Obesity and Nutrient overload
There is a rapid increase in the rate of obesity epidemic particularly in the western countries in the last decade due to the drastic rise in the calorie intake associated with increased availability of food and relatively low cost . One of the major cause of obesity is nutrient overload . The major pathogenic consequences of nutrient overload are insulin resistance and type II diabetes both of which are the main cause of cardiovascular disease . Over 50% of the patients with cardiovascular disease – a major cause of morbidity and death globally have diabetes and obesity as their risk factors according to a recent report by CDC . According to 2010 world diabetes foundation statistics, an estimate of 285 million people corresponding to 6.4% of the world adult population live with diabetes. 10.2% of the global diabetes prevalence exist in the western pacific. 70% of the current diabetic cases occur in developing countries with 50.8 million diabetic cases in India making it the diabetic capital in the world followed by China with 43.2 million cases . But according to a recent study published in the Lancet Journal , 347 million people live with diabetes which is markedly higher than the estimated projection of 285 million worldwide pointing an alarming increase in the global diabetes prevalence rate. This statistics shows the urgent need to address and prevent the disease as by 2030, 438 million people are expected to live with diabetes if the current scenario continue to exist. Diabetes mellitus is of two types. In type I diabetes, the body makes little or no insulin and daily injection of insulin is necessary for survival. While in type II diabetes, pancreas produce insulin but the fat, liver and muscle tissues in the body does not effectively respond to insulin . Of the different type of diabetes, type II diabetes (Insulin resistance) accounts for 85-95% of the total diabetic cases in developed countries and may account for an even higher percent in developing countries. Type II diabetes is strongly associated with obesity and a sedentary lifestyle. People with diabetes have inadequate blood sugar control, which can lead to serious complications like heart disease and stroke, damage to the kidneys or nerves, and to blindness. High blood glucose and diabetes cause around 3 million deaths globally each year, a number that will continue to rise as the number of people affected increases. Changing diet, increasing physical activity and improving the living environment can prevent unto 80% of type II diabetes. However, without effective prevention and treatment programs, the global occurrence of diabetes, its associated illness and death are likely to continue to rise globally. According to a recent report by the Centre for disease control and prevention (CDC), the two major causes of diabetes is obesity and lack of physical activity. Nearly 95% of diabetes cases in the United States is due to obesity and lack of physical activity . Recent studies have also shown association between obesity and several types of cancers. Cancer, undoubtedly, is among the leading causes of death worldwide. Annually, 14 million new cancer cases are diagnosed and 8.2 million cancer related deaths occur worldwide. By 2030, there will be alarming 50% increase in new cancer cases and 60% increase in cancer related death globally. Evidences point out that nearly 20% of cancer deaths are linked to obesity . These studies reveal the necessity to study the molecular mechanism of nutrient overload and its associated pathogenesis.
Although for the past 2 decades carbohydrates and fats are known as the nutrients that alter glucose and energy homeostasis, recent studies points out the role of dietary proteins and amino acids in maintaining glucose and energy homeostasis as the consumption of protein has increased by nearly 50% above the recommended amount in the western countries. Increase in the intake of dietary proteins has been shown to have important role in causing obesity and type II diabetes via its effect on gluconeogenesis and insulin resistance [10, 11].
Plasma levels of amino acids were found to be elevated in insulin resistant states of obesity [12-14]. In experimental settings, plasma AAs induced insulin resistance in skeletal muscles of healthy individuals [15, 16]. These studies suggests that nutrients also act as mediators in signaling pathways apart from being an energy source. One of the major and emerging nutrient signaling pathway is the mammalian target of Rapamycin complex 1 (mTORC1/S6K1) signaling pathway, which plays an important role in cell growth and proliferation . In 2004 Um et al., showed that excess nutrients leads to the constitutive activation of mTOR/S6K1 which subsequently desensitizes insulin signaling under nutrient overload conditions (Fig 1) .
Since excess nutrients and growth factors in obese mice lead to chronic activation of mTORC1/S6K1 and subsequently inducing insulin resistance, the effect of rapamycin (mTOR inhibitor) on metabolic profiling was tested and unexpectedly, rapamycin treatment worsened metabolic profiling by strongly inducing insulin resistance and reducing cell and tissue size for reasons unknown [19-21]. This compels the need to study and understand the molecular mechanism underlying the pathway for the development of novel therapeutic targets for the treatment and hopefully for the cure of insulin resistance and type II diabetes.
