Disturbance of oligodendrocyte function plays a key role in the pathogenesis of schizophrenia and major depressive disorder

Abstract

The major psychiatric disorders such as schizophrenia (SZ) and major depressive disorder (MDD) are thought to be multifactorial diseases related to both genetic and environmental factors. However, the genes responsible and the molecular mechanisms underlying the pathogenesis of SZ and MDD remain unclear. We previously reported that abnormalities of disrupted-in-Schizophrenia-1 (DISC1) and DISC1 binding zinc finger (DBZ) might cause major psychiatric disorders such as SZ. Interestingly, both DISC and DBZ have been further detected in oligodendrocytes and implicated in regulating oligodendrocyte differentiation. DISC1 negatively regulates the differentiation of oligodendrocytes, whereas DBZ plays a positive regulatory role in oligodendrocyte differentiation. We have reported that repeated stressful events, one of the major risk factors of MDD, can induce sustained upregulation of plasma corticosterone levels and serum/glucocorticoid regulated kinase 1 (Sgk1) mRNA expression in oligodendrocytes. Repeated stressful events can also activate the SGK1 cascade and cause excess arborization of oligodendrocyte processes, which is thought to be related to depressive-like symptoms. In this review, we discuss the expression of DISC1, DBZ, and SGK1 in oligodendrocytes, their roles in the regulation of oligodendrocyte function, possible interactions of DISC1 and DBZ in relation to SZ, and the activation of the SGK1 signaling cascade in relation to MDD.

Figure 1

DISC1 and DBZ mRNA is expressed in oligodendrocytes in the corpus callosum (CC) of the mouse brain. (a) Real-time PCR analysis of DISC1 in CC at the indicated stages. Data are expressed from at least three independent experiments. DISC1 mRNA levels at P7 were higher than at later time points. P < 0.01, as assessed via a one-way ANOVA followed by Tukey's test. (b) Primary cultured rat OPCs were harvested at indicated times after platelet-derived growth (PDGF) deprivation. mRNA was quantified by qRT-PCR. Data are expressed from at least three independent experiments. DISC1 mRNA levels at 0 h were higher than at later time points. P < 0.01, as assessed via a one-way ANOVA followed by Tukey's test. (c) Postnatal development of DBZ mRNA in the CC of mice from P0, P7, P14, P21, and adults. DBZ mRNA-expressing cells were first identified at P7. The number and intensity of the reaction of DBZ-positive cells reached to peak at P14, followed by its progressive reduction until adulthood. Scale bar: 50 μm. (d) Quantification of the proportion of DBZ+/PDGFRα+ and DBZ+/CCI+ cells. Double labeling with in situ hybridization and immunohistochemical analysis of brain sections from P14 mice with the antisense RNA probe to DBZ and antibodies against Olig2, PDGFRα, and CC1, respectively. (e) Western blot analysis of DBZ and MBP in the CC at the indicated stages. DBZ protein expression peaked at P14 before MBP expression. (b) is adapted with permission from Hattori et al. [20] and (c)–(e) are adapted with permission from Shimizu et al. [19].

Figure 2

DISC1 overexpression inhibits oligodendrocyte differentiation. (a) Primary cultured rat OPCs infected with GFP-Adv or DISC1-GFP-Adv were lysed at 0, 24, 48, 72, 96, and 120 hours after PDGF deprivation and subjected to western blot analysis. (b) Primary cultured rat OPCs were immunostained with anti-GFP and anti-CNPase antibodies. The percentage of CNPase positive cells relative to the total number of GFP positive cells is shown. Infected cells from three experiments were analyzed. ^* P < 0.05 versus GFP-Adv. Scale bar = 100 μm. (c) Primary cultured rat OPCs were immunostained with anti-GFP and anti-β-tubulin antibodies for morphological observation. Infected cells from three independent cultures were classified according to their morphology (simple, intermediate, or complex) and quantified. The percentage of cells within each category, relative to the total number of GFP positive cells, is shown. ^* P < 0.05 versus GFP-Adv. Scale bar = 50 μm. (a)–(c) are adapted with permission from Hattori et al. [20].

