- Email: [email protected]
Cortical neurons containing calretinin are selectively resistant to calcium overload and excitotoxicity in vitro
neuroscience
0306-4522(94)E0068-F
Vol. 61, No. 2, pp. 307-316, 1994 Elsevier Science Ltd Copyright 0 1994 IBRO Printed in Great Britain. All rights reserved 0306-4522/94 $7.00 + 0.00
CORTICAL NEURONS CONTAINING CALRETININ ARE SELECTIVELY RESISTANT TO CALCIUM OVERLOAD AND EXCITOTOXICITY IN JfZTRO W. Glaxo
Institute
for Molecular
LUKAS
Biology,
and K. A. JONES* 14 chemin
des Aulx, CH-1228
Geneva,
Switzerland
Abstract-Calbindin and the more recently identified protein calretinin are structurally related calciumbinding proteins having a broad distribution in the brain. Recent evidence supports a neuroprotective role for calbindin in regulating calcium homeostasis during periods of heightened Ca2+ influx. It is not known if calretinin might have a similar function. We investigated if calretinin-containing neurons have a survival advantage in rat neocortical cultures treated with a calcium ionophore or excitatory amino acids. Neuronal cultures were challenged with the calcium ionophore A23187 at different concentrations to produce a broad range of cell death. Cell loss was quantified for both the calretinin immunopositive and the calretinin
immunonegative populations of neurons. We found that 3 h after exposure to 2 PM A23187 there was a 48% loss of the calretinin immunonegative population of neurons whereas the calretinin immunopositive set of neurons was reduced by only 18%. Calretinin positive neurons were still relatively spared after treatment with 3 PM A23187. The ionophore had no cytotoxic effect when calcium ions were removed from the extracellular medium. We also studied glutamate excitotoxicity by treating the neuronal cultures with the excitatory amino acids glutamate, N-methyl-D-aspartate or kainate for Smin and examining survival three hours later. We found again that calretinin-containing neurons were relatively spared after exposure to the excitatory amino acids; at doses of N-methyl-D-aspartate and kainate that produced a 3240% loss of calretinin immunonegative neurons, only 2-10% of calretinin immunopositive neurons died. Similar results were obtained for glutamate. These results demonstrate that neurons containing calretinin are better able to survive disturbances in calcium homeostasis than cells not containing this calcium-binding protein. The fact that this effect was observed with ionophore treatment, as well as excitatory amino acids, suggests that neither the density nor distribution of glutamate receptors on the different cell types was a factor in determining selective vulnerability. We hypothesize that the neuroprotective effect of calretinin is due to the buffering capacities of the protein in a manner analogous to that suggested for calbindin.
phores,33,4~,42,45
Calcium ions play an important signalling role in neurons, and several homeostatic mechanisms are required to maintain their normally low intracellular levels.6 Malfunction or overload of the cellular systems that regulate calcium homeostasis results in rapid neurodegeneration. 9,44,62There is considerable evidence from studies of primary neuronal cultures that excessive calcium influx caused by the stimulation of glutamate receptors triggers an excitotoxic process that leads to cell death.8,‘4,‘7,39,52,65 Several types of glutamate reeeptors, including members of the Nmethyl-D-aspartate (NMDA) and non-NMDA classes, czua induce this process when over-stimulated.30s32Additionally, it is possible to observe neurotransmitter receptor-independent neurotoxicity by the use of ionophores selective for calcium.45 Not all types of neurons are equally sensitive to the toxic effects of glutamate or calcium iono-
and a robust calcium buffering capacity is one property that may enable some neurons to survive in the face of pathological disturbances.1s,38 In support of this idea it has been shown that intracellular injection of a calcium chelator can block excitotoxicity at an early stage. 59 Two calcium-binding proteins, calbindin and parvalbumin, are believed to have a physiological role in buffering intracellular calcium ions,42,“@’ and both are abundantly expressed in various populations of neurons in the brain.2.7 Indeed, there is direct evidence to show that the calcium-binding protein calbindin buffers stimulated calcium influx,35,42~48and is present in a select population of hippocampal neurons that survive excitotoxic conditions in vitro.42 The neuroprotective role for calciumbinding proteins in vivo is more difficult to demonstrate. Some populations of neurons containing calcium-binding proteins are relatively spared in animal models of neurodegeneration,34,63 but others are selectively reduced in number.‘5~‘5a~37a The relationship between calcium overload and neuronal vulnerability has encouraged studies of calcium-binding proteins in human neurodegenerative diseases. For example, a calbindin-containing subpopulation of neurons in the substantia nigra is spared
*To whom correspondence should be addressed. Abbreviations: APV. 2-amino-5-vhosvhonovaleric acid: 6-cyano-7-nitroquiioxaline-2,3-dione; HBS; CNQX, HEPES-buffered saline: HEPES. N-2-hvdroxvethvlpiperazine-N’-2-ethanes&phonic acid; I&DA, IN. methyl-D-aspartate. 307
W. LUKAS and K. A. JONES
308
in Parkinson’s disease73 and in an animal model of this disease,24 and neocortical neurons expressing high levels of calbindin or calretinin are preserved in Alzheimer’s disease.“.” On the other hand, it has been suggested that the observed decreases in calbindin mRNA and protein content in other populations of neurons may accelerate cell loss in Alzheimer’s’2.26.27.”and other neurodegenerative diseases.23 These studies support the hypothesis that maintained expression of calcium-binding proteins may be neuroprotective in some instances of human neurodegeneration. Calretinin is a relatively new member of a family of calcium-binding proteins that include calbindin and other proteins sharing the EF-hand domain.s5,56.71 Anatomical studies show that calretinin is contained in many types of neurons scattered throughout the nervous system ‘.28,54 including the neocortex,28,57 and that there is little overlap with neuronal populations containing calbindin. 43,58,70 Since calretinin displays considerable sequence homology with calbindin,5’,55,7’ and like calbindin, it is abundantly expressed in some neurons,51.70it may also be a determinant of selective neuronal vulnerability to excitotoxicity. In the present study, we investigate the possibility that calretinin-containing neurons in cultures of cerebral cortex are resistant to calcium-induced excitotoxicity induced by calcium ionophore or excitatory amino acids.
EXPERIMENTAL
PROCEDURES
Cell culture Cortical cultures containing both glia and neuronal cells were prepared as previously described.3 Briefly, pieces of visual cortex of one- to two-day-old rats were incubated with papain (Worthington Biochemical) for 30 min in a shaking bath at 32-34°C. The enzymatic solution was then replaced three times by a buffered solution containing trypsin inhibitor, and the tissue was dissociated in growth medium by trituration with a fire-polished glass pipette. The dissociated cells were plated (7.5 x lo5 cells/ml) in the central well of 35 mm plastic dishes containing a grid (Milian Instruments, SA) that aided in the counting of the same population of neurons before and after experimental treatments. The dishes were previously treated with a solution containing 200 pg/ml of poly-D-lysine and 33 pg/ml of laminin (Collaborative Research). Growth medium consisted of Minimum Essential Medium (GIBCO) supplemented with penicillin 25 U/ml, streptomycin 25 pg/ml, ducose 20mM. alutamine 0.5 mM and 10% (v/v) heat mactivated fetal calf serum (GIBCO). Cultures were maintained in a humidified incubator (5% CO,/95% air) at 37°C. Five days after plating, cytosine p-D-arabino-furanoside I
(5 PM) was added to avoid further proliferation of glial cells. Half of the medium was removed two days later and replaced by fresh medium containing 5% rat serum (Harlan Bioproducts for Science). Since excitatory amino acid toxicity develops over two weeks in cultures,*,29,M we used 12~14-day-old cultures, which exhibit robust and reproducible excitotoxicity, for all of our experiments, Treatment with excitatory amino acid.s and ionophore Before treatment cultures were washed with a HEPESbuffered saline (HBS) consisting of NaCl, 137mM; KCI, 5 mM; CaCl,, 4 mM; MgCI,, 2 mM; HEPES, 10 mM; 0.5% glucose, pH 7.2.“” Cultures were exposed to L-glutamate, NMDA or kainate dissolved in HBS, for 5 min, washed and maintained in HBS at 37°C for 3 h, and then fixed with 4% formaldehyde for 20 min. Stock solution of A23 187 (Calbiochem), dissolved in dimethylsulphoxide, was diluted in HBS and applied for 3 h at 37°C before formaldehyde fixation. In some experiments the NMDA and non-NMDA receptor antagonists, 2-amino-5-phosphonovalerate (APV) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), were incubated with A23187 at final concentrations of 50 PM. Calretinin immunohistochemistry Calretinin immunostaining was performed using the peroxidase-antiperoxidase method. Cultures were incubated overnight at 4°C with a polyclonal rabbit anti-human calretinin antibody6’ (gift of M. Celia) diluted 1: 2000 in modified tris-buffered saline3 containing 2% bovine serum albumin, 5% rat serum and 0.2% Triton X-100. This antibody does not cross-react with the related protein calbindin.’ Goat anti-rabbit secondary antibody and mouse peroxidase-antiperoxidase (Sternberger Monoclonals) were consecutively applied for 1 h at room temperature. Staining was revealed with a solution containing diaminobenzidine (0.5 mg/ml) and 0.03% H,O, in Tris-buffered saline. Cell counting and analysis
qf neuronal survival
Neuronal survival was calculated by counting neurons 3 to 24 h before treatment with excitotoxins and again after fixation of the cultures. Counting was performed at 100 x magnification with the aid of the labelled grid in the central well of the Petri dishes. Living neurons, identified as having round or oval phase-bright somata, were easily distinguishable from underlying glia. This method of quantifying living neurons correlates well with other methods that use measurements of lactate dehydrogenate release” or Trypan Blue dye exclusion. 39 Estimates of the survival of calretinin immunopositive and calretinin immunonegative neurons were obtained by counting calretinin immunopositive and calretinin immunonegative neurons in a previously determined area of each dish, and expressing these quantities as percentages of the total number of neurons counted in each respective area before treatment. These percentages were then normalized to the mean per cent of calretinin immunopositive and calretinin immunonegative neurons in the cultures treated with vehicle alone. For each experimental condition 5-10% of the total culture surface was counted in each of three Petri dishes, representing 200-400 neurons per dish. All statistical comparisons were done using the unpaired Student r-test, and values presented are the mean k S.E.M.