C.2.1 Amino acids (AAs): Classification, metabolism and physiological function
Amino acids were traditionally classified into essential (Histidine, Isoleucine, Leucine, Lysine, Methionine, Phenylalanine, Threonine, Tryptophan, Valine), non-essential (Alanine, Asparagine, Aspartic acid, Glutamic acid, Glycine, Cystein, Glutamine, Proline, Serine, Tyrosine, Taurine, Ornithine, Selenocysteine, Arginine) and conditionally essential (Cystein, Glutamine, Proline, Serine, Tyrosine, Taurine, Ornithine, Selenocysteine, Arginine) based on nitrogen balance theory and their role in growth. Essential amino acids are those that must be provided from the diet whereas non-essential amino acids are synthesized de novo. Amino acids play a role in protein synthesis, source of energy, growth and development. Although for more than two decades now the role of glucose and fat metabolism and homeostasis in the pathogenesis of obesity and diabetes is aggressively studied, recent studies and evidences suggests the possible role of amino acids in the pathogenesis of obesity induced insulin resistance [10, 22] through its role in signal transduction. Plasma levels of amino acids and especially branched chain amino acids (Leucine, Isoleucine and valine) were found to be elevated in insulin-resistant states of obesity [12-14]. In experimental settings, plasma AAs induced insulin resistance in skeletal muscles of healthy individuals by inhibiting muscle glucose transport and modulating peripheral glucose sensitivity [15, 16]. Furthermore, the infusion of Leucine (a branched chain amino acid – BCAA) alone was shown to inhibit glucose uptake in the forearm . BCAAs constitute around 35%-40% of the essential amino acids in blood plasma . These findings highlights the possible role of dietary protein excess in metabolic diseases and hence necessitates the desire to further explore the molecular mechanisms involved in the development of insulin resistance by amino acids.
With the increase in understanding of the role and mechanism by which nutrients act as mediators in signaling pathways a rapid change is occurring in the field of ‘nutrient sensing’ where traditionally nutrients such as amino acids (AAs), glucose and fatty acids have been known only as metabolic fuels for cell growth and development. Several nutrient signaling pathways that are functionally distinct are studied which have many conserved sensors and nutrient signaling pathways from yeast to mammals. One of the major and emerging nutrient signaling pathway is the mTOR/S6K1 signaling pathway which plays an important role in cell growth and proliferation . In 2004 Um et al., showed that excess nutrients leads to the constitutive activation of mTOR/S6K1 which subsequently desensitizes insulin signaling under nutrient overload conditions (Fig 1) .
Fig 1: Constitutive mTOR/S6K1 activation in obese mice due to continuous supply of nutrients especially AAs. Figure from Um et. al., 2004 .
C.2.2 Components of mTORCs:
The Ser/Thr protein kinase, target of Rapamycin (TOR) is evolutionarily conserved and forms two functionally distinct complexes, TOR complex 1 (TORC1) and TOR complex 2 (TORC2) [25, 26]. The two complexes has both shared and distinct proteins associated with each other and they control a wide variety of cellular processes in response to various stimuli. Yeast has two TOR genes, Tor1 which forms TORC1 and Tor2 forming TORC2 whereas higher eukaryotes possess only one mTOR gene [27-30]. The proteins associated with the mTOR complexes are (Fig 2)
mTORC1: mTOR, Raptor, PRAS40, mLST8/GβL, Deptor
mTORC2: mTOR, Rictor, hSIN1, mLST8/GβL, Deptor, PRR5/Protor-1
mTOR – mammalian target of rapamycin; Raptor – regulatory associated protein of mTOR; mLST8 – mammalian lethal with Sec13 protein 8 or GβL; PRAS40 – proline rich AKT substrate 40 KDa; Deptor – DEP domain containing mTOR interacting protein; Rictor – rapamycin insensitive companion of mTOR; mSIN1 – mammalian stress activated protein kinase interacting protein; Protor-1 – protein observed with Rictor-1
Fig 2: The two mTOR complexes, mTORC1 vs mTORC2. Figure from Foster et. al., 2010 .