Figure 3

DISC1 knockdown promotes oligodendrocyte differentiation. (a) Primary cultured rat OPCs transfected with control-siRNA, DISC1-siRNA-1, or DISC1-siRNA-2 were lysed 72 hours after siRNA transfection and analyzed by western blotting. (b, d) Primary cultured rat OPCs were transfected with control-siRNA, DISC1-siRNA-1, or DISC1-siRNA-2 and cultured in medium containing PDGF for 72 hours then fixed for immunostaining. Primary cultured rat OPCs were immunostained with anti-CNPase-antibody (b) or anti-β-tubulin antibody (d) and analyzed as described here. ^* P < 0.05 versus control-siRNA. Scale bars = 50 μm. (c) DISC1 knockdown mediated increase of CNPase mRNA was rescued by overexpression of DISC1. Primary cultured rat OPCs were infected with GFP- or DISC1-GFP-Adv 24 hours after control- or DISC1-siRNA transfection. Forty-eight hours after the infection, mRNA level of CNPase was quantified by qRT-PCR. Data are expressed as mean ± SEM of at least three independent experiments. ^* P < 0.05 versus control-siRNA and GFP-Adv. (a)–(d) are adapted with permission from Hattori et al. [20].

Figure 4

Myelination in the CC of DBZ-KO mice is delayed during the postnatal period but recovers by adulthood. (a–d) In situ hybridization analyses of WT and DBZ-KO littermates at P10 (a, c) or adulthood (b, d) with the antisense RNA probes to DBZ (a, b) and Mbp (c, d) mRNA. The number of cells in the CC of DBZ-KO mice expressing Mbp mRNA was significantly reduced at P10 compared with age-matched WT mice, but it was recovered to WT levels by adulthood. Cg: cingulate gyrus, Mo: motor area. (e, f) Electron microscopic analyses of the CC in P10 (e) or adult (f). (g) Western blot analyses of MBP protein in the CC of WT or DBZ-KO mice at P10 or adulthood. (h) Scatter plot of g-ratio values in the CC of WT or DBZ-KO mice at adulthood. Four hundred and four axons from four animals were analyzed for each group. Scale bar: (a–d) 50 μm, (e, f) 5 μm. (a)–(h) are adapted with permission from Shimizu et al. [19].

Figure 5

Immature oligodendrocytes are increased in DBZ-KO mouse. (a, b) Immunohistochemistry with antibodies against PDGFRα and CC1 on the CC sections from WT and DBZ-KO littermates at P10 or adult. Quantification of the proportion of PDGFRα+ and CC1+ cells at P10 (a) and adulthood (b). (c) Chromatin density was calculated as the ratio of electron dense chromatin area to the cell nucleus area using ImageJ software. One hundred and forty oligodendrocytes from four animals were analyzed for each group. ^* P < 0.0001 by Student's t-test following two-way-ANOVA. (a)–(c) are adapted with permission from Shimizu et al. [19].

Figure 6

DBZ is transiently upregulated in vitro during oligodendrocyte differentiation before myelin marker expression. (a) Primary OPC were isolated from rat postnatal cortex and differentiated by mitogen withdrawal. Total RNA was prepared from the cells 0, 2, 4, 6, or 8 days after mitogen withdrawal. DBZ mRNA expression was evaluated using qRT-PCR, with normalization to GAPDH. Data normalized to either day 0. ^* P < 0.05, ^** P < 0.01 versus day 0 or +P < 0.05, and ++P < 0.01 versus day 2 by Dunnett's test following one-way ANOVA, n = 2–4. (b) Western blot analyses of DBZ, MBP and CNPase from cells 0, 2, 4, 6, or 8 days after mitogen withdrawal. (a) and (b) are adapted with permission from Shimizu et al. [19].

Figure 7

DBZ knockdown results in an in vitro delay in oligodendrocyte differentiation. Primary cultures of OPC were transfected with DBZ-5i or CONT-si 24 h before deprivation of PDGF and bFGF to induce differentiation. (a) Lysates from primary cultured OPCs were prepared from cells harvested 48 h after mitogen deprivation and expression of DBZ, MBP, and CNPase was assessed by western blotting. (b) Primary cultured OPCs were immunostained with PDGFRα and O4 antibodies 48 h after mitogen deprivation. Quantification of the proportion of PDGFRα+/O4−, PDGFRα+/O4+, and PDGFRα−/O4+ cells. (^* P < 0.05, ^** P < 0.01 versus WT, n = 3) (c) Primary cultured OPCs were immunostained with O4 and MBP antibodies 96 h after mitogen deprivation. Quantification of the proportion of O4+/MBP−, O4+/MBP+, and O4−/MBP+ cells f (^** P < 0.01 versus WT, n = 3). For (b) and (c), more than 400 cells in total from three independent cultures were counted. (a)–(c) are adapted with permission from Shimizu et al. [19].

