Springer-Verlag 1998
Exp Brain Res (1998) 120:519±526
RESEARCH ARTICLE
Antonio Contestabile ´ Elisabetta Ciani Mauro Sparapani ´ Tiziana Guarnieri Giorgia Dell Erba ´ Fabrizio Bologna Claudio Cicognani
Activation of the ornithine decarboxylase-polyamine system and induction of c-fos and p53 expression in relation to excitotoxic neuronal apoptosis in normal and microencephalic rats Received: 2 June 1997 / Accepted: 7 January 1998
Abstract Microencephalic rats obtained by gestational treatment with the DNA alkylating agent methylazoxymethanol, show a remarkable lack of sensitivity to excitotoxic neuropathology caused by systemic injections of the convulsant neurotoxin kainic acid. Taking advantage of this, we have studied in these rats, as well as in normal rats, the relationship between the induction of cellular signals supposedly related to cell death and the neuronal apoptosis consequent to kainic acid administration. While normal rats responded to the excitatory insult with a large and relatively long lasting increase of the activity of the enzyme ornithine decarboxylase and of the concentration of putrescine in some brain regions, these alterations were much smaller in microencephalic rats. Expression of c-fos in brain regions sensitive to kainic acid was quicker but lasted a noticeably shorter time in microencephalic rats as compared to normal animals. A profusion of apoptotic neurons, labeled by an in situ technique, were observed in the olfactory cortex, amygdala and hippocampus of normal rats injected with kainic acid, in particular 48 h and 72 h after drug administration. At corresponding time intervals and with similar topographic localization, neurons expressing p53 protein were observed. By contrast, microencephalic rats displayed only in a few cases and in a small number apoptotic neurons in restricted areas of the ventral hippocampus and entorhinal cortex. Noticeably, in these cases small populations of p53-expressing neurons were also present in the same areas. The present observations clearly show that oncogenes such as c-fos and p53, as well as ornithine decarboxylase which behaves as an immediateearly gene in the brain under certain circumstances, undergo noticeably lower and/or shorter induction in microencephalic rats exposed to excitotoxic stimuli. In these rats, therefore, the cellular signalling pathways studied here and related to excitotoxic sensitivity and committ-
)
A. Contestabile ( ) ´ E. Ciani ´ M. Sparapani ´ T. Guarnieri G. Dell Erba ´ F. Bologna ´ C. Cicognani Department of Biology, University of Bologna, via Selmi 3, I-40126 Bologna, Italy e-mail:
[email protected], Fax: +39-51-251208
ment to cell death are downregulated as a probable consequence of altered brain wiring. Key words Oncogene expression ´ Polyamines ´ Neuropathology ´ Apoptosis ´ Olfactory cortex ´ Hippocampus ´ Rat
Introduction Apoptosis has been recognized as a widespread modality of neuronal death not only during normal development but also in neuropathological states (Cotman and Anderson 1995; Dragunow and Preston 1995). Owing the multiplicity of factors involved in apoptosis, it is often difficult to understand the relationship between signals expressed by the cells and their potential role in the committment to die. It is, therefore, important to have reliable experimental models in which a differential susceptibility to apoptotic neuronal death may be studied with respect to differential expression of one or more of the putative factors involved. Concerning in vivo models, in addition to transgenic or knockout rodents, animals in which the normal brain development has been altered could be potentially useful. In one such model, animals rendered microencephalic by gestational treatment with the DNA-alkylating agent methylazoxymethanol acetate (MAM; Johnston and Coyle 1982; Chen and Hillman 1986), we have recently noticed that excitotoxic neuronal death caused by systemic injection of kainic acid (Schwob et al. 1980; Ben Ari et al. 1981; Virgili et al. 1992) was essentially prevented in the most sensitive brain areas such as the hippocampus and the olfactory cortex (Virgili et al. 1996). MAM selectively kills cells with replicating DNA strands while sparing non-dividing cells (Johnston and Coyle 1982; Cattabeni and Di Luca 1997). Gestational administration at the 15th or the 16th day of gestation results in marked forebrain microencephaly of offspring, with structures such as the cortex, hippocampus and striatum undergoing 30±50% decrease in size and various degrees of alteration at the cytoarchitectonic and wiring lev-
520
els (Johnston and Coyle 1982; Chen and Hillman 1986; Virgili et al. 1989; Ramakers et al. 1993; Cattabeni and Di Luca 1997). This model has been widely characterized for structural, neurochemical and behavioural alterations and has also been tested for responses to various pharmacological manipulations (for review, see Cattabeni and Di Luca 1997). Evidence has been recently provided for a primary role of apoptosis in the excitotoxic cell death caused by kainic acid, particularly concerning delayed neural damage (Schreiber et al. 1993; Filipkowski et al. 1994; Sakhi et al. 1994; Morrison et al. 1996). In the same model of excitotoxicity, a role for inducible transcription factors, such as c-fos and p53, can be demonstrated (Dragunow and Preston 1995). The oncogene c-fos, encoding a monomer of the transcriptional regulator AP1, behaves in the brain as an immediate-early gene (Curran and Franza 1988; Morgan and Curran 1988). Its expression is, indeed, induced in neurons by a variety of physiological stimuli as well as by potentially neurodegenerative events, including excitotoxic insults (Hunt et al. 1987; Le Gal La Salle 1988; Morgan and Curran 1988; Popovici et al. 1990; Ciani et al. 1993). The p53 protein is an important negative regulator of cell proliferation, acting by inducing apoptosis in cell lines and suppressing growth of proliferating tumors (Symonds et al. 1994). Concerning neuropathology, it has recently emerged that p53 expression in neurons may be causally related to excitotoxic damage, including cell death brought about by kainic acid (Chopp et al. 1992; Sakhi et al. 1994, 1996; Morrison et al. 1996). Another protein which is potently induced by excitotoxicity is the key enzyme for polyamine synthesis, ornithine decarboxylase (ODC), which behaves as an immediate-early gene in the brain under excitotoxic challenge (Reed and de Belleroche 1990; De Vera et al. 1991; Facchinetti et al. 1992). It has been proposed that a large and relatively long-lasting increase in ODC activity, and consequently of its product putrescine, is causally related to neurodegeneration (Singh et al. 1990; De Vera et al. 1991; Ciani and Contestabile 1993). We report here the activation of the ODC-polyamine system and of c-fos and p53 expression in brain regions of normal and microencephalic rats after systemic injection of kainic acid. The altered timing and extent of the induction observed in microencephalic rats is related to the drastic decrease in the incidence of neurons dying by excitotoxic apoptosis in these rats.
