Complement C3 deficiency leads to accelerated amyloid beta plaque deposition and neurodegeneration and modulation of the microglia/macrophage phenotype in amyloid precursor protein transgenic mice

Abstract

Complement factor C3 is the central component of the complement system and a key inflammatory protein activated in Alzheimer's disease (AD). Previous studies demonstrated that inhibition of C3 by overexpression of soluble complement receptor-related protein y in an AD mouse model led to reduced microgliosis, increased amyloid beta (Abeta) plaque burden, and neurodegeneration. To further address the role of C3 in AD pathology, we generated a complement C3-deficient amyloid precursor protein (APP) transgenic AD mouse model (APP;C3(-/-)). Brains were analyzed at 8, 12, and 17 months of age by immunohistochemical and biochemical methods and compared with age-matched APP transgenic mice. At younger ages (8-12 months), no significant neuropathological differences were observed between the two transgenic lines. In contrast, at 17 months of age, APP;C3(-/-) mice showed significant changes of up to twofold increased total Abeta and fibrillar amyloid plaque burden in midfrontal cortex and hippocampus, which correlated with (1) significantly increased Tris-buffered saline (TBS)-insoluble Abeta(42) levels and reduced TBS-soluble Abeta(42) and Abeta(40) levels in brain homogenates, (2) a trend for increased Abeta levels in the plasma, (3) a significant loss of neuronal-specific nuclear protein-positive neurons in the hippocampus, and (4) differential activation of microglia toward a more alternative phenotype (e.g., significantly increased CD45-positive microglia, increased brain levels of interleukins 4 and 10, and reduced levels of CD68, F4/80, inducible nitric oxide synthase, and tumor necrosis factor). Our results suggest a beneficial role for complement C3 in plaque clearance and neuronal health as well as in modulation of the microglia phenotype.

Figure 1.

Complement C3 protein is expressed in the brain and in the plasma of 12- and 17-month-old APP mice but is undetectable in APP;C3^−/− mice. A, Complement C3 gene deletion was determined by PCR by the absence of the wild-type C3 allele PCR product (top band at ∼350 bp) and the presence of the mutated allele (bottom band at ∼280 bp). B–D, Expression of complement C3 protein in the brain (B, C) and plasma (D) of 12- and 17-month-old APP mice (n = 5 per group) was shown by Western blot (B, C) and C3 ELISA (D). C3 protein was absent in the corresponding APP;C3^−/− mice. The α/β dimer of C3 protein (∼185 kDa) in brain is indicated by the arrow in B.

Figure 2.

Neuropathological and biochemical analysis at 12 months of age showed no significant differences in plaque load, microgliosis, and astrocytosis in APP;C3^−/− (filled bars; n = 8) compared with age-matched APP control mice (open bars; n = 6). A, B, Quantitative image analysis of Aβ₄₂- (A) and Aβ₄₀- (B) specific immunoreactivity did not show any significant differences in hippocampus or midfrontal cortex. C–E, Image analysis of thioflavin S-positive plaque load (C) as well as CD45- and GFAP-specific immunoreactivity in the hippocampus (D, E) revealed no significant differences between the APP and APP;C3^−/− groups. F–H, Quantification of Aβ₄₂ and Aβ₄₀ by ELISA in TBS, TBS-T, and guanidine extracts of brain homogenates (F, G) or in plasma (H) did not reveal any significant difference between the groups. % ROI, Percentage of immunoreactivity within the region of interest.

Figure 3.

A–E, Quantitative neuropathological analysis (A–C) and biochemical analysis of brain homogenates by quantitative Aβ₄₂- and Aβ₄₀-specific ELISAs (D, E) of 17-month-old mice shows significantly higher Aβ plaque load and significantly higher Aβ₄₂ levels in guanidine brain extract in APP;C3^−/− mice (filled bars; n = 5) compared with age-matched APP controls (open bars; n = 5). A–C, Aβ₄₂- and Aβ₄₀- specific immunoreactivity (A, B), as well as thioflavin S-positive staining (C), were significantly increased in hippocampus and midfrontal cortex of APP;C3^−/− mice (A–C; p < 0.05). D, E, Aβ₄₂ and Aβ₄₀ levels in TBS-soluble extracts of brain homogenates were significantly reduced, whereas Aβ₄₂ was significantly elevated and Aβ₄₀ nonsignificantly increased in TBS-insoluble guanidine extracts in APP;C3^−/− mice. Aβ₄₂ and Aβ₄₀ levels in the membrane-bound TBS-T extracts were nonsignificantly reduced. F, A trend was evident for higher total Aβ levels in plasma samples from APP;C3^−/− mice compared with APP mice (51% increase; p = 0.06). *p < 0.05, **p < 0.01. % ROI, Percentage of immunoreactivity within the region of interest.

