Fig. 5
Microglia quantification and profiling in the MSA mice. a Sterological analysis of the number Iba-1-positive cells demonstrated no significant differences between control and MSA mice in substantia nigra after Bonferroni correction (two way ANOVA indicates a general effect of age and genotype: effect of genotype F1,23 = 6.12, p = 0.0211, effect of age F2,23 = 3.61, p = 0.0432, interaction F2,23 = 0.15, p = 0.8599). No significant differences between controls and MSA mice were detected after Bonferroni correction in the striatum (two way ANOVA indicates no general effect of age and genotype: effect of genotype F1,23 = 0.08, p = 0.7812, effect of age F2,23 = 3.04, p = 0.0671, interaction F2,23 = 0.08, p = 0.7812). In the pontine nuclei, age-related increase in the number of Iba-1-positive cells was seen in the control, but not in the MSA group after Bonferroni correction (two way ANOVA indicates a general effect of age: effect of genotype F1,21 = 1.86, p = 0.1866, effect of age F2,21 = 8.83, p = 0.0017, interaction F2,21 = 3.37, p = 0.0537). Similar effects were found in the inferior olives; however, at 15 months of age a significant difference between MSA and control mice was found after Bonferroni correction (two way ANOVA indicates a general effect of age and genotype: effect of genotype F1,22 = 9.27, p = 0.0059, effect of age F2,21 = 30.17, p < 0.0001, interaction F2,21 = 27.05, p < 0.0001) (b) Four microglia morphological profiles were distinguished (Iba1 immunohistochemistry and cresyl-violet counterstaining): Type A cells (corresponding to surveillant/homeostatic microglia, and the most abundant phenotype), characterized by a compact and regular shaped nucleus, very thin visible cytoplasm, and long and thin processes without many secondary branches. Type B cells (hyperramified) presented with ramified processes with multiple short branches, and increased amount of cytoplasm around the nucleus as compared to type A. Type C microglia (hyperthrophic) showed even larger cell body with more irregular outline, enlarged nucleus and shorter and thicker processes. Finally, type D cells (amoeboid), which resembled peripheral macrophages, were identified by their amoeboid shape, with a large nucleus and the soma merging with the processes. c The stereological counts of each microglia subtype are represented in percentages of the total Iba1+ cells counted. In the substantia nigra, the non-homeostatic activated microglia profiles (type B, C and D) were significantly more abundant in the MSA mice than in the control ones at 5 and 15 months of age. Homeostatic microglia (type A) showed a significant reduction from 2 to 5 months and from 2 to 15 months of age in the transgenic animals, while activated microglia were significantly increased from 2 to 15 months of age. In the control mice, a reduction in A cells was seen at 15 months of age (two way ANOVA indicates a general effect of age and genotype: effect of genotype F1,23 = 51.42, p < 0.0001, effect of age F2,23 = 29.42, p < 0.0001, interaction F2,23 = 19.53, p < 0.0001); (d) In the striatum, an age-related decrease in A cells, accompanied by an increase in B cells, was observed in both mouse lines. No significant differences were detected between the two at any time point, except for an increase in the percentage of D cells in the MSA mice at 15 months of age (two way ANOVA indicates a general effect of age and genotype without interaction: effect of genotype F1,23 = 8.29, p = 0.0085, effect of age F2,23 = 72.14, p < 0.0001, interaction F2,23 = 1.1, p = 0.35); (e) In the pontine nuclei, as well as in the inferior olives (f), an increase of activated microglia was visible at 15 months of age, if compared with earlier stages, in both transgenic and control mice (two way ANOVA for the pontine nuclei indicates a general effect of age: effect of genotype F1,21 = 2.65, p = 0.1184, effect of age F2,21 = 3.61, p < 0.0001, interaction F2,21 = 1.27, p = 0.3008; two way ANOVA for the inferior olives indicates a general effect of age: effect of genotype F1,21 = 0.03, p = 0.8731, effect of age F2,21 = 15.43, p < 0.0001, interaction F2,21 = 6.11, p = 0.0081)