• apoptosis;
  • Bax;
  • Bim;
  • Noxa;
  • cytosine arabinoside;
  • p53


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgements
  8. References

DNA damage activates apoptosis in several neuronal populations and is an important component of neuropathological conditions. While it is well established that neuronal apoptosis, induced by DNA damage, is dependent on the key cell death regulators p53 and Bax, it is unknown which proteins link the p53 signal to Bax. Using rat sympathetic neurons as an in vitro model of neuronal apoptosis, we show that cytosine arabinoside is a DNA damaging drug that induces the expression of the BH3-only pro-apoptotic genes Noxa, Puma and Bim. Increased expression occurred after p53 activation, measured by its phosphorylation at serine 15, but prior to the conformational change of Bax at the mitochondria, cytochrome c (cyt c) release and apoptosis. Hence Noxa, Puma and Bim could potentially link p53 to Bax. We directly tested this hypothesis by the use of nullizygous mice. We show that Puma, but not Bim or Noxa, is a crucial mediator of DNA damage-induced neuronal apoptosis. Despite the powerful pro-apoptotic effects of overexpressed Puma in Bax-expressing neurons, Bax nullizygous neurons were resistant to Puma-induced death. Therefore, Puma provides the critical link between p53 and Bax, and is both necessary and sufficient to mediate DNA damage-induced apoptosis of sympathetic neurons.

Abbreviations used

actinomycin D


cytosine arabinoside


Ataxia Telangictasia Mutated


ATM and Rad3-related


Bol-Asp (omethyl) fluoromethylketone


cerebellar granule neurons



cytochrome c

cyt c


enhanced green fluorescent protein


foetal bovine serum


haemaglutinin tag


mouse double minute 2


multiplicity of infection


nerve growth factor


phosphate-buffered saline


reverse transcriptase


superior cervical ganglion


wild type

Apoptosis is a genetically regulated form of cell death that plays a major role during the development of the nervous system (Roth and D'Sa 2001) and may participate in neurodegenerative disease processes (Akhtar et al. 2004). Hence it is crucial to understand which signalling pathways control apoptotic processes and there has been a substantial effort to identify key genes and proteins that regulate apoptosis. Neuronal apoptosis may occur upon depletion of survival factors (e.g. neurotrophins) that lead to activation of death signalling pathways (reviewed by Ham et al. 2000; Morrison et al. 2002; Putcha and Johnson 2004). However, sustained survival signals can also be overpowered by a prolonged dominant death stimulus. Recently it has been hypothesised that damage to DNA is a widespread activator of neuronal apoptosis and an important component of neuropathological conditions, as many diseases in which DNA repair or DNA damage-induced signalling are deficient result in impaired neurological function and ultimately neurodegeneration (reviewed by Rolig and McKinnon 2000). Therefore, it is likely that DNA damage is a key initiator of neuronal death. However, the mechanisms that regulate DNA damage induced neuronal death are not fully understood.

One of the hallmarks of DNA damage is the activation of the tumour suppressor gene p53. Several in vitro and in vivo studies have reported on the relevance of p53-dependent apoptosis and downstream signalling pathways in DNA damage-induced neuronal apoptosis in a wide range of neurodegenerative conditions as well as during key stages of development (reviewed by Miller et al. 2000; Morrison and Kinoshita 2000; Culmsee and Mattson 2005). Most importantly, the lack of p53 abrogated neuronal death induced by DNA damaging agents in many neuronal systems (Anderson and Tolkovsky 1999; Davies and Rosenthal 1994; Wood and Youle 1995; Enokido et al. 1996a; Enokido et al. 1996b; Xiang et al. 1996; Hughes et al. 1997; Park et al. 1998; Xiang et al. 1998; Inamura et al. 2000; Morris et al. 2001). Where studied, p53-dependent death was abrogated or greatly delayed in Bax-null neurons (Besirli et al. 2003; Xiang et al. 1998). As many different types of p53-null neurons still died normally after trophic factor deprivation, and these deaths were also inhibited in Bax-null neurons, it has been suggested that there are two separate signalling pathways that regulate Bax-dependent apoptosis in neurons (Anderson and Tolkovsky 1999; Hetman et al. 1999), although there is also evidence that the two pathways may interact (Aloyz et al. 1998).

p53 promotes apoptosis by activating the expression of a number of genes including those encoding death receptors and pro-apoptotic members of Bcl-2 family (reviewed by Sax and El-Deiry 2003). Recent data from studies where postmitotic cultured neurons were either exposed to different DNA damaging drugs or where p53 protein has been elevated by exogenous expression suggest that several BH3-only proteins are highly up-regulated. BH3-only proteins are pro-apoptotic members of the Bcl-2 family that are activated by apoptotic signals upstream of Bax (Puthalakath and Strasser 2002). Cregan et al. 2004 showed that cerebellar granule neurons (CGN) treated with camptothecin (a DNA topoisomerase I inhibitor) or infected with an adenovirus expressing transcriptionally active p53 protein increased their expression of Puma and Noxa mRNA. However, changes in the expression of Noxa and Puma protein levels were not reported. Besirli et al. 2003. found that sympathetic neurons of the superior cervical ganglion (SCG) treated with cytosine arabinoside (araC) showed increased expression of the BH3-only protein Bim. Whether there is coincident expression of Noxa, Puma and Bim, and which of these play a crucial role during DNA damage induced neuronal death, is not clear. Although it has been demonstrated that apoptosis induced by p53 over-expression is delayed in Puma-null CGN (Cregan et al. 2004).

To further examine the fundamental signals induced by DNA damage, we have used newly isolated SCG neurons exposed to the nucleoside analogue and DNA damaging drug araC. AraC is used in the treatment of hematologic malignancies and one of the major side-effects of araC chemotherapy is neurotoxicity (reviewed by Verstappen et al. 2003). AraC has been shown to induce DNA damage in postmitotic cortical neurons (Geller et al. 2001) although whether a similar mechanism occurs in other neurons has not been reported. We showed previously that SCG neurons treated with araC in the presence of the survival factor nerve growth factor (NGF) elevate the expression of the p53 protein and subsequently die by a p53-dependent and JNK-independent apoptotic process (Anderson and Tolkovsky 1999). Importantly, key survival signalling pathways such as the MEK/ERK1/2 and the PI3K/Akt modules are not altered due to araC exposure indicating that araC induces a dominant death pathway that overrides the survival sustaining actions of NGF. However, it is still not known how p53 expression is regulated and the identity of the p53-dependent molecules leading to apoptosis remain to be elucidated. Besirli et al. 2003 showed that death induced by araC in established cultures of SCG neurons is dependent on Bax. Curiously, although they did not observe increased p53 expression, p53-deficient SCG neurons had a decreased sensitivity to araC. Together these results suggested that araC induces a p53 dependent apoptotic pathway that converges onto Bax.

Here we have tested which BH3-only proteins are induced by araC in SCG neurons and whether any of these proteins are functionally important during apoptosis. Our results show that Puma is the sole BH3-only protein required for activating Bax-dependent apoptosis induced by DNA damage in sympathetic neurons.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgements
  8. References


SUPERSCRIPT™ II RNase H Reverse Transcriptase, RNAaseOUT and Taq polymerase were from Invitrogen (Paisley, Scotland), Hoechst 33342 and araC from Sigma (Poole, UK). Mouse 2.5S NGF was prepared in the lab. Antibodies were used from the following sources: Bad (1 : 500), Bcl-2 (1 : 1000), Bcl-xL (1 : 1000), cyt c (1 : 500), ERK1/2 clone MK12 (1 : 5000), Akt (1 : 1000) from BD Biosciences (Transduction/Pharmingen: San Diego, CA, USA); p-p53(ser15) (1 : 1000 for immunoblotting, 1 : 133 for immunostaining) from Cell Signalling Technologies (Beverly, MA, USA), HA Y-11 1 : 1000 and Mdm2 (SMP14, 1 : 1000) from Santa Cruz Biotech (Santa Cruz, CA, USA), Bax 1D1 (1 : 500) from Neomarkers (Laboratory Vision, Fremont, CA, USA); Bim AB17003 1 : 1000 from Chemicon (Temecula, CA, USA); Puma AB9643 1 : 200 from Abcam (Cambridge, UK); p-S139-H2AX 1 : 500 Upstate Biotech (Lake Placid, NY, USA), α-tubulin (1 : 5000) from Sigma (Poole, UK). All antimouse and antirabbit HRP-conjugated and cy3/Alexa488 conjugated secondary antibodies were from Jackson ImmuoResearch Laboratories (West Grove, PA, USA); Alexa 647 was from Molecular Probes (San Diego, CA, USA).

