Hippo and hyperplasia – TEAD promotes mTORC1 activation post-injury

Complications of atherosclerotsis remain the leading cause of death in the U.S.¹. Atherosclerotic vessels can be revascularized by angioplasty with stenting, or bypass surgery, but microtrauma to the blood vessel during these procedures can lead to a complication called intimal hyperplasia or restenosis. Restenosis results from hyper-proliferation of smooth muscle cells (SMC) in the vascular lumen². SMCs, the contractile cells that regulate vascular tone and blood pressure, exhibit the remarkable ability to dedifferentiate in response to injury. Following angioplasty, SMCs downregulate contractile apparatus proteins and migrate from the vessel wall into the lumen². SMC proliferation drives the early injury response, while SMC secretion of extracellular matrix proteins and fibrotic scar formation propagates lesion growth at later stages². Local drug delivery with drug-eluting stents (DES) is effective in reducing intimal hyperplasia in coronary arteries. In particular, mTORC1 (mechanistic target of rapamycin complex 1) inhibitors of the rapamycin family have been the most effective DES agents³. Despite these advances, intimal hyperplasia remains a major challenge in diabetic patients and in peripheral blood vessels³. Work from Osman et. al.⁴ provides new insights into the mechanisms underlying mTORC1 activation in vascular injury. The newly-identified cross-talk between the Hippo pathway, glutamine uptake, and mTORC1 suggests avenues for potential new therapeutic targets.

The mTORC1 pathway coordinates cell growth by sensing levels of key nutrients, including amino acids, energy, and growth factors. mTORC1 activity, in turn, promotes synthetic processes, including protein, lipid, and nucleotide synthesis, while repressing autophagy⁵. mTORC1 is activated by growth factors (reviewed in⁶), including PDGF, in response to vascular injury and has been shown to promote the dedifferentiated “synthetic” SMC phenotype⁷. Rapamycin inhibition of mTORC1 not only inhibits SMC proliferation and matrix synthesis, but also promotes a differentiated gene expression program in SMC^{7, 8}, which likely accounts for its improved efficacy compared to other anti-proliferative DES agents.

While mTORC1 regulates growth at the cellular level (cell size), the Hippo pathway coordinately determines organ size by regulating cell number. By integrating stimuli such as cell density, mechanical forces, cellular stress, and GPCR-mediated signaling, this pathway influences cell number by modulating proliferation and apoptosis⁹. In mammals, Mst1/2 kinases (homologues of the Drosophila Hippo protein) phosphorylate, and activate, Lats1/2 kinases. Lats1/2 subsequently phosphorylate the transcriptional coactivators YAP/TAZ, promoting their nuclear exclusion and degradation. In the absence of upstream Hippo pathway activation, YAP/TAZ bind to TEADs, among other transcription factors, to induce gene transcription. Through as-yet incompletely understood mechanisms, cell confluence activates Mst1/2, repressing YAP/TAZ/TEAD. In subconfluent conditions, YAP/TAZ/TEAD activation mediates pro-proliferative gene expression (reviewed in⁹).

Cross-talk between the mTORC1 and Hippo pathways serves to integrate cell growth/size and cell number/organ size. Each pathway has been shown to amplify the activity of the other (Figure): YAP induction of miR-29 represses PTEN, thus promoting PI3K/mTOR signaling (reviewed in⁶); similarly, mTORC1-mediated suppression of autophagy promotes YAP stability (reviewed in¹⁰). Osman et al. now reveal a new way in which these pathways intersect, with TEAD1 facilitating activation of mTORC1.

Figure: Cross-talk between mTORC1 and Hippo pathways.

Following vascular injury, growth factor signaling induces PI3K/AKT signaling to activate mTORC1, promoting biosynthetic processes and cell growth, which are necessary for proliferation. Through as-of-yet undefined stimuli, injury results in increased expression of TEAD1, a transcription factor that acts with the Hippo pathway cofactors YAP/TAZ. Cross-talk between these two pathways has previously been shown: YAP increases miR-29 expression to repress PTEN, driving AKT/mTORC1 signaling; mTORC1-mediated repression of autophagy promotes YAP activity; and new work from Osman et. al. reveals that TEAD-mediated SLC1A5 expression increases glutamine transport and, subsequently, mTORC1 activity. Positive cross-talk between mTORC1 and YAP/TEAD1 converge to promote SMC proliferation and intimal hyperplasia. AA= amino acids; Rap.= Rapamycin.

Osman et. al.⁴ show that TEAD1 is induced following endothelial denudation injury in mouse and rat models, along with mTORC1 signaling, proliferation, and SMC dedifferentiation. Notably, inducible SMC-specific knockout of TEAD1 reduced proliferation in injured vessels, resulting in smaller lesions. TEAD1 deletion also decreased mTORC1 signaling and proliferation in injured vessels, but did not rescue injury-induced contractile protein repression. The effect of TEAD1 on intimal hyperplasia was consistent with an earlier finding that YAP promotes the synthetic, dedifferentiated SMC phenotype in response to injury¹¹. While TEAD1 was essential for neointima formation, it had no apparent phenotype when deleted in adult vessels⁴. In vitro, silencing TEAD1 inhibited mTORC1 and proliferation while inducing p27^kip and contractile proteins⁴, similar to the reported effects of rapamycin treatment^{7, 8}.