Materials and methods Animals and treatment Wistar female rats were caged together with males in the evening and mating was checked by vaginal smear the following day at 8 a.m. The day of sperm-positive smear was considered as gestional day 1 (G1). In the morning of G16, dams received an i.p. injection of MAM (Aldrich Chemie), diluted in saline at a dose of 25 mg/ kg. Control dams received an equivalent injection of saline. At
birth, the pups were kept to a maximum of ten per litter. When 65±70 days old, rats were injected s.c. with kainic acid (12 mg/ kg) dissolved in phosphate buffer, pH 7.4, or with the vehicle alone. At different times after injection (from 90 min up to 96 h), rats were either killed by decapitation or deeply anaesthetized with ether and perfused through the heart with 50 ml saline followed by 300 ml 4% paraformaldehyde dissolved in 0.1 M phosphate buffer at pH 7.4. Fixed brains were removed from the skull and immersed in the same fixative overnight, washed in the buffer and left in 15% sucrose in the same buffer until sunk to the bottom. The experiments described here were conducted in accordance with the Italian law on experimental use of laboratory animals and under the control of an University bioethical committee and a veterinary commission for the supervision of animal care and comfort. ODC and polyamine assay Freshly dissected brains were cut in transverse slices, approximately 0.5 mm thick, with a Sorvall tissue chopper in the cold room. The medial prefrontal cortex, the olfactory cortex and the hippocampus were collected by microdissection under the stereomicroscope and kept in the deepfreeze until assayed. For ODC assay, the collected samples were homogenized in ice-cold 50 mM TRIS-HCl buffer (pH 7.5) containing 0.1 mM EDTA, 5 mM dithiotreitol and 0.04 mM pyridoxal-5-phosphate. After centrifugation at 20000 g for 20 min, aliquots of the supernatant were assayed essentially according to a previously described procedure (Baudry et al. 1986; Sparapani et al. 1996). For polyamine assay, samples of the olfactory cortex and the hippocampus were sonicated in 0.1 N perchloric acid and centrifuged for 15 min at 15000 g. Aliquots of the supernatant were subjected to high-performance liquid chromatography (HPLC) separation and fluorimetric detection of polyamines (Paschen et al. 1987). ODC activity and polyamine levels were referred to the protein content of the supernatant or the pellet, respectively (Lowry et al. 1951). Statistically significant differences were evaluated through Duncan s test after analysis of variance (one-way ANOVA). Immunohistochemistry Transverse sections, 40 mm thick, were obtained on the freezing microtome in the brain area comprised between the anterior olfactory cortex and the hippocampus, collected in phosphate-buffered saline (PBS) and processed as specified below. Sections were incubated for 30 min with 0.3% H2O2 in methanol to block endogenous peroxidases, washed in PBS containing 0.1% Triton X-100 and incubated for 60 min with normal goat serum (Vector) in PBS. Incubation with primary antibody was carried out overnight in the cold room using the anti-c-fos 456 (Medac) polyclonal antibody directed against residues 151±292 of the mouse Fos protein, diluted 1:5000 in PBS. Some sections were instead incubated, in some experiments, with anti-c-fos Ab-1 (Oncogene Science), a monoclonal antibody directed against residues 128±152 of the peptide, diluted 1:100, which gave essentially comparable results. For p53 immunohistochemistry, sections were incubated overnight in a monoclonal antibody (clone PAb-122; 1 mg/ml; Boheringer). After washing, the sections were incubated with a secondary biotinylated antibody for 1 h and treated with the avidin-biotin complex (Vector) in 0.05 M TRIS buffer at pH 7.6 for 1 h. After rinsing, sections were treated for 2 min with the diaminobenzidine substrate kit for horseradish peroxidase (Vector), including intensification by the nickel ammonium sulphate-cobalt chloride procedure (Adams 1981). In situ labelling of apoptotic cells The basic procedure of the terminal transferase-mediated UTP nick end-labelled (TUNEL) technique (Gavrieli et al. 1992) was used with some modification. Selected 30-mm-thick sections of the hippo-
521
Fig. 2 Levels of putrescine in the olfactory cortex and hippocampus of control and MAM-treated rats at different times after systemic injection of kainic acid. Same groups as for Fig. 1. *P
