Loss and microglia phagocytosis of synaptic proteins in frontotemporal lobar degeneration with TDP-43 proteinopathy

Abstract

Cortical synaptic loss has emerged as an early abnormality in Alzheimer's disease (AD) with a strong relationship to cognitive performance. However, the status of synapses in frontotemporal lobar degeneration (FTLD) has received meager experimental attention. The purpose of this study was to investigate changes in cortical synaptic proteins in FTLD with tar DNA binding protein-43 (TDP-43) proteinopathy. A second aim was to study phagocytosis of synaptic proteins by microglia as a surrogate for synaptic pruning. Western blot analysis in frozen tissue from the middle frontal gyrus revealed decreased levels of the presynaptic protein synaptophysin, but slightly increased levels of the postsynaptic density protein 95 (PSD95) in FTLD-TDP. Levels of the dendritic spine protein spinophilin displayed the largest decrease. Double immunofluorescent staining visualized aggregate or punctate synaptic protein immunoreactivity in microglia. Overall, the proportion of microglia containing synaptic proteins was larger in FTLD-TDP when compared with normal controls. The increase in PSD95 levels may represent reactive upregulation of this protein, as suggested in AD. While greater numbers of microglia containing synaptic proteins is consistent with loss of synapses in FTLD-TDP, it may also be an indication of abnormal synaptic pruning by microglia.

Keywords: Dendritic spines; FTLD-TDP; Microglia phagocytosis; Microglia synaptic pruning; Synaptic proteins.

Figure 1.

Western blot bands of synaptophysin immunoreactivity in the middle frontal gyrus (left) of normal control and FTLD-TDP participants, with corresponding GAPDH loading control bands. GAPDH bands were imaged after probing the membrane with a monoclonal human GAPDH antibody following stripping the membrane of synaptophysin antibody. Each individual tissue lysate was run in duplicate in adjacent lanes. Optical density obtained from duplicate bands were averaged and used as a data point for analysis. Positions of molecular markers are shown on the left of the blot. Arrows point to synaptophysin and GAPDH bands at 37 and 36 KDA respectively. The control image on the left is from a different blot than the other three. Note the lower intensity of immunoreactivity in the FTLD-TDP participants. (Right) Quantitation of Western blot bands representatives of which are shown on the left. Synaptophysin immunoreactivity in FTLD-TDP participants was 15% lower when compared with normal controls, which was nearly significant (* p = 0.059).

Figure 2.

Western blot bands of PSD95 immunoreactivity in the middle frontal gyrus (left) of normal control and FTLD-TDP participants, with corresponding GAPDH loading control bands. GAPDH bands were imaged after probing the membrane with a monoclonal human GAPDH antibody following stripping the membrane of PSD95 antibody. Each individual tissue lysate was run in duplicate in adjacent lanes. Optical density obtained from duplicate bands were averaged and used for each lysate. Positions of molecular markers are shown on the left of the blot. Arrows point to PSD95 and GAPDH bands at 95 and 36 KDA respectively. The control image on the left is from a different blot than the other three. (Right) Quantitation of Western blot bands representatives of which are shown on the left. PSD95 immunoreactivity in FTLD-TDP participants displayed an 11% increase when compared with normal controls, which was not statistically significant (* p > 0.05).

Figure 3.

Western blot bands of spinophilin immunoreactivity in the middle frontal gyrus (left) of normal control and FTLD-TDP participants, with corresponding GAPDH loading control bands. GAPDH bands were imaged after probing the membrane with a monoclonal human GAPDH antibody following stripping the membrane of spinophilin antibody. Each individual tissue lysate was run in duplicate in adjacent lanes. Optical density obtained from duplicate bands were averaged and used for each lysate. Positions of molecular markers are shown on the left of the blot. Arrows point to spinophilin and GAPDH bands at 130 and 36 KDA respectively. The control image on the left is from a different blot than the other three. Note lower intensity of immunoreactivity in the FTLD-TDP participants. (Right) Quantitation of Western blot bands representatives of which are shown on the left. Spinophilin immunoreactivity in FTLD-TDP participants was 31% lower when compared with normal controls, which was statistically significant (* p < 0.05).

Figure 4.

Fluorescent double immunohistochemistry in middle frontal gyrus demonstrating three subsets of microglia, with aggregates (A, D, and G), punctate (B, E, and H) or no immunoreactivity (C, F, and I) for spinophilin (A, B, and C), synaptophysin (D, E, and F), and PSD95 (G, H, and I). Microglia are visualized by red fluorescence and synaptic proteins by green fluorescence.

Figure 5.

Bar graphs showing significantly higher proportion of microglia in the middle frontal gyrus of both control and FTLD-TDP participants with aggregates of immunoreactivity for each synaptic protein when compared with the proportion with punctate synaptic protein immunoreactivity. * p<0.05, ** p<0.01, *** p<0.002.

Figure 6.

Bar graphs demonstrating higher proportion of microglia containing aggregates or punctate immunoreactivity for each synaptic protein in FTLD-TDP participants when compared with controls. This difference near significance for aggregates of synaptophysin (p = 0.06).

Figure 7.

When the proportion of microglia containing aggregates and punctate synaptic protein immunoreactivity were combined, a higher proportion of microglia contained immunoreactivity for each synaptic protein in FTLD-TDP participants when compared with controls, and this difference was virtually significant for synaptophysin (p = 0.05).