Preparation and culture of neurons.

Single-cell suspensions of rat SCG neurons were prepared from newborn (normally less that 24 h old) Wistar rat pups as described previously (Buckmaster et al. 1991; Virdee and Tolkovsky 1995). Mouse SCG neurons were similarly prepared and cultured. Use of animals followed Home Office guidelines and received approval by Cambridge University's ethical committee. Bax null founder mice were obtained from the late Dr. Stanley Korsmeyer (Dana Farber Institute, Boston, MA, USA) and used after extensive crossing into the CD1 background. p53-null founder mice were obtained from Dr. Alan Clarke (University of Cardiff, Cardiff, UK) and were also extensively crossed into the CD1 background. Puma (Villunger et al. 2003), Noxa (Villunger et al. 2003), and Bim (Bouillet et al. 1999) null founder mice were obtained from Dr. Andreas Strasser (Walter and Eliza Hall Institute, Parkland, Australia) and bred in the original C57Bl/6 background or after crossing them once into the CD1 background because of breeding problems.

The following primers were used for genotyping:

  • Bax: 
  • p53: 
  • Puma: 
  • Bim: 
  • Noxa: 

Neurons were purified by preplating for 30 min twice on collagen in L15-CO2 medium containing 5% foetal bovine serum (FBS). The non-adhering cells were collected by centrifugation and kept in L15 plating medium at 4°C until the start of experiments. Neurons were cultured on poly-L-lysine and laminin coated surfaces in growth medium (L15-CO2 containing 3% rat serum) and additives as indicated.

Cell counting and determination of apoptosis.

Neurons were either scored unfixed or fixed at the indicated times by the addition of an equal volume of 6% paraformaldehyde to the culture medium. Fixed cells were stored at 4°C before scoring for apoptosis or analysed immediately. Nuclei were visualised using Hoechst 33342 (5 µg/mL, final concentration; Sigma, Poole, UK). For fixed cultures, the medium and dye was replaced by phosphate-buffered saline (PBS) before counting. Between 150 and 300 neurons per well plated in duplicate or triplicate per experiment were scored for apoptosis using a Leica DMIL microscope (Leica, Wetzlar, Germany). Only those neurons that had clearly segmented and condensed chromatin were counted as apoptotic.

Semi-quantitative RT-PCR

RNA was extracted from SCG neurons using the RNeasy® mini kit (Qiagen, Crawley, UK) according to the instruction manual and dissolved in RNase-free water. Total RNA was quantified and 0.2–0.5 µg of RNA from each sample was used for the reverse transcription (RT) reaction. RNA was mixed with 5 µm random hexamers and the mix was heated to 70°C for 10 min after which 1 × first strand buffer, 10 mm dithiothreitol (DTT), 0.5 mm dNTPs and 0.75 U RNAaseOUT were added. After incubation at 20–22°C for 10 min and at 42°C for 2 min, 20 units of SUPERSCRIPT™ II were added and the reaction was further incubated at 42°C for 50 min and 70°C for 15 min. The PCR reaction mixture contained 0.4 µm forward and reverse target gene primers, 100 nm dNTPs, 1.5 mm MgCl2, 1X PCR buffer and 1.25 units of Taq polymerase. The reaction cycle was started by denaturation at 95°C for 10 min followed by amplification cycles of 1 min each at 95°C, 58°C, and 72°C with final extension at 72°C for 10 min. All the primers were 20 nucleotides long, with 60% GC content, and designed to amplify 300–600 base pairs, in order to have equivalent annealing temperatures and similar amplification efficiency. The number of amplification cycles used for each primer pair was determined after prior identification of the cycle number that yields mid-log phase amplification and carefully adhered to for each PCR reaction. PCR products were resolved in a 1.5% agarose gel containing 3 µg/mL ethidium bromide by electrophoresis and band intensities were imaged and quantified with LabWorks™ Analysis Software (UVP Products, Cambridge, UK) after determining that band intensities were within the linear range of detection. The digitised images were inverted in the figures. The intensity values obtained were normalised to the values obtained for 18SRNA. Primers were designed so that they would hybridise solely within the coding region of mRNA sequences (GenBank database, NCBI), and would span at least two exons. The absence of an amplicon of the appropriate size from genomic DNA was verified for all primers. The following primers were used:

  • p53 

  • Bax 

  • Bad 

  • BimEL/L/S 

  • Puma 

  • Noxa 

  • 18srna 

Preparation of cell extracts and immunoblotting.

Neurons were collected at the indicated times, pelleted, washed with cold PBS and then ice-cold lysis buffer was added (containing 8.6% sucrose, 50 mm Tris-Cl pH 7.4, 1 mm EDTA, 0.038% EGTA, 1% Triton X-100, 1 mm Na2VO3, 10 mm NaF, and Complete™ protease inhibitor cocktail; Roche Diagnostics, Lewes, UK). Lysis was achieved on ice over a period of 20–30 min after which the samples were stored at -80°C until further analysis. On the day of immunoblotting the samples were defrosted on ice, gel loading buffer was added, and lysates were heated at 100°C for 5–10 min before loading on gels. In most experiments, parallel cultures were set up to determine the amount of apoptosis at the time of cell collection. Cell extracts were resolved on 10–12% polyacrylamide-SDS gels before electroblotting onto nitrocellulose membranes (Schleicher & Schuell, Dassel, Germany). Blots were blocked for 1–2 h at room temperature or overnight at 4°C in Tris-buffered saline-Tween 20 (TBST) 100 mm NaCl, 10 mm Tris, pH 7.5, and 0.1% (v/v Tween 20) containing 5% or 2% skimmed milk powder (when using polyclonal or monoclonal antibodies, respectively). Blots were probed either overnight at 4°C or at room temperature for 1–2 h. After incubation with the appropriate horseradish peroxidase conjugated secondary antibody (Jackson ImmunoResearch, West Grove, PA, USA), immunoreactive bands were visualised by enhanced chemiluminescence using a kit from Amersham Biosciences (Little Chalfont, UK) kit or Pierce (Rockford, IL, USA) SuperSignal Femto kit and exposed to Kodak X-OMAT film (Eastman Kodak, Rochester, NY, USA). In some cases, blots were stripped in a low stringency buffer (0.1 m glycine/CL pH 3, 1% SDS) or high stringency buffer (containing 62.5 mm Tris, 2% SDS, and 0.7% (v/v) β-mercaptoethanol, pH 6.7) for 30 min at room temperature, followed by extensive washing in TBST before re-blocking and re-probing. Intensity of bands was analysed by optical densitometry using the NIH image analysis software. To average the data from several experiments, the intensity of bands for the proteins being investigated was divided by the intensity of the bands of total ERK1/2 (or sometimes tubulin or Bcl-2 to confirm data obtained by ERK1/2). An arbitrary value of 1 was assigned to the ratio obtained from cultures treated with NGF alone.


Cells grown on coverslips were fixed with 3% paraformaldehyde in PBS at room temperature for 20 min, rinsed twice with PBS, and permeabilised in PBS containing either 1% bovine serum albumin (BSA) and 0.1% saponin (for Bax staining) or 0.3% Triton x-100 at room temperature for 20 min. Primary antibody was diluted in the same buffer and incubated with cells at 4°C overnight or at room temperature for 1 h. Cells were washed three times and incubated with cy3/Alexa488/Alexa647-conjugated secondary antibody at room temperature for 1 h, then washed three times with buffer and mounted in VECTASHIELD®. (Vector Laboratories, Burlingame, CA, USA), and analysed by fluorescence or confocal microscopy (Olympus IX70 connected to an UltraVIEW™ LCI confocal imaging system, Perkin Elmer Life Sciences, Cambridge, UK). To quantify cells expressing p-ser139-H2AX, only neurons that showed punctate staining overlying the nucleus, and whose amount was above background measured in untreated neurons, were scored as positive. The amount of nuclear p-ser15-p53 immunoreactivity was quantified from fluorescent images by measuring the intensity within a circle covering about a third of the area of the nucleus using the UltraVIEW™ LCI software (PerkinElmerLAS, Beaconsfield, UK). In the case of Bax colocalisation with mitochondria, neurons were first infected in suspension with an adenoviral vector expressing the red fluorescent protein targeted to the mitochondria, Ad-dsRedmito (Bampton et al. 2005) and plated after 2 h in the appropriate medium as detailed in the legend to figure 4.