- mTORC1: mTOR is a central growth controller which controls cell growth by integrating various intra- and extra- cellular growth signals. All eukaryotes have been found to have TOR homologs and the mammalian homolog is called mTOR or the mammalian target of rapamycin and their functions like cell growth, translation, regulation of metabolism, autophagy, lipid synthesis and mitochondrial metabolism and biogenesis is highly conserved among different species. TOR can be activated by growth factors, cellular energy status and oxygen levels. Recently, amino acids (AA) and cellular stress have also been known to activate TOR and their pathway of activation is being extensively studied [25, 31]. mTOR is the catalytic subunit of the complex and the role of most of the proteins associated with mTORC1 is still unclear. Raptor acts as a scaffold between mTOR kinase and the substrates for mTORC1 thus recruiting them to the complex and regulates its assembly with the complex [32-34]. The function of mLST8 is still unclear albeit it has been shown to be involved in mTORC2 activity . PRAS40 and Deptor are known to negatively regulate mTOR [36-38]. Decreased mTORC1 activity recruits PRAS40 and Deptor to the complex thereby further inhibiting its activity whereas when mTORC1 is activated, it phosphorylates PRAS40 and Deptor resulting in decreased physical interaction with mTORC1 [38, 39]. Rapamycin, a bacterial macrolide binds to FKBP12 and inhibits mTORC1 activity via its interaction to the FKBP12 binding domain of mTORC1 and hence mTORC1 is called rapamycin sensitive mTOR complex [25, 40].
- mTORC2: Although initially identified as a rapamycin insensitive complex as rapamycin cannot directly bind to mTORC2, recent studies showed that prolonged rapamycin treatment can inhibit the complex by disrupting its assembly . mTORC2 is a poorly studied complex and its structure and exact functions are yet to be discovered. Rictor, the regulatory subunit of mTORC2 has been shown to be important for the activity of the complex, but since it lacks any obvious functional domains its exact role is still unclear [40, 41]. Deptor has been shown to negatively regulate mTORC2 activity while mLST8 and mSIN1 are important for maintaining the structural integrity of the complex [35, 42, 43]. Protor exists in two isoforms, protor 1 and protor 2 and they remain bound to rictor although they do not play a role in the assembly and activity of the complex . The function and signaling pathway of mTORC2 is poorly defined and is shown to play a role in cytoskeletal organization, cell survival, proliferation and cell cycle progression [25, 30].
C.2.3 mTORC1 activation by amino acids (AAs):
One of the initial studies showing the involvement of AAs in the control of metabolic pathways via its role as a signal transducer came from the study where they looked at the regulation of autophagy in hepatocytes . Autophagy is upregulated under AA deprivation conditions in order to produce AAs that are needed for cell survival (S6K1) . This finding showing the initial involvement of AAs in metabolic regulation was followed by studies which showed that in the absence of insulin, AAs caused an increase in the phosphorylation of S6K1 and 4EBP1, and that rapamycin (mTORC1 inhibitor) completely blocked the response of AAs on S6K1 phosphorylation indicating that the pathway of activation is via mTOR [47, 48]. Wortmannin (Phosphatidylinositol 3 kinase inhibitor) like rapamycin blocked the effect of AA phosphorylation of S6K whereas inhibitors against PKB/Akt did not have any effect on AA signaling to S6K, showing that the pathway of activation of S6K1 was independent of insulin signaling . Despite the extensive research that has been carried out in the field of nutrient, in particular amino acid sensing of mTORC1, it is still unclear as to how cells sense amino acids and how the signals are relayed to mTORC1.
C.2.4 Upstream signaling targets of mTORC1:
So far known upstream activators of mTORC1 by amino acids
(a) Mitogenic activation of mTOR/S6K1: Insulin and other growth factors binding to the extracellular ɑ subunit of the Insulin receptor (IR) or other receptor tyrosine kinases induces a conformational change in the β subunit of the IR which leads to the auto-phosphorylation of the receptor at the tyrosine residues and increases the kinase activity of the receptor. This triggers the recruitment and phosphorylation of insulin receptor substrates 1 and 2 (IRS 1&2) at multiple tyrosine residues [50-53] which serves as the docking site for the p85 regulatory subunit of class I phosphoinositol 3 kinase (PI3K) at the membrane. PI3K when docked phosphorylates phosphatidylinositol 4,5 bisphosphate (PIP2) to produce phosphatidylinositol 3,4,5 triphosphate (PIP3) which activates phosphoinositol dependent protein kinase 1 (PDK1) and disrupts the binding of the catalytic domain and PH domain of protein kinase B (PKB) by binding to the pleckstrin homology domain (PH) domain of PKB/Akt. Activated PDK1 phosphorylates Akt/PKB at Thr308 position thereby activating it which subsequently leads to degradation of the tuberous sclerosis complex 1/2 by phosphorylating TSC2 (Fig. 3) [54, 55].
