Neuropathologies in transgenic mice expressing human immunodeficiency virus type 1 Tat protein under the regulation of the astrocyte-specific glial fibrillary acidic protein promoter and doxycycline

Abstract

The human immunodeficiency virus type 1 (HIV-1) Tat protein is a key pathogenic factor in a variety of acquired immune deficiency syndrome (AIDS)-associated disorders. A number of studies have documented the neurotoxic property of Tat protein, and Tat has therefore been proposed to contribute to AIDS-associated neurological diseases. Nevertheless, the bulk of these studies are performed in in vitro neuronal cultures without taking into account the intricate cell-cell interaction in the brain, or by injection of recombinant Tat protein into the brain, which may cause secondary stress or damage to the brain. To gain a better understanding of the roles of Tat protein in HIV-1 neuropathogenesis, we attempted to establish a transgenic mouse model in which Tat expression was regulated by both the astrocyte-specific glial fibrillary acidic protein promoter and a doxycycline (Dox)-inducible promoter. In the present study, we characterized the phenotypic and neuropathogenic features of these mice. Both in vitro and in vivo assays confirmed that Tat expression occurred exclusively in astrocytes and was Dox-dependent. Tat expression in the brain caused failure to thrive, hunched gesture, tremor, ataxia, and slow cognitive and motor movement, seizures, and premature death. Neuropathologies of these mice were characterized by breakdown of cerebellum and cortex, brain edema, astrocytosis, degeneration of neuronal dendrites, neuronal apoptosis, and increased infiltration of activated monocytes and T lymphocytes. These results together demonstrate that Tat expression in the absence of HIV-1 infection is sufficient to cause neuropathologies similar to most of those noted in the brain of AIDS patients, and provide the first evidence in the context of a whole organism to support a critical role of Tat protein in HIV-1 neuropathogenesis. More importantly, our data suggest that the Dox inducible, brain-targeted Tat transgenic mice offer an in vivo model for delineating the molecular mechanisms of Tat neurotoxicity and for developing therapeutic strategies for treating HIV-associated neurological disorders.

Figure 1.

Transgenic constructs and Dox-dependent GFAP promoter activity. a: Transgenic constructs. pTRE-CAT was constructed as an expressor control for in vitro analysis. The GFAP promoter was cloned in place of the CMV promoter of pTeton regulator, while the Tat cDNA encoding HIV-1 HXB2 Tat (86 amino acids) was cloned into the pTRE. Both pTeton and pTRE were purchased from Clontech. b: The astrocyte-specific activity of the GFAP promoter. Astroglial U87.MG cells were transfected with pTeton and pTRE-CAT (open bars), or pTeton-GFAP and pTRE-CAT (filled bars). Epithelial-derived HeLa cells were included as a control. Transfected cells were cultured in the media supplemented with or without Dox (1 μg/ml) for 48 hours, and harvested for analysis of the CAT activity. Data represent means ± SEM of triplicate samples. c: Tat secretion from U87.MG/iTat cells. U87.MG/iTat cells were generated by sequential stable transfection of pTeton-GFAP and pTRE-Tat86, as described in the Materials and Methods section. ELISA was performed to determine Tat protein in the supernatants of U87.MG/iTat cells in the absence (open bars) or presence (filled bars) of 1 μg/ml Dox, at time points as indicated. Recombinant Tat protein was used as standards. Data represent means ± SEM of triplicate samples.

Figure 2.

Creation of Tat transgenic colonies. a: The creation scheme of the Dox-inducible and brain-targeted Tat transgenic mice. Teton-GFAP (G-tg) and TRE-Tat86 (T-tg) transgenic mice were generated independently, and then crossbred to obtain the Teton-GFAP and TRE-Tat86 bigenic mice (GT-tg). b: Representative genotyping of transgenic mice. Genomic genotyping of transgenic mice was performed by PCR of transgenes in the genomic DNAs isolated from the mouse tail. Fifty ng of genomic DNA was used as templates in the PCR reaction. PCR products were analyzed on 1% agarose gel. The expected sizes of PCR products were 250 bp for TRE-Tat86 and 420 bp for Teton-GFAP, respectively. For each transgene, genomic PCR reactions using primers specific for GAPDH were also performed as a control. Stnd, pTeton-GFAP and pTRE-Tat86 plasmids as templates; Wt, wild-type mice; t, PCR with primers specific for TRE-Tat transgene; g, PCR with primers specific for Teton-GFAP transgene.

Figure 3.