Fig. 1. Calretinin immunostaining of two-week-old neurons in primary culture. Photographs were taken of two sister cultures either untreated (HBS alone--A, B) or treated with 1 PM A23187 (C, D). Phase image (A) shows several healthy phase-bright neurons, of which three are calretinin immunopositive (asterisks) as seen under bright field illumination (B). Photomicrographs of the sister culture treated with A23187 show three phase-bright neurons (arrowheads, D), of which two are calretinin immunopositive and one calretinin immunonegative, and at least three degenerating cells (arrows, C) that are calretinin immunonegative and exhibit loss of soma integrity and extensively beaded neurites. Background shows the lettered grids that were used to relocate and quantify surviving neurons. Scale bar = 35 pm.
Fig. I.
W. LUKAS and K. A. JONFS
310
Fig. 2. High magnification view of an untreated culture grown on a glass coverslip shows calretinin immunoreactive material throughout the soma and dendrites of a calretinin immunopositive neuron and in a punctate pattern along the processes of two neighboring calretinin immunonegative neurons. Scale bar = 10,um.
RESULTS
Calretinin-immunopositive neurons represented a subpopulation of neocortical neurons in culture that averaged 28 f 3% (mean + S.E.M. of 17 different culture preparations) of the total population. Although this proportion varied from preparation to preparation, the variability within any particular experiment was always less than 6%. Staining within
Table 1. Total neuronal survival in cortical cultures following exposure to calcium ionophore or excitatory amino acids
Control A23187 A23187 A23187
(%)
n
1
65
85 k 3 53 f 4 28 i 4
8 16 29
69 + 4 63 + 3 34 + 4
20 18 26
Survival
Condition
95+ l/rM 2pM 3/1cM
NMDA 1 mM Kainate 1 mM Glutamate 1 mM
Values are mean f S.E.M. of n number of cultures from eight to nine independent experiments.
calretinin immunopositive neurons in two week-old cultures was localized to the cytoplasm of the soma and often throughout the dendritic tree (Figs 1, 2). Reaction product was not present in nuclei of calretinin immunopositive neurons nor in the underlying glial cells. A subset of calretinin immunopositive cells showed intense staining of very fine, presumably axonal processes, and it was possible to observe a punctate staining pattern on the soma and processes of many calretinin-immunonegative neurons (Fig. 2). Thus, in many cases, calretinin appears to be present throughout the entire neuronal structure, including within axon terminals. This staining pattern is very similar to that previously observed in neuronal tissue in uizlo.54 Calretinin immunoreactivity was not detected in very young cultures (
Calretinin
1 PM
2 PM
and calcium
311
overload in cortical cultures
3 FM
[A231 871
Fig. 3. Normalized percentages of calretinin immunopositive and calretinin immunonegative neurons resistant to 3 h exposure to calcium ionophore A23187. Values are normalized to the mean survival of calretinin immunopositive and calretinin immunonegative neurons in the control cultures (HBS alone) and expressed as the mean & S.E.M. (**P < 0.01; ***P < 0.001). The data are pooled from nine separate experiments; n = 8 (1 PM), 16 (2 PM), 29 (3 PM) and 33 (control). CR+, calretinin immunopositive; CR-, calretinin immunonegative.
the survival of the calretinin immunopositive and calretinin immunonegative populations of neurons. One to two hours following the period of treatment with ionophore, many neurons displayed characteristics of dead or dying cells: phase-dark, non-distinct plasma membranes, cytoplasmic vacuoles, and broken or beaded dendrites (Fig. IC, D). When cultures were allowed to survive for longer periods these phase-dark cells disappeared entirely from the underlying glial monolayer. In contrast, glia appeared to be unaffected by the addition of the ionophore. Treatment of cultures with A23187 caused a dosedependent decrease in total neuronal survival, ranging from 85% survival at 1 PM to 28% at 3 PM (Table 1) compared to control cultures (HBS alone) which exhibited a 95% survival. Immunostaining for calretinin showed that calretinin immunopositive neurons were significantly more resistant to ionophore neurotoxicity than were neurons lacking calretinin immunoreactivity (Fig. 1). This effect was observed as an increase in the proportion of surviving neurons immunopositive for calretinin relative to control cultures (Fig. 4). When the populations of calretinin immunopositive and calretinin immunonegative cells were normalized to those in control cultures, differences were found in their survival in cultures treated with varying doses of ionophore (Fig. 3). In cultures exposed to 1 PM A23187 the survival of calretinin immunopositive neurons was unchanged while the survival of calretinin immunonegative neurons decreased slightly (Fig. 3). With increasing doses of ionophore the loss of calretinin immunonegative neurons became significant;
survival fell to 52% at 2 PM and 24% at 3 PM A23187. In contrast, there was only a slight loss of calretinin immunopositive at 2 PM (82% survival), and only at the highest dose was there a substantial loss of calretinin immunopositive neurons (44% survival). Other studies have shown that the ionophore A23187 induces divalent cation-selective fluxes across membraness3*‘* and elevates intracellular Ca* + in neurons.42,45 This calcium influx appeared to be responsible for the neurotoxicity that we observed based upon the following observations. When calcium was removed from the medium, subsequent treatment with 3 PM A23187 caused no apparent neurotoxicity, although there was a slight but insignificant increase in the proportion of calretinin immunopositive cells (Fig. 4). Additionally, the neurotoxic effect of calcium ionophore was not the result of calcium-dependent release of glutamate from nerve terminals since selective antagonists of glutamate receptors did not prevent cell death caused by A23187, nor did they influence the percentage of calretinin immunopositive neurons (Fig. 4). Excitatory amino acid toxicity Excessive stimulation of glutamate receptors is known to cause a neurodegeneration that is dependent upon calcium influx.*,‘3.14,‘7,39~52,65 We therefore investigated the effects of NMDA, kainate or glutamate treatment on the survival of calretinin immunopositive and calretinin immunonegative neurons. Following a 5 min exposure to 1 mM NMDA or 1 mM kainate, total neuronal survival fell after 3 h to 0
CR+ population
m
Total suwval
100
80
20
0
Control
A23187
A23167 Ca2+ free
n
A23167 + APV/CNQX
Fig. 4. Calcium-dependence of A23187 neurotoxicity. Ordinate represents either the proportion of total neurons surviving treatment, or the proportion of neurons (not normalized) out of the total population of surviving neurons that stained positively for calretinin. A23187 was applied at 3 p M in the presence or absence of added Ca* + , or in the presence of APV (50 p M) plus CNQX (50 p M) to block glutamate neurotransmission. Data are from a single batch of cultures; n = 4 for each condition. CR+, calretinin immunopositive.
312
W.
Control
NMDA
Kainate
LUKAS
and K. A.
Glutamate
Fig. 5. ERect of excitatory amino acids on calretinin immunopositive and calretinin immunonegative neuronal survival (normalized as described for Fig. 3) 3 h after a 5 min exposure to the agonist. NMDA, kainate and glutamate were each added at I mM final concentration. Values represent mean k S.E.M. (*P < 0.05; ***P < 0.001). Data are pooled from eight experiments; n = 20 (NMDA). I8 (kainate), 26 (glutamate) and 25 (control).
69% and 63%, respectively (Table I). At the same concentration, glutamate lowered the total survival to 35%. For all three agonists, the calretinin immunopositive population of neurons was preferentially spared (Fig. 5). The most significant effects were observed for NMDA and kainate which caused a 32% and 40% loss, respectively, of calretinin immunonegative neurons and only a 2% and 10% loss of the calretinin immunopositive population (Fig. 5). Glutamate treatment (1 mM) resulted in substantial losses in both neuronal populations, although again the calretinin immunopositive population was significantly spared: 66% of calretinin immunonegative neurons died while only 47% of calretinin immunopositive neurons degenerated.
DISCUSSION
We have shown in cultures of cerebral cortex that neurons containing the calcium-binding protein calretinin are relatively resistant to cell death induced by various excitotoxins. Two independent methods were employed to induce neuronal cell death: application of the glutamate receptor agonists NMDA, kainate and glutamate, as well as the calcium ionophore A23187. As expected, we observed a range of cell loss depending upon the type of treatment and dose; in all instances where cell death occurred there was significantly higher survival within the population of neurons containing calretinin. This selective resistance of the calretinin immunopositive population is very similar to that previously shown for hippocampal neurons that contain the related protein calbindin.42 In that study cultured neurons immunopositive for calbindin were approximately six times more likely than immunoneg-
JONES
ative neurons to survive a 4 h treatment of A23187. As in our study, the threshold concentration of excitatory amino acids and A23187 required to cause neurodegeneration was greater in neurons possessing high levels of the calcium-binding protein. In the experiments with hippocampal neurons, A23187 at I PM induced an influx of calcium ions that elevated intracellular calcium to five- to six-fold above the resting level. In those neurons immunopositive for calbindin this elevation was transient while immunonegative neurons displayed a longer duration increase in intracellular Cal+ In a similar manner, clonal cell lines stably expressing calbindin were better able to buffer calcium entering through voltage-dependent Ca’ ’ channels or through ionophore pores than were cells not expressing the protein.75.4X While we cannot comment directly on a possible correlation between levels of intracellular free Ca” and calretinin, our results with a neurotoxicity assay support a similar role for calretinin as calbindin in enhancing calcium homeostasis and survival during periods of dramatically increased calcium influx. The chemical properties of both calretinin and calbindin are consistent with a role for buffering transiently elevated calcium.5h The two proteins have 59% amino acid homology5’,5’,7’ and data available for calbindin indicate an affinity for free Ca* ’ equal to approximately 0.5 ~LM.~ In addition, both calciumbinding proteins are expressed abundantly in certain areas of the brain (up to 1% of soluble protein).‘.5’,7” The added buffering capacity offered by the high level expression of these calcium-binding proteins is considerable and may have consequences for diverse cellular processes, Still, it remains unresolved if calcium-binding proteins simply buffer and promote rapid exchange of intracellular Cal +, or whether they have a more specialized catalytic function in related that dicellular processes. 47.74Further experiments rectly manipulate the level of expression of calciumbinding proteins in neurons will provide important details of their roles in calcium buffering and neuroprotection. Selective vulnerability is operationally defined by the type of chemical or pathological insult chosen, and the endogenous properties that enable select cells to survive the induced stress. Survival following receptordependent insults, such as the uniform application of glutamate to a cell culture, obviously reflects the diversity and distribution of receptor subtypes, as well as the biochemical heterogeneity within a mixed population of neurons. A differential density of NMDA and non-NMDA receptors, for example, is hypothesized to account for the selective vulnerability of some subpopulations of cortical neurons in culture.6’ On the other hand, toxins such as calcium ionophores induce a more consistent insult that is independent of the expression patterns of plasma membrane receptors.4” We found that cell death induced by A23187 was dependent on extracellular Ca2+, as previously observed,” and at the same time independent of the
