HIV immune complexes prevent excitotoxicity by interaction with NMDA receptors

Abstract

Purpose: Human immunodeficiency virus-1 (HIV)-associated neurocognitive disorder (HAND) is a neurodegenerative disease for which there is no available neuroprotective therapy. Viral proteins, such as Tat, have been implicated as agents of neurotoxicity via multiple mechanisms, including effects by directly binding to the NMDA receptor. We evaluated the ability of the immune response against Tat to modulate neurotoxicity at glutamate receptors.

Methods: Neurotoxicity was measured in primary neuronal-glial cultures and in hippocampal slice cultures. We used immunoprecipitation experiments to demonstrate interaction between Tat, NMDA receptor, and anti-Tat antibody. Using known structures of Tat and NMDA receptors, we developed a model of their interactions.

Results: Antibodies to Tat attenuated Tat-mediated neurotoxicity. Interestingly, Tat immune complexes also blocked neurotoxicity caused by NMDA receptor agonists but not kainate/AMPA receptor agonists. Neither Tat nor antibody alone blocked the excitotoxic effect, nor did an unrelated antigen-antibody complex. The protective effect of the Tat immune complexes was also lost when Tat was modified by nitrosylation or by using a deletion mutant of Tat.

Conclusions: The ability of viral immune complexes to interact with NMDA receptors and prevent excitotoxicity represents a novel host defense mechanism. Host immune responses may influence host susceptibility to various effects of viral proteins, modulating HIV complications, such as onset of HAND. These observations provide rationale for development of vaccine therapies targeting Tat for prevention of HAND.

Keywords: Dementia; Glutamate; Neuroprotection; Neurotoxicity; Neurovirology; Neutralizing antibodies.

Figure 1. Attenuation of Tat neurotoxicity by a monoclonal antibody

Mixed rat neuronal cultures were exposed to Tat_1–72 and/or C-terminal anti-Tat monoclonal antibody as indicated. Mitochondrial membrane potential was measured 15–18 hours later. Tat alone caused toxicity versus untreated controls (p<0.001). When the antibody was incubated with Tat, it provided significant protection versus the toxicity seen with Tat alone (p<0.05). Anti-p24 provided no neuroprotection. Concentrations: Tat 200nM, antibody 5ng/μl. Data represents mean ± SEM of at least six independent experiments, analyzed by ANOVA with Neumann-Keuls post-test.

Figure 2. Representative images of propidium iodide uptake in organotypic hippocampal slice cultures

following 24 hr of exposure to: (A) control cell culture medium; (B) Tat 1–72 (200nM); and (C) Tat 1–72+anti-Tat antibody (5ng/μl); and (D) anti-Tat antibody (5 ng/μl)

Figure 3. Twenty-four hours of exposure to Tat 1–72 (200nM) produced significant injury (increased uptake of propidium iodide) in dentate gyrus granule cells, as well as, pyramidal cells of the CA3 and CA1 regions

Co-exposure of slices to Tat 1–72 with C-terminal anti-Tat monoclonal antibody significantly attenuated Tat toxicity in the pyramidal cell layer of the CA1 hippocampal region, and reduced PI uptake compared to Tat alone in all three regions. All values were converted to percent control before analysis and graphing, and the dashed line indicates control values.*p<0.05 vs. control cultures; #p<0.05 vs. 200nM Tat.

Figure 4. Modulation of NMDA excitotoxicity by a Tat immune complex

Mixed rat neuronal cultures were exposed to Tat_1–72, anti-Tat antibody, NMDA, and/or kainic acid. Mitochondrial membrane potential was measured 15–18 hours later. (A) NMDA alone caused toxicity versus untreated controls (p<0.001). Antibody alone demonstrated no protective effect against NMDA excitotoxicity, but the combination of Tat and anti-Tat antibody was neuroprotective against NMDA (p<0.01). (B) Kainic acid alone caused toxicity versus untreated controls (p<0.001), but neither the anti-Tat antibody alone nor the Tat immune complex demonstrated a protective effect against kainate-mediated excitotoxicity. Concentrations: Tat 200nM, antibody 5ng/μl, NMDA 125 μM, kainic acid 50 μM. Data represents mean ± SEM of at least four independent experiments, analyzed by ANOVA with Neumann-Keuls post-test.

Figure 5. The Tat 1–72+anti-Tat antibody complex blocks NMDA toxicity in organotypic hippocampal cultures

Cultures were pre-incubated for 30 minutes in culture media containing Tat 1–72 (200nM)+anti-Tat antibody before being transferred into media containing Tat 1–72+anti-Tat antibody+20μM NMDA. Cultures were imaged 24 hours after the initial exposure to NMDA. All values were converted to percent control before analysis and graphing, and the dashed line indicates control values. **p<0.001 vs. 20μM NMDA; *p<0.001 vs. control cultures.

