Critical function of endogenous XIAP in regulating caspase activation during sympathetic neuronal apoptosis

Abstract

In sympathetic neurons, unlike most nonneuronal cells, growth factor withdrawal-induced apoptosis requires the development of competence in addition to cytochrome c release to activate caspases. Thus, although most nonneuronal cells die rapidly with cytosolic cytochrome c alone, sympathetic neurons are remarkably resistant unless they develop competence. We have identified endogenous X-linked inhibitor of apoptosis protein (XIAP) as the essential postcytochrome c regulator of caspase activation in these neurons. In contrast to wild-type neurons that are resistant to injection of cytochrome c, XIAP-deficient neurons died rapidly with cytosolic cytochrome c alone. Surprisingly, the release of endogenous Smac was not sufficient to overcome the XIAP resistance in sympathetic neurons. In contrast, the neuronal competence pathway permitted cytochrome c to activate caspases by inducing a marked reduction in XIAP levels in these neurons. Thus, the removal of XIAP inhibition appears both necessary and sufficient for cytochrome c to activate caspases in sympathetic neurons. These data identify a critical function of endogenous XIAP in regulating apoptosis in mammalian cells.

Figure 1.

The ability of exogenously microinjected Smac to inhibit IAPs is necessary for permitting cytochrome c to induce death in sympathetic neurons. NGF-maintained mouse sympathetic neurons were microinjected with either cytochrome c or wild-type AVPI-Smac alone, or cytochrome c along with wild-type AVPI or mutant MVPI-Smac. Parallel cultures of sympathetic neurons that were deprived of NGF in the presence of cycloheximide (−NGF+CHX) for 36 h were injected with cytochrome c as a positive control for competence. Viability of microinjected cells 3, 6, and 20 h after these injections is shown. Data are mean ± SEM for three experiments with ∼100 cells counted for each time point per experiment.

Figure 2.

XIAP protein levels are reduced when sympathetic neurons develop competence. (a) Protein levels of XIAP, cIAP-1, and cIAP-2 were examined in NGF-maintained neurons (+NGF) and in competent neurons that were deprived of NGF in the presence of cycloheximide for 24 h (−NGF+CHX). Levels of α-tubulin were also examined as a loading control. Quantitation of this data (±SEM) from two representative experiments is shown in b. (c) Protein levels of XIAP and Apaf-1 were examined in NGF-maintained (+NGF) and NGF- deprived (24 h; −NGF) Bax-deficient neurons. (d) Proteins levels of XIAP and Apaf-1 were examined in NGF-maintained (+NGF) and NGF-deprived, zVAD-FMK (50 uM)-treated (−NGF+zVAD) sympathetic neurons. (e) Protein levels of XIAP, cIAP-1, c-IAP-2, and Apaf-1 were examined in NGF-maintained neurons (+NGF), in NGF-maintained neurons treated with cycloheximide for 24 h (+NGF+CHX), and in NGF-deprived neurons treated with cycloheximide for 24 h (−NGF+CHX). Quantitation of this data (±SEM) from two representative experiments is shown in f.

Figure 3.

Reduction in XIAP protein levels correlates with neuronal development of competence in multiple conditions. (a) Examination of the time course of reduction in XIAP levels during development of competence. Levels of XIAP or Apaf-1 (control) protein were examined in NGF-maintained neurons (+NGF) and in parallel cultures of neurons that were deprived of NGF in the presence of cycloheximide for the indicated times after NGF deprivation (−NGF+CHX). Arrow points to the band corresponding to XIAP. (b) Levels of XIAP or tubulin (control) protein were examined in NGF-maintained neurons (+NGF) and in NGF-deprived neurons (24 h) in which the competence pathway was blocked with the addition of 35 mM KCl (−NGF+KCl) or 400 μM CPTcAMP (−NGF+cAMP). For comparison, XIAP levels were also examined in the competent, NGF-deprived, cycloheximide-treated (−NGF+CHX; 24 h) neurons. These data are representative of multiple experiments.

Figure 4.

XIAP mRNA levels are significantly reduced in neurons that develop competence. (a) Levels of XIAP and GAPDH (control) mRNAs were examined with quantitative RT-PCR analysis in sympathetic neurons that were either maintained in NGF (+NGF) or deprived of NGF in the presence of cycloheximide for 36 h to develop competence (−NGF+CHX). Levels of these mRNAs were also examined in competent neurons that were treated with NGF readdition for 24 h to reverse the competence state (−NGF+CHX → +NGF). (b) Levels of mRNAs of cIAP-1 and cIAP-2 along with those of XIAP and GAPDH were determined in the competent NGF-deprived, cycloheximide-treated neurons (−NGF+CHX; 36 h) and expressed as a percentage of the mRNA levels in NGF-maintained neurons. Data are mean ± SEM of two to three experiments.

