Increased neuroinflammatory and arachidonic acid cascade markers, and reduced synaptic proteins, in brain of HIV-1 transgenic rats

Abstract

Background: Cognitive impairment has been reported in human immune deficiency virus-1- (HIV-1-) infected patients as well as in HIV-1 transgenic (Tg) rats. This impairment has been linked to neuroinflammation, disturbed brain arachidonic acid (AA) metabolism, and synapto-dendritic injury. We recently reported upregulated brain AA metabolism in 7- to 9-month-old HIV-1 Tg rats. We hypothesized that these HIV-1 Tg rats also would show upregulated brain inflammatory and AA cascade markers and a deficit of synaptic proteins.

Methods: We measured protein and mRNA levels of markers of neuroinflammation and the AA cascade, as well as pro-apoptotic factors and synaptic proteins, in brains from 7- to 9-month-old HIV-1 Tg and control rats.

Results: Compared with control brain, HIV-1 Tg rat brain showed immunoreactivity to glycoprotein 120 and tat HIV-1 viral proteins, and significantly higher protein and mRNA levels of (1) the inflammatory cytokines interleukin-1β and tumor necrosis factor α, (2) the activated microglial/macrophage marker CD11b, (3) AA cascade enzymes: AA-selective Ca2+-dependent cytosolic phospholipase A2 (cPLA2)-IVA, secretory sPLA2-IIA, cyclooxygenase (COX)-2, membrane prostaglandin E2 synthase, 5-lipoxygenase (LOX) and 15-LOX, cytochrome p450 epoxygenase, and (4) transcription factor NF-κBp50 DNA binding activity. HIV-1 Tg rat brain also exhibited signs of cell injury, including significantly decreased levels of brain-derived neurotrophic factor (BDNF) and drebrin, a marker of post-synaptic excitatory dendritic spines. Expression of Ca2+-independent iPLA2-VIA and COX-1 was unchanged.

Conclusions: HIV-1 Tg rats show elevated brain markers of neuroinflammation and AA metabolism, with a deficit in several synaptic proteins. These changes are associated with viral proteins and may contribute to cognitive impairment. The HIV-1 Tg rat may be a useful model for understanding progression and treatment of cognitive impairment in HIV-1 patients.

Figure 1

(A) Representative immunoblot of gp120 and tat protein in HIV-1 Tg rat brain (A), detected as described in Methods. mRNA levels of brain GFAP (B) and CD11b (D) in control and HIV-1 Tg rat brain, measured using real time RT-PCR, normalized to β-globulin and relative to control level (calibrator) using the ΔΔC_T method. Representative immunoblots of (C) GFAP and (E) CD11b protein in control and HIV-1 Tg rat brain. Bar graphs are ratios of optical densities of individual protein bands to β-actin, expressed as percent of control. Data represent mean ± SEM, statistical significance: **p < 0.01, ***p < 0.001 as determined by unpaired t-test.

Figure 2

Representative immunofluorescence (gray scale) for GFAP+ astrocytes (white) in layers IV-V of somatosensory cortex of control (A) and HIV-1 Tg rats (B) and in the hippocampus dentate gyrus of control (C) and HIV-1 Tg rats (D) at 7 months of age. Scale bar = 50 microns. In 3-5 sections obtained from each of 4 animals per group, there was no evidence of astrocyte hypertrophy as represented in the higher magnification image of the astrocyte morphology in the (E, F) somatosensory cortex or (G, H) hippocampus. Images represent compiled z-stack images collected through a 50 micron section. Scale bar = 4 microns.

