Interaction of mTOR and Erk1/2 signaling to regulate oligodendrocyte differentiation

Abstract

A multitude of factors regulate oligodendrocyte differentiation and remyelination, and to elucidate the mechanisms underlying this process, we analyzed the interactions of known signaling pathways involved in these processes. Previous work from our lab and others shows that Akt, mTOR, and Erk 1/2 are major signaling pathways regulating oligodendrocyte differentiation and myelination in vitro and in vivo. However, the relative contribution of the different pathways has been difficult to establish because the impact of inhibiting one pathway in in vitro cell culture models or in vivo may alter signaling through the other pathway. These studies were undertaken to clarify the interactions between these major pathways and understand more specifically the crosstalk between them. Oligodendrocyte differentiation in vitro required Akt, mTOR, and Erk 1/2 signaling, as inhibition of Akt, mTOR, or Erk 1/2 resulted in a significant decrease of myelin basic protein mRNA and protein expression. Interestingly, while inhibition of the Erk1/2 pathway had little impact on Akt/mTOR signaling, inhibition of the Akt/mTOR pathways significantly increased Erk1/2 signaling, although not enough to overcome the loss of Akt/mTOR signaling in the regulation of oligodendrocyte differentiation. Furthermore, such crosstalk was also noted in an in vivo context, after mTOR inhibition by rapamycin treatment of perinatal pups. GLIA 2014;62:2096-2109.

Keywords: Akt; Erk1/2; mTOR; myelin basic protein; oligodendrocyte.

Figure 1. Activation of the Akt/mTOR and Erk 1/2 pathways increased as OPCs differentiated in vitro

(A) Representative western blot analysis of components of the Akt/mTOR and Erk 1/2 pathways during oligodendrocyte differentiation. OPCs derived from rat mixed glia were treated with T3 for 10 min, 30 min, 60 min, 3 hr, 6 hr, 1d, 2d, 3d, 4d, or 5d, collected and analyzed for phosphorylated or total Akt, p-mTOR S2448, S6RP, IRS1 (S318, S307, S612 and total) and phosphorylated or total Erk 1/2. MBP was quantified as a measure of oligodendrocyte differentiation, and GAPDH as the loading control. p-Akt S473 was increased while p-Akt T308 and total Akt showed little change. p-mTOR and p-S6RP expression remained constant as oligodendrocytes differentiated. p-IRS S318, S307 and total IRS1 decreased throughout differentiation, while pIRS1 S612 was only minimally detected. p-Erk1/2 decreased between 30 min and 6 h of T3 treatment, but increased from 1d until 5d; total Erk 1/2 expression remained consistent throughout differentiation. (B) Quantification of changes in phosphorylated states of Akt, Erk 1/2, mTOR and S6RP, measured as fold change relative to phosphorylation at day 0. MBP expression also significantly increased from 1-5d post mitogen withdrawal (inset, top right graph). n = 3, error bars are graphed as SEM. *p <0.05, **p <0.005, ***p <0.0005, ****p < 0.00005 and *****p < 0.000005 (One way ANOVA).

Figure 2. Pharmacological inhibition of Akt/mTOR or Erk1/2 decreased OPC differentiation in vitro

(A) Rat OPCs were incubated with pharmacological inhibitors of the Akt/mTOR and Erk 1/2 pathways, and analyzed for MBP expression. Cells were treated with DMSO or specific inhibitors (U0126 [20 μM], LY294002 [5 μM, 10 μM or 25 μM], Rapamycin [1 μM] and Torin1[10 nM, 100 nM or 250 nM]) as they were switched to T3. Five days later, MBP expression was quantified as a measure of differentiation; β-tubulin and Olig2 were the general and the oligodendrocyte-specific loading controls, respectively. (B) Quantification of MBP protein expression normalized to a DMSO control. All inhibitor treatments resulted in a significant reduction in MBP expression. (C) Antigenic markers used to identify each stage of the OPC lineage during differentiation. (D) Rat OPCs were treated with either DMSO or specific inhibitors at induction of differentiation with T3. Cells were fixed after 1-, 3- or 5d of differentiation and labeled with O4, O1 or MBP (all in green). Olig2 (red) was used to label all oligodendrocyte lineage cells. (E) The percentage of co-labeled O4-positive, O1-positive or MBP-positive/Olig2-positive cells (U0126:Red; Rapamycin: Purple; Torin1:Green). Scale bars = 200 μm. N = 3 and graphed +/- SEM (B, E). *p < 0. 05, **p < 0.005, ***p < 0.0005 (one way ANOVA).

