Research over the last decade has established that mTOR is the central component of a signaling network that controls cellular growth and a vast number of studies have been aimed at understanding how mTOR signaling is regulated. These studies have revealed that mTOR exists in two distinct multi-protein complexes and only one of them is inhibited by the FKBP12-rapamycin complex. The rapamycin-sensitive complex consists of the proteins mTOR, GβL and Raptor, and is collectively referred to as the mTORC1 complex (Huang and Manning, 2008
, Corradetti and Guan, 2006
). The other complex is referred to as the mTORC2 complex and consists of mTOR, GβL and a protein called Rictor (Guertin et al., 2006
, Corradetti and Guan, 2006
). Signaling by the mTORC2 complex is not directly inhibited by rapamycin, and thus, it seems unlikely that signaling by the mTORC2 complex contributes to the growth-regulatory effects that have been attributed to the rapamycin-sensitive pathway (Corradetti and Guan, 2006
, Jacinto et al., 2004
). Therefore, the remainder of this discussion will focus on the mechanisms that regulate signaling by the mTORC1 complex. To assist the reader, a schematic that summarizes these mechanisms is provided in .
Schematic of the General Mechanisms that Regulate mTORC1 Signaling
Phosphorylation of p70S6k
on the Thr389 residue is typically used as a read-out of mTORC1 signaling, and this marker has revealed that signaling by mTORC1 can be regulated by various stimuli including growth factors, nutrients and mechanical signals. To date, the most intensely studied regulatory mechanisms of mTORC1 signaling are those which occur following stimulation with growth factors such as insulin. These studies indicate that growth factors activate class I phosphoinositide-3 kinase (PI3K) which, in-turn, promotes the activation of protein kinase B (PKB) via the 3-phosphoinositide-dependent protein kinase-1 (PDK1) (Reiling and Sabatini, 2006
). Activated PKB can then phosphorylate and inhibit the activity of the tuberous sclerosis complex (TSC1/2) complex which is a GTPase-activating protein for a ras-family GTP-binding protein called Rheb (Huang and Manning, 2008
). Inactivated TSC1/2 allows Rheb to become charged with GTP and, in-turn, promote the activation of mTORC1 signaling (Inoki et al., 2003
). It is clear that GTP-Rheb can activate mTORC1 and it has been proposed that this might involve the sequestration of the FKBP38 protein (an endogenous inhibitor of mTORC1 signaling); however, the exact mechanism through which Rheb activates mTORC1 signaling remains a subject of intense debate (Bai et al., 2007
, Ma et al., 2008
, Avruch et al., 2006
As mentioned above, nutrients such as amino acids (AA) have also been shown to play an important role in the regulation of mTORC1 signaling. Current models indicate that, unlike growth factors, nutrients activate mTORC1 through a class I PI3K-, PKB- and TSC1/2-independent mechanism (Nobukuni et al., 2005
, Gulati and Thomas, 2007
). However, the activation of mTORC1 by nutrients is blocked by the PI3K inhibitor wortmannin, indicating that PI3K activity is still required for this event (Nobukuni et al., 2005
). This observation ultimately led to the finding that a wortmannin-sensitive class III PI3K called Vps34 might be involved in nutrient-induced mTORC1 signaling (Nobukuni et al., 2005
). For example, it was reported that nutrients could induce Vps34 activity and depletion of Vps34 with siRNA blocked nutrient-induced mTORC1 signaling (Gulati and Thomas, 2007
). It has also been suggested that nutrients activate Vps34 through a mechanism involving the Ca2+/calmodulin complex (CaM) (Gulati et al., 2008
), however, the results of other studies have not supported this conclusion (Yan et al., 2009
). The mechanism could ultimately link Vps34 to the activation of mTORC1 is even less clear, but might involve enhanced binding of Rheb to mTORC1; however, it has also been reported that nutrient stimulation does not alter the amount of GTP bound Rheb (Nobukuni et al., 2005
, Long et al., 2005
). Furthermore, genetic studies in Drosophilia
and C. elegans
have recently challenged the potential role of Vps34 in the regulation of TORC1 signaling (Juhasz et al., 2008
, Avruch et al., 2009
). Thus, there is still much to be disputed with regards to the potential role of Vps34 in amino acid induced mTORC1 signaling.
