TSC2 S1365A mutation potently regulates CD8+ T cell function and differentiation and improves adoptive cellular cancer therapy.

MTORC1 integrates signaling from the immune microenvironment to regulate T cell activation, differentiation, and function. TSC2 in the tuberous sclerosis complex tightly regulates mTORC1 activation. CD8+ T cells lacking TSC2 have constitutively enhanced mTORC1 activity and generate robust effector T cells; however, sustained mTORC1 activation prevents generation of long-lived memory CD8+ T cells. Here we show that manipulating TSC2 at Ser1365 potently regulated activated but not basal mTORC1 signaling in CD8+ T cells. Unlike nonstimulated TSC2-KO cells, CD8+ T cells expressing a phosphosilencing mutant TSC2-S1365A (TSC2-SA) retained normal basal mTORC1 activity. PKC and T cell receptor (TCR) stimulation induced TSC2 S1365 phosphorylation, and preventing this with the SA mutation markedly increased mTORC1 activation and T cell effector function. Consequently, SA CD8+ T cells displayed greater effector responses while retaining their capacity to become long-lived memory T cells. SA CD8+ T cells also displayed enhanced effector function under hypoxic and acidic conditions. In murine and human solid-tumor models, SA CD8+ T cells used as adoptive cell therapy displayed greater antitumor immunity than WT CD8+ T cells. These findings reveal an upstream mechanism to regulate mTORC1 activity in T cells. The TSC2-SA mutation enhanced both T cell effector function and long-term persistence/memory formation, supporting an approach to engineer better CAR-T cells for treating cancer.

Single serine on TSC2 exerts biased control over mTORC1 activation mediated by ERK1/2 but not Akt.

Tuberous sclerosis complex-2 (TSC2) negatively regulates mammalian target of rapamycin complex 1 (mTORC1), and its activity is reduced by protein kinase B (Akt) and extracellular response kinase (ERK1/2) phosphorylation to activate mTORC1. Serine 1364 (human) on TSC2 bidirectionally modifies mTORC1 activation by pathological growth factors or hemodynamic stress but has no impact on resting activity. We now show this modification biases to ERK1/2 but not Akt-dependent TSC2-mTORC1 activation. Endothelin-1-stimulated mTORC1 requires ERK1/2 activation and is bidirectionally modified by phospho-mimetic (S1364E) or phospho-silenced (S1364A) mutations. However, mTORC1 activation by Akt-dependent stimuli (insulin or PDGF) is unaltered by S1364 modification. Thrombin stimulates both pathways, yet only the ERK1/2 component is modulated by S1364. S1364 also has negligible impact on mTORC1 regulation by energy or nutrient status. In vivo, diet-induced obesity, diabetes, and fatty liver couple to Akt activation and are also unaltered by TSC2 S1364 mutations. This contrasts to prior reports showing a marked impact of both on pathological pressure-stress. Thus, S1364 provides ERK1/2-selective mTORC1 control and a genetic means to modify pathological versus physiological mTOR stimuli.

MTORC1-Regulated Metabolism Controlled by TSC2 Limits Cardiac Reperfusion Injury.