Figure 8

Effect of DBZ knockdown on oligodendrocytes cell morphology. Morphological features of primary cultured oligodendrocytes treated with CONT-si or DBZ-5i as determined by light microscopy (a, b) or electron microscopy (c–k). Micrographs of CONT-si-treated oligodendrocytes (a, c–g) or DBZ-5i-treated oligodendrocytes (b, h–k). CONT-si-treated oligodendrocytes had morphological characteristics of Type 2 or Type 3 cells corresponding to immature oligodendrocytes or mature oligodendrocytes, respectively, whereas DBZ-5i-treated oligodendrocytes were largely characterized by Type 1 corresponding to OPCs. Scale bar: (a, b) 50 μm, (c) 10 μm, (d, f, g, h) 5 μm, (e) 0.1 μm, (i, j) 1 μm, (k) 0.1 μm. (a)–(k) are adapted with permission from Shimizu et al. [19].

Figure 9

Interaction between DISC1 and DBZ on the oligodendrocytes development.

Figure 10

Involvement of Sox10 and/or Nkx2.2 in the regulatory pathway of oligodendrocyte differentiation by DISC1. (a, b) Expression of Sox10 and Nkx2.2 mRNA were decreased by DISC1 overexpression. Oligodendrocyte precursor cells were infected with GFP-Adv or DISC1-GFP-Adv for 12 hours and induced to differentiate by depriving PDGF for 36 hours. (c, d) Expression of Sox10 and Nkx2.2 mRNA were increased by DISC1 knockdown. Oligodendrocyte precursor cells were transfected with control-siRNA, DISC1-siRNA-1 or DISC1-siRNA-2 and cultured in medium containing PDGF for 48 hours. (e, f) Expression of Sox10 and Nkx2.2 mRNA were increased by truncated DISC1 overexpression. Oligodendrocyte precursor cells were infected with GFP-Adv or trDISC1-GFP-Adv for 12 hours and induced to differentiate by PDGF deprivation for 36 hours. (g–j) DISC1 knockdown mediated increase of CNPase mRNA was inhibited by a simultaneous knockdown of either Sox10 or Nkx2.2. Oligodendrocyte precursor cells were co-transfected with control-siRNA or DISC1-siRNA-1 and sox10-siRNA or nkx2.2-siRNA and cultured in medium containing PDGF for 24 (h) or 48 hours (g, i, j). mRNA quantification was performed 48 hours after adenovirus infection (a, b, e, f) or siRNA transfection (c, d, g, i, j) or 24 h hours after siRNA transfection (h) by qRT-PCR. Data are expressed as mean ± SEM of at least three independent experiments. ^* P < 0.05 versus GFP-Adv (a, b), ^* P < 0.05 versus control-siRNA (c, d), and ^* P < 0.05 (e–h). (a)–(j) are adapted with permission from Hattori et al. [20].

Figure 11

DBZ knockdown alters expression of transcription factors involved in OL differentiation. Primary cultures of OPCs were transfected with DBZ-5i or CONT-si 24 h before PDGF and bFGF deprivation to induce differentiation. Total RNA was prepared 24 h after mitogen deprivation. (a, b) The effect of DBZ knockdown on the expression of well-characterized transcription factors was examined using qRT-PCR. (a) Expression of the transcription factor negatively regulating oligodendrocyte differentiation. (b) Expression of the transcription factor positively regulating oligodendrocyte differentiation. ^** P < 0.01; ^* P < 0.05 versus CONT-si (n = 3). (a) and (b) are adapted with permission from Shimizu et al. [19].

Figure 12

Repeated exposure to WIRS (chronic stress) upregulated SGK1 and SGK1 is activated by PDK1. (a, b) In situ hybridization images of Sgk1 mRNA. Dark-field photomicrographs show the upregulation of Sgk1 mRNA expression in the fiber tracts after repeated exposure to WIRS ((a) controls; (b) stress). cc, corpus callosum; ac, anterior commissure. Scale bar = 5 mm. (c) Alternation of plasma corticosterone 24 h after repeated WIRS. The results are expressed as the mean ± SEM of three independent experiments. ^* P < 0.05, t-test. (d, e) The CC region images of Sgk1 mRNA in situ hybridization ((d) controls; (e) stress), respectively. Scale bar = 100 μm. (f, g) Merged images of Nissl staining and in situ hybridization images of Sgk1 mRNA. The distribution of cells expressing Sgk1 mRNA in the CC of the control (d) and repeated WIRS-exposed mice (e) in bright-filed photomicrographs. Positive grains were concentrated in the oligodendrocytes. Scale bars = 50 μm. (h, i) Western blot analysis shows SGK1 protein, its phosphorylation at positions T-256 (SGK1-256T-P) and S-422 (SGK1-422S-P), and the phosphorylation of PDK1 at position S-241 (PDK-241S-P) in the oligodendrocytes of the CC after repeated exposure to WIRS. (a)–(i) are adapted with permission from Miyata et al. [37].