(b) TSC1/2 and Rheb: Tuberous sclerosis complex is a heterodimer consisting of TSC1&2 which as an important negative regulator of mTORC1 activity. When phosphorylated, TSC1/2 activates the Ras homolog enriched in brain (Rheb), a GTPase by acting as a GTPase activating protein (GAP). The activated GTP bound Rheb interacts with and activates mTORC1 albeit the exact mechanism of activation is still unclear . In the absence of a stimuli, TSC1/2 keeps Rheb in an inactive GDP bound form thus negatively regulating cell growth. Any mutation or loss of TSC1/2 genes result in tuberous sclerosis, a pathological state associated with several benign and few malignant tumors [25, 31, 56]. In TSC1/2-/- null cells, Rheb GTPase was constitutively activated resulting in constitutive mTORC1 activation. Whereas depriving cells of amino acids in both TSC1/2+/+ cells and TSC1/2 null cells resulted in a significant decrease in the activation of mTORC1. Also, overexpression of GTP bound Rheb in TSC1/2+/+ and TSC1/2-/- cells in the absence of AA showed a significant decrease in the activation of mTORC1. All these studies suggests that the pathway responsible for the AA-induced activation of mTORC1/S6K1 is independent of TSC1/2, Rheb and on the whole the canonical insulin/PI3K/Akt pathway of mTORC1 activation [2, 57-60]. However, Rheb has been shown to be critical for AA-induced mTORC1 activation. When Rheb is knocked down, AAs-induced mTORC1 activation was not observed although Rheb may not have any direct effect on AA-induced mTORC1 activation [58, 59]. Later it was shown that AA withdrawal abolished mTORC1 activity via modulating its effect on Rheb-mTOR interaction . Recent studies in the field suggests a pathway where AAs induce translocation of mTORC1 to lysosomes containing Rheb which activates mTORC1 by working parallel to the AA pathway of mTOR activation. Thus both AA-induced translocation of mTOR to lysosomes and Insulin-induced activation of Rheb-mTOR are essential for activation of mTORC1/S6K1 (Fig. 3) [62, 63].
Fig 3: Insulin signaling pathway through mTOR. Figure from Baselga et. al., 2011 .
(c) Phospholipase D: Although many studies were done exploring the importance of TSC1/2-Rheb axis in the regulation of mTORC1, later Chen et al.  discovered the involvement of Phosphatidic acid (PA) in the activation of mTORC1. PA is a component of a phospholipid that is generated when phosphatidylcholine (PC) is hydrolyzed by phospholipase D (PLD). Inhibiting PA formation using 1-butanol or siRNA knockdown of PLD decreased mTORC1 activation. PA binds to the rapamycin binding domain of mTOR and activates it. Another study also shows that Rheb binds and activates PLD1 [66-69]. Although for long little emphasis was given on PLD and PA activation of mTORC1, recently several studies point out the role of PLD1 in the AA induced mTORC1 pathway where AA starvation significantly decreased PLD1 activity and subsequently mTORC1 activity (Fig. 4) [70-72].
(d) Rag GTPases: They are a member of the Ras superfamily of GTPases and exist as heterodimers. There are four Rag proteins in mammals – RagA, RagB, RagC and RagD. RagA and RagB are highly similar to each other like RagC and RagD and Rag heterodimers are made up of RagA or RagB bound with either RagC or RagD resulting in four possible combinations [73-75]. AA induce mTORC1 activity by altering the GTP loading state of Rag GTPases, the mechanism which is still unclear. AA promote the GTP loading of RagA or Rag B which is otherwise in the inactive or GDP bound state which binds more strongly to raptor thus aiding in mTOR-Rag protein interaction. GTP loading of RagA/B is essential for mTORC1 activity as shown by Sancak et. al., 2008, 2010 where RagA/BGTP– RagC/DGDP mutant which potently activated mTORC1 even under AA starvation conditions and displayed higher binding ability to raptor protein unlike RagA/BGDP– RagC/DGTP which even in the presence of AA displayed no mTORC1 activity and was unable to bind to raptor. Rag GTPases do not directly affect
Fig 4: mTOR signaling through PA and PLD. Figure from Foster et. al., 2009 .
mTORC1 activity. It alters the subcellular localization of mTOR where under AA starvation conditions mTOR is found to be diffuse throughout the cytoplasm and in the presence of AAs, mTOR translocates to the compartment containing Rab7, a late endosomal and lysosomal marker. Later by mass spectrometric analysis it was found that a complex of three proteins (p14, p18 and MP1) collectively called the regulator that reside on the endosomes/lysosomes to be essential for docking Rag proteins to the endosomal compartment. Thus Rag proteins together with the regulator complex is vital for AA-induced mTORC1 translocation to the endosomal compartment where Rheb proteins are also present and for its activation [62, 63]. More recently, several studies point out that lysosomes might have more critical functions in the AA pathway than merely serving as an anchor for the proteins involved in mTORC1 translocation and activity. Studies done in yeast show that AA sensing begins in the lysosomal lumen where they generate signals that activates Rag proteins, and vacuolar H+-ATPase (v-ATPase) was later identified as an essential component of the lysosomes that play a direct physical role in regulating AA dependent interactions with Rags and ragulators and thus mTORC1 signaling. v-ATPase have several functions in the AA pathway where it regulates AA transport across the lysosomes by generating a proton gradient and by physically interacting with rags and ragulators it helps in relaying signals about AA availability. A more clear understanding about the exact role v-ATPases and the transport of lysosomal AAs is yet to be discovered (Fig. 5) [20, 77].
