| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online May 20, 2004 as doi:10.1096/fj.04-1676fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Neurology;
* Department of Pathology, Division of Neuropathology; and
Department of Neurological Surgery, University of Washington, Seattle, Washington, USA
1Correspondence: Department of Neurology, Box 356465, University of Washington, Seattle, WA 98195, USA. E-mail: gagarden{at}u.washington.edu
SPECIFIC AIMS
We sought to determine whether p53 participates in the pathogenesis of HIV-associated dementia (HAD). We used p53-deficient mice to study neurotoxicity of the HIV envelope protein gp120. Immunohistochemical analysis of p53 accumulation in cortical tissue samples from HAD cases was used to determine whether HAD is associated with p53 activation in a particular subset of CNS cells.
PRINCIPAL FINDINGS
1. The neurotoxicity of gp120 requires neuronal p53
Treating mixed cerebrocortical cultures prepared from P0 mice with 200 pM gp120 led to robust p53 activation. To determine whether p53 expression is required for gp120 neurotoxicity, mixed cerebrocortical cultures were prepared from p53-deficient (p53–/–) mice. After 14 days in vitro, the portion of cells with a neuronal phenotype in p53–/– cultures was similar to that observed in cultures prepared from p53+/+ animals; a smaller portion contained GFAP immunoreactivity (Fig. 1
A). The total number of microglia present in p53–/– cultures after 14 days in vitro was dramatically increased vs. that observed in p53+/+ cultures (Fig. 1B
). Because neurotoxicity induced by gp120 requires activated microglia, increased gp120 neurotoxicity might be expected in p53–/– cultures. Despite the increased number of microglia, neurons in p53–/– cultures were protected from gp120-induced apoptosis (Fig. 1C
).
|
To determine whether neurons require p53 for gp120 neurotoxicity, we examined neuron/microglia cocultures of mixed genotype 24 h after the addition of gp120. Addition of 200 pM gp120 for 24 h to pure astrocyte, microglia (Fig. 2D
), or neuron cultures (Fig. 1D
) from either genotype did not induce apoptosis. When 
3000 p53+/+ microglia were plated atop p53+/+ pure neuronal cultures, gp120 caused a significant increase in apoptotic neurons (Fig. 1D
). In marked contrast, pure neuron cultures from p53–/– mice were resistant to toxicity induced by gp120 treatment in the presence of p53+/+ microglia (Fig. 2D
), suggesting at least some of the protection provided by p53 deficiency is intrinsic to neurons.
|
2. The microglial response to gp120 requires p53
To determine whether p53–/– microglia create a neurotoxic milieu after gp120 exposure, cocultures of p53+/+ neurons with p53–/– or p53+/+ microglia were exposed to gp120 and neuronal apoptosis was assessed 24 h later. Gp120-treated p53–/– microglia caused a statistically significant decrease in number of apoptotic neurons compared with control cultures (Fig. 2
A). This suggested that p53 affects the manner by which microglia respond to gp120.
We hypothesized that p53 may alter the surface expression of gp120 receptors on microglia. We determined that surface expression of these receptors depends on the context of attachment, so p53–/– and p53+/+ microglia were plated on top of cultured wild-type neurons for 24 h and cell surface expression of gp120 receptors was analyzed by live cell immunofluorescence microscopy. There was no significant difference in the number of p53–/– microglia with above-background expression of CXCR4 (Fig. 2B
). However, cell surface expression of CCR5 (fluorescence intensity above isotype control) was detected in a significantly larger percentage of p53–/– microglia (Fig. 2B
).
We previously reported that microglia undergo caspase-3 activation but do not progress to apoptosis after gp120 treatment. To determine whether p53–/– microglia demonstrate a similar partial activation of the apoptotic cascade after gp120 treatment, pure microglia cultures were treated with gp120 and assayed for caspase-3 activation 24 h later. Despite the presence of adequate gp120 receptors, p53–/– microglia did not undergo increased caspase-3 activation (Fig. 2C
). p53+/+ and p53–/– microglia did not undergo apoptosis 24 or 48 h after treatment (Fig. 2D
). These findings demonstrate that p53–/– microglia do not replicate the response to gp120 observed in wild-type microglia and suggest that caspase-3 enzymatic activity may participate in the process that prompts microglia to release neurotoxic components into the surrounding media.
