Microglia activation in postmortem brains with schizophrenia demonstrates distinct morphological changes between brain regions

Abstract

Schizophrenia (SCZ) is a psychiatric disorder that can include symptoms of disorganized speech and thoughts with uncertain underlying mechanisms possibly linked to over-activated microglia. In this study, we used brain samples from sixteen donors with SCZ and thirteen control donors to assess the differential activation of microglia by quantifying density and 3D reconstruction of microglia stained with ionized calcium-binding adaptor molecule-1 (Iba1). Our samples consisted of sections from the frontal, temporal, and cingulate cortical gray matter, subcortical white matter regions (SCWM), and included the anterior corpus callosum. In the first series of studies, we performed a density analysis followed by a spatial analysis to ascertain the microglial density, distribution, and soma size in SCZ brains. Second, we performed a series of morphological quantification techniques to investigate the arborization patterns of the microglia in SCZ. The results demonstrated an increase in microglia density in the cortical gray matter regions in SCZ cases, while in the SCWM, there was a significant increase in microglia density in the frontal and temporal, but not in the other brain regions of interest (ROIs). Spatial analysis using the "nearest neighbor" demonstrated that there was no effect in "clustering", but there were shorter distances between microglia seen in the SCZ cases. The morphological measures showed that there was a region-dependent increase in the microglia soma size in the SCZ cases while the Sholl analysis revealed a significant decrease in the microglia arborization in the SCZ cases across all the ROI's studied. An in-depth 3D reconstruction of microglia in Brodmann area 9 cortical region found that there was a significant association between age and reduced microglial arborization in the SCZ cases. This region-dependent age association can help determine whether longitudinal changes in microglial activation across age are brain region-dependent, which may point to potential therapeutic targets.

Keywords: microglia; morphology; neuroinflammation; postmortem; schizophrenia; spatial analysis.

FIGURE 1

The global view and the strategy for the quantification of the microglia density outlined in the hematoxylin and eosin stain. The regions of interest included the dorsal lateral frontal cortex (BA9; A), the superior temporal cortex (BA22; B), and the anterior cingulate cortex (BA32; C). The areas that were analyzed are represented by dotted lines which show the contours made in each cortical region, and the boxes represent the adjacent subcortical white matter (SCWM) regions and corpus callosum region. Cortical thickness (D) was measured in images A–C. BA, Brodmann area; Scale bar = 3mm

FIGURE 2

Photomicrographs of the ionized calcium‐binding adaptor molecule 1 (Iba1) immunoreactivity in the control cases (A, D, and G) and in the schizophrenia (SCZ) cases (B, E, and H). Graphs representing the density of microglia per mm³ are seen in (C, F, and I). The asterisks indicate the significant differences in the density of microglia counted in the SCZ cases compared to the control in the dorsal lateral frontal cortex (BA9; C), the superior temporal cortex (BA22; F), and the anterior cingulate cortex (BA32; I). BA, Brodmann area; Scale bar = 20 µm. **p < 0.01, ***p < 0.001

FIGURE 3

Examples of Iba1 immunoreactivity in the control cases (A, D, G, and H), in the schizophrenia (SCZ) cases (B, E, I, and J), in the subcortical white matter (SCWM) regions, and in the corpus callosum. Graphs representing the density of microglia per mm³ are seen in (C, F, and K). The asterisks indicate the significant differences in the density of microglia counted in the SCZ cases compared to the control in the dorsal lateral frontal SCWM (BA9; C), the superior temporal SCWM (BA22; F), and anterior cingulate SCWM and corpus callosum (BA32; K). *p < 0.05, **p < 0.01; BA, Brodmann Area; CC, corpus callosum; Scale bar = 20 µm

FIGURE 4

Photomicrographs of cortical Iba1 positive cells used to assess the general morphological differences in microglia in the control cases (A, D, and G) and in the schizophrenia (SCZ) cases (B, E, and H). Inserted above the individual radii in the Sholl plot are asterisks that indicate the significant differences in the microglia branching ramifications in the SCZ cases compared to the control in the frontal cortex (BA9; C) superior temporal cortex (BA22; F) and anterior cingulate cortex (BA32; G). *p < 0.05; BA, Brodmann area; Scale bar = 10 µm

FIGURE 5

High‐resolution images of Iba1 immunohistochemistry from the subcortical white matter (SCWM) were used to assess the general morphological differences in microglia in the control cases (A, D, and G) and in the schizophrenia (SCZ) cases (B, E, and H). In the Sholl plot, the asterisks indicate the significant differences seen in the microglia branching ramification at the individual radii in the SCZ cases compared to the control cases in the frontal SCWM (BA9; C) superior temporal SCWM (BA22; F) and anterior cingulate SCWM (BA32; G). *p < 0.05; BA=Brodmann Area; Scale bar = 10 µm (A, B); Scale bar = 5 µm (C, D, F, G)

FIGURE 6

Examples of 3D reconstructed microglia in cortical gray matter in control brain (A) and SCZ brain (E). Microglia morphology indicated less ramification by having a shorter process lengths and smaller numbers of processes in cortical gray matter (B, C) in SCZ group compared with controls. Branching pattern alterations of microglia in cortical gray matter indicated the number of branching intersections was significantly lower in SCZ cases versus the control group (D) at various distances away from the cell soma. Age‐association testing demonstrated a significant negative correlation between ramification of microglia and age of SCZ cases (F, G), **p < 0.01, ***p < 0.001

FIGURE 7

Examples of 3D reconstructed microglia in SCWM in control brain (A) and SCZ brain (E). Microglia morphology indicated less ramification by having shorter and smaller number of processes in SCWM in the SCZ group compared with controls (B, C). Branching pattern alterations of microglia in SCWM indicated the number of branching intersections at various distances away from the cell soma was significantly lower in SCZ cases versus the control group (D). Age‐associated testing demonstrated found no significant correlation between ramification of microglia in SCWM and age of SCZ cases (F, G), *p < 0.05, **p < 0.01, ***p < 0.001