HIV and FIV glycoproteins increase cellular tau pathology via cGMP-dependent kinase II activation

Abstract

As the development of combination antiretroviral therapy (cART) against human immunodeficiency virus (HIV) drastically improves the lifespan of individuals with HIV, many are now entering the prime age when Alzheimer's disease (AD)-like symptoms begin to manifest. It has been shown that hyperphosphorylated tau, a known AD pathological characteristic, is prematurely increased in the brains of HIV-infected individuals as early as in their 30s and that its levels increase with age. This suggests that HIV infection might lead to accelerated AD phenotypes. However, whether HIV infection causes AD to develop more quickly in the brain is not yet fully determined. Interestingly, we have previously revealed that the viral glycoproteins HIV gp120 and feline immunodeficiency virus (FIV) gp95 induce neuronal hyperexcitation via cGMP-dependent kinase II (cGKII; also known as PRKG2) activation in cultured hippocampal neurons. Here, we use cultured mouse cortical neurons to demonstrate that the presence of HIV gp120 and FIV gp95 are sufficient to increase cellular tau pathology, including intracellular tau hyperphosphorylation and tau release to the extracellular space. We further reveal that viral glycoprotein-induced cellular tau pathology requires cGKII activation. Taken together, HIV infection likely accelerates AD-related tau pathology via cGKII activation.

Keywords: Alzheimer's disease; CGMP-dependent kinase II; Feline immunodeficiency virus; Human immunodeficiency virus; Tau pathology.

Fig. 1.

HIV gp120 or FIV gp95 treatment significantly increases neuronal Ca²⁺ activity in cultured mouse cortical neurons via cGKII activation. (A) Representative traces of GCaMP7s fluorescence intensity and a summary graph of normalized mean±s.d. of total Ca²⁺ activity in cultured WT cortical neurons in each condition showing that 1 nM HIV gp120 or 1 nM FIV gp95 treatment significantly increases neuronal Ca²⁺ activity, which is reversed by inhibition of cGKII activity using 1 μM Rp8-Br-PET-cGMPS (RP) [n=number of neurons from two independent cultures, control (CTRL)=48, RP=49, HIV gp120=50, HIV gp120+RP=49, FIV gp95=47, and FIV gp95+RP=46]. *P<0.05, ***P<0.001, ****P<0.0001 (two-way ANOVA with Tukey test). (B) Representative traces of GCaMP7s fluorescence intensity and a summary graph of normalized mean±s.d. of total Ca²⁺ activity in cultured cGKII KO cortical neurons in each condition showing that 1 nM HIV gp120 or 1 nM FIV gp95 treatment has no effect on neuronal Ca²⁺ activity in KO cells (n=number of neurons from two independent cultures, control (CTRL)=60, HIV gp120=56, and FIV gp95=56). Results are not significantly different (one-way ANOVA with Tukey test).

Fig. 2.

HIV gp120 or FIV gp95 treatment significantly increases extracellular tau levels via cGKII activation. (A) A summary graph of extracellular tau levels in cultured WT cortical neurons in the absence or presence of 1 mM glutamate (Glu) showing that an increase in neuronal activity significantly increases extracellular tau concentration [mean±s.d.; n=number of ELISA measurements, control (CTRL)=4 and Glu=4 from two independent cultures]. **P<0.01 (Mann–Whitney test). (B) A summary graph of extracellular tau levels in cultured WT cortical neurons in each condition demonstrating that viral glycoproteins, HIV gp120 and FIV gp95, significantly elevates tau release to the extracellular space, which is reversed by cGKII inhibition using 1 μM Rp8-Br-PET-cGMPS (RP) [mean±s.d.; n=number of ELISA measurements from three or four independent cultures, control (CTRL)=8, RP=6, HIV gp120=8, HIV gp120+RP=6, FIV gp95=8, and FIV gp95+RP=6]. *P<0.05, **P<0.01, ***P<0.001 (two-way ANOVA with Tukey test). (C) A summary graph of extracellular tau levels in cultured cGKII KO cortical neurons treated with 1 nM HIV gp120 or 1 nM FIV gp95 showing that viral glycoproteins are unable to alter extracellular tau levels in KO cells [mean±s.d.; n=number of ELISA measurements from 6 independent cultures, control (CTRL)=12, HIV gp120=12, and FIV gp95=12]. Results are not significantly different (one-way ANOVA with Tukey test).

Fig. 3.