Figure 6. Interaction of Tat immune complexes with the NMDA receptor, but not kainate receptors

(A) Lanes 1–6: Protein G sepharose beads were prepared with a monoclonal anti-NR1 antibody. HEK 293 cells expressing NR1A and NR2A were incubated for 1 hour in conditioned media with the indicated agent(s): untreated (lane 1), Tat protein (lane 2), nitrosylated Tat (lane 3), Tat_Δ31–61 (lane 4), Tat and anti-Tat antibody (lane 5), Tat and anti-Tat antibody and NMDA (lane 6). The cells were then harvested for immunoprecipitation over the anti-NR1 loaded protein G sepharose. Eluted proteins were separated by SDS-PAGE and immunoblotted with anti-Tat antibody. As controls, NR1 (lane 7) and Tat (lane 8) were run on the gel. (B) Top panel: Tat protein or vehicle was incubated with HEK 293 cells which had been transfected with NMDA receptor proteins, NR1A and NR2A. Immunoprecipitation was performed with anti-Tat antibody. Eluted proteins were immunoblotted with anti-NR1A antibody. Bottom panel: Tat protein or vehicle was incubated with HEK 293 cells which had been transfected with an AMPA receptor-GFP fusion protein, GluR1-GFP. Immunoprecipitation was performed with anti-Tat antibody. Eluted proteins were immunoblotted with an antibody against the GluR1-GFP. As a control, the third lane, shows cell lysate from HEK 293 cells transfected with NR1A and NR2A (top panel) and with GluR1-GFP (bottom panel). The arrows indicate the monomeric forms of NR1A (top) and GluR1 (bottom). (C) Mixed rat neuronal cultures were exposed to NMDA and/or anti-NR2b antibody. Mitochondrial membrane potential was measured 18 hours later. NMDA alone caused toxicity versus untreated controls (p<0.001). When the cells were pre-treated with anti-NR2b antibody for 30 minutes prior to addition of NMDA, it provided significant protection versus the toxicity seen with NMDA alone (*p<0.05, **p<0.01). Concentrations: NMDA 125 μM, ab1 5ng/μl, ab2 10ng/μl. Data represents mean ± SEM of at least three independent experiments, analyzed by ANOVA with Neumann-Keuls post-test.

Figure 7. Tat immune complexes which do not interact with the NMDA receptor do not neuroprotect

Mixed rat neuronal cultures were exposed to Tat_Δ31–61, nitrosylated Tat, anti-Tat antibody, and/or NMDA. Mitochondrial membrane potential was measured 15–18 hours later. (A) NMDA caused toxicity versus untreated controls (p<0.01), which the mutant Tat-antiTat immune complex could not attenuate. (B) Nitrosylated Tat-antiTat immune complex could not attenuate toxicity caused by NMDA. Concentrations: Tat 200nM, antibody 5ng/μl, NMDA 125 μM. Data represents mean ± SEM of at least five independent experiments, analyzed by ANOVA with Neumann-Keuls post-test.

Figure 8. Neuroprotection by Tat immune complexes at NMDA receptors

Model of the interaction between Tat and NMDA. Tat (aa 1–48, PDBid 3MIA) was manually docked to a cleft at the interface between the NR1/NR2A complex (PDBid 2AT5). The close proximity of NR1’s cysteine-744 and Tat’s cysteine-31 and Zn-site (coordinated by cysteines 25, 27 and 30) suggests the possibility of a Zn-mediated or disulfide bridge between Tat and the receptor. A similar interaction was observed between HIV-Tat and P-TEFb(Tahirov et al., 2010) (A) Tat (colored magenta) bound to the ligand binding domain of the NMDA receptor (cyan and green). N- and C-term of Tat are exposed to the solvent, and available for binding to anti-Tat antibody. The interaction between the molecules may stabilize the receptor in an active conformation, leading to excitotoxicity. When the anti-Tat antibody binds to the exposed N- or C-terminal of Tat, Tat is still able to bind to the NMDA receptor, but the receptor must then be stabilized in an inactive confirmation. Furthermore, in the presence of this Tat-antiTat immune complex, small molecule agonists are not able to stimulate the receptor either. (B) Protein cartoon of Tat bound to NR1-NR2A heterodimer (NR1 in cyan, NR2A in green, and Tat in magenta.) (C) Solvent accessible surface of the NR1/NR2A heterodimer colored by surface-charge distribution (blue positively charged, red negatively). The interaction between Tat and cysteine-744 on the NMDA receptor is also depicted. (D) Tat, represented as a solvent accessible surface colored in magenta, is docked in the NR1-NR2A interface cleft. The surface corresponding to the Zn²⁺ ion and cysteine-744 are colored yellow and red respectively. The molecular modeling program Pymol^™(Schrodinger, LLC) was used for the modeling and figures.