Fig 5: AAs induce inactive v-ATPase-Rag-Ragulator complex (left) to active complex (right). Figure from Efeyan et. al., 2012 .
(e) hVps34: Human vacuolar protein sorting 34 (hVps34) is a class III PI(3)K, initially identified in yeast and is the only class III PI(3)K in yeast. It is known as a positive regulator of mTORC1. In mammals, hVps34 interacts with hVps15 and regulates vesicular trafficking, endosomal sorting and autophagy . Vps34 phosphorylates phosphophatidylinositol to phosphatidylinositol 3 phosphate (PI3P), a phospholipid involved in membrane/vesicular trafficking. PI3P is generally locally produced and it serves as a docking platform for proteins containing FYXE or PX domain to build a trafficking complex [79-81]. Vps34 overexpression increased mTORC1 activity and Vps34 activity is inhibited by amino acid and glucose starvation, highlighting its possible involvement in AA pathway of mTORC1 activation. Rapamycin, mTOR inhibitor showed no effect on hVps34 activity indicating that hVps34 acts upstream of mTOR.
hVps34 and Ca2+: Hannan et el., have previously shown that activation of S6K1 requires the initial Ca2+ dependent priming event which involves the disruption of the auto-inhibitory interaction between the N- and C- termini of S6K1 . Further study from our lab  have suggested a possible role of Ca2+ in AA-induced mTORC1 activation (Fig. 6). On analysis of hVps34, a potential Ca2+/CaM binding motif in the PI3K accessory domain have been identified and it is conserved from yeast to mammals . Binding of Ca2+ to hVps34 increased the activity of hVps34 and AA increased intracellular Ca2+ levels, and removal of Ca2+ in the buffer or by using chemical chelators greatly decreased the activity of mTORC1. The AA-induced increase in Ca2+ enhances the binding ability of CaM to hVps34 leading to the activation of mTORC1 , although the exact mechanism of how AA increase intracellular Ca2+ levels and subsequently activate mTORC1 is a big question of interest in the field.
Fig 6: A proposed model showing AA induced Ca2+/CaM activation of mTORC1. Figure from Gulati et.al., 2008 
hVps34 and PLD: Many studies have emphasized on the role of PLD and hVps34 in regulating mTORC1 activity independently, while a more recent study  points out the relationship between PLD1-hVps34 together in mTORC1 activation by AAs. The study shows that AA induce hVps34 activity which generates PI3P. The PX domain of PLD1 interacts with PI3P thereby catalytically activating PLD1 and subsequently inducing its subcellular translocation. Furthermore, translocation of PLD1 to the lysosomal membrane which is dependent on hVps34 activity was shown to be critical for the translocation of mTORC1 to the lysosomal membrane by AAs (Fig. 7). Hence, AA-induced mTORC1 activation could involve two parallel pathways (i) AA-hVps34-PLD1-mTORC1 pathways and (ii) AA-RagGTPase-ragulator-vATPAse-mTORC1. Both the pathways function independent of each other and yet both are necessary for the activation of mTORC1. Thus, knockdown of hVps34 and PLD1 decreased mTORC1 activity but not its translocation to the lysosomes, whereas knockdown of Rag-ragulator complex decreased both the translocation and hence the activity of the complex [71, 72].
