Le, Nhat Tu

Background: D-flow’s impact on metabolic reprogramming, particularly glycolysis, needs reevaluation due to its conflicting role in atherosclerosis. We show that d-flow reduces LATS1/2 expression, leading to both EC proliferation and senescence. However, the exact molecular mechanisms remain elusive. Methods: We utilized EC-specific Lats1 +/- /Lats2 -/- knockout (LATS1/2-EKO) mice in a partial left carotid ligation (PLCL) model under hypercholesterolemia with AAV-PCSK9 injection to mimic d-flow. Anti-CD38 antibody (Ab68) and IgG2a control were administered intraperitoneally every 5 days at a dose of 5 mg/kg starting one day before PLCL. Plaque characterization was performed using imaging mass cytometry and sequential immunofluorescence (COMET TM ), integrating spatial metabolite data from Bruker timsTOF Flex Mass Spectrometer, and spatial single-cell metabolite data via the Visiopharm platform. Results: EC LATS1/2 depletion induced a unique atherothrombosis lesion with induced EC proliferation and senescence-associated secretory phenotype (SASP). Utilizing spatial single-cell metabolomics, we cataloged 134 distinct metabolites within the ECs. Subsequent overrepresentation analysis revealed that pathways related to glutamate and the citric acid cycle were enhanced in LATS1/2-EKO mice. Specifically, CD38-dependent downregulation of the pyruvate dehydrogenase (PDH) substrate lipoamide decreased PDH enzymatic activity, linking to CD38 to senescence induction. This inhibition led to a metabolic shift towards YAP-dependent glutaminolysis, and upregulated the citric acid cycle. The use of a glutaminase 1 inhibitor (CB-839) significantly curtailed senescence-associated stemness (SAS), underscoring the pivotal role of CD38 in driving glutaminolysis that contributes to d-flow-induced SAS. Moreover, the application of Ab68 but not IgG2a control decreased atherothrombosis formation following PLCL in LATS1/2-EKO mice. However, the effectiveness of Ab68 in attenuating atherosclerosis in wild-type mice was less pronounced. Conclusions: In the presence of d-flow, upregulated CD38, resulting from reduced LATS1/2 expression, triggers EC responses characterized by increased glutaminolysis and the emergence of an EC SAS phenotype. These processes play a crucial role in a unique form of atherothrombosis. Additionally, CD38’s impact on reducing PDH activity, along with YAP and CD38-mediated enhancement of glutaminolysis, constitutes critical pathways for inducing EC SAS by d-flow.

Background: Atherosclerotic plaques, especially high-risk ones, form in blood vessel regions exposed to d-flow. EC dysfunction induced by d-flow likely contributes to plaque development. In contrast, regions with laminar flow (l-flow) are less susceptible to plaque formation. Although the Hippo pathway is implicated in mechano-transduction, the exact role and molecular mechanisms of LATS1/2 in response to d-flow remain incompletely understood. Methods: We employed endothelial cell (EC)-specific Lats1 and Lats2 knockout mice within a partial left carotid ligation (PLCL) model to simulate disturbed flow conditions. We characterized plaques using imaging mass cytometry (IMC) and sequential immunofluorescence via COMET TM . Results: We investigated the effects of d-flow on YAP activity and LATS1/2 expression. While both l-flow and d-flow activate YAP activity, only d-flow leads to decreased LATS1/2 expression. We induced Lats1/2 deletion in tamoxifen-inducible Lats1 -/- /Lats2 -/- EKO mice. Within 14 days, all mice (27/27) died due to severe systemic edema and increased vascular permeability. In Lats1 +/- /Lats2 -/- EKO mice (LATS1/2-EKO), we observed atherothrombotic lesions characterized by increased EC proliferation and a senescence-associated secretory phenotype (SASP) in vivo . Mechanistically, d-flow reduces LATS1/2 expression, leading to increased CD38 expression. CD38 triggers not only SASP, but also senescence-associated stemness (SAS) via NAD + depletion, dependent on Lamin A but independent of YAP. We also explored the relationship between CD38 and Ki67 in plaques in vivo. We calculated the logarithm of the Ki67:CD38 expression ratio at the single-cell level using IMC/COMET data. Based on the trimodal density distribution of this parameter, we found that in one group, CD38 and Ki67 were linearly co-expressed, and the percentage of LATS1/2-depleted cells in this group exceeded that of wild type (WT) cells, suggesting that CD38’s pro-proliferative effects dominate under LATS1/2 depletion conditions, despite its role in inducing senescence. Lastly, inhibiting CD38 mitigated d EC SAS/SASP, and atherothrombosis under LATS1/2 depletion conditions. Conclusions: Reduced LATS1/2-mediated CD38 expression in response to d-flow leads to an EC response marked by increased SASP/SAS, ultimately contributing to atherothrombosis. While this phenomenon is common in human advanced coronary atherosclerosis, it is not observed in commonly used mouse models of atherosclerosis.

