Pharmacological induction of CCL5 in vivo prevents gp120-mediated neuronal injury

Abstract

The human immunodeficiency virus (HIV) envelope protein gp120 promotes neuronal injury which is believed to cause HIV-associated neurocognitive disorders. Therefore, blocking the neurotoxic effect of gp120 may lead to alternative strategies to reduce the neurotoxic effect of HIV. In vitro, the neurotoxic effect of M-tropic gp120BaL is reduced by the chemokine CCL5, the natural ligand of CCR5 receptors. To determine whether CCL5 reduces the toxic effect of gp120BaL in vivo, animals were intrastriatally injected with lentiviral vectors overexpressing CCL5 prior to an intrastriatal injection of gp120BaL (400 ng). Neuronal injury was determined by silver staining, cleaved caspase-3 and TUNEL. Overexpression of CCL5 decreased gp120-mediated neuronal injury. CCL5 expression can be up-regulated by chronic morphine. Therefore, we examined whether morphine reduces the neurotoxic effect of gp120BaL. Rats stereotaxically injected with gp120BaL into the striatum received saline or chronic morphine for five days (10 mg/kg escalating to 30 mg/kg twice a day). Morphine-treated rats showed a decrease in all markers used to determine neuronal degeneration compared to saline-treated rats. The neuroprotective effect of morphine was significantly attenuated by expressing CCL5 shRNA. Our results suggest that compounds that increase the endogenous production of CCL5 may be used to reduce the pathogenesis of HIV-associated neurocognitive disorders.

Keywords: CCL5-lentivirus; Caspase-3; IL-1β; Morphine withdrawal; Neurodegeneration; Neuroprotection.

Figure 1. Preparation and testing of a pCDH-CCL5 pseudovirus

(A) pCDH-Empty vector or pCDH-CCL5 were transfected into HEK293FT cells using Lipo293 for 24 hr. Representative Western blot analysis of cell lysates using a CCL5 specific antibody. (B) Pseudovirus particles were created using established methods and primary neuronal cultures were infected with either pCDH-Empty vector or pCDH-CCL5 (100 MOI). Media was collected from the cells 24 and 48 hr post infection and CCL5 levels were measured by ELISA. Data are ± SEM (n=3 separate experiments), *p< 0.001 vs untreated and pCDH-Empty vector, **p< 0.001 vs pCDH-CCL5 at 24 hr. (C) Left panel: pCDH-CCL5 pseudovirus was injected into the rat striatum (AP = +0.7mm, ML = +3.0mm, DV = -6.0mm). Middle panel: example of viral infection detected by GFP. Bar=500 μm. Right panel. Protein production by pCDH-CCL5 infection was visualized using the anti-FLAG antibody (red). Arrows indicate CCL5-FLAG production (yellow= overlay). Bar= 20 μm. Levels of CCL5 (D) and IL-1β (E) were measured in the striatum by ELISA two weeks after the injection of pCDH-Empty vector or pCDH-CCL5 pseudovirus. Data are expressed as mean ± SEM (n = 5 per group), *p< 0.05.

Figure 2. pCDH-CCL5 decreases gp120BaL-induced neuronal injury in the rat striatum

pCDH-Empty vector or pCDH-CCL5 packaged into pseudoviruses were injected into the rat striatum two weeks prior to gp120BaL or boiled gp120BaL. Serial sections were then processed for silver staining, cleaved caspase-3 and NeuN (green), and TUNEL and DAPI (blue) five days after gp120BaL injection. Representative images of striata from pCDH-Empty vector + boiled gp120BaL (A, B, C), pCDH-CCL5 + boiled gp120BaL (D, E, F), pCDH-Empty vector + gp120BaL (G, H, I) and pCDH-CCL5 + gp120BaL (J, K, L) - treated rats. Black arrows indicate silver impregnated, degenerating neuronal processes. White arrows indicate either cleaved caspase-3 positive neurons (yellow, green + red) or TUNEL (purple, blue + red) positive cells. Scale bar in J = 10 μm, L = 100 μm. Silver staining, cleaved caspase-3 and TUNEL positive cells, respectively were quantified by ImageJ in serial sections of treated animals (M, N, O). Data are the mean ± SEM (n = rats 6 per group). *p< 0.01 vs boiled gp120BaL. #p< 0.05 vs pCDH-empty vector+gp120BaL.

