Modulation of ABCA1 by an LXR agonist reduces β-amyloid levels and improves outcome after traumatic brain injury

Abstract

Traumatic brain injury (TBI) increases brain beta-amyloid (Aβ) in humans and animals. Although the role of Aβ in the injury cascade is unknown, multiple preclinical studies have demonstrated a correlation between reduced Aβ and improved outcome. Therefore, therapeutic strategies that enhance Aβ clearance may be beneficial after TBI. Increased levels of ATP-binding cassette A1 (ABCA1) transporters can enhance Aβ clearance through an apolipoprotein E (apoE)-mediated pathway. By measuring Aβ and ABCA1 after experimental TBI in C57BL/6J mice, we found that Aβ peaked early after injury (1-3 days), whereas ABCA1 had a delayed response (beginning at 3 days). As ABCA1 levels increased, Aβ levels returned to baseline levels-consistent with the known role of ABCA1 in Aβ clearance. To test if enhancing ABCA1 levels could block TBI-induced Aβ, we treated TBI mice with the liver X-receptor (LXR) agonist T0901317. Pre- and post-injury treatment increased ABCA1 levels at 24 h post-injury, and reduced the TBI-induced increase in Aβ. This reduction in Aβ was not due to decreased amyloid precursor protein processing, or a shift in the solubility of Aβ, indicating enhanced clearance. T0901317 also limited motor coordination deficits in injured mice and reduced brain lesion volume. These data indicate that activation of LXR can reduce Aβ accumulation after TBI, and is accompanied by improved functional recovery.

FIG. 1.

Traumatic brain injury (TBI) increases amyloid precursor protein (APP) processing, beta-amyloid (Aβ), and ATP-binding cassette A1 (ABCA1) levels in non-transgenic C57BL/6 mice. (A) Soluble Aβ₄₀ and Aβ₄₂ peptides were measured in ipsilateral cortex 1, 3, and 7 days post-TBI. (B) TBI-induced changes in protein levels of full-length APP (flAPP) and APP-C-terminal fragments (APP-CTF) in mouse cortex. The panel on the left shows mouse brain compared to an APP-overexpressing mouse neuroblastoma cell line treated with a γ-secretase inhibitor to accumulate APP-CTF as a positive control. The asterisk denotes a non-specific band in mouse brain that does not occur in cell lines. The panel on the right shows changes in flAPP, APP-CTF, and β-secretase enzyme 1 (BACE1) after injury. (C) ABCA1 and low-density lipoprotein receptor (LDLR) protein levels in the ipsilateral cortex of controlled cortical impact (CCI)-injured mice after 1, 3, or 7 days compared to sham animals. Representative Western blots are shown (*p < 0.05, **p < 0.01 versus sham animals by analysis of variance and Neuman-Keuls post-hoc analysis; mean ± standard error of the mean; n = 3).

FIG. 2.

T0901317 pretreatment prevents the accumulation of beta-amyloid (Aβ) following traumatic brain injury (TBI). (A) The effects of TBI and T0901317 (T0) on APP processing, BACE1, and ABCA1 protein levels. The asterisk denotes a non-specific band. (B) Bar graph representing quantification of ABCA1 after TBI, and the effect of T0901317 on these changes. (C) Percent increases in Aβ₄₀ and Aβ₄₂ following TBI, and the effects of T0901317 on the Aβ response, as measured by ELISA. Baseline levels are indicated in the results section. Bar graphs representing the effects of T0901317 on the TBI-induced changes in (D) full-length APP (flAPP), (E) APP-CTF, and (F) BACE1 protein levels (*p < 0.05, **p < 0.01 versus sham animals; +p < 0.05 versus TBI + vehicle by analysis of variance and Neuman-Keuls post-hoc analysis; mean ± standard error of the mean; n = 6; ELISA, enzyme-linked immunosorbent assay; BACE1, β-secretase enzyme 1; ABCA1, ATP-binding cassette A1; APP-CTF, APP-C-terminal fragment; APP, amyloid precursor protein).

FIG. 3.

T0901317 pretreatment attenuates inflammation markers following traumatic brain injury (TBI). (A) Time course of IL-1β levels after TBI in mice as quantified by ELISA (*p < 0.05 versus sham animals). (B) Effect of T0901317 on the TBI-induced increase in IL-1β at 1 day post-trauma (**p < 0.01 versus sham vehicle animals; ++p < 0.01 versus TBI + vehicle animals). (C) Time course of COX-2 protein levels after TBI in mice (*p < 0.05 versus sham animals). (D) Effect of T0901317 (T0) on the TBI-induced increase in COX-2 protein levels at 1 day post-trauma (*p < 0.05 versus sham vehicle animals; +p < 0.05 versus TBI + vehicle animals; n = 6; ELISA, enzyme-linked immunosorbent assay; COX-2, cyclooxygenase-2; IL-1β, interleukin-1β).

FIG. 4.

Post-injury administration of T0901317 reduces Aβ, but not IL-1β, following traumatic brain injury (TBI). T0901317 was administered either for 3 days prior to TBI (pre), or as a single injection 15 min after TBI (post). The ipsilateral cortex was collected 24 h after injury. (A) The effects of these regimens on protein levels of ABCA1 and LDLR (*p < 0.05; **p < 0.01 versus TBI + vehicle animals). (B) The effects of T0901317 treatment regimens on TBI-induced Aβ₄₀ levels. Data are shown as percent increase over sham animals. Sham Aβ₄₀ levels were 5.7 ± 0.54 fmol/mg protein (*p < 0.05; **p < 0.01 versus TBI + vehicle animals). (C) The differential effects of pre- and post-injury treatment regimens on TBI-induced IL-1β levels (*p < 0.05 versus sham + vehicle animals; + p < 0.05 versus TBI + vehicle animals). (D) Effects of TBI and T0901317 treatments on APP processing. The asterisk denotes a non-specific band (mean ± standard error of the mean; n = 5; IL-1β, interleukin-1β; ABCA1, ATP-binding cassette A1; LDLR, low-density lipoprotein receptor; Aβ, beta-amyloid; APP, amyloid precursor protein; APP-CTF, APP-C-terminal fragment; flAPP, full-length APP).

FIG. 5.

T0901317 attenuates behavioral deficits and reduces lesion volume following traumatic brain injury (TBI). (A) Fine motor coordination deficits were quantified using a modified beam-walking test. Hindlimb foot placement was recorded and the number of mistakes (footfaults) were recorded from 50 steps (***p < 0.001 versus TBI + vehicle animals by repeated-measures two-way analysis of variance [ANOVA] and Bonferonni post-hoc analysis; n = 13–14 for the TBI + vehicle and TBI + pretreatment [pre] groups; n = 6 for all other groups; post, post-treatment). (B) The lesions were imaged using a 7-Tesla magnetic resonance imaging (MRI) device. 2T-weighted images were generated, and the lesion area can be seen as a hyperintense white region in the upper left cortical area. Representative images from the TBI groups are shown. (C) Lesion quantification by MRI (**p < 0.01 versus vehicle animals by ANOVA and Neuman-Keuls post-hoc analysis; mean ± standard error of the mean; n = 8 for the TBI + T0901317 pretreatment group and the TBI + vehicle group; n = 5 for the TBI + T0901317 treatment group).