Platelet-derived growth factor-induced Akt phosphorylation requires mTOR/Rictor and phospholipase C-γ1, whereas S6 phosphorylation depends on mTOR/Raptor and phospholipase D

Abstract

Mammalian target of rapamycin (mTOR) can be found in two multi-protein complexes, i.e. mTORC1 (containing Raptor) and mTORC2 (containing Rictor). Here, we investigated the mechanisms by which mTORC1 and mTORC2 are activated and their downstream targets in response to platelet-derived growth factor (PDGF)-BB treatment. Inhibition of phosphatidylinositol 3-kinase (PI3K) inhibited PDGF-BB activation of both mTORC1 and mTORC2. We found that in Rictor-null mouse embryonic fibroblasts, or after prolonged rapamycin treatment of NIH3T3 cells, PDGF-BB was not able to promote phosphorylation of Ser473 in the serine/threonine kinase Akt, whereas Thr308 phosphorylation was less affected, suggesting that Ser473 in Akt is phosphorylated in an mTORC2-dependent manner. This reduction in Akt phosphorylation did not influence the phosphorylation of the S6 protein, a well established protein downstream of mTORC1. Consistently, triciribine, an inhibitor of the Akt pathway, suppressed PDGF-BB-induced Akt phosphorylation without having any effect on S6 phosphorylation. Thus, mTORC2 does not appear to be upstream of mTORC1. We could also demonstrate that in Rictor-null cells the phosphorylation of phospholipase Cγ1 (PLCγ1) and protein kinase C (PKC) was impaired, and the PKCα protein levels strongly reduced. Furthermore, interfering with the PLCγ/Ca2+/PKC pathway inhibited PDGF-BB-induced Akt phosphorylation. In addition, PDGF-BB-induced activation of mTORC1, as measured by phosphorylation of the downstream S6 protein, was dependent on phospholipase D (PLD). It has been shown that Erk1/2 MAP-kinase directly phosphorylates and activates mTORC1; in partial agreement with this finding, we found that a Mek1/2 inhibitor delayed S6 phosphorylation in response to PDGF-BB, but it did not block it. Thus, whereas both mTORC1 and mTORC2 are activated in a PI3K-dependent manner, different additional signaling pathways are needed. mTORC1 is activated in a PLD-dependent manner and promotes phosphorylation of the S6 protein, whereas mTORC2, in concert with PLCγ signaling, promotes Akt phosphorylation.

Figure 1

PDGF-BB-mediated S6 phosphorylation does not require Akt phosphorylation. NIH3T3 cells (A, B, D) and Rictor-null MEFs (C) were serum-starved for 24 h and then stimulated for indicated time periods with PDGF-BB (20 ng/ml) without or with pretreatment for 1 hours (otherwise specified) in the presence of the inhibitors NPV-BKM120 (A), rapamycin (Rap, 10 nM) (B), or tricribine (20 μM) (D) for indicated time periods. Total cell lysates (TCL) were prepared, and the levels of Akt phosphorylation at S473 and T308, as well as S6 phosphorylation and the expression of total protein were assayed by immunoblotting (Ib). The relative protein phosphorylations were quantified for a representative experiment.

Figure 2

PDGF-BB-mediated S6 phosphorylation is regulated by PLD/Ca²⁺ signaling. NIH3T3 cells (A, B) and Rictor-null MEFs (C) were serum-starved for 24 h and then stimulated for indicated time periods with PDGF-BB (20 ng/ml) with or without pretreatment with 1-butanol (1-B, 0.3%), 2-butanol (2-B, 0.3%), and Ca²⁺ chelators BAPTA-AM (BA, 10 μM) or EDTA (ED, 2 mM) for 30 min, as indicated. Total cell lysates (TCL) were prepared, and the levels of Akt phosphorylation at S473 and T308, S6 phosphorylation, as well as the total protein expression, were assayed by immunoblotting (Ib). The relative protein phosphorylations were quantified for a representative experiment.

Figure 3

PDGF-BB-induced Akt phosphorylation involves the PLC/PKC pathway. NIH3T3 cells (A and E), dnPLCγ PAE cells (B), PLCγ-null (C) and Rictor null (D and F) MEFs were serum-starved for 24 h and then stimulated with PDGF-BB (20 ng/ml) with or without pretreatment with the inhibitor U73122 (5 μM) and SB203580 (10 μM) for 1 h, or with or without PMA (1 μM) for 24 h, as indicated. Akt phosphorylation level at S473 and T308, as well as mTOR, S6, PLCγ, PKCα and pMarcks phsphorylation and the total protein expression were assayed by immunoblotting (Ib) of total cell lysate. β-actin immunoblotting served as a loading control. The relative protein phosphorylations were quantified for a representative experiment.

Figure 4

PDGF-BB-induced Erk1/2 signalling affects the kinetics of S6 phosphorylation. NIH3T3 cells (A, B, D, E, F and G) and Rictor-null MEFs (C) were serum-starved for 24 h and then stimulated with PDGF-BB (20 ng/ml) in the absence or presence of CI-1040 (0.5 μM) and rapamycin (10 nM) for 1 h, 1-butanol (1-B, 0.3%), 2-butanol (2-B, 0.3%), BAPTA-AM (10 μM) and EDTA (2 mM) for 30 min, and PMA (1 μM) for 24 h, as indicated. The levels of phosphorylation of Erk1/2, mTOR, Akt and S6, as well as the total protein, were assayed by immunoblotting (Ib) of total cell lysates.

Figure 5

Effect of mTOR signaling on caspase 3 cleavage, apoptosis, migration and proliferation upon PDGF-BB stimulation. Rictor-null or control MEFs were serum-starved for 24 h and then treated with PDGF-BB for 24 h; activation of caspase 3 was measured thereafter by immunoblotting against cleaved caspase 3 (A). Internucleosomal DNA fragmentation was quantitatively determined by assaying for cytoplasmic mononucleosome- and oligonucleosome-associated histone accumulated in apoptotic cell (B), data represent three separate experiments each performed in duplicate ± SEM. Cell migration experiments were carried out in a 96-well ChemoTX cell migration microplate. The wells of the microplate were filled with medium containing combinations of PDGF-BB with Rictor-null or control MEFs (C), as well as NIH3T3 cells with or without longterm treatment with rapamycin (E), as indicated. The amounts of migrated cells are given as index units; data represent three separate experiments, each performed in quadruplicates ± SEM. In separate experiments, NIH3T3 cells were serum-starved and then stimulated for 24 h with PDGF-BB in medium containing [³H] thymidine. The fold increase of PDGF-induced [³H]thymidine incorporation over the respective positive control values is shown. Values are means ± S.E of three independent experiments each performed in triplicate. Statistical significant differences (Students T-test) are indicated by *P < .05 compared with unstimulated or control cells (B &D).

Figure 6

Schematic representation of PDGF-BB-mediated regulation of Akt and S6. Initially, PDGF-BB-mediated activation of Akt involves both mTORC2 and the PLCγ/PKC pathways. Activation of S6 downstream of mTORC1 depends on PLD activation, independent of mTORC2 and Akt signaling. Ca²⁺ is required for regulation of Akt and S6 activity.