Molecular mechanisms of HIV TAT-induced neuronal toxicity : investigating the role of Polyamine oxidases as mediators of ROS production and the activation of the antioxidant cell response

Abstract

Chronic oxidative stress plays an important role in the pathogenesis of HIVassociated dementia (HAD). Like many viruses, HIV-1 initiates oxidative stress in infected cells. However, HIV-1-derived proteins can also induce oxidative stress in uninfected cell types (e.g., neurons, endothelia and astrocytes), through the release of cytotoxic factors, including nitric oxide (NO) and reactive oxygen species (ROS) (Mattson et al., 2005). These two molecules generates peroxinitrite, the main cause of a chronic oxidative/nitrosative stress often observed in the pathogenesis of many neurodegenerative diseases, including HAD (Allan Butterfield et al., 2006; Steiner et al., 2006). Moreover, HIV proteins such as gp120 and Tat are able to activate cell surface receptors such as the N-methyl-D-aspartate receptor (NMDAR) leading to excitotoxicity (Mattson et al., 2005). Literature data indicated that HIV-Tat is an important neurotoxic effector (Eliseo et al., 2007) being able to activate NMDAR and induce apoptotic neuronal death. In particular, the stimulation of the polyamine-sensitive type of NMDAR is associated with the neurotoxic effects induced by HIVTat (Li et al., 2008). It should be reminded that Tat-induced NMDAR activation stimulate NO production in astrocytes and neuronal cultures (Eliseo et al., 2007). Polyamines are known to be essential for normal cell growth and differentiation, but their oxidative products may act as negative regulators of cell growth and survival. Indeed, many evidences link the products of polyamine degradation to excitotoxicity (Wood et al., 2006) and neurodegeneration (Wood et al., 2007). We hypothesize that HIV-1-Tat may induce oxidative stress and cell death through the production of both NO and H2O2, the latter by a mechanism involving polyamine oxidation. A chronic oxidative/nitrosative stress occurs in cells when the production of oxidants exceeds the intracellular antioxidant defences. In the cell, a fundamental antioxidant defence system is represented by the induction of cytoprotective (phase-2) enzymes, such as Glutathione transferases (GST), Glutamate cysteine lygase (GCL), and Heme oxygenase-1 (HO-1). Under basal conditions, these enzymes are present as a fraction of their full capacity, but the transcription of their cognate genes can be co-ordinately upregulated by exposure to a variety of stimuli, including oxidants, through the activation of a cis-acting enhancer termed antioxidant response element (ARE). ARE-mediated gene activation is coordinated by the Nrf-2 transcription factor, which is considered a major player orchestrating the antioxidant defence response in the cell. It has been recently reported that Nrf2 can interfere with the activity of Tat in inducing HIV-1 LTR transactivation (Hong-Sheng et al., 2009). Thus, we hypothesize that Tat iii may directly affect Nrf2-dependent gene expression, likely preventing the neuronal antioxidant cell response (Jeffrey et al., 2007; de Vries et al., 2008). This project was aimed at clarifying the molecular mechanisms of HIV-Tatinduced neuronal toxicity, focusing on the role of polyamine metabolism in Tat-induced oxidative stress and on the activation of the antioxidant response induced by HIV-Tat in astroglial and neuronal cells. The experiments were performed on human astrocytoma (U373-MG), neuroblastoma (SH-SY5Y) and mouse neuroblastoma (NIE-115) cell lines. In the first step of the study we carried out experiments to evaluate the role of polyamine oxidases as mediators of oxidative stress in cells treated with HIV-Tat recombinant protein. We found that Tat was able to induce ROS production mainly in neuroblastoma rather than in glioma cells. Accordingly, Tat affected cell viability of only neuronal cells. Astrocytoma cells, indeed appeared completely resistant to the cytotoxic effect of Tat. To evaluate the involvement of polyamine metabolism in the neurotoxicity elicited by Tat we measured ROS content and cell viability in Tatstimulated cells in absence or presence of the polyamine oxidases inhibitors Chlorhexidine (CHX), a structural spermine analogue, affecting the enzymatic activity of SMO (spermine oxidase) and APAO (N1-acetyl polyamine oxidase). The results showed that CHX inhibited ROS formation and restored cell viability, suggesting that polyamine degradation products such as H2O2 may be involved in neuronal cell death induced by Tat. In addition, by RT-PCR we evaluated the gene expression of the enzymes involved in polyamine catabolism (SMO, APAO, ODC and SSAT), after treating SH-SY5Y with Tat. We found that mRNA levels of enzymes SMO, APAO and SSAT were up-regulated and that SMO activity was higher in HIV-Tat stimulated SH-SY5Y cells, respect to the untreated cells. Next, to assess whether the activation of NMDA receptor might play a role in triggering the up-regulation of polyamine metabolism we performed experiments with the NMDA receptor antagonist MK-801. We observed that MK-801 pre-treatment of Tat-stimulated cells prevented ROS generation and restored cell viability. Moreover, the NMDA-dependent ROS generation was abolished by CHX pretreatment in SH-SY5Y cells. As the neurotoxic effect of HIV-Tat is likely due to the overstimulation of NMDA receptor and concomitant nitric oxide increasing, we tested the effect of the NOS (Nitric Oxidase Sinthase) inhibitor L-NAME on the survival of SH-SY5Y cells treated with Tat, demonstrating that L-NAME pre-treatment restored the viability of Tat-treated cells. Our results strongly suggest that the origin of ROS generation could be related to iv spermine/spermidine metabolism. We provided also clearly evidence of the involvement of polyamine-derived hydrogen peroxide in Tat induced neuronal cell death since the specific inhibition of SMO/APAO completely restored cell viability. In regard to the involvement of the NMDA receptor in this pathway we demonstrated that inhibiting polyamine metabolism by specific inhibitor completely prevented NMDA-induced cell death as well as ROS production. HIV is known to progressively deplete GSH content in patients. In our cell culture models we found that Tat treatment caused a significant decrease of GSH levels only in neuroblastoma cells, evidencing an increase of oxidative stress in neurons. Thus in the second step of the project, we analyzed the activation of the antioxidant Nrf2/ARE pathway. In particular TransAM assays (DNA-binding ELISA) has been performed to evaluate Nrf2 transcription factor activation. In this respect, the treatment with Tat was able to activate Nrf2 in both cell lines. Further, we analyzed the mRNA expression of the ARE-genes coding for HO-1 and GCL after Tat treatment, by RT-PCR. The results underlined the up-regulation of HO-1 and GCL-C mRNA levels in U373-MG, indicating that the activation of the antioxidant response occurred in astrocytes, but not in neurons. Finally, coimmunoprecipitation experiments indicated a physical interaction between Nrf2 and Tat proteins occurred almost exclusively in SH-SY5Y. Our findings clearly show the inability of neuroblastoma cells to activate an efficient antioxidant response even in the presence of activated Nrf2. Moreover, our preliminary data indicate a different ability of astroglial and neuronal cell lines in releasing Tat. This could result in the accumulation of the viral protein within neuronal cells thus increasing its toxic effect, likely interfering with the cell antioxidant response machinery. On the contrary astrocytes, being able to release the viral protein into the extracellular space, can activate the antioxidant pro-survival response. In conclusion, the obtained results suggest a mechanism by which Tat could modulate the antioxidant response in neurons thus leading to cell death. However, additional studies will be required to obtain a detailed understanding of the specific proteins involved in regulation of this pathway.