Infection of adenoviral vectors expressing HA-Puma wildtype (wt), Puma ΔBH3, or dsRedmito

Adenoviral vectors expressing HA-Puma [wildtype (wt) or ΔBH3] and EGFP under a separate CMV promoter (Yu et al. 2001; Yu et al. 2003) were kindly provided by Dr. J. Yu and Dr. B. Vogelstein (Howard Hughes Medical Institute, Baltimore, MD, USA). Viruses were propagated in 911 or 293 cells, purified on a caesium chloride (CsCl2) gradient, desalted by chromatography on a Sepharose PD-10 column (Amersham Biosciences) and stored in 10% glycerol at -80°C. SCG neurons were plated for 4 h and then infected for 30–60 min at 37°C at the indicated multiplicity of infection (MOI) (plaque forming units per neuron) and fixed after the indicated time using 3% paraformaldehyde. Ad-dsRedmito was prepared in the lab by Dr. C. Goemans. Neurons were infected in suspension with a low titre of Ad-dsRedmito to obtain partial infection. After 1 h, neurons were plated in the media as indicated.

Statistical analysis

Data were analysed using Student's unpaired t-test for conditions obtained with and without araC. The value of n signifies the number of independent experiments (plating of neurons), the results from multiple wells (2–3) being averaged for each experiment. Means and standard deviations are shown in the figures. In some cases where the estimate of the mean varied considerably between samples due to experimental variability, the 95% confidence intervals of the fold changes are given.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgements
  8. References

AraC induces p53 phosphorylation at serine 15, which correlates with DNA damage

We previously showed that araC caused a ∼3-fold increase in p53 protein 4-hours after its addition to rat sympathetic neurons (Anderson and Tolkovsky 1999) but it remained unclear whether this increase was due to a transcriptional induction of p53 and/or stabilisation of the protein due to post-translational modification (e.g. phosphorylation). To resolve this question we first examined whether p53 mRNA was increased after araC treatment using semi-quantitative RT-PCR (Fig. 1a). We found a consistent, though small, 1.2-fold increase in p53 mRNA at 4 h (± 0.05, SD, n = 5) after araC addition which was reduced to unstimulated levels by 8 h, suggesting that a mechanism in addition to transcription contributed to p53 elevation in these neurons.


Figure 1. AraC induces p53 expression and phosphorylation at serine 15 which are under transcriptional and translational control. A. mRNA expression of p53 after araC (1 mm) treatment for 4 or 8 h compared to untreated neurons. The mean fold change of p53 transcript is shown with respective standard deviation (SD) (unpaired t-test; 4 h: p < 0.0001, n = 5; 8 h: p = 0.2, n = 6). B. Expression of p-ser15-p53 in neurons treated with araC (1 mm) or etoposide (100 µm) for 4 or 8 h. Blots were probed for p-ser15-p53 and total ERK1/2 as a loading control. Mean fold change (Fold) and (SD), 5–10 experiments (araC) and 3–4 experiments (etoposide). Unpaired t-test, araC: 4 h, p < 0.001, 8 h, p < 0.01; etoposide: 4 h, p = 0.09, 8 h, p = 0.003). C. Expression of p-ser15-p53 after araC treatment in the presence of the transcription inhibitor actinomycin D (actD, 1 µg/mL), the translational elongation inhibitor cycloheximide (CHX, 10 µg/mL) or the proteasome inhibitor lactacystin (lact, 10 µm). Mean fold changes ± SD, actD: n = 3; CHX: n = 3; lact: n = 2 (95% confidence intervals: araC, 1.9–5.7; araC and CHX, 1.1–2.9; araC and actD, 0.9–2.3). ERK1/2 or Akt used as loading controls for normalisation did not change significantly. D. Mdm2 expression after 8 h of exposure to araC. Mean fold change ± SD (n = 8), p < 0.001. No change was detected after 4 h (0.85 ± 0.2, n = 4).

Download figure to PowerPoint

Because phosphorylation of p53 at serine 15 can stabilise the protein after DNA damage (Khanna et al. 1998; Lakin et al. 1999; Nakagawa et al. 1999) and weak phosphorylation of p53 was detected in cerebellar granule neurons after UV radiation (Romero et al. 2003), we examined whether p53 is phosphorylated using an antibody raised against human p-ser15-p53. We detected an increased signal of about 3-fold 4 h after araC addition whose intensity increased to ∼5-fold after 8 h (Fig. 1b). Etoposide, which causes DNA damage by inhibiting topoisomerase II (Tomkins et al. 1994; Besirli and Johnson 2003) also induced a time dependent increase in p-ser15-p53 immunoreactivity (Fig. 1b). Interestingly, the increase in p-ser15-p53 immunoreactivity induced by etoposide was consistently higher compared to that induced by araC, although it induced a similar amount of apoptosis after 15 h (araC: 30.2 ± 6% (as in Fig. 3b below), etoposide 29.5 ± 5.8%, n = 8).


Figure 3. Bim, Noxa and Puma gene expression is increased after araC treatment before the onset of apoptosis. A. RT-PCR (as described in Methods) was performed on RNA extracted from neurons 8 h after plating in medium containing NGF without or with araC (1 mm) as described in Fig. 1.18(S) RNA was amplified as the internal control. The numbers to the right are mean ± 95% confidence intervals from 4 to 6 independent neuron preparations. B. Apoptosis was determined by scoring neurons with fragmented nuclei after staining with Hoechst 33342 (see Fig. 4 A as an example). A significant rise in apoptosis occurs between 10 and 12 h after addition of araC. Unpaired t-test, * = p < 0.05, **p < 0.001, ns = non-significant. Proteins from cultures treated as in A immunoblotted for the proteins indicated. Each column in C shows a representative blot obtained from independent experiments where p-53-Ser15 was detected prior to stripping and re-probing for the other indicated proteins. D. Fold change in expression is presented as mean ± 95% confidence interval (n = 3–6).

Download figure to PowerPoint

To test whether the increase in p-ser15-p53 requires de novo transcription and translation, we used actinomycin D (actD) and cycloheximide (CHX) at concentrations previously shown to completely ablate both processes in these neurons (Edwards and Tolkovsky 1994). Blocking transcription or translation reduced the amount of p-ser15-p53 after araC treatment by about 50%, consistent with there being a partial requirement for de novo synthesis for p-ser15-p53 accumulation (Fig. 1c). No change of p-ser15-p53 after treatment with CHX or ActD in the absence of araC was observed (Fig. 1c) suggesting that the residual amount of p-ser15-p53 was not due to a signal emitted by these compounds. Etoposide-induced p-ser15-p53 was also significantly reduced by CHX treatment (data not shown) showing that this response was not araC-specific. However, given that the inhibitors did not completely suppress the elevation of p-ser15-p53, it is apparent that pre-existing p53 can also be stabilised by phosphorylation at serine 15.