The viral glycoproteins HIV gp120 and FIV gp95 significantly enhance tau hyperphosphorylation via cGKII activation. (A) Representative immunoblots and a summary graph of normalized phosphorylated tau levels in cultured WT cortical neurons in each condition showing that 1 nM HIV gp120 or 1 nM FIV gp95 treatment significantly increases AT8-positive tau hyperphosphorylation, which is reversed by inhibition of cGKII activity using 1 μM Rp8-Br-PET-cGMPS (RP) [mean±s.d.; n=number of immunoblots from seven independent cultures, control (CTRL)=24, RP=22, HIV gp120=12, HIV gp120+RP=10, FIV gp95=14, and FIV gp95+RP=14]. **P<0.01, ****P<0.0001 (two-way ANOVA with Tukey test). (B) Representative immunoblots and a summary graph of normalized phosphorylated tau levels in cultured cGKII KO cortical neurons demonstrating that 1 nM HIV gp120 or 1 nM FIV gp95 treatment has no effect on tau hyperphosphorylation in KO cells [mean±s.d.; n=number of immunoblots from two independent cultures, control (CTRL)=4, HIV gp120=4, and FIV gp95=4]. Results are not significantly different (one-way ANOVA with Tukey test). The position of molecular mass markers (kDa) is shown on the right of the blots.

Fig. 4.

Viral glycoprotein-induced tau hyperphosphorylation is mediated by cGKII-induced p38K activation. (A) Representative immunoblots and a summary graph of normalized phosphorylated p38K (pp38K) levels in cultured WT cortical neurons in each condition showing that 1 nM HIV gp120 or 1 nM FIV gp95 treatment significantly increases p38K activation, which is reversed by inhibition of cGKII activity using 1 μM Rp8-Br-PET-cGMPS (RP) [mean±s.d.; n=number of immunoblots from four independent cultures, control (CTRL)=8, RP=8, HIV gp120=8, HIV gp120+RP=8, FIV gp95=8, and FIV gp95+RP=8]. **P<0.01, ***P<0.001, ****P<0.0001 (two-way ANOVA with Tukey test). (B) Representative immunoblots and a summary graph of normalized phosphorylated p38K levels in cultured cGKII KO cortical neurons demonstrating that 1 nM HIV gp120 or 1 nM FIV gp95 treatment has no effect on p38K activation in KO cells [mean±s.d.; n=number of immunoblots from six independent cultures, control (CTRL)=12, HIV gp120=12, and FIV gp95=12]. Results are not significantly different (one-way ANOVA with Tukey test). (C) Representative immunoblots and a summary graph of normalized phosphorylated tau levels in cultured WT cortical neurons in each condition showing that 1 nM HIV gp120 or 1 nM FIV gp95 treatment significantly increases AT8-positive tau hyperphosphorylation, which is reversed by inhibition of p38K activity using 10 μM SB203580 (SB) [mean±s.d.; n=number of immunoblots from four independent cultures, control (CTRL)=8, SB=8, HIV gp120=8, HIV gp120+SB=8, FIV gp95=8, and FIV gp95+SB=8]. **P<0.01, ***P<0.001, ****P<0.0001 (two-way ANOVA with Tukey test). The position of molecular mass markers (kDa) is shown on the right of the blots.

Fig. 5.

Glutamate treatment is sufficient to induce tau hyperphosphorylation, but this is not mediated by cGKII activation. (A) Representative immunoblots and a summary graph of normalized phosphorylated tau levels in cultured WT cortical neurons treated with 1 mM glutamate (Glu) or 1 mM glutamate and 1 μM Rp8-Br-PET-cGMPS (Glu+RP) showing that glutamate treatment significantly increases AT8-positive tau hyperphosphorylation, while cGKII inhibition is unable to prevent such an increase [mean±s.d.; n=number of immunoblots from three to five independent cultures, control (CTRL)=14, Glu=14, and Glu+RP=6]. *P<0.05, **P<0.01 (two-way ANOVA with Tukey test). (B) Representative immunoblots and a summary graph of normalized phosphorylated tau levels in cultured cGKII KO cortical neurons treated with 1 mM glutamate (Glu) showing that glutamate treatment significantly increases tau hyperphosphorylation in cGKII KO cells [mean±s.d.; n=number of immunoblots from three independent cultures, control (CTRL)=6 and Glu=6]. *P<0.05 (Wilcoxon signed rank test). The position of molecular mass markers (kDa) is shown on the right of the blots.

Fig. 6.

A schematic model of viral glycoprotein-induced cellular tau pathology. FIV gp95 or HIV gp120 interaction with chemokine receptor CXCR4 enhances Ca²⁺-regulating systems through glutamate NMDA receptors (NMDARs). Ca²⁺ fluxes through NMDARs promote the production of nitric oxide (NO) by neuronal nitric oxide synthase (nNOS), which is tethered by the scaffolding protein postsynaptic density 95 (PSD95; also known as DLG4). NO induces the production of cGMP, in turn activates cGMP-dependent protein kinase II (cGKII), which can phosphorylate the glutamate AMPA receptor (AMPAR) subunit GluA1 to increase AMPAR synaptic trafficking, which leads to synaptic hyperexcitation. cGKII-induced increase in glutamatergic excitation elicits extracellular tau release, resulting in tau propagation. Additionally, cGKII can phosphorylate p38Ks, leading to tau PHF formation. Therefore, HIV- and FIV-induced stimulation of cGKII is critical for AD-related cellular tau pathology. Rp8-Br-PET-cGMPS (RP) and SB203580 (SB) are inhibitors for cGKII and p38Ks, respectively.