Tat expression in the brain of the GT-tg bigenic mice. a: Dox-dependent Tat expression. GT-tg bigenic mice, 21 days old, were fed with Dox (6 mg/ml)-containing water for 7 consecutive days, and the brain was harvested. Age-matched wild-type mice were also included as a control. Total RNA was then isolated from the brain using the Trizol Reagent (Life Technologies, Inc.). Tat expression was analyzed by RT-PCR of the total RNA using Tat-specific primers. RT-PCR using primers specific for GAPDH, and in the absence of RT were performed as controls. *, Heterozygous GT-tg bigenic mice. b: Tat expression is Dox dose-dependent. GT-tg bigenic mice were fed with drinking water containing Dox at the concentrations as indicated, and Tat expression was analyzed as described above. c: Exclusive Tat expression in the brain. GT-tg bigenic mice were treated with Dox (6 mg/ml) as above, and sacrificed for tissue harvest. Tat expression was then determined by RT-PCR as above. The amplification products in the tissues except brain were determined to be nonspecific, because no hybridization signals were detected on the Southern blot using a Tat-specific probe (data not shown).

Figure 4.

Tat expression-induced growth failure. a: Appearance of the GT-tg bigenic mice treated with or without Dox. b: Effects of Tat expression on body growth of mice. GT-tg bigenic mice, 21 days of age (filled bars), were fed with Dox-containing water (6 mg/ml) for 7 consecutive days and monitored for body growth at days 1, 3, and 7. Age-matched wild-type (open bars), G-tg transgenic (dotted bars), and T-tg transgenic (hatched bars) mice were included as controls. Tat expression in the brain of the mice were confirmed by RT-PCR. c: Dox dose dependence of Tat-induced growth failure. GT-tg bigenic mice were fed with drinking water containing Dox at 1.5 mg/ml (open triangle), 3 mg/ml (open circle), and 6 mg/ml (filled square). The drinking water containing no Dox was included in the GT-tg bigenic mice (open diamond), while 6 mg/ml of Dox was added in the drinking water for the age-matched wild-type mice (open square). Body weight represents means ± SEM of mice in each group.

Figure 5.

Disruption of cerebella and cortex and neuronal death in the brain of Tat-expressing mice. H&E staining was performed in the sections of brain from wild-type mice treated with Dox (6 mg/ml) for 7 days, GT-tg bigenic mice treated with or without Dox (6 mg/ml) for 7 days, and the GT-tg bigenic mice that died 5 days after initiation of Dox treatment (shown as GT-tg + Dox*). The representative images of A–D were taken from the whole brain section, while the representative images of E–L were taken in the cortex region of the brain.

Figure 6.

Astrocytosis, loss of neuron dendrite, and neuronal apoptosis in the brain of Tat-expressing mice. The serial sections of the brain of the wild-type mice treated with Dox and GT-tg bigenic mice treated with or without Dox were immunolabeled with anti-Tat antibody (A–C), anti-GFAP antibody (D–F), and anti-MAP2 antibody (G–I). TUNEL staining was also performed in the brain sections (J–L). The representative images were taken in the cortex region of the brain. Original magnifications: ×20 (A–F and J–L); ×64 (G–I).

Figure 7.

The CNS infiltration induced by Tat expression. a: Increased staining of CD14-, CD4-, and CD8-positive, but not CD68-positive cells in the brain of Tat-expressing mice. The serial sections of the brain of the wild-type mice treated with Dox, the GT-tg bigenic mice treated with or without Dox, and the dead GT-tg bigenic mice resulting from Dox treatment (shown as GT-tg + Dox*) were immunolabeled with anti-CD14 antibody (A–D), anti-CD68 antibody (E–H), anti-CD4 antibody (I–L), and anti-CD8 antibody (M–P). The representative images were taken in the cortex region of the brain. b: Quantitation of CD14-, CD68-, CD4-, and CD8-positive cells in the cortex region of immunolabeled sections from A. Cell countings of positively stained cells were performed as described in the Materials and Methods section. Open bars, wild-type mice treated with Dox; dotted bars, GT-tg bigenic mice treated without Dox; filled bars, GT-tg bigenic mice treated with Dox; hatched bars, dead GT-tg bigenic mice 5 days after initiation of Dox treatment. c: IFN-γ production in the brain of Tat-expressing mice. Brain homogenates were prepared from the other hemibrain of the mice as described in the Materials and Methods section. The levels of IFN-γ production in the brain homogenates were determined by ELISA. Data represent means ± SEM of mice in each group (b and c).