Fig 7: PLD1 interaction with hVps34 is critical for AA induced mTORC1 activation. Figure from Yoon et.al., 2011 
(f) RalA: RalA and RalB are members of Ras superfamily of small G proteins and it is involved in vesicle sorting, gene expression, cell morphology and cell proliferation. For long Ras GTP was considered to be the only activator of Ral where Ras GTP binds to Ral GEFs and delivers them to Ral leading to its activation. Later PDK1 activated by PI3K after growth factor binding was also shown to activate RalGDS (a member of Ral GEF) via Ras activation [83, 84]. Following this several Ras independent mechanisms of RalGDS/Ral activation were also discovered. (i) β-arrestin exists in an inactive complex with RalGDS and the dissociation of this complex by a chemoattractant peptide formyl-Met-Leu-Phe (fMLP) resulted in activation of Ral via RalGDS. (ii) Ca2+ influx independent of Ras is shown to activate RalGDS/RalA . RalA also contains a calmodulin binding domain, but the involvement of it in the activation of Ral is yet to be uncovered . A more recently discovered role of RalA is in nutrient, particularly AA, sensing of mTORC1, where it was shown that AA induce GTP loading of RalA and not RalB and thus regulate mTORC1 activation with response to AAs. Knockdown of RalA significantly decreased AA- and glucose- induced mTORC1 activity while insulin-stimulated mTOR activity had no effect on RalA knockdown. Also knockdown of RalA GEF and RalA GDS produced similar effect, concluding that the effect of RalA on mTORC1 is specific to nutrients and not to insulin (Fig. 8). mTORC1 activation by hyperactive mutant of Rheb was also suppressed when RalA was knocked down. Overexpression of constitutively activated RalA partially rescued the inhibition on S6K caused by Rheb knockdown [19, 31, 72, 87]. Although so much is known on nutrient sensing by RalA the exact mechanism by which RalA integrates with the Rheb-mTORC1 pathway is yet to be discovered.
Fig 8: A model of action of RalGDS and RalA in nutrient sensing of mTORC1. Figure from Maehama et. al., 2008 
C.2.5 Downstream targets of mTORC1:
The two best known downstream targets of mTORC1 are the ribosomal protein S6 kinase 1 (S6K1) and eukaryotic translation initiation factor 4E binding protein 1 (4E-BP1) both of which are important translation regulators [31, 88, 89].
(a) S6K1: S6K is a member of AGC family of serine/threonine protein kinases like PKB and PKC. S6K exists in two forms, S6K1 and S6K2. S6K2 is recently identified and shares almost 82% homology identity to S6K1 and is a poorly studied gene. S6K1 gene have two isoforms, a long and a short isoform. The long isoform (p85S6K1) has 23 aminoacids extra in its amino terminal that codes for the nuclear localization signal while the short isoform (p70S6K/p70S6K1) is predominantly present in the cytoplasm. S6K2 gene similar to S6K1 has two isoforms, long (p85S6K2) and short (p54S6K2). mTORC1 activates p70S6K as one of its downstream targets. Activation of p70S6K requires phosphorylation at eight residues where phosphorylation at three residues (Thr229 – in the catalytic domain, Thr389 and Ser371 – in the linker domain) have been shown to be vital for the kinase activity. Thr229 is phosphorylated by active PDK1 which also phosphorylates PI3K. But for this phosphorylation to take place, p70S6K must first be phosphorylated at Thr389 and Ser371 by mTORC1 in order to provide access to Thr229 [2, 90-94]. Activated p70S6K1 phosphorylates ribosomal protein S6 (a component of 40S ribosomal subunit) and aid in translation.
(b) Ribosomal protein S6 (rpS6): Ribosomal S6 protein is a downstream substrate for S6K1 protein kinase. Mammalian ribosomes are composed of 40S (small) and 60S (large) subunits. 40S subunit contains 18S rRNA and 33 proteins whereas 60S subunit contains 5S, 5.8S, 28S rRNA and 46 proteins. rpS6 is evolutionary conserved protein from yeast to vertebrates. Conditional knockout of rpS6 in mouse hepatocytes failed to synthesize 40S subunit and did not progress through cell cycle and proved to be an indispensable ribosomal protein. rpS6 was the first ribosomal protein known to undergo inducible phosphorylation. There are several known physiological, pathological and pharmacological stimuli that has been demonstrated to phosphorylate rpS6. The phosphorylation of the 32KDa protein has been noticed in both cytoplasm and nucleus. Phosphorylation of ribosomal S6 proteinhappens at 5 evolutionarily conserved serine residues in response to various stimuli. The five serine clustered residues are S235, S236, S240, S244, and S247. Phosphorylation of all five residues is essential for full activation of rpS6. MAPK pathway has been shown to induce partial phosphorylation of rpS6 at S235 and S236 residues, whereas mTORC1 pathway has shown to phosphorylated rpS6 at all 5 residues. Recently it is proposed to use phospho-rpS6 (Ser240/244) as a diagnostic biomarker for PI3K/TORC1/S6K in tissue samples from tumor biopsies or transplants .