Figure 3

Representative immunofluorescence (gray scale) for Iba-1+ microglia (white) in layers IV-V of the somatosensory cortex of control (A, C) and HIV-1 Tg rats (B, D) and within the dentate gyrus of the hippocampus of control (E, G) and HIV-1 Tg (F, H) rats at 7 months of age. Images represent compiled z-stack images collected throughout a 50 micron section. Higher magnification of individual representative cells demonstrates diminished arborization of Iba-1+ microglia primarily within the hippocampus. Microglia within defined regional areas were randomly selected (10/section/animal) and the projection distance of the processes was determined using a modified Sholl analysis. In the control brain, 90% (± 10%) of the processes projected past the 4^th Sholl while in the HIV-1 Tg rat this was decreased to only 40% (± 18%). Estimates of complexity of the dendritic branching were generated by counting the number of processes originating at cell body. The number of processes was not statistically different from the number in the HIV-1 Tg rat hippocampus, ranging between 5 and 6 in Tg rats and between 7 and 8 in controls, although complexity and secondary branching appeared lower.

Figure 4

mRNA levels of brain IL-1β (A) and TNFα (B) in control and HIV-1 Tg rats, measured using real time RT-PCR. Data are levels of brain IL-1β and TNFα in the HIV-1 Tg rat normalized to β-globulin and represented relative to control level (calibrator) using the ΔΔC_T method. Representative immunoblots of (C) IL-1β and (D) TNFα protein in control and HIV-1 Tg rat brain. Bar graphs are ratios of optical densities of immunoblots to β-actin, expressed as percent of control (mean ± SEM). Representative brain transcription factor binding activities (DNA-protein complex) of NF-κBp50 (E) and NF-κBp65 (F) in control and HIV-1 Tg rats. DNA binding activity was measured in brain nuclear extracts as described in Methods. Data represent mean ± SEM. Statistical significance: **p < 0.01, ***p < 0.001 as determined by unpaired t-test.

Figure 5

mRNA levels of brain cPLA₂-VIA (A), sPLA₂-II (B), COX-2 (C), and mPGES (D) in control and HIV-1 Tg rats, determined using real time TaqMan RT-PCR. Data are levels of brain cPLA₂-VIA, sPLA₂-II, COX-2 and mPGES in the HIV-1 Tg rat normalized to the endogenous control (β-globulin) and relative to control level (calibrator) using the ΔΔC_T method. Representative immunoblots of (E) cPLA₂-VIA, (F) sPLA₂- IIA (G) COX-2, and (H) mPGES protein in control and HIV-1 Tg rats. Bar graphs represent ratios of optical densities of each individual protein band relative to β-actin, expressed as percent of control mean ± SEM. Mean ± SEM. Data were analyzed by individual unpaired t-tests, statistical significance: **p < 0.01, ***p < 0.001.

Figure 6

mRNA levels of brain 5-LOX _BB(A), 15-LOX (B) and cytochrome p450 epoxygenase (C) in control and HIV-1 Tg rats, measured using real time TaqMan RT-PCR. Data represent individual transcript levels normalized to β-globulin, in HIV-1 Tg rat brain relative to control level (calibrator) using the ΔΔC_T method. Representative immunoblots of (D) 5-LOX, (E) 15-LOX, and (F) cytochrome p450 epoxygenase protein in control and HIV-1 Tg rats. Bar graphs display ratios of optical densities of individual protein bands to β-actin, expressed as percent of control. Mean ± SEM, statistical significance: **p < 0.01, ***p < 0.001 as determined by an unpaired t-test.

Figure 7

(A) Representative brain active caspase-3 level in control and HIV-1 Tg rats. The active caspase-3 level was measured in brain cytosolic fractions as described in Methods. Bar graphs are relative to control and were compared using an unpaired t-test, mean ± SEM. Representative immunoblots of BDNF (B), neurofilament-L (D) and drebrin (F) protein levels in control and HIV-1 Tg rats. Bar graphs display mean ± SEM optical densities of individual protein bands relative to β-actin, expressed as percent of control. mRNA levels of neurofilament-L (C) and drebrin (E) in control and HIV-1 Tg rat brain, measured using real time TaqMan RT-PCR. Data represent mean ± SEM mRNA levels in the HIV-1 Tg rat brain normalized to β-globulin, relative to control level (calibrator) using the ΔΔC_T method. statistical significance: **p < 0.01, ***p < 0.001 as determined by an unpaired t-test.