Calretinin and calcium overload in cortical cultures release of glutamate from synaptic terminals. Since it is known that glutamate release following chronic depolarization contributes to excitotoxicity in cultures,6 it was important to show that the trigger for cell death was mediated directly by Ca2+ entry through the ionophore and not by the possible subsequent stimulation of postsynaptic glutamate receptors. The distinction between glutamate receptor-dependent and receptor-independent toxicity is significant if the experimental goal is to determine the neuroprotective potential of an intracellular entity such as a calcium-binding protein. The fact that calretinin immunopositive neurons preferen.tially survive the application of excitatory amino acids could be attributed to a low density of glutamate receptors on their surfaces, however, they also resist the neurotoxicity of A23187. These findings strongly suggest that the subpopulation of calretinin immunopositive neurons survive because they have an increased ability to control Ca2 + -induced toxicity.. While the evidence from this and other in vitro studies supports a role for calcium-binding proteins in reducing the toxic effects of elevated intracellular Ca2+, the situation in viuo is more complex. For example, some populations of neurons containing calcium-binding proteins in the hippocampus are relatively spared in animal models of neurodegeneration 334,63 but others in hippocampus and elsewhere are selectively reduced in number.15~‘5a*37a Differences in survival patterns within a biochemically characterized class of neurons may be due to diversity in their afferent inputs and the nature of the neurotoxic insult. Neurons in the dentate gyrus, for instance, are exceptionally vulnerable to ischemia, and in a recent study Freund and Mag16czky15” show that calretinin immunopositive cells are not spared this insult. In this region many neurons, including those containing calretinin, receive a rich glutamatergic innervation and appear to have an exceptionally high density of glutamate receptors, ‘6.59features which are unlikely to be preserved in dissociated cultures. These observations suggest that patterns of afferent input and receptor density are additional important factors in determining survival from neurotoxicity in viuo. Therefore, while calretinin may have an important calcium-buffering capacity during normal electrical activity, in certain pathological states this capacity may be rapidly over-loaded. The role of calciumbinding proteins with regard to excitotoxicity in oivo may be clarified by studying neuronal survival in transgenic animals lacking expression of one or more of these proteins. Expression of calbindin in the hippocampus can be upregulated following administration of corticosterone,22 nerve growth factoti’ or stimulation of the perforant path. ” It is interesting to speculate, therefore, that certain stressful stimuli will induce potentially neuroprotective reactions such as an increased production of calcium-binding proteins. These regulatory events typically occur many hours or days after possible secondary
313
the stimuli. In the present study we interpret the increase in the proportion of calretinin immunopositive neurons as a relative sparing phenomenon and not as a dramatic upregulation of calretinin in response to noxious chemical stimuli. This interpretation is based on the finding that the calculated number of calretinin immunopositive neurons either remained unchanged or decreased following treatments, and on the assumption that significantly increased production of calretinin protein would likely take more than 3 h. Emerging details of the kinetics of changing intracellular calcium concentrations in response to glutamate is illuminating mechanisms of excitotoxic cell death. Previously it has been shown that cell death is triggered primarily by an influx of extracellular Ca2+ 8~14~‘7,39 and, at least for some agonists, an additional release of Ca2+ from internal stores.‘3,‘4 More recently using long duration microfluorimetry, it has been revealed that glutamate exposure leads to a transient rise in intracellular Ca2+ that lasts 5-10 min.65 Depending upon the individual cell and the dose of agonist, this primary calcium response is followed, minutes or hours later, by a secondary sustained rise in calcium that immediately precedes or coincides with cell death.52,65Little is known about the sequence of events that occurs during the period of time between the primary and secondary increases in intracellular Ca* + . Pretreatment with gangliosides or sphingosine, inhibitors of protein kinase C, can block the development of the secondary response without effecting the primary increase induced by glutamate’0~“~68or A23 187.49The key subcellular disturbances initiated by the activation of calciumdependent enzymes have not yet been identified, but likely candidates include fragmentation of DNA, disruption of plasma membrane cytoskeleton, and inhibition of mitochondrial function.50 It will be important for future studies to determine where in the cascade of events leading to cell death calcium-binding proteins may interfere.
CONCLUSION
Our results in vitro demonstrate that neurons containing the calcium-binding protein calretinin are more resistant to cell death caused either by excitatory amino acids or by calcium ionophore. We hypothesize that this protective effect is due to the Ca2+ -buffering capacities of calretinin in a manner analogous to that suggested for calbindin. Acknowledgements-We would like to thank M. Celio for helpful discussions during the early phases of this work and for his generous gift of calretinin antibody, and J. Knowles and S. Cats&s for their support and encouragement. We also wish to thank N. Hussy, J.-C. Martinou and S. Rabacchi for their comments on the manuscript, and C. Hebert for help in preparing the photographs.
314
W. LUKAS and K. A. JONES REFERENCES
1. Arai R., Winsky L., Arai M., Jacobowitz D. M. (1991) Immunohistochemical localization of calretinin in the rat hindbrain. J. camp. Neurol. 310, 21.-44. 2. Baimbridge K. G., Celia M. R. and Rogers J. H. (1992) Calcium-binding proteins in the nervous system. Trends Neurosci. 15, 303-308. 3. Baughman R. W., Huettner J. E., Jones K. A. and Khan A. (1991) Cell culture of neocortex and basal forebrain from postnatal rats. In Culturing Neroe Cells (eds Banker G. and Goshn K.), pp. 2277250. MIT Press, Cambridge. 4. Bredderman P. J. and Wasserman R. H. (1974) Chemical composition, affinity for calcium, and some related properties of the vitamin D-dependent CaBP. Biochemisfry 13, 1687-1694. 5. Brown D. A. and Griffith W. H. (1983) Persistent slow inward calcium current in voltage-clamped hippocampal neurones of the guinea-pig. J. Physiol. 337, 303-320. 6. Carat-oh E. (1987) Intracellular calcium homeostasis. A. Rar. Biochem. 56, 395-433. 7. Celio M. R. (1990) Calbindin D-28k and parvalbumin in the rat nervous system. Neuroscience 35, 375-475. 8. Choi D. W. (1985) Glutamate neurotoxicity in cortical cell culture is calcium dependent. Neurosci. Left. 58, 293-297. 9. Choi D. W. (1988) Glutamate neurotoxicity and diseases of the nervous system. Neuron 1, 6233634. IO. Connor J. A., Wadman W. J., Hockberger P. E. and Wong R. K. (1988) Sustained dendritic gradients of Ca’+ induced by excitatory amino acids in CA1 hippocampal neurons.-Science 240, 649-653. G. A.. Manev H.. Guidotti A.. Costa E. and Brooker G. (1990) Ganaliosides normalize distorted II. de Erausauin single-cell’intracellular free Ca *+ dynamics after toxic doses of glutamate in derebdllar g&mle cells. Proc. n&n. Acad. Sci. U.S.A. 87, 8017-8021. 12. Ferrer I., Tufion T., Soriano E., de1 Rio A., Iraizos I., Fonseca M., Guionnet N. (1993) Calbindin D-28k immunoreactivity in the temporal neocortex in patients with Alzheimer’s disease. Clin. Neuropathol. 12, 53-58. 13. Frandsen A. and Schousboe A. (1992) Mobilization of dantrolene-sensitive intracellular calcium pools is involved in the cytotoxicity induced by quisqualate and N-methyl-o-aspartate but not by 2-amino-3-(3-hydroxy-5-methylisoazol-4yl)propionate and kainate in cultured cerebral cortical neurons. Proc. nafn. Acad. Sci. U.S.A. 89, 2590-2594. 14. Frandsen A. and Schousboe A. (1993) Excitatory amino acid-mediated cytotoxicity and calcium homeostasis in neurons. J. Neurochem. 60, 1202~1211. 15. Freund T. F., Buzsaki G., Leon A., Baimbridge K. G. and Somogyi P. (1990) Relationship of neuronal vulnerability and calcium binding protein immunoreactivity in ischemia. Expl Bruin Res. 83, 55-66. 15a. Freund T. F. and Magloczky 2s. (1993) Early degeneration of calretinin-containing neurons in the rat hippocampus after ischemia. NeuroscGence 56, 58 I -596. D. M. and Freund T. F. (1992) Calretinin is present in non-pyramidal cells 16. Gulyas A. I., Miettinen R., Jacobowitz of the rat hippocampus-I. A new type of neuron specifically associated with the mossy fibre system. Neuroscience 48, 1-27. receptor-induced “%Za” 17. Hartley D. M., Kurth M. C., Bjekness L., Weiss J. H. and Choi D. W. (1993) Glutamate accumulation in cortical cell culture correlates with subsequent neuronal degeneration. J. Neurosci. 13, 1993.-2000. disorders. Trend~y 18. Heizmann C. W. and Braun K. (1992) Changes in Ca 2+-binding proteins in human neurodegenerative Neurosci. 15, 259-264. of P-type calcium 19. Hillman D., Chen S., Aung T. T., Cherksey B., Sugimori M. and Llinas R. R. (1991) Localization channels in the central nervous system. Proc. nafn. Acad. Sci. U.S.A. 88, 7076G7080. J. H. (1991) Neocortical neuronal subpopulation labeled by a monoclonal antibody to 20. Hof P. R. and Morrison calbindin exhibit differential vulnerability in Alzheimer’s disease. Expl Neural. 111, 293-301. neocortical 21. Hof P. R., Nimchinsky E. A., Ceho M. R., Bouras C. and Morrison J. H. (1993) Calretinin-immunoreactive interneurons are unaffected in Alzheimer’s disease. Neurosci. Letf. 152, 145-149. 22. Iacooino A. M. and Christakos S. (1990) Corticosterone regulates calbindin-D28k mRNA and protein levels in rat hippbcampus. J. biol. Chem. 265, ltJ,177~10,180. 23. Iacopino A. M. and Christakos S. (1990) Specific reduction of calcium-binding protein (28-kilodalton calbindin-D) gene expression in aging and neurodegenerative diseases. Proc. n&n. Acad. Sci. U.S.A. 87, 4078-4082. 24. Iacopino A., Christakos S., German D., Sonsalla P. K. and Altar C. A. (1992) Calbindin-D28k-containing neurons in animal models of neurodegeneration: possible protection from excitotoxicity. Molec. Bruin Res. 13, 251-261. 25. Iacopino A. M., Christakos S., Modi P. and Altar C. A. (1992) Nerve growth factor increases calcium binding protein (CalbindinD28k) in rat olfactory bulb. Brain Res. 578, 305S310. 26. Ichimiya Y., Emson P. C., Mountjoy C. Q., Lawson D. E. M. and Heizmann C. W. (1988) Loss of calbindin-28k immunoreactive neurones from the cortex in Alzheimer-type dementia. Brain Res. 475, 156-159. 27. Inaguma Y., Shinohara H.. Inagaki T. and Kato K. (1992) Immunoreactive parvalbumin concentrations in parahippocampal gyrus decrease in patients with Alzheimer’s disease. J. neurol. Sci. 110, 57761. 28. Jacobowitz D. M. and Winsky L. (1991) Immunocytochemical localization of calretinin in the forebrain of the rat. J. cotnp. Neurol. 304, 198-218. 29. Keilhoff G. and Erdii S. L. (1991) Parallel development of excitotoxic vulnerability to N-methyl-D-aspartate and kainate in dispersed cultures of the rat cerebral cortex. Neuroscience 43, 35-40. of non-NMDA receptors does not block rapidly triggered 30. Koh J. and Choi D. W. (1991) Selective blockade glutamate-induced neuronal death. Brain Res. 548, 318-321. of cultured cortical neurons to damage by excitotoxins: Differential 31 Koh J. and Choi D. W. (1988) Vulnerability susceptibility of neurons containing NADPH-diaphorase. J. Neurosci. 8, 2153-2163. receptor-mediated neurotoxicity in 32 Koh J. Y., Goldberg M. P., Hartley D. M. and Choi D. W. (1990) Non-NMDA cortical culture. J. Neurosci. 10, 693-705. are selectively resistant to quinohnate 33 Koh J., Peters S. and Choi D. W. (1986) Neurons containing NADPH-diaphorase toxicity. Science 234, 73376. proteins are concentrated in the CA2 field of the monkey 34 Leranth C. and Ribak C. E. (1991) Calcium-binding hippocampus: a possible key to this region’s resistance to epileptic damage. Expl Brain Res. 85, 129-136. B., Morton A. J., Emson P. C. and Mason W. T. (1992) Stable transfection of 35 Lledo P. M., Somasundaram calbindin-D28k into the GH3 cell line alters calcium currents and intracellular calcium homeostasis. Neuron 9,943-954.