Currently, the most widely accepted mechanism for explaining how amino acids regulate mTORC1 signaling involves the Rag subfamily of Ras small GTPases. Mammalian cells express four Rag GTPases (RagA, RagB, RagC and RagD) and two independent groups have demonstrated that Rag GTPases are necessary for the activation of mTORC1 signaling by amino acids (Kim et al., 2008
, Sancak et al., 2008
). Furthermore, both groups demonstrated that overexpression of constitutively GTP bound RagA or RagB is sufficient to induce an increase in mTORC1 signaling and an increase in the size of both Drosophilia
and mammalian cells (Kim et al., 2008
, Sancak et al., 2008
). More recently, amino acids have been shown to induce translocation of mTORC1 to lysosomal membranes where both the Rag GTPases and Rheb reside (Sancak et al.
). Furthermore, the lysosomal membrane recruitment of mTORC1 has been shown to be dependent on a complex of proteins encoded by the MAPKSP1, ROBLD3 and c11orf59 genes, and this complex has been termed the Ragulator complex (Sancak et al.
). Of particular significance, constitutive targeting of mTORC1 to the lysosomal surface renders mTORC1 signaling independent of amino acids, Rag GTPases and the Ragulator complex, but it remains dependent on Rheb (Sancak et al.
). Thus, the interaction of mTORC1 with Rheb, via a Rag GTPase-Ragulator dependent translocation of mTORC1 to lysosomal membranes, appears to be a critical event for the activation of mTORC1 signaling in response to amino acids and possibly several other stimuli.
Another important regulator of mTORC1 signaling is the AMP-activated protein kinase (AMPK). During periods of energy deprivation (i.e. exercise, hypoxia and nutrient deprivation), AMP levels rise and promote the activation of AMPK (Kimball, 2006
, Musi et al., 2003
, Fujii et al., 2006
, Kim et al., 2004
). Activated AMPK can phosphorylate and activate the TSC1/2 complex which represses the Rheb-induced activation of mTORC1 signaling. For this reason, it has been proposed that excess nutrients may indirectly activate mTORC1 signaling through repression of the inhibitory signals provided by the AMPK → TSC1/2 -| Rheb → mTORC1 pathway (Hahn-Windgassen et al., 2005
). Furthermore, it was recently demonstrated that AMPK can directly phosphorylate raptor, and that this event inhibits mTORC1 signaling (Gwinn et al., 2008
). Thus, AMPK appears to play an important role in the regulation of mTORC1 signaling and the exact mechanisms involved in this process continue to be defined.
An additional regulatory mechanism of mTORC1 signaling involves the lipid second messenger phosphatidic acid (PA) (Foster, 2007
). The potential role of PA in the regulation of mTORC1 signaling was revealed in a study which demonstrated that PA can directly bind to mTOR in a region that lies within the FRB domain and, in doing so, activates mTORC1 signaling (Fang et al., 2001
). Furthermore, there is evidence that PA competes with the FKBP12-rapamycin complex for binding to the FRB domain (Fang et al., 2001
, Chen et al., 2003
, Veverka et al., 2008
). Based on these observations, it has been proposed that rapamycin inhibits mTORC1 signaling by removing PA from the FRB domain. Since these initial reports, several additional studies have provided evidence that PA plays a fundamental role in the regulation of mTORC1 signaling. For example, over-expression of diacylglycerol kinase, an enzyme that converts diacylglycerol into PA, induces the activation of mTORC1 signaling and this event is blocked when the PA binding domain of mTOR is mutated (Avila-Flores et al., 2005
). Over-expression of lysophosphatidic acid acyltransferase, an enzyme that converts lysophosphatidic acid into PA, has also been shown to activate mTORC1 signaling (Tang et al., 2006
). Finally, it was recently suggested that PA might be the primary effector of Rheb that promotes the activation of mTORC1 signaling (Sun et al., 2008
). This hypothesis was based on the findings that GTP-bound Rheb binds to, and activates, phospholipase D (PLD), an enzyme that generates PA from phosphotidylcholine (Sun et al., 2008
). Furthermore, it was reported that knocking down the expression of PLD prevented the activation of mTORC1 by Rheb (Sun et al., 2008
). Thus, several lines of evidence suggest that PA is a direct regulator of mTORC1 signaling, and therefore, understanding the exact role that PA plays in the regulation of mTORC1 will be an important area for future studies.
The work cited above demonstrates that the regulation of mTORC1 signaling is controlled by a complex network of signaling events, and our understanding of this network is rapidly evolving. Due to page limitations, several other potentially important regulatory mechanisms have not been discussed and hence, the reader is referred to Yang et al. (2007), Foster (2007)
, Kim and Guan (2009)
and Avruch et al. (2009)
for more comprehensive reviews on the mechanisms that regulate mTORC1 signaling (Yang and Guan, 2007
, Foster, 2007
, Kim and Guan, 2009
, Avruch et al., 2009