Impaired αVβ8 and TGFβ signaling lead to microglial dysmaturation and neuromotor dysfunction

Abstract

Microglia play a pivotal role in the coordination of brain development and have emerged as a critical determinant in the progression of neurodegenerative diseases; however, the role of microglia in the onset and progression of neurodevelopmental disorders is less clear. Here we show that conditional deletion of αVβ8 from the central nervous system (Itgb8ΔCNS mice) blocks microglia in their normal stepwise development from immature precursors to mature microglia. These "dysmature" microglia appear to result from reduced TGFβ signaling during a critical perinatal window, are distinct from microglia with induced reduction in TGFβ signaling during adulthood, and directly cause a unique neurodevelopmental syndrome characterized by oligodendrocyte maturational arrest, interneuron loss, and spastic neuromotor dysfunction. Consistent with this, early (but not late) microglia depletion completely reverses this phenotype. Together, these data identify novel roles for αVβ8 and TGFβ signaling in coordinating microgliogenesis with brain development and implicate abnormally programmed microglia or their products in human neurodevelopmental disorders that share this neuropathology.

Graphical abstract

Figure 1.

αVβ8 and TGFβ signaling to microglia. (A) Immunostaining for pSMAD3 (red) in microglia (F4/80, green) reveals reduced pSMAD3 staining intensity (yellow arrows) in most microglia from Itgb8ΔCNS mice compared with intense staining (red arrows) in controls; quantified on right: pSMAD3 staining intensity within individual microglia (arbitrary units) documents reduced microglia-specific pSMAD3. See Fig. S1 for pSMAD staining intensity in other CNS cell types. (B) Transcriptional profiling (quantitative PCR) of sorted microglia documents alterations in several TGFβ-dependent genes, confirmed by immunostaining (C). See Fig. S2 for sorting strategy. Note expression of APOE in IBA1-negative astrocytes. (D) Increased number and proliferation (KI67⁺) and activated morphology in IBA1⁺PU.1⁺ microglia from Itgb8ΔCNS and Tgfbr2ΔMG mice; quantified on right (we did not observe KI67⁺IBA1⁺ cells in the cortex of control animals at this time point). (E) Developmental brain hemorrhage in Itgb8ΔCNS but not Tgfbr2ΔMG mice, as evidenced by gross examination and immunostaining for red blood cells (TER119, red) outside of blood vessels (CD31, blue). Bars, 50 µm (A); 100 µm (C–E). Error bars indicate SE. **P < 0.005; ***P < 0.0005. Student’s t test (A) or ANOVA with Tukey’s post hoc test (B and D). n = 4 animals (P60) for all groups.

Figure 2.

Motor deficits and glial abnormalities in mice with deficient αVβ8 or TGFβ signaling to microglia. (A) Identical neuromotor symptoms in adult (P30–60) Itgb8ΔCNS and Tgfbr2ΔMG mice including waddling unstable gait with shortened stride length, reduced time on rotarod, unkempt appearance, and tremors. See also Videos 1 and 2. (B) Persistent expression of APOE and KI67 and reduced expression of P2RY12 and TMEM119 over time in microglia from Itgb8ΔCNS and Tgfbr2ΔMG mice. See accompanying Fig. S4. (C) Overlapping astrogliosis, microgliosis, and reduced MBP staining in P30 Itgb8ΔCNS and Tgfbr2ΔMG mice compared with controls. (D) Increased percentage of GFAP⁺SOX9⁺ astrocytes in Itgb8ΔCNS and Tgfbr2ΔMG mice over time; quantified on right. See accompanying Fig. S4. (E) Increased percentage of OLIG2⁺NG2-DSR⁺ OPC, and reduced mature OLIG2⁺CC1⁺ oligodendrocytes over time (quantified on right) and reduced staining for mature myelin marker TMEM10 in Itgb8ΔCNS and Tgfbr2ΔMG mice compared with controls. See accompanying Fig. S4. Bars, 100 µm. Error bars indicate SE. *P < 0.05; **P < 0.005; ***P < 0.0005; ****P < 0.0001. Student’s t test. n = 4 animals for all groups. Behavioral analysis: ANOVA with Tukey’s post hoc test; n = 6 animals for all groups. a.u., arbitrary units.