As the amount of p-ser15-p53 is increased by inhibition of proteasomal degradation (Nakagawa et al. 1999), we tested whether inhibition of the proteasome alters the amount of p-ser15-p53 in the neurons. We found that the amount of p-ser15-p53 was further increased when araC was added in the presence of the irreversible proteasome inhibitor lactacystin (10 µm) (araC: 2.9 ± 1.4; araC + lactacystin: 4.6 ± 1.9; mean+-range, n = 2) but there was no change in p-ser15-p53 when lactacystin was added in the presence of NGF alone (Fig. 1c). The changes in p53 were relatively specific as there was no significant change in the amount of total ERK1/2 (Fig. 1c) used as loading control, or in Bcl-xL or α-tubulin (Fig. 3 and data not shown). Mouse double minute 2 (Mdm2) is a p53-dependent gene that interacts with p53 (Momand et al. 1992; Barak et al. 1993; Oliner et al. 1993) and regulates its activity by numerous mechanisms (reviewed by Kohn and Pommier 2005). Amongst it activities, Mdm2 has ubiqutin protein ligase (E3) activity for p53 and so can target p53 to degradation by the proteasome, but p53 phosphorylation at serine 15 can prevent this activity (Shieh et al. 1997). Indeed, Mdm2 mRNA (data not shown) and protein expression were increased significantly by araC within 8 h (Fig. 1d), perhaps accounting for the difference in the amounts of p-ser15-p53 observed in the absence and presence of lactacystin. Thus it appears that the amount of p53 induced by DNA damaging drugs in SCG neurons is determined by de novo induction of gene expression as well as by phosphorylation and degradation via the proteasome.

AraC can affect lipid signalling pathways, reviewed in Grant 1998, and mitochondrial permeability transition (Xue et al. 2002) in addition to causing DNA damage. To confirm that DNA damage is induced by araC, we examined whether nuclei display Histone 2AX (H2AX) phosphorylation at serine 139. H2AX can be phosphorylated at ser139 by Ataxia Telangiectasia Mutated (ATM) (Burma et al. 2001), the same kinase implicated in phosphorylation of p53 at serine 15 (Khanna et al. 1998; Nakagawa et al. 1999). In NGF-maintained neurons, no p-ser139-H2AX immunoreactivity could be observed (Fig. 2a and data not shown). However, 4 h or 8 h after araC addition about 40% and 80% of neurons, respectively, showed increased nuclear immunoreactivity for p-ser139-H2AX (Fig. 2b). Preliminary experiments showed that etoposide also induced p-ser139-H2AX immunoreactivity (data not shown). Consistent with H2AX and ser-15 p53 phosphorylation being dependent on ATM, the xanthine derivative caffeine, which inhibits both ATM and ATM and Rad3-related (ATR) kinases (Sakaria et al. 1999), inhibited the increase in phosphorylation and nuclear localisation of p-ser15-p53, determined immunocytochemically in mouse SCG neurons (Fig. 2c). Together these results suggest that DNA damage in sympathetic neurons is correlated with increased expression and phosphorylation of p53 via ATM (and/or ATR). We then examined how the p53 signal might be transmitted to downstream effectors.


Figure 2. AraC treatment leads to increased p-ser139-H2AX and p-ser15-p53 immunoreactivity within nuclei of sympathetic neurons. A. Neurons were plated for 8 h in medium containing NGF without or with araC, fixed with 3% paraformaldehyde, immunostained for H2AX phosphorylation at serine 139 (primary antibody) and visualised with a cy3-conjugated secondary antibody (left panels). Right hand panels show nuclei stained with Hoechst 33342. B. Quantification of the proportion of neurons showing p-ser139-H2AX staining at 4 and 8 h after of araC addition. Mean ± SD, four independent experiments in duplicate. Unpaired t-test, **p < 0.001, ***p < 0.0001. C. Mouse SCG neurons were plated for 8 h in medium containing NGF without araC, with araC (1 mm) or with araC and 3 mm caffeine, immunostained for p-ser15-p53 and visualised with Alexa-488-conjugated secondary antibody. The numbers in each panel indicate mean ± SD of nuclear intensity measured in 3 random fields (n = 30–38 neurons) as detailed in Methods. Unpaired t-test, untreated vs. araC-treated, p < 0.001; araC-treated vs. araC- and caffeine-treated, p < 0.05. Experiment was repeated using neurons from Puma null and Puma heterzygote animals with similar results.

Download figure to PowerPoint

Increased expression of Puma, Bim and Noxa mRNA and Puma and Bim protein precedes araC-induced apoptosis

Bim, Puma and Noxa have been previously shown to be highly inducible and transcriptionally regulated in neurons in response to various insults (Putcha et al. 2001; Whitfield et al. 2001; Reimertz et al. 2003; Cregan et al. 2004), while Bad appears to be mainly regulated at the post-transcriptional level in neurons, including SCG neurons (Datta et al. 1997; Datta et al. 2000; Roberts et al. 2000). Bax has also been reported to be regulated transcriptionally in SCG neurons in a p53-dependent manner after NGF deprivation (Aloyz et al. 1998) although others found no such change (Putcha et al. 2002). As p53 is a transcription factor, we first examined which of these genes are regulated after araC treatment using semiquantitative RT-PCR. SCG neurons treated with araC for 4 h showed no significant changes in mRNA expression for Bad, Bax, Bim, Noxa, and Puma (data not shown). However, 8 h after araC addition we found a significant elevation of Bim, Noxa, and Puma mRNA while Bax and Bad mRNA levels were unchanged (Fig. 3a). These changes in mRNA preceded the onset of apoptosis (as measured by nuclear fragmentation), which significantly increased only between 10 and 15 h after onset of araC treatment (Fig. 3b).

To examine whether the levels of the respective proteins were changed after araC treatment, lysates were analysed by immunoblotting after 4 and 8 h. No changes were detected after 4 h (Fig. 3d and data not shown), but a notable increase in Puma and Bim expression was detected after 8 h (Fig. 3c & d). In contrast, there were no significant changes in the expression of Bad or Bax, nor were there any significant decreases in the expression of the antiapoptotic proteins Bcl-2 or Bcl-xL (Fig. 3C). We were unable to detect Noxa protein using many different antibodies and hence it was not possible to confirm if the increase in Noxa mRNA also occurred at the protein level.

If Puma, Bim or Noxa are important upstream regulators of Bax, their up-regulation would be predicted to occur prior to Bax activation during apoptosis. Since we did not detect any changes in Bax mRNA and protein expression upon araC treatment we speculated that pre-existing Bax within neurons was activated and translocated to mitochondria. To test this idea we made use of the conformationally sensitive monoclonal anti-Bax antibody (1D1), which detects a conformational change in the N-terminus of rat Bax after its insertion into the mitochondrial membrane (Hsu and Youle 1997). We observed that neurons treated with araC and undergoing apoptosis, detected by nuclear fragmentation, exhibited a punctate 1D1 staining pattern that colocalised with mitochondria, detected by expressing a mitochondrially targeted red fluorescent protein (dsRedmito) as detailed in the Methods section (Fig. 4a and data not shown). Concomitantly with Bax translocation (1D1 positive neurons, Fig. 4b lower panel), a proportion of araC-treated neurons also released cytochrome c (cyt-c) from the mitochondria leading to loss of cyt-c immunoreactivity (cyt-c-negative neurons, Fig. 4b upper panel). When the pan-caspase inhibitor Boc-Asp-(Omethyl)FMK (BAF) was added to prevent neuronal loss from the culture dish due to apoptosis, nuclear fragmentation was prevented but we could still observe the neurons exhibiting Bax translocation or cyt-c loss (Fig. 4b).


Figure 4. Bax translocation to the mitochondria and cyt c release occur coincidentally in overtly apoptotic cells, and after activation of p53, Puma, Noxa and Bim expression. A. Neurons were infected in suspension with the vector Ad-dsRedmito, and then plated in medium containing NGF and araC (1 mm). Cells were fixed after 15 h and stained with anti-Bax 1D1 (which recognises an N-terminal epitope that is exposed after Bax translocation to the mitochondria) followed by a secondary antibody coupled to Alexa647. Fluorescent images were captured with a CCD camera. In the merged image, Bax is represented by a green colour, dsRedmito by a red colour, and nuclei stained with Hoechst 33342 are in blue. Note Bax expression, and the coincidence of Bax and dsRedmito, only in cells with apoptotic nuclei (indicated by arrows). Arrow head indicates a live neuron expressing dsRedmito but no Bax. B. Neurons treated with araC for 15 h in the presence of the caspase inhibitor BAF were stained for cyt c or Bax 1D1. Arrows indicate loss of cyt (upper panel) or presence of 1D1 staining (lower panel). Nuclear fragmentation as measured by Hoechst staining (see A) is prevented by BAF. Confocal z series were captured and merged into a single image. C. The proportion of neurons lacking cyt c staining (cyt c negative) or gaining 1D1 staining (1D1 positive) (as in B) was scored at different times after araC addition. Results show mean ± SD of three independent experiments performed in duplicate. Unpaired t-test, * = p < 0.05, **p < 0.001.