Figure 3. Chronic morphine and morphine withdrawal differentially affect gp120BaL-induced neuronal injury

Rats were injected with gp120BaL (400 ng) into the striatum two days prior to receiving saline or escalating doses (see Materials and Methods) of morphine (Mor.) over the course of five days. Another group of animals were underwent spontaneous morphine withdrawal. Boiled gp120 was used as a control. Animals were perfused and neuronal injury was examined in serial striatal sections by silver staining, cleaved caspase-3 (red) and NeuN (green), and TUNEL (red) and DAPI (blue). Representative images from striata of gp120BaL + saline (A, B, C), gp120BaL + chronic mor. (D, E, F), and gp120BaL + mor. withdrawal (G, H, I) - treated animals. Black arrows indicate silver impregnated, degenerating neuronal processes. White arrows indicate either cleaved caspase-3 positive neurons or TUNEL positive cells. Scale bar in G = 10 μm, I = 100 μm. J, K, L. Silver staining, cleaved caspase-3 (yellow) and TUNEL (purple) positive cells, respectively were quantified by ImageJ in serial sections. Data are the mean ± SEM (n = 6 rats per group). *p< 0.01 vs boiled gp120BaL, #p< 0.05 vs saline + gp120BaL, **p<0.01 vs saline + gp120BaL.

Figure 4. Characterization of CCL5 shRNA lentivirus for in vivo silencing

(A) HEK293FT cells were transfected with plasmids containing 1 = no treatment, 2 = pCDH CCL5, 3 = pCDH CCL5 + scramble shRNA, 4 = pCDH CCL5 + CCL5 shRNA-1, 5 = pCDH CCL5 + CCL5 shRNA-2, 6 = pCDH CCL5 + CCL5 shRNA-3. 24 hr later lysates were prepared and analyzed by Western blot. Please note that CCL5 shRNA-3 exhibited the highest amount of knockdown and was used for the subsequent experiments. (B) Primary rat astrocytes were infected with pseudovirus particles carrying either scramble shRNA or CCL5 shRNA-3 for 24 hr prior to stimulation with 10 μM morphine. Media was collected 24 and 48 hr following morphine stimulation and CCL5 levels were measured by ELISA. Data are expressed as the mean ± SEM (n=3 separate experiments), *p< 0.05 vs untreated, #p <0.001 vs 10μM morphine and 10 μM morphine + scramble shRNA. (C) Left panel. Example of CCL5 shRNA-3 infection (green). Middle panel. Representative image showing cells positive for endogenous CCL5 (red). Right panel. Image shows that shRNA-3 infected cells are CCL5 negative. Scramble shRNA or CCL5 shRNA-3 pseudovirus was injected into the rat striatum and levels of CCL5 (D) and IL-1β (E) were measured by ELISA two weeks later. Data are expressed as the mean ± SEM (n = 5 per group), *p< 0.05.

Figure 5. CCL5 shRNA-3 decreases morphine-mediated neuroprotection against gp120BaL

Scramble shRNA or CCL5 shRNA-3 packaged into pseudoviruses were injected into the rat striatum two weeks prior to intrastriatal injection of gp120BaL (400 ng). Two days later, rats were treated with chronic morphine (Mor). Serial striatal sections were then processed for (A, D) silver stain, (B, E) cleaved caspase-3 and NeuN (green), and (C, F) TUNEL and DAPI (blue). A, B and C are representative striatal images of scramble shRNA + chronic morphine + gp120BaL treated rats. D, E and F are representative striatal images of CCL5 shRNA-3 + chronic morphine + gp120BaL–treated animals. Black arrows indicate silver impregnated, degenerating neuronal processes. White arrows indicate either cleaved caspase-3 positive neurons (yellow, red + green) or TUNEL positive cells (purple, blue + red). Scale bar in D = 10 μm, F = 100 μm. Silver stained area (G), cleaved caspase-3 positive neurons (H) and TUNEL (I) positive cells were quantified by ImageJ in serial sections. Data are the mean ± SEM (n = 6 rats per group). *p< 0.05 vs scramble shRNA + chronic morphine + gp120BaL.