Calretinin
and calcium
overload
in cortical
cultures
315
36. Llinas R. and Sugimori M. (1980) Electrophysiological properties of in vitro Purkinje cell dendrites in mammalian cerebellar slices. J. Physiol. 305, 197-213. 37. Lowenstein D. H., Miles M. F., Hatam F. and McCabe T. (1991) Up regulation of calbindin-D28k mRNA in the rat hippocampus following focal stimulation of the perforant path. Neuron 6, 627-633. 37a. Maeloczkv Zs. and Freund T. F. (1993) Selective neuronal death in the contralateral hitmocamnus _. . followine: unilateral kainite injections into the CA3 subfield. Neuroscience 56, 317-336. 38. Mattson M. P. (1992) Calcium as sculptor and destroyer of neural circuitry. Expl Gerontol. 27, 29-49. messengers in the generation and degeneration of 39. Mattson M. P., Guthrie P. B. and Kater S. B. (1988) Intracellular hippocampal neuroarchitecture. J. Neurosci. Res. 21, 447-464. 40. Mattson M. P., Guthrie P. B. and Kater S. B. (1989) A role for Na+-dependent Ca’+ extrusion in protection against neuronal excitotoxicity. Fedn Am. Sots exp. Biol. J. 3, 2519-2526. 41. Mattson M. P. and Kater S. B. (1989) Development and selective neurodegeneration in cell cultures from different hippocampal regions. Bruin Res. 110-125. 42. Mattson M. P., Rychlik B., Chu C. and Christakos S. (1991) Evidence for calcium-reducing and excitoprotective roles for the calcium-binding protein calbindinD28k in cultured hippocampal neurons. Neuron 6, 41-51. 43. Miettinen R., Gulyas A. I., Baimbridge K. G., Jacobowitz D. M. and Freund T. F. (1992) Calretinin is present in non-pyramidal cells of the rat hippocampus-II. Co-existence with other calcium binding proteins and GABA. Neuroscience 48, 29-43. 44. Miller R. J. (1991) The control of neuronal Ca2+ homeostasis. Prog. Neurobiol. 37, 255-285. 45. Mills L. R. and Kater S. B. (1990) Neuron-specific and state-specific differences in calcium homeostasis regulate the generation and degeneration of neuronal architecture. Neuron 2, 1499163. 46. Monyer H., Giffard R. G., Hartley D. M., Dugan L. L., Goldberg M. P. and Choi D. W. (1992) Oxygen or glucose deprivation-induced neuronal injury in cortical cultures is reduced by tetanus toxin. Neuron 8, 967-973. 47. Morgan D. W., Welton A. F., Heick A. E., Christakos S. (1986) Specific in vitro activation of Ca, Mg-ATPase by vitamin D-dependent rat renal calcium binding protein (calbindin-D28K). Biochem. biophys. Res. Commun. 138, 547-553. 48. Muir E. M., Lawson D. E. M., Sepulveda F. V., O’Brien J. A. and Harding M. (1993) Calbindin D28K expression in transfected mouse NIH3T3 cells. Cell Calcium 14. 349-358. 49. Nakamura K., Wu G. and Ledeen R. W. (1992) Protection of Neuro-2a cells against calcium ionophore cytotoxicity by gangliosides. J. Neurosci. Res. 31, 245-253. 50. Nicotera P., Bellomo G. and Orrenius S. (1992) Calcium-mediated mechanisms in chemically induced cell death. A. Rev. Pharmac. Toxicol. 32, 449-470. 51. Parmentier M. and Lefort A. (1991) Structure of the human brain calcium-binding protein calretinin and its expression in bacteria. Eur. J. Biochem. 196, 79-85. 52. Randall R. D. and Thayer S. A. (1992) Glutamate-induced calcium transient triggers delayed calcium overload and neurotoxicity in rat hippocampal neurons. J. Neurosci. 12, 1882-1895. 53. Reed P. W. and Lardy H. A. (1972) A23187: a divalent cation ionophore. J. biol. Chem. 247, 6970-6977. 54. Resibois A. and Rogers J. H. (1992) Calretinin in rat brain: an immunohistochemical study. Neuroscience 46, 101-134. 55. Rogers J. H. (1987) Calretinin: a gene for a novel calcium-binding protein expressed principally in neurons. J. Cell Biol. 105, 134331353. 56. Rogers J. H. (1991) Calretinin. In Novel Calcium-Binding Proteins (ed. Heizmann C. W.), pp. 251-276. Springer, Berlin. 57. Rogers J. H. (1992) Immunohistochemical markers in rat cortex: co-localization of calretinin and calbindin-D28k with neuropeptides and GABA. Bruin Res. 587, 147-157. 58. Rogers J. H. and Resibois A. (1992) Calretinin and calbindin-D28k in rat brain: patterns of partial co-localization. Neuroscience 51, 843-865. 59. Scharfman H. E. and Schwartzkroin P. A. (1989) Protection of dentate hilar cells from prolonged stimulation by intracellular calcium chelation. Science 246, 257-260. 60. Schousboe A., Frandsen A., Wahl P., Drejer J. and Honore T. (1991) Excitotoxicity in cultured neurons: Role of second messenger systems and intracellular Ca2+ homeostasis. In Excitatory Amino A&Is (eds Meldrum B. S., Moroni F., Simon R. P. and Woods J. H.), pp. 5933599. Raven Press, New York. 61. Schwaller B., Buchwald P., Bliimcke I., Hunziker W. and Celia M. R. (1994) Characterization of a polyclonal antiserum against the purified human recombinant calcium binding protein calretinin. Cell Calcium (in press). 62 Siesjo B. K. and Bengtsson F. (1989) Calcium fluxes, calcium antagonists, and calcium-related pathology in brain ischemia, hypoglycemia, and spreading depression: a unifying hypothesis. J. cerebr. Blood Flow Met. 9, 127-140. 63 Sloviter R. S. (1989) Calcium-binding protein (calbindin-D28k) and parvalbumin immunocytochemistry: localization in the rat hippocampus with specific reference to the selective vulnerability of hyppocampal neurons to seizure activity. J. coma. Neurol. 280. 183-196. M. K., Wong L., Somerville M. J., Yoong L. K. K., Bergeron C., Parrnentier M. and McLachlan D. R. 64 Sutherland (1993) Reduction of calbindin-28k mRNA levels in Alzheimer as compared to Huntington hippocampus. Molec. Brain Res. 18, 32-42. 65. Tymianski M., Charlton M. P., Carlen P. L. and Tator C. H. (1993) Source specificity of early calcium neurotoxicity in cultured embryonic spinal neurons. J. Neurosci. 13, 1085-2104. 66. Van Eldik L. J., Zendegui J. G., Marshak D. R. and Watterson D. M. (1982) Calcium-binding proteins and the molecular basis of calcium action. Inf. Rev. Cyrol. 77, I-61. 67. Vickers J. C., Huntley G. W., Edwards A. M., Moran T., Rogers S. W., Heinemann S. F. and Morrison J. H. (1993) Quantitative localization of AMPA/kainate and kainate glutamate receptor subunit immunoreactivity in neurochemitally identified subpopulations of neurons in the prefrontal cortex of the macaque monkey. J. Neurosci. 13,2982-2992. 68. Wadman W. J. and Connor J. A. (1992) Persisting modification of dentritic calcium influx by excitatory amino acid stimulation in isolated CA1 neurons. Neuroscience 48, 293-305. 69. Weiss J. H., Koh J. Y., Baimbridge K. G. and Choi D. W. (1990) Cortical neurons containing somatostatinor parvalbumin-like immunoreactivity are atypically vulnerable to excitotoxic injury in vitro. Neurol. 40, 1288-1292. 70. Winsky L. and Jacobowitz D. M. (1991) Radioimmunoassay of calretinin in the rat brain. Neurochem. Int. 19, 517-522.
W.
316
LUKAS
and K. A.
