicity (BeasZa´rate et al., 2002a; Rivera-Cervantes et al., 2009), the pathological/plastic modifications in dendritic spines seen in this study could be related to changes in NMDA receptor. This hypothesis must be examined in further experiments. References Amaral, D.G., Witter, M.P., 1995. Hippocampal formation. In: Paxinos, G. (Ed.), The Rat Nervous System. Academy Press, London, pp. 443–493. Beas-Za´rate, C., Flores-Soto, M.E., Armendariz-Borunda, J., 2002a. NMDAR-2C and 2D subunits gene expresio´n is induced in brain by neonatal exposure of monosodium L-glutamate to adult rats. Neurosci. Lett. 321, 9–12. Beas-Za´rate, C., Pe´rez-Vega, M.I., Gonza´lez-Burgos, I., 2002b. Neonatal exposure to monosodium L-glutamate induces loss of neurons and cytoarchitectural alterations in hippocampal pyramidal neurons of adult rats. Brain Res. 952, 275– 281.
I. Gonza´lez-Burgos et al. / Int. J. Devl Neuroscience 27 (2009) 741–745 Bliss, T.V., Lømo, T., 1973. Long-lasting potentiation of synaptic transmission in the dentate area of the anesthetized rabbit following stimulation of the perforant path. J. Physiol. 232, 331–356. Block, F., 1999. Global ischemia and behavioral deficits. Prog. Neurobiol. 58, 279–295. Bourne, J., Harris, K.M., 2007. Do thin spines learn to be mushroom spines that remember? Curr. Opin. Neurobiol. 17, 1–6. Calabrese, B., Wilson, M.S., Halpain, S., 2006. Development and regulation of dendritic spine synapses. Physiology 21, 38–47. Choi, D.W., 1988. Calcium-mediated neurotoxicity: relationship to specific channel types and role in ischemic damage. Trends Neurosci. 11, 465–469. De Roo, M., Klauser, P., Muller, D., 2008. LTP promotes a selective long-term stabilization and clustering of dendritic spines. PLoS Biol. 6, 1850–1860. Feria-Velasco, A., Karnovsky, M., 1970. Preservacio´n o´ptima del sistema nervioso central por perfusio´n con glutaraldehı´do para estudio ultraestructural. Arch. Invest. Med. 1, 201–220. Fitch, J.M., Juraska, J.M., Washington, L.M., 1989. The dendritic morphology of pyramidal neurons in the rat hippocampal CA3 area. l. Cell types. Brain Res. 479, 105–114. ˜ a, A., Dı´az-Guerra, M., 2008. Gasco´n, S., Sobrado, M., Roda, J.M., Rodrı´guez-Pen Excitotoxicity and focal cerebral ischemia induce truncation of the NR2A and NR2B subunits of the NMDA receptor and cleavage of the scaffolding protein PSD-95. Mol. Psychiatry 13, 99–114. Geinisman, Y., de Toledo-Morrell, L., Morrell, F., Heller, R.E., Rossi, M., Parshall, R.F., 1993. Structural synaptic correlate of long-term potentiation: formation of axospinous synapses with multiple, completely partitioned transmission zones. Hippocampus 3, 435–446. Gonza´lez-Burgos, I., Alejandre-Go´mez, M., Cervantes, M., 2005. Spine-type densities of hippocampal CA1 neurons vary in proestrus and estrus rats. Neurosci. Lett. 379, 52–54. Gonza´lez-Burgos, I., Pe´rez-Vega, M.I., Beas-Za´rate, C., 2001. Neonatal exposure to monosodium glutamate induces cell death and dendritic hypotrophy in rat prefrontal pyramidal neurons. Neurosci. Lett. 297, 69–72. Gonza´lez-Burgos, I., Tapia-Arizmendi, G., Feria-Velasco, A., 1992. Golgi method without osmium tetroxide for the study of the central nervous system. Biotech. Histochem. 