Figure 4.

A–N, Immunohistochemical analysis of hippocampus of representative sections of 17-month-old APP (left column; 2 sets of serial sections: A, C, E and G, I, K, M) and APP;C3^−/− (right column; 2 sets of serial sections: B, D, F and H, J, L, N) mice. A, B, Relative to complement-sufficient APP Tg mice (A), the total plaque load demonstrated using the pan-specific Aβ antibody R1282 was increased in the complement-deficient APP;C3^−/− mice (B). C, D, In parallel, the number of CD45-positive microglia was also increased in APP;C3^−/− mice, and the microglia were mostly associated with compacted plaques (arrows). E, F, GFAP immunoreactivity, a marker for astrogliosis, was comparable between both groups, although modest increases were observed in APP;C3^−/− mice. G–N, In parallel to the increase of total (G, H; R1282 antibody) and fibrillar, compact plaque load (M, N; thioflavin S), the number of Iba1-immunoreactive microglia/macrophage (I, J) and dystrophic neurites (K, L) observed in APP;C3^−/− mice (right column) was increased compared with that of APP mice (left column). Scale bars, 200 μm.

Figure 5.

Quantitative neuropathological and biochemical analysis of gliosis at 17 months of age. A, B, Total CD45 immunoreactivity was significantly increased in APP;C3^−/− mice (A; *p < 0.05, **p < 0.01), whereas the CD45/thioflavin S ratio (B) only showed a trend to be elevated in APP;C3^−/− mice (p = 0.10). C, Iba1 immunoreactivity was modestly and nonsignificantly elevated in APP;C3^−/− compared with APP mice. D, GFAP immunoreactivity was only slightly higher in APP;C3^−/− compared with APP mice, but the difference was nonsignificant. E, F, In contrast to CD45 and Iba1, the levels of microglial markers CD68 and F4/80 by Western blot were nonsignificantly lower in TBS-soluble brain homogenates of APP;C3^−/− compared with APP mice. The levels in the APP;C3^−/− mice are presented as a percentage of the levels in the APP transgenic mice. β-Actin was used as a protein loading control.

Figure 6.

Differential activation of microglia/macrophage in the brains of APP;C3^−/− mice toward a more alternative activation/M2 phenotype. A, iNOS levels were significantly reduced in TBS-soluble, but not TBS-T-soluble, membrane-bound brain homogenates by Western blot in APP;C3^−/− compared with APP mice (*p < 0.05). B, TBS-T-soluble, membrane-bound and TBS-soluble TNF levels by ELISA were lower in APP;C3^−/− compared with APP mice, but only the difference in membrane-bound TNF reached significance (*p < 0.05). C, D, IL-4 (C) and IL-10 (D) levels in TBS and TBS-T brain homogenates by ELISA were elevated in APP;C3^−/− compared with APP mice, but only the difference in IL-4 in TBS-soluble brain homogenates reached significance (*p < 0.05).

Figure 7.

APP processing was not altered in C3-deficient APP mice (filled bars) at either 12 or 17 months of age. Quantification of β-actin normalized total APP protein (A, B; average of 2 experiments, n = 4 mice per group) and αAPPs fragment (C, D; average of 2 experiments, n = 4 mice per group) in brain homogenates by Western blot revealed no difference in protein levels (p > 0.05).

Figure 8.

At 17 months of age, C3-deficient APP transgenic mice showed a reduction in the number of NeuN-positive neurons that correlated with the thioflavin S plaque load. A, B, Seventeen-month-old APP;C3^−/− mice (B; n = 5) had fewer NeuN-positive neurons in the CA3 region of the hippocampus compared with the APP control group (A; n = 5). Neuronal counting was performed in CA3 between the black lines indicated in A and B. C, Quantification of NeuN-positive neurons in the CA3 region of the hippocampus by stereological counting showed a significant reduction in APP;C3^−/− mice (filled bars) compared with APP mice (open bars) (*p < 0.05). No difference in neuronal counts was detected between APP and APP;C3^−/− mice at 12 months of age. A significant reduction in neurons was observed in APP;C3^−/− mice, but not APP mice, from 12 to 17 months of age (p = 0.015). D, The number of NeuN-positive neurons of individual mice correlated significantly with their thioflavin S plaque load (r = −0.636; p < 0.05; n = 5 mice per group; open diamonds indicate APP mice, and black circles indicate APP;C3^−/− mice). E, MAP2 immunoreactivity was nonsignificantly reduced in complement C3-deficient APP mice compared with complement-sufficient APP mice. F, Synaptophysin levels in TBS-T extracts of brain homogenate were nonsignificantly reduced in APP;C3^−/− mice by Western blot compared with APP mice. Scale bar, 100 μm.