Cell Growth

The mTORC1 Pathway

Essential to connecting cell growth control with cancer pathogenesis was the identification of intracellular signaling molecules that coordinate signals from nutrient availability, growth factors, and hormones with autonomous cell growth. In cells, these inputs converge upon a Ser/Thr protein kinase called TOR (target of rapamycin), which has emerged as a central controller of eukaryotic cell growth. TOR was discovered in the 1970s to be the molecular target of an antifungal macrolide produced by a soil bacterium (Streptomyces hygroscopicus) that was isolated on Easter Island (Rapa Nui). Today rapamycin is recognized for its immunosuppressive function, ability to prevent restenosis after angioplasty, and limited anticancer properties. Mechanistically, rapamycin binds an intracellular protein called FKBP12 (an immunophilin), and the rapamycin-FKBP12 complex binds potently and specifically to TOR. Extensive studies across multiple eukaryotic model systems, all spawned by the identification of TOR as the target of rapamycin, have unveiled a complex TOR-centric signaling network responsible for integrating numerous growth signals into a metabolic program that drives biomass production (i.e., cell growth).

About a decade after the discovery of TOR, biochemical studies revealed that TOR (named mTOR in mammals in which the m now officially denotes “ m echanistic” TOR) associates into at least two distinct multisubunit protein complexes called mTOR complex 1 (mTORC1) and mTORC2 . Growth control by mTOR is largely attributed to the best understood mTOR complex—mTORC1. In addition to the catalytic mTOR subunit, mTORC1 contains Raptor (regulatory associated protein of mammalian target of rapamycin), PRAS40 (proline-rich AKT substrate 40 kDa), mLST8 (mammalian lethal with sec-13 protein), and DEPTOR (DEP domain containing mTOR interacting protein) ( Figure 12-1 ). Of these interacting proteins, Raptor and PRAS40 are unique to mTORC1, whereas mLST8 and DEPTOR are shared with mTORC2 (discussed later). Raptor is both a regulator of mTORC1 catalytic activity and a scaffold for recruiting mTORC1 substrates. PRAS40 and DEPTOR inhibit mTORC1 activity by undefined mechanisms. The function of mLST8 is unknown, but it does not appear to be required for mTORC1 activity. Rapamycin-FKBP12 directly binds mTORC1 through a domain in the mTOR catalytic subunit called the FRB domain ( F KBP12- r apamycin- b inding domain), and although the drug destabilizes the association between mTOR and Raptor, it does not dissociate any components of the complex. In fact, rapamycin’s exact mechanism of action remains a mystery despite years of research.

It is now widely accepted that mTORC1 positively controls an array of cellular processes critical for growth, including protein synthesis, ribosome biogenesis, and metabolism, and negatively influences catabolic processes such as autophagy—all of which have roles in cancer pathogenesis. Elucidating the key downstream targets of mTORC1 driving these events is an intense area of research. Originally, much of the study of mTOR relied on experiments in which rapamycin was used acutely to inhibit mTOR (which we now know was mTORC1) in cultured cells. This led to extensive characterization of the best known mTORC1 substrates eiF-4E-binding protein 1(4E-BP1) and S6 kinase 1 (S6K1), both of which regulate protein synthesis. In the unphosphorylated state, 4E-BP1 binds and inhibits the cap-binding protein and translational regulator eIF4E. When phosphorylated by mTOR, 4E-BP1 is relieved of its inhibitory duty, promoting eIF4E interaction with the eIF4F complex and the translation of capped nuclear transcribed mRNA. Following co-regulatory phosphorylation by mTORC1 and another kinase called phosphatidylinositol 3-dependent kinase 1 (PDK1), S6K1 positively affects mRNA synthesis at multiple steps including initiation and elongation by phosphorylating several translational regulators. Although the preponderance of evidence indicates that S6K1 and 4E-BP1 are directly phosphorylated by mTOR, an unidentified phosphatase activity may also be involved in their regulation. For example, the rapamycin-sensitive phosphorylation site on S6K1 is rapidly dephosphorylated (i.e., within minutes) of exposure to the drug.

For many years studies using rapamycin were at odds with the model that mTOR is an essential controller of protein synthesis because rapamycin has only modest effects on translation in most mammalian cells. Rapamycin is an allosteric inhibitor that binds mTOR outside of the kinase domain; it was therefore suspected that rapamycin might incompletely inhibit mTORC1. In 2009, the first mTOR kinase inhibitors became available (discussed later), which, unlike rapamycin, are ATP-competitive inhibitors that target the ATP-binding pocket of mTOR. The first studies employing mTOR catalytic site inhibitors confirmed what had previously been suspected: that rapamycin only partially inhibits mTORC1 activity. The mTOR catalytic site inhibitors exposed rapamycin-resistant functions of mTORC1 in translational control, and consequently the mTOR ATP-competitive inhibitors have more profound effects on mRNA translation and cell proliferation than does rapamycin. Researchers are now using these new-generation mTOR inhibitors to identify novel mTORC1 substrates (many of which are rapamycin resistant) and beginning to fill in the gaps between mTOR and the myriad cellular processes it regulates.