JONES
71. Winsky L., Nakata H., Martin B. M. and Jacobowitz D. M. (1989) Isolation, partial amino acid sequence, and immunohistochemical localization of a brain-specific calcium-binding protein. Proc. natn. Acad. Sci. U.S.A. 86, 10,139s10,143. 72. Wulf J. and Pohl W. G. (1977) Calcium ion-flux across phosphatidylcholine membranes mediated by ionophore A23187. Biochim. biophys. Acta 465, 471-485. 73. Yamada T., McGeer P. L., Baimbridge K. G. and McGeer E. G. (1990) Relative sparing in Parkinson’s disease of substantia nigra dopamine neurons containing calbindin-D28k. Brain Rex 526, 303-307. 74. Yamaguchi T., Winsky L. and Jacobowitz D. M. (1991) Calretinin, a neuronal calcium binding protein, inhibits
phosphorylation
of a 39 kDa synaptic membrane protein from rat brain cerebral cortex. Neurosci. Letf. 131, 79-82. (Accepted 19 January 1994)
0306-4522(94)E0068-F
Vol. 61, No. 2, pp. 307-316, 1994 Elsevier Science Ltd Copyright 0 1994 IBRO Printed in Great Britain. All rights reserved 0306-4522/94 $7.00 + 0.00
CORTICAL NEURONS CONTAINING CALRETININ ARE SELECTIVELY RESISTANT TO CALCIUM OVERLOAD AND EXCITOTOXICITY IN JfZTRO W. Glaxo
Institute
for Molecular
LUKAS
Biology,
and K. A. JONES* 14 chemin
des Aulx, CH-1228
Geneva,
Switzerland
Abstract-Calbindin and the more recently identified protein calretinin are structurally related calciumbinding proteins having a broad distribution in the brain. Recent evidence supports a neuroprotective role for calbindin in regulating calcium homeostasis during periods of heightened Ca2+ influx. It is not known if calretinin might have a similar function. We investigated if calretinin-containing neurons have a survival advantage in rat neocortical cultures treated with a calcium ionophore or excitatory amino acids. Neuronal cultures were challenged with the calcium ionophore A23187 at different concentrations to produce a broad range of cell death. Cell loss was quantified for both the calretinin immunopositive and the calretinin
immunonegative populations of neurons. We found that 3 h after exposure to 2 PM A23187 there was a 48% loss of the calretinin immunonegative population of neurons whereas the calretinin immunopositive set of neurons was reduced by only 18%. Calretinin positive neurons were still relatively spared after treatment with 3 PM A23187. The ionophore had no cytotoxic effect when calcium ions were removed from the extracellular medium. We also studied glutamate excitotoxicity by treating the neuronal cultures with the excitatory amino acids glutamate, N-methyl-D-aspartate or kainate for Smin and examining survival three hours later. We found again that calretinin-containing neurons were relatively spared after exposure to the excitatory amino acids; at doses of N-methyl-D-aspartate and kainate that produced a 3240% loss of calretinin immunonegative neurons, only 2-10% of calretinin immunopositive neurons died. Similar results were obtained for glutamate. These results demonstrate that neurons containing calretinin are better able to survive disturbances in calcium homeostasis than cells not containing this calcium-binding protein. The fact that this effect was observed with ionophore treatment, as well as excitatory amino acids, suggests that neither the density nor distribution of glutamate receptors on the different cell types was a factor in determining selective vulnerability. We hypothesize that the neuroprotective effect of calretinin is due to the buffering capacities of the protein in a manner analogous to that suggested for calbindin.
phores,33,4~,42,45
Calcium ions play an important signalling role in neurons, and several homeostatic mechanisms are required to maintain their normally low intracellular levels.6 Malfunction or overload of the cellular systems that regulate calcium homeostasis results in rapid neurodegeneration. 9,44,62There is considerable evidence from studies of primary neuronal cultures that excessive calcium influx caused by the stimulation of glutamate receptors triggers an excitotoxic process that leads to cell death.8,‘4,‘7,39,52,65 Several types of glutamate reeeptors, including members of the Nmethyl-D-aspartate (NMDA) and non-NMDA classes, czua induce this process when over-stimulated.30s32Additionally, it is possible to observe neurotransmitter receptor-independent neurotoxicity by the use of ionophores selective for calcium.45 Not all types of neurons are equally sensitive to the toxic effects of glutamate or calcium iono-
and a robust calcium buffering capacity is one property that may enable some neurons to survive in the face of pathological disturbances.1s,38 In support of this idea it has been shown that intracellular injection of a calcium chelator can block excitotoxicity at an early stage. 59 Two calcium-binding proteins, calbindin and parvalbumin, are believed to have a physiological role in buffering intracellular calcium ions,42,“@’ and both are abundantly expressed in various populations of neurons in the brain.2.7 Indeed, there is direct evidence to show that the calcium-binding protein calbindin buffers stimulated calcium influx,35,42~48and is present in a select population of hippocampal neurons that survive excitotoxic conditions in vitro.42 The neuroprotective role for calciumbinding proteins in vivo is more difficult to demonstrate. Some populations of neurons containing calcium-binding proteins are relatively spared in animal models of neurodegeneration,34,63 but others are selectively reduced in number.‘5~‘5a~37a The relationship between calcium overload and neuronal vulnerability has encouraged studies of calcium-binding proteins in human neurodegenerative diseases. For example, a calbindin-containing subpopulation of neurons in the substantia nigra is spared
*To whom correspondence should be addressed. Abbreviations: APV. 2-amino-5-vhosvhonovaleric acid: 6-cyano-7-nitroquiioxaline-2,3-dione; HBS; CNQX, HEPES-buffered saline: HEPES. N-2-hvdroxvethvlpiperazine-N’-2-ethanes&phonic acid; I&DA, IN. methyl-D-aspartate. 307
W. LUKAS and K. A. JONES
308
in Parkinson’s disease73 and in an animal model of this disease,24 and neocortical neurons expressing high levels of calbindin or calretinin are preserved in Alzheimer’s disease.“.” On the other hand, it has been suggested that the observed decreases in calbindin mRNA and protein content in other populations of neurons may accelerate cell loss in Alzheimer’s’2.26.27.”and other neurodegenerative diseases.23 These studies support the hypothesis that maintained expression of calcium-binding proteins may be neuroprotective in some instances of human neurodegeneration. Calretinin is a relatively new member of a family of calcium-binding proteins that include calbindin and other proteins sharing the EF-hand domain.s5,56.71 Anatomical studies show that calretinin is contained in many types of neurons scattered throughout the nervous system ‘.28,54 including the neocortex,28,57 and that there is little overlap with neuronal populations containing calbindin. 43,58,70 Since calretinin displays considerable sequence homology with calbindin,5’,55,7’ and like calbindin, it is abundantly expressed in some neurons,51.70it may also be a determinant of selective neuronal vulnerability to excitotoxicity. In the present study, we investigate the possibility that calretinin-containing neurons in cultures of cerebral cortex are resistant to calcium-induced excitotoxicity induced by calcium ionophore or excitatory amino acids.
EXPERIMENTAL
PROCEDURES
Cell culture Cortical cultures containing both glia and neuronal cells were prepared as previously described.3 Briefly, pieces of visual cortex of one- to two-day-old rats were incubated with papain (Worthington Biochemical) for 30 min in a shaking bath at 32-34°C. The enzymatic solution was then replaced three times by a buffered solution containing trypsin inhibitor, and the tissue was dissociated in growth medium by trituration with a fire-polished glass pipette. The dissociated cells were plated (7.5 x lo5 cells/ml) in the central well of 35 mm plastic dishes containing a grid (Milian Instruments, SA) that aided in the counting of the same population of neurons before and after experimental treatments. The dishes were previously treated with a solution containing 200 pg/ml of poly-D-lysine and 33 pg/ml of laminin (Collaborative Research). Growth medium consisted of Minimum Essential Medium (GIBCO) supplemented with penicillin 25 U/ml, streptomycin 25 pg/ml, ducose 20mM. alutamine 0.5 mM and 10% (v/v) heat mactivated fetal calf serum (GIBCO). Cultures were maintained in a humidified incubator (5% CO,/95% air) at 37°C. Five days after plating, cytosine p-D-arabino-furanoside I
(5 PM) was added to avoid further proliferation of glial cells. Half of the medium was removed two days later and replaced by fresh medium containing 5% rat serum (Harlan Bioproducts for Science). Since excitatory amino acid toxicity develops over two weeks in cultures,*,29,M we used 12~14-day-old cultures, which exhibit robust and reproducible excitotoxicity, for all of our experiments, Treatment with excitatory amino acid.s and ionophore Before treatment cultures were washed with a HEPESbuffered saline (HBS) consisting of NaCl, 137mM; KCI, 5 mM; CaCl,, 4 mM; MgCI,, 2 mM; HEPES, 10 mM; 0.5% glucose, pH 7.2.“” Cultures were exposed to L-glutamate, NMDA or kainate dissolved in HBS, for 5 min, washed and maintained in HBS at 37°C for 3 h, and then fixed with 4% formaldehyde for 20 min. Stock solution of A23 187 (Calbiochem), dissolved in dimethylsulphoxide, was diluted in HBS and applied for 3 h at 37°C before formaldehyde fixation. In some experiments the NMDA and non-NMDA receptor antagonists, 2-amino-5-phosphonovalerate (APV) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), were incubated with A23187 at final concentrations of 50 PM. Calretinin immunohistochemistry Calretinin immunostaining was performed using the peroxidase-antiperoxidase method. Cultures were incubated overnight at 4°C with a polyclonal rabbit anti-human calretinin antibody6’ (gift of M. Celia) diluted 1: 2000 in modified tris-buffered saline3 containing 2% bovine serum albumin, 5% rat serum and 0.2% Triton X-100. This antibody does not cross-react with the related protein calbindin.’ Goat anti-rabbit secondary antibody and mouse peroxidase-antiperoxidase (Sternberger Monoclonals) were consecutively applied for 1 h at room temperature. Staining was revealed with a solution containing diaminobenzidine (0.5 mg/ml) and 0.03% H,O, in Tris-buffered saline. Cell counting and analysis
qf neuronal survival
Neuronal survival was calculated by counting neurons 3 to 24 h before treatment with excitotoxins and again after fixation of the cultures. Counting was performed at 100 x magnification with the aid of the labelled grid in the central well of the Petri dishes. Living neurons, identified as having round or oval phase-bright somata, were easily distinguishable from underlying glia. This method of quantifying living neurons correlates well with other methods that use measurements of lactate dehydrogenate release” or Trypan Blue dye exclusion. 39 Estimates of the survival of calretinin immunopositive and calretinin immunonegative neurons were obtained by counting calretinin immunopositive and calretinin immunonegative neurons in a previously determined area of each dish, and expressing these quantities as percentages of the total number of neurons counted in each respective area before treatment. These percentages were then normalized to the mean per cent of calretinin immunopositive and calretinin immunonegative neurons in the cultures treated with vehicle alone. For each experimental condition 5-10% of the total culture surface was counted in each of three Petri dishes, representing 200-400 neurons per dish. All statistical comparisons were done using the unpaired Student r-test, and values presented are the mean k S.E.M.
Fig. 1. Calretinin immunostaining of two-week-old neurons in primary culture. Photographs were taken of two sister cultures either untreated (HBS alone--A, B) or treated with 1 PM A23187 (C, D). Phase image (A) shows several healthy phase-bright neurons, of which three are calretinin immunopositive (asterisks) as seen under bright field illumination (B). Photomicrographs of the sister culture treated with A23187 show three phase-bright neurons (arrowheads, D), of which two are calretinin immunopositive and one calretinin immunonegative, and at least three degenerating cells (arrows, C) that are calretinin immunonegative and exhibit loss of soma integrity and extensively beaded neurites. Background shows the lettered grids that were used to relocate and quantify surviving neurons. Scale bar = 35 pm.
