CNS expression of anti-inflammatory cytokine interleukin-4 attenuates Alzheimer's disease-like pathogenesis in APP+PS1 bigenic mice

Abstract

Cytokines play an emerging role as neurotransmitters, neuromodulators, and neurohormones in the brain. This paradigm shift in cytokine function offers a new framework to understand their roles in ameliorating neurodegenerative disorders, such as Alzheimer's disease (AD). Molecular adjuvant therapy of AD animal models with glatiramer acetate induces anti-inflammatory responses and therapeutic effects. Although these effects are potentially mediated through anti-inflammatory cytokine signaling, the exact molecular identities and pathways are poorly understood. Here, we show that virus-mediated expression of the mouse interleukin (IL)-4 gene in beta-amyloid precursor protein + presenilin-1 (APP+PS1) bigenic mice attenuates AD pathogenesis. Introduction of an adeno-associated viral (AAV) vector encoding IL-4 into the hippocampus resulted in sustained expression of IL-4, reduced astro/microgliosis, amyloid-beta peptide (Abeta) oligomerization and deposition, and enhanced neurogenesis. Moreover, increased levels of IL-4 improved spatial learning, promoted phosphorylation of N-methyl-D-aspartate receptor subunit 2B at Tyr 1472, and enhanced its cell surface retention both in vivo and in vitro. Our data suggest that neuronal anti-inflammatory cytokine signaling may be a potential alternative target for non-Abeta-mediated treatment of AD.

Figure 1

AAV injection and strategy. A) AAV-mediated somatic gene transfer of IL-4. Non-Tg mice at 3 mo of age received bilateral hippocampal injection of AAV-GFP (2×10⁹ viral particles (VP)/hippocampus, dotted line) or AAV-IL-4 (2×10⁹ VP/hippocampus, solid line), were sacrificed 4, 12, and 24 wk postinjection. The amounts of IL-4 in mouse hippocampal protein extracts were quantified by enzyme-linked immunosorbent assay, showing sustained expression of IL-4. Error bars represent se (n=3). B) Experimental design. APP+PS1 mice received bilateral hippocampal injection of AAV-GFP or AAV-IL-4 at 3 mo of age and were tested by 2-d RAWM task at 7 mo of age. Non-Tg mice served as a positive control for the spatial learning task. Mice were intraperitoneally injected with BrdU 5 times for 2.5 d, 21 d before being sacrificed at 8 mo of age.

Figure 2

Gene delivery of IL-4 suppresses glial accumulation in APP+PS1 mice. A–D) APP+PS1 mice injected with AAV-GFP (A, B) or AAV-IL-4 (C, D) at 3 mo of age were sacrificed at 8 mo of age. Hippocampal frozen sections were immunostained for GFAP (astrocyte; A, C) or IBA1 (microglia; B, D), and counterstained by TS. Scale bars = 200 μm (low magnification; left); 40 μm (high magnification; right). E, F) Quantification of GFAP⁺ (E) or IBA1⁺ (F) cells found within the circle surrounding TS⁺ Aβ plaques. Radii of outer concentric circles in GFAP⁺ cells were 100 μm greater than the inner circles that surrounded the compact plaques (A, C), and 50 μm greater in IBA1⁺ cells (B, D). Error bars = se (n=5/group, 10 sections/brain). *P < 0.05, **P < 0.01 vs. AAV-GFP group; Student’s t test.

Figure 3

Aβ deposition in the hippocampal region of gene-delivered APP+PS1 mouse brain. A–D) Frozen brain sections of APP+PS1 mice injected with AAV-GFP (A, B) or AAV-IL-4 (C, D) were immunostained with anti-Aβ antibody (A, C) and counterstained by TS for compact plaque (B, D). E, F) Total Aβ load (E) and TS⁺ area (F) in hippocampal regions were quantified in APP+PS1 mice injected with AAV-GFP (GFP) or AAV-IL-4 (IL-4). Error bars = se (n=5/group, 10 sections/brain). *P < 0.05, **P < 0.01 vs. AAV-GFP group; Student’s t test. Scale bar = 500 μm.

Figure 4

Aβ oligomer formation in the hippocampus of gene-delivered APP+PS1 mouse brain. A) Immunoblotting using 6E10 antibody showed specific Aβ oligomers in the extracellular-enriched fraction of AAV-injected APP+PS1 mouse hippocampus. Arrowheads indicate respective migration positions of multimers. Arrow indicates sAPPα fragments processed from full-length APP. Asterisk indicates nonspecific band. B) Band luminescent intensities for 1-, 8-, and 9-mers were quantified by Image J software. Bars represent average values (n=6/group). *P < 0.05, **P < 0.01 vs. AAV-GFP group; Student’s t test.

Figure 5

Gene delivery of IL-4 enhances neurogenesis in SGZ of APP+PS1 mice. A–C) Representative images of Dcx staining in the dentate gyrus of non-Tg (A) or APP+PS1 mice injected with AAV-GFP (B) or AAV-IL-4 (C) into the hippocampus at 3 mo of age, and intraperitoneally injected with BrdU 3 wk prior to euthanasia at 8 mo of age. D–F) Immunofluorescence of BrdU (green) and NeuN (red) in the dentate gyrus of non-Tg (D) or APP+PS1 mice injected with AAV-GFP (E) or AAV-IL-4 (F). G, H) Quantification of Dcx⁺ (G) or BrdU⁺Neu⁺ cells (H) in SGZ. Error bars = se (n=5/group, 10 sections/brain). *P < 0.05, ***P < 0.001 vs. non-Tg group; ^#P < 0.05, ^##P < 0.01 vs. AAV-GFP group; 1-way ANOVA and Newman-Keuls posttest. Scale bars = 200 μm (C); 100 μm (F).

Figure 6

Gene delivery of IL-4 improves memory function of APP+PS1 mice. APP+PS1 mice received bilateral hippocampal injection of AAV-GFP or AAV-IL-4 at 3 mo of age and were tested by the 2-d RAWM task at 7–8 mo of age. Non-Tg and uninjected APP+PS1 mice (Tg/uninjected group) serve as positive/negative controls for the spatial learning task. Error bars = se (n=8/group). *P < 0.05, **P < 0.01 vs. AAV-GFP group; 2-way repeated-measures ANOVA and Bonferroni post hoc test.

Figure 7

IL-4 promotes the expression and phosphorylation of NR2B. A) Immunoblotting (top) and quantification (bottom) of NR2A and NR2B in the membrane-enriched fraction of the mouse hippocampus in non-Tg (NTg), uninjected APP+PS1 (uninj.), APP+PS1 mice injected with AAV-GFP (GFP), and APP+PS1 mice injected with AAV-IL-4 (IL-4). B) Immunoblotting (top) and quantification (bottom) of phospho-Tyr1472 NR2B (pNR2B) in the same fraction of the hippocampus. *P < 0.05, **P < 0.01, ***P < 0.001 vs. NTg group; ^#P < 0.05, ^###P < 0.001 vs. AAV-GFP group; 1-way ANOVA and Newman-Keuls post hoc test. Error bars = se (n=6/group). C) Immunoblotting (top) and quantification (bottom) of pNR2B, NR2B, and β-actin in IL-4-treated primary cultured neurons. D) Immunoblotting (top) and quantification (bottom) of biotinylated NR2B at the cell surface level, total NR2B in IL-4 and/or oligomeric Aβ42-treated primary cultured neurons. **P < 0.01, ***P < 0.001 vs. control; ^##P < 0.01 vs. Aβ42 group; 1-way ANOVA and Newman-Keuls post hoc test. Error bars represent se (n=3/group). Con, control; N.D., no significant difference.