(c) 4EBP-1: Under unphosphorylated conditions 4E-BP1 binds to eIF4E (eukaryotic translation initiation factor 4E) thereby preventing the recruitment of 40S ribosomal subunit to 5‘capped mRNAs. Once mTORC1 phosphorylates 4E-BP1, it releases its interaction with eIF4E, thus aiding in translation [28, 96].
C.3 Calcium signaling:
Calcium is necessary for most cellular processes and calcium ions impact nearly every aspect of cellular life. They play a critical role in gene regulation, proliferation, secretion, differentiation, contraction, metabolism, trafficking, cell death, cell motility, fertilization, hormone responses etc. All these processes are executed by cytosolic calcium variations which acts as signals for complex signaling system. Cytosolic calcium concentrations are precisely controlled in space and time. Calcium signaling system comprises of an extensive toolkit of signaling components which can be used to create a store house of signaling units. Each cell type has their own set of calcium signaling components in order to control calcium signals precisely in space and time [97, 98]. Calcium signaling process consists of various steps:
1) Generation of calcium signals – Cytosolic calcium is maintained at a very low concentration of around 100nM when compared to extracellular surface which has around 1.5mM calcium. In order to generate a calcium signal, cell uses both its internal and external sources of calcium (Table 1). Calcium signals can be transient or in the form of a wave. This is generated by the activation or opening of plasma membrane calcium channel or calcium release channels from intracellular stores, which causes an initial increase in cytoplasmic calcium levels. This is called the rising phase of calcium signals. The shape and duration of the calcium signals depends on the calcium buffers and the calcium extrusion signals in the cell. At any point, calcium concentration depends on calcium influx, calcium efflux from the cell and the action of calcium buffers .
2) Decoding calcium signals – Upon increase in intracellular calcium, it binds to calcium binding proteins like Calmodulin (CaM) which acts like calcium sensor. Typically calcium binding to its sensor causes a conformational change in the sensor, which targets series of downstream signaling proteins. In few other cases, calcium can also bind some protein kinases like PKC, which has a calcium binding region in its structure and hence transmits signal from calcium directly to its downstream effectors .
|Cytosol||~ 100 nM|
|Extracellular surface||~ 1.5 – 2 mM|
|Endoplasmic reticulum||~ 0.5 – 2 mM|
|Endolysosomes||~ 0.5 mM|
Table 1: Table describing calcium concentration in different regions of the cell.
Types of calcium signals:
1) Elementary calcium signals – Elementary calcium signals or fundamental mechanism of calcium signals is produced by very short opening of the calcium entry or release channels leading to tiny cloud of calcium around the mouth of the channel. These signals have very small spatial range due to high buffering of cytoplasmic signals. The signals generated by ryanodine receptors are called sparks and the signals generated by IP3 receptors are called puffs. These elementary calcium signals control several local events like, release of secretory vesicles, activation of ion channels, generation of nuclear calcium signal, mitochondrial metabolism etc .
2) Calcium oscillations – Calcium oscillations deliver calcium in the cell more globally and continuously in the form of oscillations over a long period of time. Calcium oscillations has the potential to transmit different regulatory signals to different regions of the cell at the same time. Calcium oscillations may present themselves as a wave that spreads throughout the cell or as repetitive calcium spikes restricted to specific subcellular locations. Calcium oscillations and repetitive calcium spikes are important for activation of transcription factors like NFAT, NF-κB etc. The calcium oscillations predominantly depend on calcium release from the ER and its propagation depends on calcium induced calcium release from plasma membrane. The calcium waves can also be propagated via gap junctions [97, 98, 102].
Properties of calcium signals:
Extracellular stimuli like hormones, neurotransmitters and growth factors generate intracellular calcium increase and these calcium signals are very versatile. Decoding their response can happen in various subcellular levels. The calcium signals can vary in amplitude, kinetics, frequency and localization. Calcium signals can act like a global signal that spread throughout the cytoplasm or can act like a local signal only around the intracellular store of origin. These signals can last anywhere from microseconds to hours and the signals can also appear as transient signals or pulsating signals. Regulation of a calcium signal in the cell depends on calcium influx, release, buffering/binding proteins, and efflux [97, 98].