Fig. I.
W. LUKAS and K. A. JONFS
310
Fig. 2. High magnification view of an untreated culture grown on a glass coverslip shows calretinin immunoreactive material throughout the soma and dendrites of a calretinin immunopositive neuron and in a punctate pattern along the processes of two neighboring calretinin immunonegative neurons. Scale bar = 10,um.
RESULTS
Calretinin-immunopositive neurons represented a subpopulation of neocortical neurons in culture that averaged 28 f 3% (mean + S.E.M. of 17 different culture preparations) of the total population. Although this proportion varied from preparation to preparation, the variability within any particular experiment was always less than 6%. Staining within
Table 1. Total neuronal survival in cortical cultures following exposure to calcium ionophore or excitatory amino acids
Control A23187 A23187 A23187
(%)
n
1
65
85 k 3 53 f 4 28 i 4
8 16 29
69 + 4 63 + 3 34 + 4
20 18 26
Survival
Condition
95+ l/rM 2pM 3/1cM
NMDA 1 mM Kainate 1 mM Glutamate 1 mM
Values are mean f S.E.M. of n number of cultures from eight to nine independent experiments.
calretinin immunopositive neurons in two week-old cultures was localized to the cytoplasm of the soma and often throughout the dendritic tree (Figs 1, 2). Reaction product was not present in nuclei of calretinin immunopositive neurons nor in the underlying glial cells. A subset of calretinin immunopositive cells showed intense staining of very fine, presumably axonal processes, and it was possible to observe a punctate staining pattern on the soma and processes of many calretinin-immunonegative neurons (Fig. 2). Thus, in many cases, calretinin appears to be present throughout the entire neuronal structure, including within axon terminals. This staining pattern is very similar to that previously observed in neuronal tissue in uizlo.54 Calretinin immunoreactivity was not detected in very young cultures (
Calretinin
1 PM
2 PM
and calcium
311
overload in cortical cultures
3 FM
[A231 871
Fig. 3. Normalized percentages of calretinin immunopositive and calretinin immunonegative neurons resistant to 3 h exposure to calcium ionophore A23187. Values are normalized to the mean survival of calretinin immunopositive and calretinin immunonegative neurons in the control cultures (HBS alone) and expressed as the mean & S.E.M. (**P < 0.01; ***P < 0.001). The data are pooled from nine separate experiments; n = 8 (1 PM), 16 (2 PM), 29 (3 PM) and 33 (control). CR+, calretinin immunopositive; CR-, calretinin immunonegative.
the survival of the calretinin immunopositive and calretinin immunonegative populations of neurons. One to two hours following the period of treatment with ionophore, many neurons displayed characteristics of dead or dying cells: phase-dark, non-distinct plasma membranes, cytoplasmic vacuoles, and broken or beaded dendrites (Fig. IC, D). When cultures were allowed to survive for longer periods these phase-dark cells disappeared entirely from the underlying glial monolayer. In contrast, glia appeared to be unaffected by the addition of the ionophore. Treatment of cultures with A23187 caused a dosedependent decrease in total neuronal survival, ranging from 85% survival at 1 PM to 28% at 3 PM (Table 1) compared to control cultures (HBS alone) which exhibited a 95% survival. Immunostaining for calretinin showed that calretinin immunopositive neurons were significantly more resistant to ionophore neurotoxicity than were neurons lacking calretinin immunoreactivity (Fig. 1). This effect was observed as an increase in the proportion of surviving neurons immunopositive for calretinin relative to control cultures (Fig. 4). When the populations of calretinin immunopositive and calretinin immunonegative cells were normalized to those in control cultures, differences were found in their survival in cultures treated with varying doses of ionophore (Fig. 3). In cultures exposed to 1 PM A23187 the survival of calretinin immunopositive neurons was unchanged while the survival of calretinin immunonegative neurons decreased slightly (Fig. 3). With increasing doses of ionophore the loss of calretinin immunonegative neurons became significant;
survival fell to 52% at 2 PM and 24% at 3 PM A23187. In contrast, there was only a slight loss of calretinin immunopositive at 2 PM (82% survival), and only at the highest dose was there a substantial loss of calretinin immunopositive neurons (44% survival). Other studies have shown that the ionophore A23187 induces divalent cation-selective fluxes across membraness3*‘* and elevates intracellular Ca* + in neurons.42,45 This calcium influx appeared to be responsible for the neurotoxicity that we observed based upon the following observations. When calcium was removed from the medium, subsequent treatment with 3 PM A23187 caused no apparent neurotoxicity, although there was a slight but insignificant increase in the proportion of calretinin immunopositive cells (Fig. 4). Additionally, the neurotoxic effect of calcium ionophore was not the result of calcium-dependent release of glutamate from nerve terminals since selective antagonists of glutamate receptors did not prevent cell death caused by A23187, nor did they influence the percentage of calretinin immunopositive neurons (Fig. 4). Excitatory amino acid toxicity Excessive stimulation of glutamate receptors is known to cause a neurodegeneration that is dependent upon calcium influx.*,‘3.14,‘7,39~52,65 We therefore investigated the effects of NMDA, kainate or glutamate treatment on the survival of calretinin immunopositive and calretinin immunonegative neurons. Following a 5 min exposure to 1 mM NMDA or 1 mM kainate, total neuronal survival fell after 3 h to 0
CR+ population
m
Total suwval
100
80
20
0
Control
A23187
A23167 Ca2+ free
n
A23167 + APV/CNQX
Fig. 4. Calcium-dependence of A23187 neurotoxicity. Ordinate represents either the proportion of total neurons surviving treatment, or the proportion of neurons (not normalized) out of the total population of surviving neurons that stained positively for calretinin. A23187 was applied at 3 p M in the presence or absence of added Ca* + , or in the presence of APV (50 p M) plus CNQX (50 p M) to block glutamate neurotransmission. Data are from a single batch of cultures; n = 4 for each condition. CR+, calretinin immunopositive.
312
W.
Control
NMDA
Kainate
LUKAS
and K. A.
Glutamate
Fig. 5. ERect of excitatory amino acids on calretinin immunopositive and calretinin immunonegative neuronal survival (normalized as described for Fig. 3) 3 h after a 5 min exposure to the agonist. NMDA, kainate and glutamate were each added at I mM final concentration. Values represent mean k S.E.M. (*P < 0.05; ***P < 0.001). Data are pooled from eight experiments; n = 20 (NMDA). I8 (kainate), 26 (glutamate) and 25 (control).
69% and 63%, respectively (Table I). At the same concentration, glutamate lowered the total survival to 35%. For all three agonists, the calretinin immunopositive population of neurons was preferentially spared (Fig. 5). The most significant effects were observed for NMDA and kainate which caused a 32% and 40% loss, respectively, of calretinin immunonegative neurons and only a 2% and 10% loss of the calretinin immunopositive population (Fig. 5). Glutamate treatment (1 mM) resulted in substantial losses in both neuronal populations, although again the calretinin immunopositive population was significantly spared: 66% of calretinin immunonegative neurons died while only 47% of calretinin immunopositive neurons degenerated.
DISCUSSION
We have shown in cultures of cerebral cortex that neurons containing the calcium-binding protein calretinin are relatively resistant to cell death induced by various excitotoxins. Two independent methods were employed to induce neuronal cell death: application of the glutamate receptor agonists NMDA, kainate and glutamate, as well as the calcium ionophore A23187. As expected, we observed a range of cell loss depending upon the type of treatment and dose; in all instances where cell death occurred there was significantly higher survival within the population of neurons containing calretinin. This selective resistance of the calretinin immunopositive population is very similar to that previously shown for hippocampal neurons that contain the related protein calbindin.42 In that study cultured neurons immunopositive for calbindin were approximately six times more likely than immunoneg-
JONES
ative neurons to survive a 4 h treatment of A23187. As in our study, the threshold concentration of excitatory amino acids and A23187 required to cause neurodegeneration was greater in neurons possessing high levels of the calcium-binding protein. In the experiments with hippocampal neurons, A23187 at I PM induced an influx of calcium ions that elevated intracellular calcium to five- to six-fold above the resting level. In those neurons immunopositive for calbindin this elevation was transient while immunonegative neurons displayed a longer duration increase in intracellular Cal+ In a similar manner, clonal cell lines stably expressing calbindin were better able to buffer calcium entering through voltage-dependent Ca’ ’ channels or through ionophore pores than were cells not expressing the protein.75.4X While we cannot comment directly on a possible correlation between levels of intracellular free Ca” and calretinin, our results with a neurotoxicity assay support a similar role for calretinin as calbindin in enhancing calcium homeostasis and survival during periods of dramatically increased calcium influx. The chemical properties of both calretinin and calbindin are consistent with a role for buffering transiently elevated calcium.5h The two proteins have 59% amino acid homology5’,5’,7’ and data available for calbindin indicate an affinity for free Ca* ’ equal to approximately 0.5 ~LM.~ In addition, both calciumbinding proteins are expressed abundantly in certain areas of the brain (up to 1% of soluble protein).‘.5’,7” The added buffering capacity offered by the high level expression of these calcium-binding proteins is considerable and may have consequences for diverse cellular processes, Still, it remains unresolved if calcium-binding proteins simply buffer and promote rapid exchange of intracellular Cal +, or whether they have a more specialized catalytic function in related that dicellular processes. 47.74Further experiments rectly manipulate the level of expression of calciumbinding proteins in neurons will provide important details of their roles in calcium buffering and neuroprotection. Selective vulnerability is operationally defined by the type of chemical or pathological insult chosen, and the endogenous properties that enable select cells to survive the induced stress. Survival following receptordependent insults, such as the uniform application of glutamate to a cell culture, obviously reflects the diversity and distribution of receptor subtypes, as well as the biochemical heterogeneity within a mixed population of neurons. A differential density of NMDA and non-NMDA receptors, for example, is hypothesized to account for the selective vulnerability of some subpopulations of cortical neurons in culture.6’ On the other hand, toxins such as calcium ionophores induce a more consistent insult that is independent of the expression patterns of plasma membrane receptors.4” We found that cell death induced by A23187 was dependent on extracellular Ca2+, as previously observed,” and at the same time independent of the
