Colchicine induces apoptosis in organotypic hippocampal slice cultures

Abstract

The microtubule-disrupting agent colchicine is known to be particular toxic for certain types of neurons, including the granule cells of the dentate gyrus. In this study we investigated whether colchicine could induce such neuron-specific degeneration in developing (1 week in vitro) and mature (3 weeks in vitro) organotypic hippocampal slice cultures and whether the induced cell death was apoptotic and/or necrotic. When applied to 1-week-old cultures for 48 h, colchicine induced primarily apoptotic, but also a minor degree of necrotic cell death in the dentate granule cells, as investigated by cellular uptake of the fluorescent dye propidium iodide (PI), immunostaining for active caspase 3 and c-Jun/AP-1 (N) and fragmentation of nuclei as seen in Hoechst 33342 staining. All four markers appeared after 12 h of colchicine exposure. Two of them, active caspase 3 and c-Jun/AP-1 (N) displayed a similar time course and reached a maximum after 24 h of exposure, 24 h ahead of both PI uptake and Hoechst 33342 staining, which together displayed similar time profiles and a close correlation. In 3-week-old cultures, colchicine did not induce apoptotic or necrotic cell death. Attempts to interfere with the colchicine-induced apoptosis in 1-week-old cultures showed that colchicine-induced PI uptake and formation of apoptotic nuclei were temporarily prevented by coapplication of the protein synthesis inhibitor cycloheximide. Application of the pancaspase inhibitor z-VAD-fmk almost completely abolished the formation of active caspase 3 protein and apoptotic nuclei induced by colchicine, but the formation of necrotic nuclei increased correspondingly and the PI uptake was unaffected. We conclude that colchicine induces caspase 3-dependent apoptotic cell death of dentate granule cells in hippocampal brain slice cultures, but the apoptotic cell death is highly dependent on the developmental stage of the cultures.

Introduction

Colchicine is a well known neurotoxin, which by binding to tubulin inhibits microtubule assembly, and thereby blocks mitosis and disrupts axoplasmic transport [51]. Neuronal function is dependent on an intact cytoskeleton and cytoskeletal alterations are thought to be associated with neurological disorders, like for example Alzheimer’s disease [67]. Colchicine administered directly into the hippocampus of rats results in preferential destruction of dentate gyrus granule cells [23]. This lesion produces impairment in spatial learning [47], epileptiform activity [62] and wet dog shakes [51].

Colchicine has previously been shown to be neurotoxic to dentate granule cells in organotypic rat hippocampal slice cultures grown for 10 days before colchicine exposure [52]. The cell death induced by colchicine in dentate granule cells has been suggested to be apoptotic in vivo [14], [58] and in vitro [34]. In most in vitro studies on apoptosis, including studies using colchicine in hippocampal slice cultures [34], cultures from immature tissues have been used [21], [22], [54], [55]. This does, however, complicate the extrapolation of results to the in vivo situation in adult rats (or humans). The amount of apoptosis related enzymes procaspase 3 and active caspase 3 in the rat brain are for example high during the first 2 postnatal weeks followed by a decrease in several brain areas [50], and this might in fact explain why older dispersed cortical cultures displayed a lower susceptibility to staurosporine- or cyclosporine-induced apoptosis than younger cultures [46]. In this study we therefore compared the neurotoxic effect of colchicine on both developing (1 week in vitro) and matured (3 weeks in vitro) hippocampal slice cultures, derived from 7-day-old rats.

For detection of cell death in general, we used cellular uptake of propidium iodide (PI), a DNA-binding fluorescent dye, which only enters dead or dying cells with a damaged or leaky cell membrane. Previously, PI has been used as a marker for cell death including both apoptotic and necrotic cells [18], [74]. Propidium iodide has been widely used in hippocampal slice cultures exposed to oxygen-glucose deprivation [1], [5], [9], [42], ethanol [59], β-amyloid toxicity in relation to Alzheimer’s disease [13] and in cultures used for induction of epileptic seizures [57]. For visualization of the characteristic apoptotic nuclear fragmentation [10], [14], we stained histological sections with the DNA-binding dye Hoechst 33342 [70], [77].

A well-known biochemical feature of apoptosis is the activation of a family of cysteine proteases called caspases [8], [16], [45]. Caspases are normally present in the cell in an inactive pro-form and during apoptosis they are proteolytically processed to the active form, usually by other caspases. In the present study dependence on activation of caspase 3 was investigated by active caspase 3 immunostaining and treatment with the pancaspase inhibitor z-VAD-fmk. The dependence of apoptosis on de novo protein synthesis in general was tested by treatment with the protein synthesis inhibitor cycloheximide [2].

De novo protein synthesis, which is required for apoptosis to occur [8], [19], is partly induced by c-Jun, which belongs to a family of transcriptional factors called AP-1 (activator protein 1) [43]. The association between c-Jun and cell demise is seen in such different disorders as ischemia, Alzheimer’s disease, amyotrophic lateral sclerosis (ALS) and multiple sclerosis, and it has even been possible to block cell death by anti-c-Jun antibodies [8], [27], [58]. In rat brains after hypoxia–ischemia, c-Jun/AP-1 (N) immunoreactivity is found in apoptotic cells but not in necrotic areas [20]. Based on this knowledge we also considered c-Jun/AP-1 (N) an important marker to include in this study.

Results from part of this study have appeared in abstract form [40], [80].