This study found that the glutamine transporter solute carrier family 1 member 5 (SLC1A5) is induced following vascular injury in a TEAD1-dependent manner, with TEAD1 directly binding and transactivating the promoter in vitro⁴. SLC1A5 is required for L-glutamine-dependent activation of mTORC1¹². Furthermore, glutamine is essential to cancer cells as its metabolites provide not only a source of energy, but also nitrogen for nucleic and amino acid biosynthesis, allowing for rapid proliferation¹³. Notably, the TEAD1-SLC1A5-glutamine uptake signaling axis was shown to regulate SMC mTORC1 activity, de-differentiation and proliferation in vitro. Overexpression of TEAD1 activated mTORC1 to a level comparable to PDGF-BB stimulation and potentiated the effects of PDGF-BB. Treatment with the SLC1A5 inhibitor GPNA demonstrated that this transporter is required for TEAD1-induced SMC mTORC1 activation and proliferation. The major novel mechanistic finding is that TEAD1 provides a direct transcriptional link between the Hippo pathway and glutamine-driven activation of mTORC1 signaling, adding new depth to our understanding of the interplay between these pathways. This work reveals a new metabolic mechanism by which rapidly proliferating and highly synthetic SMC obtain a key nutrient, glutamine, which coordinately regulates and provides fuel for the biosynthetic demands of vascular repair and neointima formation.

Several questions arise from this study for future research. TEAD1 activity amplifies the mTORC1 signaling that is likely initiated by growth factors such as PDGF at sites of vascular injury, but the stimuli that promote repression of upstream Hippo kinases in vascular injury are unknown. Depending on cell type and context, growth factors, cytokines, GPCR ligands, cellular stresses, and disruption of cell-cell contacts can activate YAP/TAZ (reviewed in⁹). The complexity of potential stimuli and lack of clearly defined agonists and receptors necessitates reliance on overexpression and knockdown approaches to study Hippo pathway functions, which raises the possibility of artifacts due to high level overexpression, and/or lack of other concomitant signaling interactions in the absence of native stimuli. The mechanisms that mediate TEAD1 upregulation post-injury remain to be determined, but TEAD factors can be regulated by phosphorylation, palmitoylation, and, similar to YAP/TAZ, TEADs can be excluded from the nucleus under conditions of high cell density¹⁴. The specific cofactors with which TEAD1 partners to regulate SLC1A5 and SMC phenotype are also not yet known. The similar phenotypes shared between YAP and TEAD1, as well as YAP regulation of SLC1A5 in cancer cells¹⁵ suggests that they likely act in concert in vascular injury response. TEADs, however, can additionally partner with Hippo-independent cofactors¹⁴. The full spectrum of TEAD1-dependent target genes in the injury setting is not yet known, but RNA-seq, ideally paired with ChIP-seq, may provide future insights.

From a translational standpoint, this work suggests that inhibition of TEAD1 activity and/or of downstream glutamine transport, may synergize with mTORC1 inhibition, representing a novel combinatorial strategy for treating vasculopathies. Targeting glutamine metabolism is an area of intensive research as “glutamine addiction” can confer tumor resistance to mTOR inhibitors. Inhibition of glutamine uptake, however, has been problematic, as GPNA and other SLC1A5 inhibitors have failed in cancer clinical trials due to adverse effects of glutamine deprivation in healthy cells (reviewed in¹³). Inhibitors of glutaminase (GLS), the enzyme that converts glutamine to glutamate, were in early clinical trials as of 2018, and inhibitors of glutamate dehydrogenase (GLUD) are in development. A more detailed understanding of glutamine metabolism in SMC in response to injury will help determine the therapeutic utility of targeting this process in intimal hyperplasia or other SMC pathologies. Glutamine-dependent import of leucine through SLC7A5/SLC3A2 is another key mechanism by which glutamine promotes mTORC1 activity¹² and another potential cancer drug target¹³. While SLC7A5, SLC3A2, and other solute carrier family members were not regulated at the mRNA level in TEAD1-deficient SMC⁴, it remains to be determined whether these or other components of cellular glutamine metabolism (GLS, GLUD, etc) may be regulated by other mechanisms post-injury to influence mTORC1 activity. Finally, in order to determine whether targeting TEAD1 and/or glutamine metabolism could be a viable therapeutic strategy for next generation DES, further studies on the expression and functions of these factors in other relevant cell types, as well as in comorbid conditions such as diabetes, will be required.

The Hippo pathway has been implicated in other pathologies, including cancer and fibrosis, and is a target in regenerative medicine⁹. The identification of a new level of interplay between these pathways will likely have implications beyond the vascular injury response and may suggest novel combinatorial therapies.