Calretinin and calcium overload in cortical cultures release of glutamate from synaptic terminals. Since it is known that glutamate release following chronic depolarization contributes to excitotoxicity in cultures,6 it was important to show that the trigger for cell death was mediated directly by Ca2+ entry through the ionophore and not by the possible subsequent stimulation of postsynaptic glutamate receptors. The distinction between glutamate receptor-dependent and receptor-independent toxicity is significant if the experimental goal is to determine the neuroprotective potential of an intracellular entity such as a calcium-binding protein. The fact that calretinin immunopositive neurons preferen.tially survive the application of excitatory amino acids could be attributed to a low density of glutamate receptors on their surfaces, however, they also resist the neurotoxicity of A23187. These findings strongly suggest that the subpopulation of calretinin immunopositive neurons survive because they have an increased ability to control Ca2 + -induced toxicity.. While the evidence from this and other in vitro studies supports a role for calcium-binding proteins in reducing the toxic effects of elevated intracellular Ca2+, the situation in viuo is more complex. For example, some populations of neurons containing calcium-binding proteins in the hippocampus are relatively spared in animal models of neurodegeneration 334,63 but others in hippocampus and elsewhere are selectively reduced in number.15~‘5a*37a Differences in survival patterns within a biochemically characterized class of neurons may be due to diversity in their afferent inputs and the nature of the neurotoxic insult. Neurons in the dentate gyrus, for instance, are exceptionally vulnerable to ischemia, and in a recent study Freund and Mag16czky15” show that calretinin immunopositive cells are not spared this insult. In this region many neurons, including those containing calretinin, receive a rich glutamatergic innervation and appear to have an exceptionally high density of glutamate receptors, ‘6.59features which are unlikely to be preserved in dissociated cultures. These observations suggest that patterns of afferent input and receptor density are additional important factors in determining survival from neurotoxicity in viuo. Therefore, while calretinin may have an important calcium-buffering capacity during normal electrical activity, in certain pathological states this capacity may be rapidly over-loaded. The role of calciumbinding proteins with regard to excitotoxicity in oivo may be clarified by studying neuronal survival in transgenic animals lacking expression of one or more of these proteins. Expression of calbindin in the hippocampus can be upregulated following administration of corticosterone,22 nerve growth factoti’ or stimulation of the perforant path. ” It is interesting to speculate, therefore, that certain stressful stimuli will induce potentially neuroprotective reactions such as an increased production of calcium-binding proteins. These regulatory events typically occur many hours or days after possible secondary
313
the stimuli. In the present study we interpret the increase in the proportion of calretinin immunopositive neurons as a relative sparing phenomenon and not as a dramatic upregulation of calretinin in response to noxious chemical stimuli. This interpretation is based on the finding that the calculated number of calretinin immunopositive neurons either remained unchanged or decreased following treatments, and on the assumption that significantly increased production of calretinin protein would likely take more than 3 h. Emerging details of the kinetics of changing intracellular calcium concentrations in response to glutamate is illuminating mechanisms of excitotoxic cell death. Previously it has been shown that cell death is triggered primarily by an influx of extracellular Ca2+ 8~14~‘7,39 and, at least for some agonists, an additional release of Ca2+ from internal stores.‘3,‘4 More recently using long duration microfluorimetry, it has been revealed that glutamate exposure leads to a transient rise in intracellular Ca2+ that lasts 5-10 min.65 Depending upon the individual cell and the dose of agonist, this primary calcium response is followed, minutes or hours later, by a secondary sustained rise in calcium that immediately precedes or coincides with cell death.52,65Little is known about the sequence of events that occurs during the period of time between the primary and secondary increases in intracellular Ca* + . Pretreatment with gangliosides or sphingosine, inhibitors of protein kinase C, can block the development of the secondary response without effecting the primary increase induced by glutamate’0~“~68or A23 187.49The key subcellular disturbances initiated by the activation of calciumdependent enzymes have not yet been identified, but likely candidates include fragmentation of DNA, disruption of plasma membrane cytoskeleton, and inhibition of mitochondrial function.50 It will be important for future studies to determine where in the cascade of events leading to cell death calcium-binding proteins may interfere.
CONCLUSION
Our results in vitro demonstrate that neurons containing the calcium-binding protein calretinin are more resistant to cell death caused either by excitatory amino acids or by calcium ionophore. We hypothesize that this protective effect is due to the Ca2+ -buffering capacities of calretinin in a manner analogous to that suggested for calbindin. Acknowledgements-We would like to thank M. Celio for helpful discussions during the early phases of this work and for his generous gift of calretinin antibody, and J. Knowles and S. Cats&s for their support and encouragement. We also wish to thank N. Hussy, J.-C. Martinou and S. Rabacchi for their comments on the manuscript, and C. Hebert for help in preparing the photographs.
314
W. LUKAS and K. A. JONES REFERENCES
1. Arai R., Winsky L., Arai M., Jacobowitz D. M. (1991) Immunohistochemical localization of calretinin in the rat hindbrain. J. camp. Neurol. 310, 21.-44. 2. Baimbridge K. G., Celia M. R. and Rogers J. H. (1992) Calcium-binding proteins in the nervous system. Trends Neurosci. 15, 303-308. 3. Baughman R. W., Huettner J. E., Jones K. A. and Khan A. (1991) Cell culture of neocortex and basal forebrain from postnatal rats. In Culturing Neroe Cells (eds Banker G. and Goshn K.), pp. 2277250. MIT Press, Cambridge. 4. Bredderman P. J. and Wasserman R. H. (1974) Chemical composition, affinity for calcium, and some related properties of the vitamin D-dependent CaBP. Biochemisfry 13, 1687-1694. 5. Brown D. A. and Griffith W. H. (1983) Persistent slow inward calcium current in voltage-clamped hippocampal neurones of the guinea-pig. J. Physiol. 337, 303-320. 6. Carat-oh E. (1987) Intracellular calcium homeostasis. A. Rar. Biochem. 56, 395-433. 7. Celio M. R. (1990) Calbindin D-28k and parvalbumin in the rat nervous system. Neuroscience 35, 375-475. 8. Choi D. W. (1985) Glutamate neurotoxicity in cortical cell culture is calcium dependent. Neurosci. Left. 58, 293-297. 9. Choi D. W. (1988) Glutamate neurotoxicity and diseases of the nervous system. Neuron 1, 6233634. IO. Connor J. A., Wadman W. J., Hockberger P. E. and Wong R. K. (1988) Sustained dendritic gradients of Ca’+ induced by excitatory amino acids in CA1 hippocampal neurons.-Science 240, 649-653. G. A.. Manev H.. Guidotti A.. Costa E. and Brooker G. (1990) Ganaliosides normalize distorted II. de Erausauin single-cell’intracellular free Ca *+ dynamics after toxic doses of glutamate in derebdllar g&mle cells. Proc. n&n. Acad. Sci. U.S.A. 87, 8017-8021. 12. Ferrer I., Tufion T., Soriano E., de1 Rio A., Iraizos I., Fonseca M., Guionnet N. (1993) Calbindin D-28k immunoreactivity in the temporal neocortex in patients with Alzheimer’s disease. Clin. Neuropathol. 12, 53-58. 13. Frandsen A. and Schousboe A. (1992) Mobilization of dantrolene-sensitive intracellular calcium pools is involved in the cytotoxicity induced by quisqualate and N-methyl-o-aspartate but not by 2-amino-3-(3-hydroxy-5-methylisoazol-4yl)propionate and kainate in cultured cerebral cortical neurons. Proc. nafn. Acad. Sci. U.S.A. 89, 2590-2594. 14. Frandsen A. and Schousboe A. (1993) Excitatory amino acid-mediated cytotoxicity and calcium homeostasis in neurons. J. Neurochem. 60, 1202~1211. 15. Freund T. F., Buzsaki G., Leon A., Baimbridge K. G. and Somogyi P. (1990) Relationship of neuronal vulnerability and calcium binding protein immunoreactivity in ischemia. Expl Bruin Res. 83, 55-66. 15a. Freund T. F. and Magloczky 2s. (1993) Early degeneration of calretinin-containing neurons in the rat hippocampus after ischemia. NeuroscGence 56, 58 I -596. D. M. and Freund T. F. (1992) Calretinin is present in non-pyramidal cells 16. Gulyas A. I., Miettinen R., Jacobowitz of the rat hippocampus-I. A new type of neuron specifically associated with the mossy fibre system. Neuroscience 48, 1-27. receptor-induced “%Za” 17. Hartley D. M., Kurth M. C., Bjekness L., Weiss J. H. and Choi D. W. (1993) Glutamate accumulation in cortical cell culture correlates with subsequent neuronal degeneration. J. Neurosci. 13, 1993.-2000. disorders. Trend~y 18. Heizmann C. W. and Braun K. (1992) Changes in Ca 2+-binding proteins in human neurodegenerative Neurosci. 15, 259-264. of P-type calcium 19. Hillman D., Chen S., Aung T. T., Cherksey B., Sugimori M. and Llinas R. R. (1991) Localization channels in the central nervous system. Proc. nafn. Acad. Sci. U.S.A. 88, 7076G7080. J. H. (1991) Neocortical neuronal subpopulation labeled by a monoclonal antibody to 20. Hof P. R. and Morrison calbindin exhibit differential vulnerability in Alzheimer’s disease. Expl Neural. 111, 293-301. neocortical 21. Hof P. R., Nimchinsky E. A., Ceho M. R., Bouras C. and Morrison J. H. (1993) Calretinin-immunoreactive interneurons are unaffected in Alzheimer’s disease. Neurosci. Letf. 152, 145-149. 22. Iacooino A. M. and Christakos S. (1990) Corticosterone regulates calbindin-D28k mRNA and protein levels in rat hippbcampus. J. biol. Chem. 265, ltJ,177~10,180. 23. Iacopino A. M. and Christakos S. (1990) Specific reduction of calcium-binding protein (28-kilodalton calbindin-D) gene expression in aging and neurodegenerative diseases. Proc. n&n. Acad. Sci. U.S.A. 87, 4078-4082. 24. Iacopino A., Christakos S., German D., Sonsalla P. K. and Altar C. A. (1992) Calbindin-D28k-containing neurons in animal models of neurodegeneration: possible protection from excitotoxicity. Molec. Bruin Res. 13, 251-261. 25. Iacopino A. M., Christakos S., Modi P. and Altar C. A. (1992) Nerve growth factor increases calcium binding protein (CalbindinD28k) in rat olfactory bulb. Brain Res. 578, 305S310. 26. Ichimiya Y., Emson P. C., Mountjoy C. Q., Lawson D. E. M. and Heizmann C. W. (1988) Loss of calbindin-28k immunoreactive neurones from the cortex in Alzheimer-type dementia. Brain Res. 475, 156-159. 27. Inaguma Y., Shinohara H.. Inagaki T. and Kato K. (1992) Immunoreactive parvalbumin concentrations in parahippocampal gyrus decrease in patients with Alzheimer’s disease. J. neurol. Sci. 110, 57761. 28. Jacobowitz D. M. and Winsky L. (1991) Immunocytochemical localization of calretinin in the forebrain of the rat. J. cotnp. Neurol. 304, 198-218. 29. Keilhoff G. and Erdii S. L. (1991) Parallel development of excitotoxic vulnerability to N-methyl-D-aspartate and kainate in dispersed cultures of the rat cerebral cortex. Neuroscience 43, 35-40. of non-NMDA receptors does not block rapidly triggered 30. Koh J. and Choi D. W. (1991) Selective blockade glutamate-induced neuronal death. Brain Res. 548, 318-321. of cultured cortical neurons to damage by excitotoxins: Differential 31 Koh J. and Choi D. W. (1988) Vulnerability susceptibility of neurons containing NADPH-diaphorase. J. Neurosci. 8, 2153-2163. receptor-mediated neurotoxicity in 32 Koh J. Y., Goldberg M. P., Hartley D. M. and Choi D. W. (1990) Non-NMDA cortical culture. J. Neurosci. 10, 693-705. are selectively resistant to quinohnate 33 Koh J., Peters S. and Choi D. W. (1986) Neurons containing NADPH-diaphorase toxicity. Science 234, 73376. proteins are concentrated in the CA2 field of the monkey 34 Leranth C. and Ribak C. E. (1991) Calcium-binding hippocampus: a possible key to this region’s resistance to epileptic damage. Expl Brain Res. 85, 129-136. B., Morton A. J., Emson P. C. and Mason W. T. (1992) Stable transfection of 35 Lledo P. M., Somasundaram calbindin-D28k into the GH3 cell line alters calcium currents and intracellular calcium homeostasis. Neuron 9,943-954.
