HIV-1 Tat-induced cerebrovascular toxicity is enhanced in mice with amyloid deposits

Abstract

HIV-1-infected brains are characterized by elevated depositions of amyloid beta (Aβ); however, the interactions between Aβ and HIV-1 are poorly understood. In the present study, we administered specific HIV-1 protein Tat into the cerebral vasculature of 50-52-week-old double transgenic (B6C3-Tg) mice that express a chimeric mouse/human amyloid precursor protein (Mo/HuAPP695swe) and a mutant human presenilin 1 (PS1-dE9) and are characterized by increased Aβ depositions in the brain. Exposure to Tat increased permeability across cerebral capillaries, enhanced disruption of zonula occludens (ZO)-1 tight junction protein, and elevated brain expression of matrix metalloproteinase-9 (MMP-9) in B6C3-Tg mice as compared with age-matched littermate controls. These changes were associated with increased leukocyte attachment and their transcapillary migration. The majority of Tat-induced effects were attenuated by treatment with a specific Rho inhibitor, hydroxyfasudil. The results of animal experiments were reproduced in cultured brain endothelial cells exposed to Aβ and/or Tat. The present data indicate that increased brain levels of Aβ can enhance vascular toxicity and proinflammatory responses induced by HIV-1 protein Tat.

Figure 1. Brain Aβ deposits potentiate Tat-induced disruption of the BBB integrity

(A) Young (3 month old) C57BL/6 mice were injected with the indicated doses of Tat (5, 25, 50 μg), vehicle (Veh) or bovine serum albumin (BSA) into the internal carotid artery (ICA). The BBB integrity was evaluated 3 and 24 h post Tat administration by injecting 100 μl 2% sodium fluorescein (NaF) into the ICA. Fifteen minutes post NaF injection, permeability across the BBB was calculated as the ratio of fluorescence in the brain tissue to plasma. Mannitol (20% in PBS, 200 μl; 15 min exposure) was used as positive control. (B) Tat-induced disruption of the BBB integrity was compared in aged (12 month old) B6C3-Tg mice overexpressing Aβ, age-matched littermate wild type (wt) controls, and young (3 month old) wt C57BL/6 mice. Tat (25 μg) was administered as in (A) and heat-inactivated Tat (Tat_hi) was used as negative control. Selected groups of mice were also injected with hydroxyfasudil (HF; Rho inhibitor; 100 μM, i.p.) 12 h prior to Tat administration. Permeability across the BBB was assessed as in (A). (C) Integrity of the cerebral microvasculature as determined by leakage of FITC-albumin. Young and aged mice were injected with 25 μg Tat as in (A). Twenty four hours later, the animals were perfused with saline, followed by 4 ml 5% FITC-albumin. Control brains show normal staining pattern in which fluorescence is retained within the cerebral vessels. In contrast, the staining is diffused in the ipsilateral hemisphere of Tat-injected brains, indicating albumin leakage. Images shown are the representative data from 4 experiments; the arrows indicate albumin leakage. (D) A typical amyloid plaque in the brain of 12 month old B6C3-Tg mice. Vessel integrity was evaluated as in (C). An advanced amyloid plaque is surrounded by a diffused microvessel network with substantial capillary leakage (arrows), indicating compromised BBB integrity. Data in A and B are mean ±SEM, n= 4–10 in each group; *, p< 0.05; ^†, p< 0.01. The scale bar, 20 μm.

Figure 2. Tat-induced alterations of tight junction (TJ) protein expression in young and aged mice

Aged (12 month old) B6C3-Tg mice, age-matched littermate controls, and young (3 month old) C57BL/6 mice were injected with Tat, Tat_hi, or hydroxyfasudil (HF) as in Figure 1. Expression of ZO-1 (A) and claudin-5 (B) was analyzed 24 h post Tat administration by Western blotting (left panels) and fluorescent microscopy (right panels). In immunostaining studies, ZO-1 and claudin-5 revealed continuous and linear staining in the control mice. In contrast, immunoreactivity of these TJ proteins was decreased and exhibited a dotted and fragmented staining pattern in Tat-injected mice. Images were taken using a 40x lens and are representative from 4 experiments. Data in (A) are mean ±SEM, n= 4–7 in each groups; *, p< 0.05; ^†, p< 0.01.