67, 288–296. Goodrich-Hunsaker, N.J., Hunsaker, M.R., Kesner, R.P., 2005. Dissociating the role of the parietal cortex and dorsal hippocampus for spatial information processing. Behav. Neurosci. 119, 1307–1315. Goodrich-Hunsaker, N.J., Hunsaker, M.R., Kesner, R.P., 2008. The interactions and dissociations of the dorsal hippocampus subregions: how the dentate gyrus, CA3, and CA1 process spatial information. Behav. Neurosci. 122, 16–26. Grishin, A.A., Gee, C.E., Gerber, U., Benquet, P., 2004. Differential calcium-dependent modulation of NMDA currents in CA1 and CA3 hippocampal pyramidal cells. J. Neurosci. 24, 350–355. Gundersen, H.J.G., 1977. Notes on the estimation of the numerical density of arbitrary profiles: the edge effect. J. Microsc. 111, 219–223. Gundersen, H.J.G., Jensen, E.B., 1987. The efficiency of systematic sampling in stereology and its prediction. J. Microsc. 147, 229–263. Harris, K.M., Jensen, F.E., Tsao, B.H., 1989. Ultrastructure, development and plasticity of dendritic spine synapses in area CA1 of the rat hippocampus: extending our vision with serial electron microscopy and three-dimensional analyses. In: Chan-Palay, V., Koˆhler, Ch. (Eds.), The Hippocampus—New Vistas. Alan R. Liss, pp. 33–52. Harris, K.M., Kater, S.B., 1994. Dendritic spines: cellular specializations imparting both stability and flexibility to synaptic function. Annu. Rev. Neurosci. 17, 341– 371. Jaffe, D.V., Brown, T.H., 1997. Calcium dynamics in thorny excrescences of CA3 pyramidal neurons. J. Neurophysiol. 78, 10–18. Kasai, H., Matsuzaki, M., Noguchi, J., Yasumatsu, N., Nakahara, H., 2003. Structurestability-function relationships of dendritic spines. Trends Neurosci. 26, 360– 368. Kesner, R.P., 2007. A behavioural analysis of dentate gyrus function. Prog. Brain Res. 163, 567–576. Kesner, R.P., Lee, I., Gilbert, P., 2004. A behavioral assessment of hippocampal function based on a subregional analysis. Rev. Neurosci. 15, 333–351.
745
Letechipı´a-Vallejo, G., Lo´pez-Loeza, E., Espinoza-Gonza´lez, V., Gonza´lez-Burgos, I., Olvera-Corte´s, M.E., Moralı´, G., Cervantes, M., 2007. Long-term morphological and functional evaluation of the neuroprotective effects of post-ischemic treatment with melatonin in rats. J. Pineal Res. 42, 139–146. Martel, M.A., Sorianom, F.X., Baxter, P., Rickman, C., Duncan, R., Wyllie, D.J., Hardingham, G.E., 2009. Inhibiting pro-death NMDA receptor signaling dependent on the NR2 PDZ ligand may not affect synaptic function or synaptic NMDA receptor signaling to gene expression. Channels (Austin) 3, 12–15. Matsuzaki, M., Honkura, N., Ellis-Davies, G.C.R., Kasai, H., 2004. Structural basis of long-term potentiation in single dendritic spines. Nature 429, 761–766. Morris, R.G., Anderson, E., Lynch, G.S., Baurdy, M., 1986. Selective impairment of learning and blockade of long-term potentiation by an N-methyl D-aspartate receptor antagonist, AP5. Nature 319, 774–776. Moser, M.B., Trommald, M., Andersen, P., 1994. An increase in dendritic spine density on hippocampal CA1 pyramidal cells following spatial learning in adult rats suggests the formation of new synapses. Proc. Natl. Acad. Sci. U.S.A. 91, 12673–12675. Muller, D., Toni, N., Buchs, P.A., 2000. Spine changes associated with long-term potentiation. Hippocampus 10, 596–604. O’Keefe, J., Nadel, L., 1978. The Hippocampus as a Cognitive Map. Oxford University Press, Oxford. Olvera-Corte´s, E., Lo´pez-Va´zquez, M.A., Beas-Za´rate, C., Gonza´lez-Burgos, I., 2005. Neonatal exposure to monosodium glutamate disrupts place learning ability in adult rats. Pharmacol. Biochem. Behav. 