RATIONALE: The mTORC1 (mechanistic target of rapamycin complex-1) controls metabolism and protein homeostasis and is activated following ischemia reperfusion (IR) injury and by ischemic preconditioning (IPC). However, studies vary as to whether this activation is beneficial or detrimental, and its influence on metabolism after IR is little reported. A limitation of prior investigations is their use of broad gain/loss of mTORC1 function, mostly applied before ischemic stress. This can be circumvented by regulating one serine (S1365) on TSC2 (tuberous sclerosis complex) to achieve bidirectional mTORC1 modulation but only with TCS2-regulated costimulation. OBJECTIVE: We tested the hypothesis that reduced TSC2 S1365 phosphorylation protects the myocardium against IR and is required for IPC by amplifying mTORC1 activity to favor glycolytic metabolism. METHODS AND RESULTS: Mice with either S1365A (TSC2SA; phospho-null) or S1365E (TSC2SE; phosphomimetic) knockin mutations were studied ex vivo and in vivo. In response to IR, hearts from TSC2SA mice had amplified mTORC1 activation and improved heart function compared with wild-type and TSC2SE hearts. The magnitude of protection matched IPC. IPC requited less S1365 phosphorylation, as TSC2SE hearts gained no benefit and failed to activate mTORC1 with IPC. IR metabolism was altered in TSC2SA, with increased mitochondrial oxygen consumption rate and glycolytic capacity (stressed/maximal extracellular acidification) after myocyte hypoxia-reperfusion. In whole heart, lactate increased and long-chain acylcarnitine levels declined during ischemia. The relative IR protection in TSC2SA was lost by lowering glucose in the perfusate by 36%. Adding fatty acid (palmitate) compensated for reduced glucose in wild type and TSC2SE but not TSC2SA which had the worst post-IR function under these conditions. CONCLUSIONS: TSC2-S1365 phosphorylation status regulates myocardial substrate utilization, and its decline activates mTORC1 biasing metabolism away from fatty acid oxidation to glycolysis to confer protection against IR. This pathway is also engaged and reduced TSC2 S1365 phosphorylation required for effective IPC. Graphic Abstract: A graphic abstract is available for this article.

Phosphorylation Modifications Regulating Cardiac Protein Quality Control Mechanisms.

Many forms of cardiac disease, including heart failure, present with inadequate protein quality control (PQC). Pathological conditions often involve impaired removal of terminally misfolded proteins. This results in the formation of large protein aggregates, which further reduce cellular viability and cardiac function. Cardiomyocytes have an intricately collaborative PQC system to minimize cellular proteotoxicity. Increased expression of chaperones or enhanced clearance of misfolded proteins either by the proteasome or lysosome has been demonstrated to attenuate disease pathogenesis, whereas reduced PQC exacerbates pathogenesis. Recent studies have revealed that phosphorylation of key proteins has a potent regulatory role, both promoting and hindering the PQC machinery. This review highlights the recent advances in phosphorylations regulating PQC, the impact in cardiac pathology, and the therapeutic opportunities presented by harnessing these modifications.

Targeting Protein Kinase G to Treat Cardiac Proteotoxicity.

Impaired or insufficient protein kinase G (PKG) signaling and protein quality control (PQC) are hallmarks of most forms of cardiac disease, including heart failure. Their dysregulation has been shown to contribute to and exacerbate cardiac hypertrophy and remodeling, reduced cell survival and disease pathogenesis. Enhancement of PKG signaling and PQC are associated with improved cardiac function and survival in many pre-clinical models of heart disease. While many clinically used pharmacological approaches exist to stimulate PKG, there are no FDA-approved therapies to safely enhance cardiomyocyte PQC. The latter is predominantly due to our lack of knowledge and identification of proteins regulating cardiomyocyte PQC. Recently, multiple studies have demonstrated that PKG regulates PQC in the heart, both during physiological and pathological states. These studies tested already FDA-approved pharmacological therapies to activate PKG, which enhanced cardiomyocyte PQC and alleviated cardiac disease. This review examines the roles of PKG and PQC during disease pathogenesis and summarizes the experimental and clinical data supporting the utility of stimulating PKG to target cardiac proteotoxicity.

PKG1α Cysteine-42 Redox State Controls mTORC1 Activation in Pathological Cardiac Hypertrophy.