Figure 5.

XIAP is necessary for the postcytochrome c regulation of caspase activation in sympathetic neurons. (a) NGF-maintained sympathetic neurons from XIAP-deficient (−/−) or wild-type (+/+) littermate mice were microinjected with either bovine or yeast cytochrome c. Survival of the injected neurons was assessed at the indicated times. Data shown are mean ± SEM of three independent experiments. (b) Phase contrast micrographs of wild-type (XIAP+/+) and XIAP-deficient (XIAP−/−) neurons are shown. Neurons that were injected with bovine cytochrome c (and rhodamine dextran) were identified by fluorescence micrographs for the same field and are marked with arrowheads. Bar, 10 μm. (c) Reintroduction of XIAP into XIAP-deficient neurons restores resistance to cytosolic microinjection of cytochrome c. NGF-maintained, XIAP-deficient sympathetic neurons were coinjected with plasmids expressing either XIAP or vector alone and EGFP. After 24 h to allow for expression, the injected neurons were reinjected with cytochrome c and survival of the cytochrome c–injected cells was determined 12 h after the injections. Data are mean ± SEM of three independent experiments.

Figure 6.

Wild-type and XIAP-deficient sympathetic neurons show no differences in the time course of death after NGF withdrawal. Equal numbers of wild-type and XIAP-deficient sympathetic neurons were either maintained in NGF or deprived of NGF for 12, 24, and 48 h. The number of neurons that became committed to die at those times was determined by replacing the media with NGF-containing media and counting the number of neurons that could be rescued after 7 d of NGF readdition. Rescued neurons show healthy, phase bright cell bodies. Data shown are mean ± SD of three experiments.

Figure 7.

Release of endogenous cytochrome c and Smac are not sufficient to overcome the XIAP-mediated postcytochrome c inhibition in NGF-maintained sympathetic neurons. (a) NGF-maintained neurons were either left untreated or treated with 3 mM H₂O₂ for 1 h. The status of cytochrome c and Smac in these neurons was examined 3 h after the H₂O₂ exposure by immunohistochemical techniques. Micrographs show that whereas untreated, NGF-maintained neurons show a punctate, mitochondrial pattern of staining for both cytochrome c (red) and Smac (green), H₂O₂ exposure induces the loss of both cytochrome c and Smac staining from the mitochondria in these neurons. Bizbenzimide staining shows the nuclei (blue). (b) Quantitation of the loss of mitochondrial cytochrome c and Smac from these neurons in response to the transient H₂O₂ treatment. Approximately 100 neurons were counted to determine the status of cytochrome c and Smac in untreated or H₂O₂-treated neurons. (c) Wild-type (XIAP+/+) or XIAP-deficient (XIAP−/−) sympathetic neurons were maintained in NGF and either left untreated or transiently exposed to H₂O₂ as described in panel a. The status of cytochrome c (red) and activated caspase-3 (green) in these neurons was examined by immunohistochemical techniques. Bizbenzimide staining shows the nuclei (blue). For untreated neurons, the staining patterns for both XIAP+/+ and XIAP−/− conditions are identical but only the image for XIAP−/− is shown. The percentage of neurons showing activated caspase-3 immunostaining after transient H₂O₂ exposure in the wild-type (XIAP+/+) or XIAP-deficient (XIAP−/−) is quantitated in panel d. (e) Sympathetic neurons were either maintained in NGF (+NGF) or made competent by depriving them of NGF in the presence of cycloheximide for 24 h (−NGF+CHX). These neurons were either left untreated or treated with H₂O₂, and the status of cytochrome c (red) and activated caspase-3 (green) in these neurons was examined. For untreated neurons, the staining patterns for both +NGF and –NGF+CHX conditions are identical but only the image for −NGF+CHX is shown. The percentage of neurons showing activated caspase-3 immunostaining after transient H₂O₂ exposure in the NGF-maintained (+NGF) or competent, (−NGF+CHX) conditions is quantitated in panel f. Data shown in b, d, and f are mean ± SD of three experiments. Bars: (a, c, and e) 10 μm.