In addition to synthesizing protein, growing and dividing cells need to synthesize lipids (to build the plasma membrane and the membranes of intracellular organelles) and nucleic acids (to make RNA and DNA). There is growing evidence that mTORC1 regulates lipid metabolism (and storage in the case of adipocytes) through SREBP1. SREBPs are transcription factors that facilitate the expression of lipid and sterol biosynthesis enzymes. Several recent studies suggest that mTORC1 is an upstream regulator of SREBP1 activation. In addition to driving lipid metabolism genes, SREBP1 can also increase the expression of genes involved in the pentose phosphate pathway, which produces important metabolites needed for both lipogenesis and nucleotide biosynthesis. How mTORC1 activates SREBP1 is still unclear, although it appears to function through multiple substrates including S6K1 and another direct mTORC1 substrate called Lipin1. In an mTORC1-dependent manner, S6K1 also controls pyrimidine biosynthesis by directly phosphorylating and regulating the activity of CAD (carbamoyl-phosphate synthetase 2, aspartate transcarbamoylase, dihydroorotase), which catalyzes the first three steps of de novo pyrimidine synthesis. Thus, mTORC1 truly is a master controller of growth, as it significantly influences the intracellular synthesis of the major macromolecules required by growing (and dividing) cells: proteins, lipids, and nucleic acids.

Upstream Signaling to mTORC1

The connection between cell growth control by mTORC1 and cancer solidified with the identification of upstream mTORC1 regulators. Building biomass requires adequate building material, sufficient energy, and favorable environmental conditions. Therefore, it is not surprising that mTORC1 activity is controlled by numerous factors such as amino acid and glucose availability, ATP level, mitochondrial activity, growth factor signaling, and oxygen levels, all of which affect cancer cell growth. The discovery that the TSC1 (hamartin) and TSC2 (tuberin) tumor suppressors function together in a complex (the TSC complex) to negatively regulate mTORC1 provided a key first step in unraveling the biochemical mechanism of how upstream signals control mTORC1 activity. The TSC complex contains GTPase-activating protein (GAP) activity, which suppresses the activity of a small GTPase called Rheb . Rheb directly activates mTORC1 by an unknown mechanism, and therefore by promoting Rheb-GTP hydrolysis, the TSC complex suppresses mTORC1 (see Figure 12-1 ). Mutation in either the TSC1 or TSC2 gene results in aberrant upregulation of mTORC1 activity and causes a tumor-prone syndrome called tuberous sclerosis complex (described later).

The TSC complex integrates many positive and negative signals responsible for modulating mTORC1 activity. For example, extracellular growth factors such as insulin and insulin growth factor 1 (IGF-1) can activate an intracellular signaling pathway that inhibits TSC2. Through activation of receptor tyrosine kinases, growth factors stimulate PI3K to phosphorylate membrane-associated phosphatidylinositol 4,5-bisphosphate (PtdIns(4,5)P ₂ ) to generate phosphatidylinositol 3,4,5-triphosphate PtdIns(3,4,5)P ₃ . PtdIns(3,4,5)P ₃ serves as a docking site for the membrane recruitment and activation of the AKT kinase, one of the most versatile kinases in the human kinome. Among its many functions, AKT phosphorylates and inhibits TSC2 and thereby potentiates mTORC1 signaling. AKT can also directly phosphorylate the PRAS40 subunit of mTORC1, which relieves a negative effect of this subunit on mTORC1 activity. The phosphatase and tumor suppressor PTEN, a major tumor suppressor in human cancer, balances PI3K activity by dephosphorylating PtdIns(3,4,5)P ₃ and thus negatively regulates AKT activity. Other growth factor pathways such as the MAPK/ERK pathway and WNT pathway can also activate mTORC1 by inhibiting TSC as can proinflammatory cytokines such as tumor necrosis factor-α (TNFα) through the IκB kinase β (IKKβ).

Although extracellular growth factors provide the external (systemic) cues for growth, intracellular energy, oxygen, and nutrients must also be available. An intracellular ratio of ATP to ADP and AMP is sensed by the AMP-activated protein kinase (AMPK). Under conditions of energy stress, high levels of ADP/AMP bind and activate AMPK, which suppresses anabolic processes and stimulates catabolic reactions. Active AMPK directly phosphorylates and activates TSC2 to suppress mTORC1 signaling when energy supplies drop. AMPK also directly inhibits mTORC1 by phosphorylating the Raptor subunit, emphasizing a theme that positive and negative regulators of growth often impinge on the mTORC1 pathway at multiple steps. Oxygen stress (i.e., hypoxia) inactivates mTORC1 through TSC, but by a different mechanism. In hypoxic conditions, hypoxia-inducible factor α (HIF1α) is stabilized, and its accumulation leads to a transcriptional response that promotes a metabolic shift away from oxidative phosphorylation toward oxygen-independent glycolysis, increased vascularization, and decreased mTORC1 activity by inducing expression of REDD1 and REDD2. REDD1/2 release TSC2 from an inhibitory interaction with 14-3-3 proteins. Notably, mTORC1 can also potentiate HIF1α by increasing its transcription and translation, possibly functioning as part of a feedback regulatory circuit. Finally, DNA damage can reduce mTORC1 activity in part by increasing expression of TSC2 and PTEN.