Atherosclerotic and high-risk plaques are predominantly localized to vessel wall regions with non-laminar disturbed blood flow (d-flow), leading to the principle that d-flow-, but not laminar flow (l-flow)-driven endothelial cells (EC) dysfunction plays a role in this process. LATS1/2 and YAP are components of the Hippo signaling pathway that are involved in cellular responses to l-flow. The l-flow-mediated inhibition of LATS1/2 activity triggers YAP activation and maintains vascular structure. However, the role of Hippo signaling under d-flow remains unclear. Although both l- and d-flow increased YAP activity, only d-flow decreased LATS1/2 expression. Since the effects of l- and d-flow on EC differ, we studied the role of EC LATS1/2 using EC-specific Lats1 and 2 knock-out (EKO) mice. Within 14 days after inducing Lats1/2 deletion in tamoxifen-inducible Lats1 homo / Lats2 homo EKO mice, all the mice (27/27) died of severe systemic edema, accompanied by increased EC apoptosis and vascular permeability. We generated Lats1 he t / Lats2 homo -EKO mice ( Lats1 het /2 homo -EKO) and induced hypercholesteremia with AAV8-PCSK9 injections and a high-fat diet (HFD). After HFD, we detected plaque erosion-like plaque lesions at the d-flow areas, which revealed strongly fibrin/fibrinogen-positive organized thrombus (OTh) formation without sizeable necrotic cores. In Lats1 het /2 homo -EKO, we found significant increases in 1) EC proliferation, 2) EC apoptosis with senescent phenotype, 3) tissue factor (TF) expression, and 4) inflammation. Therefore, the phenotype observed in Lats1 het /2 homo -EKO represents a unique senescence-associated stemness (SAS) phenotype, which allows EC to escape senescence-induced cell cycle arrest. The EC proliferation induced by depletion of LATS1/2 was YAP-dependent. In contrast, EC senescence induced by CD38-mediated NAD + depletion was YAP-independent. These findings indicate that downregulating LATS1/2 expression-induced SAS leads to OTh via both YAP-dependent and independent mechanisms. OTh plays a key role in human atherosclerosis but is rarely observed in the mouse atherosclerosis model. Lats1 het /2 homo -EKO can provide a valuable platform for studying the mechanism and unique role of EC SAS-induced OTh formation.

Senescence induced by various stresses reprograms cancer cells to acquire senescence-associated (SA) cell growth (SACG), which allows them to escape senescence-induced cell cycle arrest with enhanced growth potential. ERK5 is a dual kinase-transcription factor, containing an N-terminal kinase domain and transactivation domains in the C-terminal half. Recent data have raised questions regarding the functional role of ERK5 kinase inhibitors in cancer and inflammatory diseases. We aimed to investigate how ERK5 reprograms myeloid cells (MC) to the SA secretory phenotype (SASP) and SACG, consequently leading to atherosclerosis (AthS). AthS was inhibited in ERK5 S496A knock-in (KI) mice. Single-cell analysis of p53 and Ki67 with imaging mass cytometry revealed three groups of MCs with different p53 and Ki67 expression patterns in the plaque. One group showed that p53 and Ki67 were co-expressed lineally, suggesting that this group of cells escaped from the growth-suppression effects of p53, which is also known as a senescence marker. The % of ERK5 S496A KI cells in this group was significantly lower than in wild-type (WT). Also, we found an increase of SA β-gal (SAβ-gal) + Ki67 + and 53BP1 + Ki67 + cells in bone narrow-derived macrophage (BMDMs) isolated from hypercholesterolemia (H-chol) WT mice, suggesting H-chol-induced MC SACG. MC SACG were significantly inhibited in ERK5 S496A KI mice, supporting the role of ERK5 S496 phosphorylation on SACG. RNA sequencing analysis identified the ERK5 S496 phosphorylation-dependent AHR induction by H-chol, and AHR depletion significantly inhibited SACG. We also discovered a critical effect of ERK5 S496 phosphorylation on SUMOylation at a novel site of NRF2 (i.e., K518), which inhibited NRF2 transcriptional activity without affecting ERK5 kinase activity and antagonized oxidized LDL (oxLDL)-induced SASP. Specific ERK5 kinase inhibitors (AX15836 and XMD8-92) inhibited oxLDL-induced ERK5 S496 phosphorylation, suggesting that ERK5 S496 phosphorylation was involved at least in part of the effects of these inhibitors. We have discovered a novel mechanism in which ERK5 S496 phosphorylation directly inhibits NRF2 activity by upregulating NRF2 K518 SUMOylation and AHR induction, subsequently inducing SACG and AthS.