Figure 2. Tat-induced alterations of tight junction (TJ) protein expression in young and aged mice

Aged (12 month old) B6C3-Tg mice, age-matched littermate controls, and young (3 month old) C57BL/6 mice were injected with Tat, Tat_hi, or hydroxyfasudil (HF) as in Figure 1. Expression of ZO-1 (A) and claudin-5 (B) was analyzed 24 h post Tat administration by Western blotting (left panels) and fluorescent microscopy (right panels). In immunostaining studies, ZO-1 and claudin-5 revealed continuous and linear staining in the control mice. In contrast, immunoreactivity of these TJ proteins was decreased and exhibited a dotted and fragmented staining pattern in Tat-injected mice. Images were taken using a 40x lens and are representative from 4 experiments. Data in (A) are mean ±SEM, n= 4–7 in each groups; *, p< 0.05; ^†, p< 0.01.

Figure 3. Tat-induced MMP-9 expression is potentiated in B6C3-Tg mice

Mice were injected with Tat, Tat_hi, or hydroxyfasudil (HF) as in Figure 1 and analyses were performed 24 h post Tat injection. (A) Plasma MMP-9 activity was determined by zymography. (B) Expression of MMP-9 protein in the brain tissue was assessed by Western blotting. Data are mean ±SEM, n= 4–5; *, p< 0.05; ^†, p< 0.01. (C) In situ zymography was performed to visualize gelatinase activity in the hippocampal sections of aged B6C3-Tg and control mice. Images are representative from 4 experiments.

Figure 4. Exposure to Tat induces leukocyte attachment to the brain endothelium and the presence of inflammatory cells in brain parenchyma

(A) Young (3 month old) mice were operated to install cranial window, followed by injection with Tat, Tat_hi, or hydroxyfasudil (HF), as in Figure 1. Circulatory leukocytes were labeled with rhodamine 6G and the interaction of labeled leukocytes with cerebral vessels was visualized 3 h post Tat injection via the cranial window under a confocal microscope. Arrows indicate leukocytes attached to the brain endothelium. Images are representative from 4 experiments. (B) Aged B6C3-Tg and age-matched wild type control mice were exposed to Tat for 24 h as in Figure 1. Brain slices were stained for factor VIII (endothelial cell marker, red staining) and for F4/80 antigen (macrophage marker, green staining). The areas of colocalization of endothelial cells with macrophages are depicted in yellow.

Figure 5. Aβ potentiates Tat-induced toxicity in brain endothelial cells

(A) Confluent HCMEC/D3 cells were serum deprived for 12 h, followed by treatment with Tat (100 nM), Tat_hi, and/or Aβ (1 μM) for 24 h. Additional cultures were also exposed to Rho inhibitor hydroxyfasudil (HF, 100 μM, pretreatment for 12 h). Then, permeability was measured by transendothelial flux of FITC-dextran 20 kDa. (B) HCMEC/D3 cells were treated as in (A), followed by determination of ZO-1 and claudin-5 protein levels by Western blotting. (C) Confluent HCMEC/D3 cells were treated as in (A). Then, 1×10⁴ fluorescently labeled THP-1 monocytic cells were added for 30 min to HCMEC/D3 cultures and the attachment was assessed using a fluorescence plate reader. (D) Confluent HCMEC/D3 cells cultured on Transwell inserts were treated as in (A). CCL-2 (50 ng/ml) was added to the lower chamber of the Transwell system. THP-1 monocytic cells were added into the upper chamber, and cell migration across HBMEC/D3 monolayers was measured 2 h later. Data are mean ±SEM, n=5–7; *, p< 0.05; ^†, p< 0.01.