3. CNS HIV infection produces increased p53 protein in multiple cell types
Cortical tissue sections were immunohistochemically labeled for p53. We hypothesized that if p53 participates in HAD, then CNS tissue from patients with HAD would contain increased nuclear p53 immunoreactivity. Tissue obtained from patients with the premorbid diagnosis of HAD contained numerous cortical neurons and non-neuronal cells with strong nuclear immunoreactivity for p53, whereas p53 immunoreactivity comparable with that observed in HAD cases was rarely present in control tissue. In sections double-labeled for p53 and GFAP or the microglial marker CD68, nuclear p53 accumulation was present in a subpopulation of astrocytes and microglia. These findings suggest that HIV infection in the CNS is associated with nuclear accumulation of p53 protein in neurons and non-neuronal cells.
The frequency of cases containing >10% p53-labeled nuclei was significantly higher in HAD cases than non-HAD controls for neurons (P<0.03) and non-neuronal cells (P<0.01). To determine whether there was evidence to suggest that p53 activation was observable in the cohort of AIDS patients that had not received a premorbid diagnosis of neurocognitive impairment, distribution of p53 scores in the three separate cohorts was evaluated by ANOVA, revealing no statistically significant increase in mean p53 score in the AIDS without HAD cohort vs. the HIV-negative control group.
CONCLUSIONS AND SIGNIFICANCE
The findings reported here demonstrate that p53 protein accumulates in the nuclei of neurons and glia in patients with HAD. Evidence for activation of p53 in a wide variety of neurodegenerative diseases as well as acute CNS injury is reported, but the role of p53 in HAD has not been extensively studied. The present study expands on a recent report demonstrating activation of p53 in a small number of HAD cases. The observation that p53 accumulates in multiple CNS cell types suggests the proinflammatory environment associated with HAD may promote the expression of genes leading to apoptosis in postmitotic cells or cell cycle arrest in glia that might otherwise continue to proliferate.
Several lines of prior research supported our hypothesis that gp120 would lead to p53-dependent neurotoxicity. 1) gp120 is shed into the extracellular environment by HIV infected cells and soluble gp120 has been shown to cause neuronal apoptosis. 2) gp120 has been observed to cause p53 activation and apoptosis in immune cells and cell lines. 3) Transgenic mice expressing and secreting gp120 in the CNS develop caspase-mediated dendrite degeneration. However, it is possible that other HIV proteins may also modulate p53-mediated events. Since neurons, astrocytes, and the majority of activated microglia in HAD patients are not actually infected by HIV, p53 activation in bystander cells is most likely secondary to the effects of factors released into the extracellular environment by HIV-infected perivascular macrophages and microglia. In conjunction with reports that tat, nef, and vpr can all modulate p53 activation in infected or bystander cell populations, the data presented here support the hypothesis that widespread p53 activation seen in the majority of HAD cases may result from the convergence of toxic effects from multiple HIV proteins on one signaling pathway in multiple CNS cell types.
Through use of an in vitro model of HIV-induced neuronal injury, we report two potentially fundamental observations regarding the role of p53 in HAD. First, p53 is required for gp120-mediated neuronal apoptosis. Perhaps this is not surprising in light of data demonstrating that gp120 toxicity is mediated in part by excitotoxins, and p53–/– neurons are resistant to several models of excitotoxicity. However, we previously reported that gp120 neurotoxicity requires activation of both a mitochondrial apoptotic pathway and a TNF
-caspase-8-dependent pathway in mixed cerebrocortical cultures. Those studies could not determine whether TNF acted directly on neurons or was involved in the gp120-mediated release of excitotoxins from microglia or astrocytes. Since caspase-8-dependent apoptosis should not require p53-mediated gene expression, the hypothesis that neuronal p53 activation is necessary for gp120-induced neurotoxicity required experimental confirmation. Results presented above definitively demonstrate that intrinsic neuronal p53 is required for gp120-mediated toxicity through the use of neuron/microglia cocultures of mixed genotype.