Calretinin
and calcium
overload
in cortical
cultures
315
36. Llinas R. and Sugimori M. (1980) Electrophysiological properties of in vitro Purkinje cell dendrites in mammalian cerebellar slices. J. Physiol. 305, 197-213. 37. Lowenstein D. H., Miles M. F., Hatam F. and McCabe T. (1991) Up regulation of calbindin-D28k mRNA in the rat hippocampus following focal stimulation of the perforant path. Neuron 6, 627-633. 37a. Maeloczkv Zs. and Freund T. F. (1993) Selective neuronal death in the contralateral hitmocamnus _. . followine: unilateral kainite injections into the CA3 subfield. Neuroscience 56, 317-336. 38. Mattson M. P. (1992) Calcium as sculptor and destroyer of neural circuitry. Expl Gerontol. 27, 29-49. messengers in the generation and degeneration of 39. Mattson M. P., Guthrie P. B. and Kater S. B. (1988) Intracellular hippocampal neuroarchitecture. J. Neurosci. Res. 21, 447-464. 40. Mattson M. P., Guthrie P. B. and Kater S. B. (1989) A role for Na+-dependent Ca’+ extrusion in protection against neuronal excitotoxicity. Fedn Am. Sots exp. Biol. J. 3, 2519-2526. 41. Mattson M. P. and Kater S. B. (1989) Development and selective neurodegeneration in cell cultures from different hippocampal regions. Bruin Res. 110-125. 42. Mattson M. P., Rychlik B., Chu C. and Christakos S. (1991) Evidence for calcium-reducing and excitoprotective roles for the calcium-binding protein calbindinD28k in cultured hippocampal neurons. Neuron 6, 41-51. 43. Miettinen R., Gulyas A. I., Baimbridge K. G., Jacobowitz D. M. and Freund T. F. (1992) Calretinin is present in non-pyramidal cells of the rat hippocampus-II. Co-existence with other calcium binding proteins and GABA. Neuroscience 48, 29-43. 44. Miller R. J. (1991) The control of neuronal Ca2+ homeostasis. Prog. Neurobiol. 37, 255-285. 45. Mills L. R. and Kater S. B. (1990) Neuron-specific and state-specific differences in calcium homeostasis regulate the generation and degeneration of neuronal architecture. Neuron 2, 1499163. 46. Monyer H., Giffard R. G., Hartley D. M., Dugan L. L., Goldberg M. P. and Choi D. W. (1992) Oxygen or glucose deprivation-induced neuronal injury in cortical cultures is reduced by tetanus toxin. Neuron 8, 967-973. 47. Morgan D. W., Welton A. F., Heick A. E., Christakos S. (1986) Specific in vitro activation of Ca, Mg-ATPase by vitamin D-dependent rat renal calcium binding protein (calbindin-D28K). Biochem. biophys. Res. Commun. 138, 547-553. 48. Muir E. M., Lawson D. E. M., Sepulveda F. V., O’Brien J. A. and Harding M. (1993) Calbindin D28K expression in transfected mouse NIH3T3 cells. Cell Calcium 14. 349-358. 49. Nakamura K., Wu G. and Ledeen R. W. (1992) Protection of Neuro-2a cells against calcium ionophore cytotoxicity by gangliosides. J. Neurosci. Res. 31, 245-253. 50. Nicotera P., Bellomo G. and Orrenius S. (1992) Calcium-mediated mechanisms in chemically induced cell death. A. Rev. Pharmac. Toxicol. 32, 449-470. 51. Parmentier M. and Lefort A. (1991) Structure of the human brain calcium-binding protein calretinin and its expression in bacteria. Eur. J. Biochem. 196, 79-85. 52. Randall R. D. and Thayer S. A. (1992) Glutamate-induced calcium transient triggers delayed calcium overload and neurotoxicity in rat hippocampal neurons. J. Neurosci. 12, 1882-1895. 53. Reed P. W. and Lardy H. A. (1972) A23187: a divalent cation ionophore. J. biol. Chem. 247, 6970-6977. 54. Resibois A. and Rogers J. H. (1992) Calretinin in rat brain: an immunohistochemical study. Neuroscience 46, 101-134. 55. Rogers J. H. (1987) Calretinin: a gene for a novel calcium-binding protein expressed principally in neurons. J. Cell Biol. 105, 134331353. 56. Rogers J. H. (1991) Calretinin. In Novel Calcium-Binding Proteins (ed. Heizmann C. W.), pp. 251-276. Springer, Berlin. 57. Rogers J. H. (1992) Immunohistochemical markers in rat cortex: co-localization of calretinin and calbindin-D28k with neuropeptides and GABA. Bruin Res. 587, 147-157. 58. Rogers J. H. and Resibois A. (1992) Calretinin and calbindin-D28k in rat brain: patterns of partial co-localization. Neuroscience 51, 843-865. 59. Scharfman H. E. and Schwartzkroin P. A. (1989) Protection of dentate hilar cells from prolonged stimulation by intracellular calcium chelation. Science 246, 257-260. 60. Schousboe A., Frandsen A., Wahl P., Drejer J. and Honore T. (1991) Excitotoxicity in cultured neurons: Role of second messenger systems and intracellular Ca2+ homeostasis. In Excitatory Amino A&Is (eds Meldrum B. S., Moroni F., Simon R. P. and Woods J. H.), pp. 5933599. Raven Press, New York. 61. Schwaller B., Buchwald P., Bliimcke I., Hunziker W. and Celia M. R. (1994) Characterization of a polyclonal antiserum against the purified human recombinant calcium binding protein calretinin. Cell Calcium (in press). 62 Siesjo B. K. and Bengtsson F. (1989) Calcium fluxes, calcium antagonists, and calcium-related pathology in brain ischemia, hypoglycemia, and spreading depression: a unifying hypothesis. J. cerebr. Blood Flow Met. 9, 127-140. 63 Sloviter R. S. (1989) Calcium-binding protein (calbindin-D28k) and parvalbumin immunocytochemistry: localization in the rat hippocampus with specific reference to the selective vulnerability of hyppocampal neurons to seizure activity. J. coma. Neurol. 280. 183-196. M. K., Wong L., Somerville M. J., Yoong L. K. K., Bergeron C., Parrnentier M. and McLachlan D. R. 64 Sutherland (1993) Reduction of calbindin-28k mRNA levels in Alzheimer as compared to Huntington hippocampus. Molec. Brain Res. 18, 32-42. 65. Tymianski M., Charlton M. P., Carlen P. L. and Tator C. H. (1993) Source specificity of early calcium neurotoxicity in cultured embryonic spinal neurons. J. Neurosci. 13, 1085-2104. 66. Van Eldik L. J., Zendegui J. G., Marshak D. R. and Watterson D. M. (1982) Calcium-binding proteins and the molecular basis of calcium action. Inf. Rev. Cyrol. 77, I-61. 67. Vickers J. C., Huntley G. W., Edwards A. M., Moran T., Rogers S. W., Heinemann S. F. and Morrison J. H. (1993) Quantitative localization of AMPA/kainate and kainate glutamate receptor subunit immunoreactivity in neurochemitally identified subpopulations of neurons in the prefrontal cortex of the macaque monkey. J. Neurosci. 13,2982-2992. 68. Wadman W. J. and Connor J. A. (1992) Persisting modification of dentritic calcium influx by excitatory amino acid stimulation in isolated CA1 neurons. Neuroscience 48, 293-305. 69. Weiss J. H., Koh J. Y., Baimbridge K. G. and Choi D. W. (1990) Cortical neurons containing somatostatinor parvalbumin-like immunoreactivity are atypically vulnerable to excitotoxic injury in vitro. Neurol. 40, 1288-1292. 70. Winsky L. and Jacobowitz D. M. (1991) Radioimmunoassay of calretinin in the rat brain. Neurochem. Int. 19, 517-522.
W.
316
LUKAS
and K. A.
JONES
71. Winsky L., Nakata H., Martin B. M. and Jacobowitz D. M. (1989) Isolation, partial amino acid sequence, and immunohistochemical localization of a brain-specific calcium-binding protein. Proc. natn. Acad. Sci. U.S.A. 86, 10,139s10,143. 72. Wulf J. and Pohl W. G. (1977) Calcium ion-flux across phosphatidylcholine membranes mediated by ionophore A23187. Biochim. biophys. Acta 465, 471-485. 73. Yamada T., McGeer P. L., Baimbridge K. G. and McGeer E. G. (1990) Relative sparing in Parkinson’s disease of substantia nigra dopamine neurons containing calbindin-D28k. Brain Rex 526, 303-307. 74. Yamaguchi T., Winsky L. and Jacobowitz D. M. (1991) Calretinin, a neuronal calcium binding protein, inhibits
phosphorylation
of a 39 kDa synaptic membrane protein from rat brain cerebral cortex. Neurosci. Letf. 131, 79-82. (Accepted 19 January 1994)