Figure 3.

Interneuron abnormalities in mice with deficient αVβ8 or TGFβ signaling to microglia. (A) Delayed loss of Lhx6-GFP⁺ cortical interneurons in Tgfbr2ΔMG; quantification below. (B) Reduced PV expression in Lhx6-GFP⁺ interneurons at P15 and P30, before the reduction in the numbers of these cells; quantification below. See also Fig. S4. Bars, 100 µm. Error bars indicate SE. *P < 0.05; **P < 0.005; ***P < 0.0005. Student’s t test. n = 4 animals for all groups. MFI, mean fluorescence intensity.

Figure 4.

Presence of abnormal microglia (and not absence of mature microglia) drives neuromotor phenotype in Itgb8ΔCNS mice. (A–D) Itgb8ΔCNS pups and controls were identified at birth, randomized to receive either ICD or PLX5622, then evaluated for neuromotor symptoms (A) and associated neuropathology (B–D) 90 d later (P90). Compared with Itgb8ΔCNS mice treated with control drug and littermate control mice treated with either ICD or PLX5622, Itgb8ΔCNS mice treated with PLX5622 had normalization of oligodendrocyte abnormalities (B, MBP staining; C, Olig2/NG2/CC2 staining) and interneuron deficiencies (C, PV/SST staining), but persistence of reactive astrocytes (C, Sox9/GFAP staining). Note overlapping staining of GFAP and MBP in B, which shows normalization of MBP staining despite persistent GFAP staining in Itgb8ΔCNS mice treated with PLX5622. (D) Quantification of cellular phenotypes in Itgb8ΔCNS and control mice treated with either ICD or PLX5622. Bars, 100 µm. Error bars indicate SE. *P < 0.05; **P < 0.005; ***P < 0.0005. Behavioral analysis and cell counting: ANOVA with Tukey’s post hoc test; n = 4 animals. a.u., arbitrary units; PLX, PLX5622.

Figure 5.

Microglial dysmaturation in the absence of TGFβ signaling. (A) Volcano plots of differentially expressed genes (RNA sequencing) in microglia from Tgfbr2ΔMG versus litter mate controls at indicated time points (n = 3–4 per genotype). Green, P < 0.05 and log₂ fold change >1; red, P < 0.05; orange, log₂ fold change >1; black, P > 0.05. (B) Venn diagrams of overlaps of differentially expressed genes from E16 (yellow), P15 (blue), and P60 (pink) time points. (C) Heat map representation of differential expression Z statistics for indicated datasets. Genes are divided into immature, mature, and reactive groups based on similarity to published data (see Table S1).

Figure 6.

Microglial dysmaturation due to loss of TGFβ signaling is critically time dependent. (A) Diagram of Tgfbr2 gene inactivation and mutant analysis. Tamoxifen was injected at P1 for early induction in B or at P30 for adult induction in C. Mice were sacrificed and analyzed at indicated time points after tamoxifen injection. (B and C) Staining of cortical brain sections from Tgfbr2iΔMG and Cre⁺ control littermates 120 d after tamoxifen administration following early (B) or adult (C) induction. Left panels: tdTomato recombination reporter (tdTom, red) marks all microglia and immature (APOE, green) and mature (P2RY12, blue) microglia markers to identify type A (dysmature) or type B microglia, respectively. Right panels (insets from Fig. S5): pSMAD3 (green) coimmunostaining reveals reduction in pSMAD3 staining intensity in type A (dysmature) microglia compared with type B microglia and controls. Percent (%) recombination (upper right graphs in B and C) based on % F4/80⁺, CD45⁺, CD11b⁺ cells that are also tdTomato positive (cells isolated and analyzed by flow cytometry as in Fig. S3). Mean pSMAD3 per nucleus (lower right graphs in B and C) based on fluorescent intensity of individual recombined (tdTom⁺) microglia coexpressing P2RY12 (blue, control and type B cells), or lacking P2RY12 expression (type A cells). Bars, 100 µm or 25 µm (pSMAD3, right panels). Error bars indicate SE. *P < 0.05; ***P < 0.005; ****P < 0.0005. ANOVA with Tukey’s post hoc test (B); Student’s t test (C); n = 4 animals for each group.