82, 247–251. Pakkenberg, B., Gundersen, J.G., 1988. Total number of neuron and glial cells in human brain nuclei estimated by the dissector and the fractionator. J. Microsc. 150, 1–20. Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates, 2nd ed. Oxford University Press, USA. W.A. Pulsinelli, 1985. Selective neuronal vulnerability: morphological and molecular characteristics. In: K. Kogure, K.A. Hossman, B.K. Siesjo¨, F.A. Welsch (Eds.), Molecular Mechanisms of Ischemic Brain Damage. Prog. Brain Res., vol. 63. Elsevier, New York, pp. 29–37. Reid, C.A., Fabian-Fine, R., Fabian, A., 2001. Postsynaptic calcium transients evoked by activation of individual hippocampal mossy fiber synapses. J. Neurosci. 21, 2206–2214. Rivera-Cervantes, M.C., Torres, J.S., Feria-Velasco, A., Armendariz-Borunda, J., BeasZarate, C., 2004. NMDA and AMPA receptor expression and cortical neuronal death are associated with p38 in glutamate-induced excitotoxicity in vivo. J. Neurosci. Res. 76, 678–687. Rivera-Cervantes, M.C., Flores-Soto, M.E., Chaparro-Huerta, V., Reyes-Go´mez, J., Feria-Velasco, A., Schliebs, R., Beas-Za´rate, C., 2009. Changes in hippocampal NMDA-R subunit composition induced by exposure of neonatal rats to Lglutamate. Int. J. Dev. Neurosci. 27, 197–204. Rogers, J.L., Kesner, R.P., 2006. Hippocampal-parietal cortex interactions: evidence from a disconnection study in the rat. Behav. Brain Res. 179, 19–27. Scheetz, A.J., Constantine-Paton, M., 1994. Modulation of NMDA receptor function: implications for vertebrate neuronal development. FASEB J. 8, 745–752. Schmidt-Kastner, R., Ophoff, B.G., Hossmann, K.A., 1990. Pattern of neuronal vulnerability in the cat hippocampus after one hour of global cerebral ischemia. Acta Neuropathol. 19, 444–455. Schubert, V., Dotti, C.G., 2007. Transmitting on actin: synaptic control of dendritic architecture. J. Cell Sci. 120 (Pt 2), 205–212. Siegel, S.J., Janssen, W.G., Tullai, J.W., Rogers, S.W., Moran, T., Heinesmann, S.F., Morrison, J.H., 1995. Distribution of the excitatory amino acid receptor subunits GluR2(4) in monkey hippocampus and colocalization with subunits GluR5-7 and NMDAR1. J. Neurosci. 15, 2707–2719. ˜ a, L., Olvera-Corte´s, E., Gonza´lez-Burgos, I., 2000. Prenatal and postnatal Tarelo-Acun exposure to ethanol induces changes in the shape of the dendritic spines from hippocampal CA1 pyramidal neurons of the rat. Neurosci. Lett. 286, 13–16. Toni, N., Buchs, P.A., Nikonenko, I., Povilaitite, P., Parisi, L., Muller, D., 2001. Remodeling of synaptic membranes after induction of long-term potentiation. J. Neurosci. 21, 6245–6251. Yang, Y., Wang, X., Frerking, M., Zhou, Q., 2008. Spine expansion and stabilization associated with long-term potentiation. J. Neurosci. 28, 5740–5751.