Figure 3. The Akt/mTOR and Erk 1/2 pathways regulate distinct stages of oligodendrocyte differentiation in vitro

Representative western blots of MBP expression in rat OPCs shifted to differentiation media (A) with or without LY294002 (25 μM) or U0126 (20 μM), or (C) with or without rapamycin (1 uM) or Torin1 (250 nM) over a 5 day time course. Drugs were added for the full time course (+5d), after 1d in T3 (-1d+4d), after 2d in T3 (-2d+3d), after 3d in T3 (-3d+2d) or after 4d in T3 (-4d+1d). β-tubulin was the loading control. (B, D) Quantification of western blot analysis for MBP protein with treatments (B:LY294002, gray bars; U0126, white bars; D:Rapamycin, gray bars; Torin1, white bars) standardized to a DMSO control. Data represent the mean ± SEM from three independent experiments. *p < 0.05, **p < 0.005, ***p < 0.0005 (One way ANOVA).

Figure 4. Inhibition of the Akt/mTOR or Erk 1/2 pathway did not result in bidirectional crosstalk

Single and combinatorial treatments of inhibitors were added to rat OPCs upon the addition of T3 and cells were collected 1 day post treatment. (A) Representative western blot analysis of phosphorylated and total Erk 1/2, Akt and S6RP expression levels. Increased pErk1/2 was observed after 24 h treatment with LY294002, Rapamycin or Torin1 (red asterisks). Combined treatment with Rapamycin and LY294002 reduced Rapamycin-mediated Erk1/2 activation (green asterisk); in contrast, LY294002 and Torin1 increased the Torin1-mediated Erk pathway activation (green asterisk). (B) A schematic of the crosstalk between the pathways upon inhibitor treatments. Single or multiple drug treatments resulting in inhibited protein kinases (red) or activated protein kinases (green) are represented. Increased size of the kinase represents an increase in activation (larger circle = dramatic increases) and the thick arrows represent a strong feedback mechanism (e.g., arrow from S6K to IRS-1).

Figure 5. The Akt, mTOR and Erk 1/2 signaling pathways coordinately regulate both MBP mRNA and protein expression in vitro

Rat OPCs were treated with different kinase inhibitors, individually or in combination, and allowed to differentiate for 5 days. Cells were then collected for mRNA and protein analysis. (A) Combination treatments of U0126 and rapamycin, Torin1 or LY all resulted in greater reduction of MBP mRNA compared to single treatments. (B) Representative western blot analysis of MBP protein expression. (C) Quantification of MBP protein in treated cells. When correlated to the reduction in MBP mRNA, some combination treatments (treatments including LY294002; rapamycin or Torin1) had a significantly greater impact reducing MBP protein than MBP RNA. Data represent the mean ± SEM from three independent experiments. *p < 0.05, **p < 0.005, ***p < 0.0005, ****p < 0.00005 (Two way ANOVA).

Figure 6. In vivo, Akt, mTOR and Erk 1/2 signaling are dynamically regulated during oligodendrocyte maturation from P1 to P10

(A) Representative PLP-EGFP coronal brain section outlining the area of interest (red). (B) Representative images at the midline of the PLP-EGFP corpus callosum with EGFP-positive oligodendrocyte lineage cells (green) co-stained with PDGFRα (red) for P1 section or Olig2 (red) for P4, P7 or P10 section. Arrows represent OPCs (P1 and P4), premyelinating oligodendrocytes (P7) or myelinating oligodendrocytes (P10). (C) Immunostaining of P1, P4, P7 and P10 mouse brains with p-Erk1/2, p-mTOR, p-Akt S473, p-Akt T308 and p-S6RP (green), Sox10 (red) and DAPI (blue). Arrowheads (white) identify specific oligodendrocytes co-labeled with the phosphorylated protein. (D) Quantification of the percentage of phosphorylated protein-positive cells/Sox10-positive cell. pErk 1/2 (black), pAkt 473 (red) and pS6RP (green) expression peaked in oligodendrocytes at P7, which correlates to the transition of OPCs to premyelinating oligodendrocytes (B), while pAkt 308 peaks earlier. Scale bars = 50 μm (B) and 20 μM (C). Data represents the mean +/- SEM for three independent experiments.

Figure 7. Pharmacological inhibition of mTORC1 in vivo elevated the level of phosphorylated Erk1/2 signaling in both oligodendrocytes and neurons

Mice were injected IP at P7 with either vehicle or rapamycin, and tissue for immunohistochemistry was collected 24 hours post injection. (A) Representative images of the lateral corpus callosum of oligodendrocytes stained with pS6RP or pErk 1/2 (green), Sox10 (red) and DAPI (blue) of the vehicle (left panel) or rapamycin injected animals (right). Effective inhibition of mTORC1 was seen by complete loss of pS6RP detection in the rapamycin injected animals (top, right). Arrows represent pErk 1/2+/Sox10+ cells. (B) Quantification of oligodendrocyte co-expressing pErk 1/2 (black) or pS6RP (red)/Sox10 positive cells in vehicle-injected compared to rapamycin-injected animals. Note that there is essentially complete loss of pS6RP in rapamycin-injected animals (i.e., absence of far right column). Scale bars = 20 μM. **p < 0.05 (Student's t test).