Download figure to PowerPoint

To examine the time course of Bax translocation and cyt c release in relation to the other events described above, we quantified the proportion of neurons treated with araC that released cyt-c or those showing Bax translocation in the presence of BAF (Fig. 4c). The first significant increase of 1D1-positive or cyt-c-negative neurons was detected at 12 h after araC treatment, similar to the time when apoptosis begins to manifest as measured by nuclear fragmentation (Fig. 3b). Thus, both these events occurred after the increase in expression of Puma, Noxa and Bim mRNA and/or Puma and Bim proteins that occurred after 8 h (Fig. 3) suggesting that these genes could indeed be positive stress sensors for DNA damage in these neurons. We then tested which, if any, of these BH3-only proteins are functionally involved in mediating apoptosis by performing similar experiments with SCG neurons from Noxa, Bim and Puma nullizygous mice.

P53, Puma and Bax, but not Bim and Noxa nullizygous neurons are protected against araC induced apoptosis

Exposure of Puma, p53 and Bax nullizygous (–/–) neurons to araC did not induce significant cell death after 15 h, the extent of apoptosis in araC treated neurons being similar to that of the respective untreated neurons (Fig. 5a-c). The lack of apoptosis was not due to lack of signalling, as immunoreactivity for p-ser15-p53 increased similarly in nuclei of heterozygote and Puma -/– SCG neurons treated for 6 h with araC (data not shown). In preliminary experiments, resistance of Puma -/– and Bax-/– neurons to araC persisted up to 48 h (data not shown) showing that the protection afforded by the absence of these genes is not just due to a short delay. Apoptosis increased in heterozygous (–/+) and wildtype (+/+) neurons in a gene dosage dependent manner though statistical significance was not reached for all classes of heterozygous neurons. In contrast, a similar number of Bim-/– and Noxa-/– neurons died compared to wildtype neurons after araC treatment suggesting that neither Bim nor Noxa are relevant during apoptosis induced by DNA damage (at least in the presence of NGF) (Fig. 5d-e).


Figure 5. Puma, but not Bim or Noxa, is necessary for apoptosis induced by araC. Percent apoptosis of newly isolated araC-treated (1 mm) (black bars) mouse neurons vs. untreated neurons (white bars) derived from individual mice, littermate progeny of het × het crosses. A. p53, B. Bax, C. Puma, D. Bim, E. Noxa. Results from wildtype, heterozygous and nullizygous neurons are shown for each line. Apoptosis was scored after 15 h. Results are from three independent platings obtained from three separate crossings (mean ± SD). Note the complete inhibition of death in Puma nullizygous neurons, similar to p53 and Bax nullizygous neurons. F. Het or null neurons from Puma, Bim, or Noxa lines were plated in the absence of NGF and scored for apoptosis after 20 h. The amount of apoptosis in the het animals was normalised to 100% so as to be able to compare the efficacy of the respective gene ko in preventing apoptosis. Results are mean ± SD from three independent experiments performed in duplicate wells. Unpaired t-test comparing differences between araC-treated vs. untreated neurons, * = p < 0.05, **p < 0.001, ***p < 0.0001, ns =-non-significant. Comparing between genotypes, significant differences between araC-treated wt and ko neurons were found for p53, Bax and Puma lines and between wt and het neurons for p53 and Bax. No significant difference was found between any of the Bim and Noxa genotypes.

Download figure to PowerPoint

In parallel to these experiments, we also tested whether Bim, Noxa and/or Puma played any role during apoptosis induced by NGF deprivation by comparing the responses of heterozygous and nullizygous neurons. While apoptosis in Puma-/– or Bim-/– neurons was reduced by 50% compared to Puma-/+ or Bim-/+ neurons in the absence of NGF, we did not detect such an effect in Noxa -/– neurons (Fig. 5f). Together, our results suggest that while both Puma and Bim play a role in NGF-deprivation induced apoptosis, and confirm that Bim-/– neurons can be distinguished from Bim-/+ neurons in some apoptotic settings, as shown previously (Putcha et al. 2001; Whitfield et al. 2001), only Puma is necessary for apoptosis induced by DNA damage. Noxa appears to play no part in either paradigm. We then tested if Puma was sufficient to mediate a dominant apoptotic response in SCG neurons and whether Puma-mediated apoptosis was Bax dependent, as Bax-null neurons are resistant to araC-induced apoptosis.

Puma mediated apoptosis is dependent on the presence of Bax

When rat SCG neurons were infected with an adenoviral construct expressing human wildtype HA (haemaglutinin)-Puma α (Ad-Puma wt) concomitantly with enhanced green fluorescent protein (EGFP) expressed from a separate CMV promoter, Puma expression induced apoptosis in a dose and time dependent manner (Fig. 6a). A BH3 deletion (ΔBH3) mutant of Puma applied at similar MOI had little or no death inducing capability. After 30 h, most of the neurons (which were maintained in the presence of BAF to prevent cell loss) were EGFP-positive indicating that they had been infected by Ad-Puma wt (data not shown). These neurons were also immunopositive for the HA tag (Fig. 6b), confirming coexpression of Puma. While EGFP was distributed in the cytoplasm and nuclear compartment, HA immunoreactivity was excluded from the nucleus (Fig. 6b). HA-Puma is much smaller than EGFP so the difference is not due to size exclusion. This expression pattern might be expected since over-expressed Puma has previously been shown to localise to mitochondria (Nakano and Vousden 2001; Yu et al. 2001). Interestingly, adenovirus-mediated expression of human Noxa wt protein did not induce apoptosis (data not shown).


Figure 6. HA-Puma, but not the BH3 deletion mutant, mediates Bax-dependent neuronal apoptosis. A. Rat neurons were cultured in the presence of NGF and infected with 50 or 150 MOI of Ad HA-Puma or Ad HA-PumaΔBH3, a mutant lacking the BH3 domain. Cells were fixed after 30 h and apoptosis scored using the DNA dye Hoechst 33342. Control neurons were left uninfected (mean ± SD, three independent experiments performed in duplicate). B. Example of a neuron expressing HA-Puma. Neurons were infected with Ad HA-Puma at low titre in the presence of 100 µm BAF to preserve the cells. EGFP expressed independently can be seen to fill the neuron, while HA-Puma, detected using an anti-HA antibody followed by a secondary antibody conjugated to Cy3, is restricted to the cytoplasm. C. Neurons from Bax-/+and Bax-/– genotypes were cultured in the presence of NGF and infected with the indicated MOI of either Ad HA-Puma wt or Ad HA-PumaΔBH3. Neurons were scored for apoptosis after 20 h. Results are mean ± SD from 2 to 3 independent experiments performed in duplicate wells. For A and C: unpaired t-test, * = p < 0.05, **p < 0.001, ***p < 0.0001, ns =-non-significant.

Download figure to PowerPoint

To directly test whether Puma mediated apoptosis is Bax-dependent, we infected heterozygous and nullizygous Bax mouse neurons with Ad-Puma wt or the ΔBH3 mutant vector. Whereas Ad-Puma wt expression induced a significant amount of apoptosis in Bax heterozygous neurons, death was abrogated in the Bax-null neurons to a level similar to that of untreated neurons (Fig. 6c). Similar to rat SCG neurons, expression of Puma ΔBH3 had a reduced efficacy (Fig. 6c). Thus, although Puma containing an intact BH3 domain is highly toxic, its apoptotic effects are entirely Bax-dependent.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgements
  8. References

In this paper we have traced the mechanism of p53 regulation by araC and investigated how it is linked to activation of Bax in sympathetic neurons. We show that DNA damage measured as p-ser139-H2AX immunoreactivity is detectable in the neurons 4 h after araC exposure, a time when p53 transcription, translation and activation of kinases that induce p53 phosphorylation at ser15 (thus stabilising p-ser15-p53) appear to be activated. Based on the differences in the timing of each step, we propose that DNA damage is the first signal sensed by the cells, which results in the changes in p53 mentioned above. Following on, p53 (and/or p-ser15-p53) then activates expression of several BH3-only proteins such as Bim, Puma and Noxa. However, only Puma mediates the activation of Bax required for cyt c release and apoptosis.