Second, experiments described here demonstrate that p53 is also required for gp120-exposed microglia to create a neurotoxic milieu. The finding that microglia in microglia/neuron cocultures require p53 to effectively transmit the toxicity of gp120 to neurons, in conjunction with the observation that gp120 exposure promotes caspase-3 activity in microglia, suggests a possible link between the gene expression programs leading to apoptosis and microglial activation. Others have reported that activating stimuli such as lipopolysaccharide and trauma leads to up-regulation of p53-dependent genes in microglia. Together with data presented here, these studies suggest that concomitant induction of apoptotic pathways may be required for microglial activation to ensue.
We observed that p53 deficiency was associated with increased microglia surface expression of CCR5. Previous reports that the CCR5 ligands, ß-chemokines, provide protection against HIV-induced neurotoxicity suggest that up-regulation of CCR5 surface expression in p53–/– microglia may tilt the balance away from the release of cytokines and excitotoxins induced by -chemokines. However, the small portion of p53+/+and p53–/–microglia with detectable expression of CCR5 suggests that additional phenotypic differences may also participate in generating a neurotoxic microglial response to gp120.
The data presented here support the model of HAD pathogenesis displayed in Fig. 3
. The function of p53 activation in astrocytes from HAD patients remains to be determined, but many possibilities exist, including altered chemokine receptor surface expression as seen in microglia. The altered extracellular milieu produced by activated astrocytes and microglia will induce p53 activation in neurons, leading to synaptic and dendritic degradation and eventual apoptosis.
|
We have identified p53 activation as a novel component of the molecular phenotype of HAD. The findings reported here revealed that p53-dependent signaling is required in neurons and microglia for gp120-induced neuronal apoptosis. Taken together, these results justify further study of p53-dependent signaling in HAD with the hope of identifying specific molecular targets for the prevention of neurodegeneration.
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-1676fje; doi: 10.1096/fj.04-1676fje
This article has been cited by other articles:
|
|
 |
|
  J. Li, C. A. Ghiani, J. Y. Kim, A. Liu, J. Sandoval, J. DeVellis, and P. Casaccia-Bonnefil Inhibition of p53 Transcriptional Activity: A Potential Target for Future Development of Therapeutic Strategies for Primary Demyelination J. Neurosci., June 11, 2008; 28(24): 6118 - 6127. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Alirezaei, D. D. Watry, C. F. Flynn, W. B. Kiosses, E. Masliah, B. R. G. Williams, M. Kaul, S. A. Lipton, and H. S. Fox Human Immunodeficiency Virus-1/Surface Glycoprotein 120 Induces Apoptosis through RNA-Activated Protein Kinase Signaling in Neurons J. Neurosci., October 10, 2007; 27(41): 11047 - 11055. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Amini, G. Mameli, L. Del Valle, A. Skowronska, K. Reiss, B. B. Gelman, M. K. White, K. Khalili, and B. E. Sawaya p73 Interacts with Human Immunodeficiency Virus Type 1 Tat in Astrocytic Cells and Prevents Its Acetylation on Lysine 28 Mol. Cell. Biol., September 15, 2005; 25(18): 8126 - 8138. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Nardacci, A. Antinori, L. M. Larocca, V. Arena, A. Amendola, J.-L. Perfettini, G. Kroemer, and M. Piacentini Characterization of Cell Death Pathways in Human Immunodeficiency Virus-Associated Encephalitis Am. J. Pathol., September 1, 2005; 167(3): 695 - 704. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Noorbakhsh, N. Vergnolle, J. C. McArthur, C. Silva, M. Vodjgani, P. Andrade-Gordon, M. D. Hollenberg, and C. Power Proteinase-Activated Receptor-2 Induction by Neuroinflammation Prevents Neuronal Death during HIV Infection J. Immunol., June 1, 2005; 174(11): 7320 - 7329. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-L. Perfettini, M. Castedo, R. Nardacci, F. Ciccosanti, P. Boya, T. Roumier, N. Larochette, M. Piacentini, and G. Kroemer Essential role of p53 phosphorylation by p38 MAPK in apoptosis induction by the HIV-1 envelope J. Exp. Med., January 18, 2005; 201(2): 279 - 289. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