RATIONALE: Stimulated PKG1α (protein kinase G-1α) phosphorylates TSC2 (tuberous sclerosis complex 2) at serine 1365, potently suppressing mTORC1 (mechanistic [mammalian] target of rapamycin complex 1) activation by neurohormonal and hemodynamic stress. This reduces pathological hypertrophy and dysfunction and increases autophagy. PKG1α oxidation at cysteine-42 is also induced by these stressors, which blunts its cardioprotective effects. OBJECTIVE: We tested the dependence of mTORC1 activation on PKG1α C42 oxidation and its capacity to suppress such activation by soluble GC-1 (guanylyl cyclase 1) activation. METHODS AND RESULTS: Cardiomyocytes expressing wild-type (WT) PKG1α (PKG1αWT) or cysteine-42 to serine mutation redox-dead (PKG1αCS/CS) were exposed to ET-1 (endothelin 1). Cells expressing PKG1αWT exhibited substantial mTORC1 activation (p70 S6K [p70 S6 kinase], 4EBP1 [elF4E binding protein-1], and Ulk1 [Unc-51-like kinase 1] phosphorylation), reduced autophagy/autophagic flux, and abnormal protein aggregation; all were markedly reversed by PKG1αCS/CS expression. Mice with global knock-in of PKG1αCS/CS subjected to pressure overload (PO) also displayed markedly reduced mTORC1 activation, protein aggregation, hypertrophy, and ventricular dysfunction versus PO in PKG1αWT mice. Cardioprotection against PO was equalized between groups by co-treatment with the mTORC1 inhibitor everolimus. TSC2-S1365 phosphorylation increased in PKG1αCS/CS more than PKG1αWT myocardium following PO. TSC2S1365A/S1365A (TSC2 S1365 phospho-null, created by a serine to alanine mutation) knock-in mice lack TSC2 phosphorylation by PKG1α, and when genetically crossed with PKG1αCS/CS mice, protection against PO-induced mTORC1 activation, cardiodepression, and mortality in PKG1αCS/CS mice was lost. Direct stimulation of GC-1 (BAY-602770) offset disparate mTORC1 activation between PKG1αWT and PKG1αCS/CS after PO and blocked ET-1 stimulated mTORC1 in TSC2S1365A-expressing myocytes. CONCLUSIONS: Oxidation of PKG1α at C42 reduces its phosphorylation of TSC2, resulting in amplified PO-stimulated mTORC1 activity and associated hypertrophy, dysfunction, and depressed autophagy. This is ameliorated by direct GC-1 stimulation.

PKG1-modified TSC2 regulates mTORC1 activity to counter adverse cardiac stress.

The mechanistic target of rapamycin complex-1 (mTORC1) coordinates regulation of growth, metabolism, protein synthesis and autophagy1. Its hyperactivation contributes to disease in numerous organs, including the heart1,2, although broad inhibition of mTORC1 risks interference with its homeostatic roles. Tuberin (TSC2) is a GTPase-activating protein and prominent intrinsic regulator of mTORC1 that acts through modulation of RHEB (Ras homologue enriched in brain). TSC2 constitutively inhibits mTORC1; however, this activity is modified by phosphorylation from multiple signalling kinases that in turn inhibits (AMPK and GSK-3ß) or stimulates (AKT, ERK and RSK-1) mTORC1 activity3-9. Each kinase requires engagement of multiple serines, impeding analysis of their role in vivo. Here we show that phosphorylation or gain- or loss-of-function mutations at either of two adjacent serine residues in TSC2 (S1365 and S1366 in mice; S1364 and S1365 in humans) can bidirectionally control mTORC1 activity stimulated by growth factors or haemodynamic stress, and consequently modulate cell growth and autophagy. However, basal mTORC1 activity remains unchanged. In the heart, or in isolated cardiomyocytes or fibroblasts, protein kinase G1 (PKG1) phosphorylates these TSC2 sites. PKG1 is a primary effector of nitric oxide and natriuretic peptide signalling, and protects against heart disease10-13. Suppression of hypertrophy and stimulation of autophagy in cardiomyocytes by PKG1 requires TSC2 phosphorylation. Homozygous knock-in mice that express a phosphorylation-silencing mutation in TSC2 (TSC2(S1365A)) develop worse heart disease and have higher mortality after sustained pressure overload of the heart, owing to mTORC1 hyperactivity that cannot be rescued by PKG1 stimulation. However, cardiac disease is reduced and survival of heterozygote Tsc2S1365A knock-in mice subjected to the same stress is improved by PKG1 activation or expression of a phosphorylation-mimicking mutation (TSC2(S1365E)). Resting mTORC1 activity is not altered in either knock-in model. Therefore, TSC2 phosphorylation is both required and sufficient for PKG1-mediated cardiac protection against pressure overload. The serine residues identified here provide a genetic tool for bidirectional regulation of the amplitude of stress-stimulated mTORC1 activity.