That p53 is most likely induced and stabilised by phosphorylation as a result of DNA damage induced by araC is suggested by our detection of p-ser139-H2AX in araC-treated neurons. Our data are consistent with those of Geller et al. 2001 who demonstrated that araC induces single DNA strand breaks in cortical neurons using a comet assay. Moreover, similar results with regard to p53 induction were obtained with etoposide, which produces double strand breaks in DNA by stabilising a cleavable complex between topoisomerase II and DNA in the neurons (Tomkins et al. 1994). Previously, phosphorylation of p53 at serine 15 (actually serine 18, the murine equivalent of serine 15 in rat and human p53) induced by camptothecin or etoposide in cortical neurons was shown to require ATM (Keramaris et al. 2003), the same kinase that phosphorylates H2AX at ser139 (Nur-E-Kamal et al. 2003). Consistent with the notion that DNA damage, induced by araC, is also sensed by ATM, and is the underlying cause of p53 phosphorylation at serine 15, the xanthine derivative caffeine, which inhibits cellular ATM (and ATR) kinase activity at millimolar concentrations (Sakaria et al. 1999), reduced the amount of araC-induced p-ser15-p53 in murine SCG neurons. The early increase in p-ser15-p53 was of a similar magnitude to that which we reported previously for total p53 protein in SCG neurons (Anderson and Tolkovsky 1999) though the link between the amount of phosphorylation and the extent of apoptosis is not direct, as etoposide induced much higher amounts of p-ser15-p53 immunoreactivity, but the rate of apoptosis was similar.

Interestingly, our demonstration of p53 and p-ser15-p53 elevation are in contrast to the findings of (Besirli et al. 2003), who found no increase in p53 mRNA and protein expression and no p-ser15-p53 immunoreactivity after exposure of SCG neurons to araC though some p53 was detected in untreated SCG neurons. These data show that the precise mechanism of neuronal apoptosis can be very sensitive to subtle changes within the neurons. While we used newly isolated (axotomised and regenerating) neurons which had been exposed to NGF only during the course of the experiment (up to 24 h), Besirli et al. (2003) performed their studies on established cultures which had been grown in the presence of NGF for several days. Several differences between newly isolated and mature SCG neurons have already been noted (Fletcher et al. 2000; Wright et al. 2004). Hence, it is possible that newly isolated neurons are either more susceptible to p53 activation or that p53 activation after DNA damage is developmentally regulated. Romero et al. 2003 found that CGN neurons are more resistant to genotoxic stress due to maturation in vitro and hypothesised that this could be due to the concurrent up-regulation of the DNA repair machinery in these neurons. This hypothesis could also explain the increased resistance of mature (3 weeks old) SCG neurons to araC induced death compared to neurons only grown for several days in vitro (Besirli et al. 2003) though other changes, such as those due to araC metabolism into lipids and altered lipid signalling, cannot de dismissed. Together, these findings would support a scenario where mature neurons are less able to induce p53, but in order to exclude the hypothesis that axotomy sensitises neurons to p53 induction further experiments need to be performed.

Because there was more p-ser15-p53 in response to etoposide than to araC yet the neurons underwent apoptosis with similar kinetics, these data suggest that the rate limiting steps that lead to apoptosis are determined further downstream of p53. It is unlikely that changes in Bax expression account for the rate of apoptosis as we found no changes in Bax expression during the course of araC-induced apoptosis. However, in keeping with the data of Besirli et al. 2003 it is clear that araC induced apoptosis is Bax dependent as Bax-/– neurons were protected from araC-induced death. The most likely link between p53 and Bax is the BH3-only protein Puma for the following reasons: (i) its de novo expression is required for araC-induced apoptosis; (ii) its genetic ablation prevents araC-induced apoptosis; (iii) its overexpression induces a dominant apoptotic response that nevertheless depends on Bax. Although Noxa expression was also elevated by p53, it may not play a role in SCG neurons due to its propensity to interact specifically with MCl-1 (Chen et al. 2005), but not Bcl-2 or Bcl-xL, the major antiapoptotic proteins expressed in these neurons. Given that a Puma construct that is deficient in the BH3 domain was relatively non-toxic, our data are consistent with the two current models of Puma action: (i) that it activates apoptosis by neutralising the activity of antiapoptotic Bcl-2 family members that prevent Bax activation and (b) that it directly chaperones Bax into an active conformation (Letai et al. 2002; Harada et al. 2004; Chen et al. 2005).

It is interesting that Bim was also up-regulated in response to araC treatment although it seemed to have no role in mediating araC-induced death. That Bim can play a role in the death of these neurons is demonstrated by the rescue of about 50% of the Bim-/– neurons compared to Bim-/+neurons, neurons that would normally die during NGF deprivation (as shown previously, Putcha et al. 2001; Whitfield et al. 2001). As NGF retains its capacity to signal via the PI3-K and ERK pathways in the presence of araC (Anderson and Tolkovsky 1999) it may be that NGF can protect against Bim's apoptotic function, perhaps due to its being phosphorylated by ERKs (Biswas and Greene 2002), which render it ineffective in binding to antiapoptotic Bcl2-family proteins (Harada et al. 2004). It is interesting to note that the degree of protection against NGF withdrawal was similar (c. 50%) in the Puma and Bim nullizygous neurons. Assuming that these two mechanisms of death protection are additive, we would predict that the Puma/Bim double null neurons would show a complete resistance to NGF withdrawal.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgements
  8. References

We have shown that DNA damaging agents activate p53 with a subsequent induction of the BH3 only members Bim, Noxa and Puma in sympathetic neurons. However, only Puma is necessary for mediating apoptosis in SCG neurons after DNA damage. Because Puma is unable to induce apoptosis in Bax null neurons, we conclude that Puma transmits the pro-apoptotic signal elicited by p53 to Bax. As NGF is unable to combat this type of signal effectively, more focus should be put on this type of dominant signalling as a possible cause of degeneration in brain disorders characterised by apoptotic events.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgements
  8. References