Figure 13

Repeated exposure to WIRS upregulates NDRG1 phosphorylation in the fiber tracts via SGK1 activation. (a) In situ hybridization images of Ndrg1 mRNA. Dark-field photomicrographs show the distribution of Ndrg1 mRNA-expressing cells in the mouse brain on the sagittal plane. Sections were hybridized with ³⁵S-labeled antisense RNA probe for Ndrg1 mRNA. As controls, adjacent sections were hybridized with ³⁵S-labeled sense RNA probe (inset). Scale bar = 5 mm. (b) Immunoprecipitation and western blot analysis show that repeated exposure to WIRS elevated the interaction between SGK1 and NDRG1 (second column). However, NDRG1 expression did not increase in the CC (first column). (c) Western blot analysis shows that repeated exposure to WIRS elevated phosphorylated NDRG1 levels in the CC. (a)–(c) are adapted with permission from Miyata et al. [37].

Figure 14

Repeated exposure to WIRS upregulates adhesion molecules expression levels in oligodendrocytes of the fiber tracts. (a) Immunoprecipitation and western blot analysis show that repeated exposure to WIRS elevated the interaction between NDRG1 and β-catenin (second panel) and that the expression levels of β-catenin, N-cadherin, and α-catenin were elevated in the corpus callosum. (b) Immunohistochemical analysis of β-catenin in the CC demonstrates increased labeling of the processes of the oligodendrocytes (i.e., greater number and intensity) in mice exposed to repeated WIRS. Scale bar = 50 μm. (c) Immunocytochemical analysis of β-catenin in SK-N-SH cells overexpressing nonphosphorylated (NP-NDRG1) or constitutively phosphorylated NDRG1 (CP-NDRG1). Increased expression of β-catenin due to the overexpression of the CP-NDRG1 was mostly observed on the surfaces of the cell bodies and the processes of SK-N-SH cells. Scale bar = 20 μm. (a)–(c) are adapted with permission from Miyata et al. [37].

Figure 15

Repeated exposure to WIRS causes morphological alterations in oligodendrocytes in the CC. (a) Representative transverse electron micrographs of the CC from control (upper panel of (a)) and repeated WIRS-exposed mice (lower panels of (a)). Scale bar = 2 μm. (b) Results of the quantification of the sum of oligodendrocytes in the cross-sectional area. The results are expressed as the mean ± SEM of 3 independent experiments. ^* P < 0.05, t-test. (c, d) The distributions of the axon diameters (c) and myelin thicknesses (d) of the CC in the control and repeated WIRS-exposed mice were assessed. The results are expressed as the mean ± SEM of 3 independent experiments. ^* P < 0.05, t-test. (a)–(d) are adapted with permission from Miyata et al. [37].

Figure 16

Activation of the SGK1-NDRG1 pathway also causes morphological changes in primary cultured oligodendrocytes. (a) Morphology of primary cultured oligodendrocytes treated with (+DEX) ((a) right column) or without DEX (−DEX) for 2 days. (b) Overexpression of the constitutively active form of SGK1 (CA-SGK1) or kinase inactive form of SGK1 (KI-SGK1). (a, b) Morphometric measurements of oligodendrocyte diameters were performed using ImageJ software. The results are expressed as the mean ± SEM of 3 independent experiments. ^* P < 0.05, t-test. (a) and (b) are adapted with permission from Miyata et al. [37].

Figure 17

Activation of the PDKI-SGKI-NDRG1-adhesion molecule pathway returns to the control level after 3 weeks of recovery. (a) Quantification of the sum of the cross-sectional areas of the oligodendrocytes is measured from transverse electron micrographs of the CC of no-stress controls (18-week-old mice) (Cont) and after 3 weeks of recovery after exposure to WIRS (18-week-old mice) (stress and recover). Morphometric measurements were made using ImageJ software. The results are expressed as the mean ± SEM of 3 independent experiments. ^* P < 0.05, t-test. (b) Western blot analysis of the CC of mice exposed to repeated WIRS (15-week-old mice) and after 3 weeks of recovery (18-week-old mice). (c) Effects of a chronic stress, that is, 3 weeks of recovery, on mouse behavior. After 3 weeks of recovery mice (18-week-old mice) were discontinued showing no significant difference in the tail-suspension test compared to the control mice (18-week-old mice). The results are expressed as the mean ± SEM of 3 independent experiments. ^* P < 0.05, t-test. (a)–(c) are adapted with permission from Miyata et al. [37].

Figure 18

Activation of HPA axis-PDK1-SGK1-NDRG1-adhesion molecules by repeated stress induces excess arborization of the oligodendrocyte processes.