(D) RATIONALE & PREVIOUS STUDIES:
Ever since the initial discovery on the role of AAs in regulating mTORC1, several studies have been conducted to unravel the pathway of this very intricate circuitry. Some of the important pathways like the insulin/PI3K/TSC1/2 pathway of mTORC1 activation and more recently the RagGTPase pathway and PLD1 pathway of mTORC1 activation are being extensively explored. Despite the enormous amount of research in the area of AA sensing that has been done so far there are still several gaps and uncertainties that needs to be addressed. One such pathway is the calcium pathway of mTORC1 activation. The involvement of calcium in S6K activation is known since 1998 where Conus et.al., 1998 showed that in Balb/c-3T3 fibroblast cells, the activation of p70S6K is totally dependent on the intracellular calcium levels. Depletion of intracellular calcium stores by EGTA completely suppressed p70S6K activation while an increase in intracellular calcium levels using Ionomycin (Io) or Thapsigargin (Tg) resulted in complete activation of p70S6K. Also the activation of p70S6K was shown to be sensitive to Wortmannin (PI3K inhibitor) but was independent of class I PI3K, suggesting that a different class of PI3K might be involved in the process . Following this, a study from the laboratory of Dr. George Thomas showed that activation of S6K1 requires an initial calcium dependence which releases the auto-inhibitory interaction between the N- and C- terminus of the kinase allowing it to get phosphorylated . Few years later our laboratory in collaboration with Dr. George Thomas group have shown that AAs activate S6K1 by increasing the intracellular calcium levels and that this process is wortmannin sensitive and more precisely is dependent on class III PI3K also called hVps34 . However the mechanism by which AAs cause a rise in [Ca2+]i and subsequently activate mTORC1 is yet to be determined, which will be the main focus of this proposal.
There are exactly five papers that suggests a possible role for calcium in mTORC1 activation. Work done by Dr. Anton Bennett group from Yale University showed that in C2C12 myoblast cells, AA Leucine caused a robust increase in intracellular calcium levels and that this response was sensitive to chelation of intracellular calcium by BAPTA-AM. They also showed that this leucine-induced intracellular calcium increase was from ER calcium store . Following this, a study done to understand the role of taste receptors T1R1/T1R3 (GPCR) in AA sensing showed that in MIN6 pancreatic β cells, T1R1/T1R3 receptors sense AAs which then causes calcium influx into the cytosol from extracellular surface. This increase in intracellular calcium was suggested to activate mTORC1 via stimulation of the Erk/MAPK pathway . Similar to this was a study done on orexin neuropeptide, where orexin A and orexin B binding to GPCR’s namely orexin 1 & 2 receptors, activate mTORC1 which was independent of Akt and Erk signaling pathway in mouse neuron and neuronal cell model. They also suggested that, orexin binding to orexin receptors triggered increase in intracellular calcium by calcium influx from extracellular surface which showed to activate mTORC1 pathway much strongly when compared to chelating extracellular calcium using EGTA. The study also suggests a role for v-ATPase in mTORC1 activation . Another study done by Dr. Zhang group in Johns Hopkins University showed that, in NIH 3T3 cells, PDGF-induced mTORC1 activation was dependent on intracellular calcium as chelation of intracellular calcium using BAPTA-AM blocked PDGF-induced S6K1 activation. Also, treating cells with PDGF caused intracellular calcium transients which was not dependent on extracellular calcium . One of the most recent work published in 2016 in eLife journal by a group in Johns Hopkins suggested that mTORC1 activation is dependent on lysosomal calcium and Calmodulin. They suggested that lysosomal Transient receptor potential Mucolipin 1 (TRPML1) channels were necessary for mTORC1 activation. Consistent with all previous work, they showed calcium dependence of mTORC1 pathway as chelation of intracellular calcium suppressed mTORC1 pathway under basal, AA replete and insulin replete conditions. The work also supported calmodulin dependence similar to Gulati et.al., . Although these are like five different studies done to understand the role of calcium in mTORC1 activation, none of these studies are technically challenged to study calcium signaling. All of their support for calcium involvement in mTORC1 activation comes from biochemical and molecular biology studies. Some of these studies done to understand the role of calcium by using drugs targeting calcium channels in the plasma membrane, ER and lysosomes with hours of treatment, that in our experimental settings we find it to be destructive to the cell morphology and physiology. Hence, there is a compelling need to perform a comprehensive study to understand the role of calcium in mTORC1 activation and in our interest to understand the role of calcium in nutrient signaling to mTORC1.
Overall aim and Central hypothesis:
The overall aim of this present study is to investigate the role of calcium in nutrient stimulation of mTORC1. The central hypothesis was that there is a calcium dependency for mTORC1 activation which involves an increase in cytosolic calcium and this cytosolic calcium increase in response to AA treatment is mediated by calcium release from internal calcium store.
1) To determine if there is a role for calcium in AA stimulation for mTORC1 in breast cancer cell line (MDA-MB-231 cells)
2) To determine the source of calcium
3) To touch base on calcium regulation of AA-induced mTORC1 activation in primary rat hepatocytes.