We are grateful to Drs. Ewa Michalak, Andreas Strasser and Andreas Villunger for providing the Bim, Noxa, and Puma heterozygous lines of founder mice, Dr. Alan Clarke for providing the heterozygous p53 founder mice, the late Dr. Stanley Korsmeyer for providing the heterozygous Bax founder mice, and Drs. Jian Yu and Bert Vogelstein for the provision of the Puma adenoviral vectors. We thank the Biotechnology and Biological Sciences Research Council (AMT, AW) the Wellcome Trust (AMT) and Southampton University (AW) for financial support.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Conclusion
  7. Acknowledgements
  8. References
  • Akhtar R. S. , Ness J. M. and Roth K. A. ( 2004 ) Bcl-2 family regulation of neuronal development and neurodegeneration . Biochim. Biophys. Acta 1644 , 189203 .
  • Aloyz . R. S. , Bamji S. X. , Pozniak C. D. , Toma J. G. , Atwal J. , Kaplan D. R. and Miller F. D. ( 1998 ) p53 is essential for developmental neuron death as regulated by the TrkA and p75 neurotrophin receptors . J. Cell Biol. 143 , 16911703 .
  • Anderson C. N. and Tolkovsky A. M. ( 1999 ) A role for MAPK/ERK in sympathetic neuron survival: protection against a p53-dependent, JNK-independent induction of apoptosis by cytosine arabinoside . J. Neurosci. 19 , 664673 .
  • Bampton E. T. W. , Goemans C. G. , Niranjan D. , Mizushima N. and Tolkovsky A. M. ( 2005 ) The dynamics of autophagy visualised in live cells; from autophagosome formation to fusion with endo/lysosomes . Autophagy 1 , 2336 .
  • Barak Y. , Juven T. , Haffner R. and Oren M. ( 1993 ) mdm2 expression is induced by wild type p53 activity . EMBO J. 12 , 461468 .
  • Besirli C. G. , Deckwerth T. L. , Crowder R. J. , Freeman R. S. and Johnson E. M. Jr ( 2003 ) Cytosine arabinoside rapidly activates Bax-dependent apoptosis and a delayed Bax-independent death pathway in sympathetic neurons . Cell Death Differ 10 , 10451058 .
  • Besirli C. G. and Johnson E. M. Jr ( 2003 ) JNK-independent activation of c-Jun during neuronal apoptosis induced by multiple DNA-damaging agents . J. Biol. Chem. 278 , 22 35722 366 .
  • Biswas S. C. and Greene L. A. ( 2002 ) Nerve growth factor (NGF) down-regulates the Bcl-2 homology 3 (BH3) domain-only protein Bim and suppresses its proapoptotic activity by phosphorylation . J. Biol. Chem. 277 , 49 51149 516 .
  • Bouillet P. , Metcalf D. , Huang D. C. , Tarlinton D. M. , Kay T. W. , Kontgen F. , Adams J. M. and Strasser A. ( 1999 ) Proapoptotic Bcl-2 relative Bim required for certain apoptotic responses, leukocyte homeostasis, and to preclude autoimmunity . Science 286 , 17351738 .
  • Buckmaster A. , Nobes C. D. , Edwards S. N. and Tolkovsky A. M. ( 1991 ) Nerve Growth Factor is Required for Induction of c-Fos Immunoreactivity by Serum, Depolarization, Cyclic AMP or Trauma in Cultured Rat Sympathetic Neurons . Eur J. Neurosci. 3 , 698707 .
  • Burma S. , Chen B. P. , Murphy M. , Kurimasa A. and Chen D. J. ( 2001 ) ATM phosphorylates histone H2AX in response to DNA double-strand breaks . J. Biol. Chem. 276 , 42 46242 467 .
  • Chen L. , Willis S. N. , Wei A. , Smith B. J. , Fletcher J. I. , Hinds M. G. , Colman P. M. , Day C. L. , Adams J. M. and Huang D. C. ( 2005 ) Differential targeting of prosurvival Bcl-2 proteins by their BH3-only ligands allows complementary apoptotic function . Mol Cell 17 , 393403 .
  • Cregan S. P. , Arbour N. A. , MacLaurin J. G. , Callaghan S. M. , Fortin A. , Cheung E. C. C. , Guberman D. S. , Park D. S. and Slack R. S. ( 2004 ) p53 activation domain 1 is essential for PUMA upregulation and p53-mediated neuronal cell death . J. Neurosci. 24 , 10 00310 012 .
  • Culmsee C. and Mattson M. P. ( 2005 ) p53 in neuronal apoptosis . Biochem. Biophys. Res. Comm 331 , 761777 .
  • Datta S. R. , Dudek H. , Tao X. , Masters S. , Fu H. , Gotoh Y. and Greenberg M. E. ( 1997 ) Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery . Cell 91 ( 2 ), 231241 .
  • Datta S. R. , Katsov A. , Hu L. , Petros A. , Fesik S. W. , Yaffe M. B. and Greenberg M. E. ( 2000 ) 14-3-3 proteins and survival kinases cooperate to inactivate BAD by BH3 domain phosphorylation . Mol. Cell. 6 , 4151 .
  • Davies A. M. and Rosenthal A. ( 1994 ) Neurons from mouse embryos with a null mutation in the tumour suppressor gene p53 undergo normal cell death in the absence of neurotrophins . Neurosci. Lett. 182 , 112114 .
  • Edwards S. N. and Tolkovsky A. M. ( 1994 ) Characterization of apoptosis in cultured rat sympathetic neurons after nerve growth factor withdrawal . J. Cell Biol. 124 , 537546 .
  • Enokido Y. , Araki T. , Aizawa S. and Hatanaka H. ( 1996a ) p53 involves cytosine arabinoside-induced apoptosis in cultured cerebellar granule neurons . Neurosci. Lett. 203 , 14 .
  • Enokido Y. , Araki T. , Tanaka K. , Aizawa S. and Hatanaka H. ( 1996b ) Involvement of p53 in DNA strand break-induced apoptosis in postmitotic CNS neurons . Eur J. Neurosci. 8 , 18121821 .
  • Fletcher G. C. , Xue L. , Passingham S. K. and Tolkovsky A. M. ( 2000 ) Death commitment point is advanced by axotomy in sympathetic neurons . J. Cell Biol. 150 , 741754 .
  • Geller H. M. , Cheng K. Y. , Goldsmith N. K. , Romero A. A. , Zhang A. L. , Morris E. J. and Grandison L. ( 2001 ) Oxidative stress mediates neuronal DNA damage and apoptosis in response to cytosine arabinoside . J. Neurochem. 78 , 265275 .
  • Grant S. ( 1998 ) Ara-C: cellular and molecular pharmacology . Adv. Cancer Res. 72 , 197233 .
  • Ham J. , Eilers A. , Whitfield J. , Neame S. J. and Shah B. ( 2000 ) c-Jun and the transcriptional control of neuronal apoptosis . Biochem. Pharmacol. 60 , 10151021 .
  • Harada H. , Quearry B. , Ruiz-Vela A. and Korsmeyer S. J. ( 2004 ) Survival factor-induced extracellular signal-regulated kinase phosphorylates BIM, inhibiting its association with BAX and proapoptotic activity . Proc. Natl Acad. Sci. U S A 101 , 15 31315 317 . Epub 12004 October 15314.
  • Hetman M. , Kanning K. , Cavanaugh J. E. and Xia Z. ( 1999 ) Neuroprotection by brain-derived neurotrophic factor is mediated by extracellular signal-regulated kinase and phosphatidylinositol 3-kinase . J. Biol. Chem. 274 , 2256922580 .
  • Hsu Y. T. and Youle R. J. ( 1997 ) Nonionic detergents induce dimerization among members of the Bcl-2 family . J. Biol. Chem. 272 , 13 82913 834 .
  • Hughes P. E. , Alexi T. and Schreiber S. S. ( 1997 ) A role for the tumour suppressor gene p53 in regulating neuronal apoptosis . Neuroreport 8 , v–xii.
  • Inamura N. , Araki T. , Enokido Y. , Nishio C. , Aizawa S. and Hatanaka H. ( 2000 ) Role of p53 in DNA strand break-induced apoptosis in organotypic slice culture from the mouse cerebellum . J. Neurosci. Res. 60 , 450457 .
  • Keramaris E. , Hirao A. , Slack R. S. , Mak T. W. and Park D. S. ( 2003 ) Ataxia telangiectasia-mutated protein can regulate p53 and neuronal death independent of Chk2 in response to DNA damage . J. Biol. Chem. 278 , 37 78237 789 .
  • Khanna K. K. , Keating K. E. , Kozlov S. , Scott S. , Gatei M. , Hobson K. , Taya Y. , Gabrielli B. , Chan D. , Lees-Miller S. P. and Lavin M. F. ( 1998 ) ATM associates with and phosphorylates p53: mapping the region of interaction . Nat Genet 20 , 398400 .
  • Kohn K. W. and Pommier Y. ( 2005 ) Molecular interaction map of the p53 and Mdm2 logic elements, which control the Off-On switch of p53 in response to DNA damage . Biochem. Biophys. Res. Commun. 331 , 816827 .
  • Lakin N. D. , Hann B. C. and Jackson S. P. ( 1999 ) The ataxia-telangiectasia related protein ATR mediates DNA-dependent phosphorylation of p53 . Oncogene 18 , 39893995 .
  • Letai A. , Bassik M. C. , Walensky L. D. , Sorcinelli M. D. , Weiler S. and Korsmeyer S. J. ( 2002 ) Distinct BH3 domains either sensitize or activate mitochondrial apoptosis, serving as prototype cancer therapeutics . Cancer Cell 2 , 183192 .
  • Miller F. D. , Pozniak C. D. and Walsh G. S. ( 2000 ) Neuronal life and death: an essential role for the p53 family . Cell Death Differ. 7 , 880888 .
  • Momand J. , Zambetti G. P. , Olson D. C. , George D. and Levine A. J. ( 1992 ) The mdm-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation . Cell 69 , 12371245 .
  • Morris E. J. , Keramaris E. , Rideout H. J. , Slack R. S. , Dyson N. J. , Stefanis L. and Park D. S. ( 2001 ) Cyclin-dependent kinases and P53 pathways are activated independently and mediate Bax activation in neurons after DNA damage . J. Neurosci. 21 , 50175026 .
  • Morrison R. S. and Kinoshita Y. ( 2000 ) The role of p53 in neuronal cell death . Cell Death Differ. 7 , 868879 .
  • Morrison R. S. , Kinoshita Y. , Johnson M. D. , Ghatan S. , Ho J. T. and Garden G. ( 2002 ) Neuronal survival and cell death signaling pathways . Adv. Exp Med. Biol. 513 , 4186 .
  • Nakagawa K. , Taya Y. , Tamai K. and Yamaizumi M. ( 1999 ) Requirement of ATM in phosphorylation of the human p53 protein at serine 15 following DNA double-strand breaks . Mol Cell Biol. 19 , 28282834 .
  • Nakano K. and Vousden K. H. ( 2001 ) PUMA, a novel proapoptotic gene, is induced by p53 . Mol Cell 7 , 683694 .
  • Nur-E-Kamal A. , Li T. K. , Zhang A. , Qi H. , Hars E. S. and Liu L. F. ( 2003 ) Single-stranded DNA induces ataxia telangiectasia mutant (ATM) /p53-dependent DNA damage and apoptotic signals . J. Biol. Chem. 278 , 12 47512 481 .
  • Oliner J. D. , Pietenpol J. A. , Thiagalingam S. , Gyuris J. , Kinzler K. W. and Vogelstein B. ( 1993 ) Oncoprotein MDM2 conceals the activation domain of tumour suppressor p53 . Nature 362 , 857860 .
  • Park D. S. , Morris E. J. , Stefanis L. , Troy C. M. , Shelanski M. L. , Geller H. M. and Greene L. A. ( 1998 ) Multiple pathways of neuronal death induced by DNA-damaging agents, NGF deprivation, and oxidative stress . J. Neurosci. 18 , 830840 .
  • Putcha G. V. , Harris C. A. , Moulder K. L. , Easton R. M. , Thompson C. B. and Johnson E. M. Jr ( 2002 ) Intrinsic and extrinsic pathway signaling during neuronal apoptosis: lessons from the analysis of mutant mice . J. Cell Biol. 157 , 441453 .
  • Putcha G. V. and Johnson E. M. J. ( 2004 ) Men are but worms: neuronal cell death in C elegans and vertebrates . Cell Death Differ. 11 , 3848 .
  • Putcha G. V. , Moulder K. L. , Golden J. P. , Bouillet P. , Adams J. A. , Strasser A. and Johnson E. M. ( 2001 ) Induction of BIM, a proapoptotic BH3-only BCL-2 family member, is critical for neuronal apoptosis . Neuron 29 , 615628 .
  • Puthalakath H. and Strasser A. ( 2002 ) Keeping killers on a tight leash: transcriptional and post-translational control of the pro-apoptotic activity of BH3-only proteins . Cell Death Differ 9 , 505512 .
  • Reimertz C. , Kogel D. , Rami A. , Chittenden T. and Prehn J. H. ( 2003 ) Gene expression during ER stress-induced apoptosis in neurons: induction of the BH3-only protein Bbc3/PUMA and activation of the mitochondrial apoptosis pathway . J. Cell Biol. 162 , 587597 .
  • Roberts M. L. , Virdee K. , Sampson C. P. , Gordon I. and Tolkovsky A. M. ( 2000 ) The combination of bcl-2 expression and NGF-deprivation facilitates the selective destruction of BAD protein in living sympathetic neurons . Mol Cell Neurosci. 16 , 97110 .
  • Rolig R. L. and McKinnon P. J. ( 2000 ) Linking DNA damage and neurodegeneration . Trends Neurosci. 23 , 417424 .
  • Romero A. A. , Gross S. R. , Cheng K. Y. , Goldsmith N. K. and Geller H. M. ( 2003 ) An age-related increase in resistance to DNA damage-induced apoptotic cell death is associated with development of DNA repair mechanisms . J. Neurochem. 84 , 12751287 .
  • Roth K. A. and D' , SaC. ( 2001 ) Apoptosis and brain development . Ment Retard Dev Disabil Res. Rev. 7 , 261266 .
  • Sakaria J. N. , Busby E. C. , Tibbetts R. S. , Roos P. , Taya Y. , KarnitZ. L. M. and Abraham R. T. ( 1999 ) Inhibition of ATM and ATR kinase activities by radiosensitizing Agent caffeine . Cancer Res. 59 , 43754382 .
  • Sax J. K. and El-Deiry W. S. ( 2003 ) p53 downstream targets and chemosensitivity . Cell Death Differ 10 , 413417 .
  • Shieh S. Y. , Ikeda M. , Taya Y. and Prives C. ( 1997 ) DNA damage-induced phosphorylation of p53 alleviates inhibition by MDM2 . Cell 91 , 925934 .
  • Tomkins C. E. , Edwards S. N. and Tolkovsky A. M. ( 1994 ) Apoptosis is induced in post-mitotic rat sympathetic neurons by arabinosides and topoisomerase II inhibitors in the presence of NGF . J. Cell Sci. 107 , 14991507 .
  • Verstappen C. C. , Heimans J. J. , Hoekman K. and Postma T. J. ( 2003 ) Neurotoxic complications of chemotherapy in patients with cancer: clinical signs and optimal management . Drugs 63 , 15491563 .
  • Villunger A. , Michalak E. M. , Coultas L. , Mullauer F. , Bock G. , Ausserlechner M. J. , Adams J. M. and Strasser A. ( 2003 ) p53- and Drug-Induced Apoptotic Responses Mediated by BH3-Only Proteins Puma and Noxa . Science 302 , 10361038 .
  • Virdee K. and Tolkovsky A. M. ( 1995 ) Activation of p44 and p42 MAP kinases is not essential for the survival of rat sympathetic neurons . Eur J. Neurosci. 7 , 21592169 .
  • Yu , J. , Wang Z. , Kinzler K. W. , Vogelstein B. and Zhang L. ( 2003 ) PUMA mediates the apoptotic response to p53 in colorectal cancer cells . Proc. Natl Acad. Sci. U S A 100 , 19311936 .
  • Whitfield J. , Neame S. J. , Paquet L. , Bernard O. and Ham J. ( 2001 ) Dominant-negative c-Jun promotes neuronal survival by reducing BIM expression and inhibiting mitochondrial cytochrome c release . Neuron 29 , 629643 .
  • Wood K. A. and Youle R. J. ( 1995 ) The role of free radicals and p53 in neuron apoptosis in vivo . J. Neurosci. 15 , 58515857 .
  • Wright K. M. , Linhoff M. W. , Potts P. R. and Deshmukh M. ( 2004 ) Decreased apoptosome activity with neuronal differentiation sets the threshold for strict IAP regulation of apoptosis. J Cell Biol 2004 October 25; 167 (2) : 303–13. 167 , 303313 .
  • Xiang H. , Hochman D. W. , Saya H. , Fujiwara T. , Schwartzkroin P. A. and Morrison R. S. ( 1996 ) Evidence for p53-mediated modulation of neuronal viability . J. Neurosci. 16 , 67536765 .
  • Xiang H. , Kinoshita Y. , Knudson C. M. , Korsmeyer S. J. , Schwartzkroin P. A. and Morrison R. S. ( 1998 ) Bax involvement in p53-mediated neuronal cell death . J. Neurosci. 18 , 13631373 .
  • Xue L. , Borutaite V. and Tolkovsky A. M. ( 2002 ) Inhibition of mitochondrial permeability transition and release of cytochrome c by anti-apoptotic nucleoside analogues . Biochem. Pharmacol. 64 , 441449 .
  • Yu , J. , Zhang L. , Hwang P. M. , Kinzler K. W. and Vogelstein B. ( 2001 ) Puma induces the rapid apoptosis of colorectal cancer cells . Mol. Cell 7 , 673682 .