Endothelial Cu Uptake Transporter CTR1 Senses Disturbed Flow to Promote Atherosclerosis through Cuproptosis

Abstract

Endothelial cells (ECs) lining blood vessels sense disturbed blood flow (D-flow), which drives mitochondrial dysfunction and atherosclerosis. Copper (Cu) is an essential micronutrient, and its disruption of homeostasis has been implicated in atherosclerosis. Cellular Cu levels are tightly controlled by Cu transport proteins including the Cu importer CTR1. Cuproptosis is a recently discovered form of regulated cell death triggered by mitochondrial Cu accumulation, but its endogenous stimulants and role in atherosclerosis remain unknown. Using EC-specific CTR1-deficient mice and cultured ECs, we show that endothelial CTR1 responds to D-flow by increasing mitochondrial Cu levels through its interaction with the mitochondrial Cu transporter SLC25A3 at caveolae/lipid rafts. This leads to the aggregation of lipoylated mitochondrial proteins, mitochondrial dysfunction, and cuproptosis, thereby exacerbating atherosclerosis. Importantly, mitochondria-targeted Cu-chelating nanoparticles effectively mitigate D-flow-induced cuproptosis and atherosclerosis, highlighting the endothelial CTR1-SLC25A3-mitochondrial Cu axis as a potential therapeutic target.

Figure 1.. D-flow, but not L-flow, induces Cu accumulation via CTR1 in ECs, which drives inflammation and atherosclerosis.

A. Apoe^−/− mice were fed a high fat diet (HFD) for 3 weeks. Synchrotron X-ray fluorescence microscopy (XFM) analysis of the lesser curvature (D-flow, DF) and greater curvature (L-flow, LF) of aorta. The minimal and maximal metal content displayed in micrograms per square centimeter is shown. Scale bars, 20 μm. Quantification of Copper (Cu), Iron (Fe) and Zinc (Zn) accumulation are shown in right (n=4–6). B-C. Human aortic endothelial cells (HAECs) were exposed to laminar flow (LF, 15 dyne/cm²) or disturbed flow (DF, ± 5 dyne/cm², 1 Hz frequency) conditions for various times (B). HAECs were treated with Cu chelator, Bathocuproinedisulfonic acid (BCS, 200 μM for 48 hrs) and exposed to LF and DF for 24 hrs (C). Cells incubated with the Cu-specific fluorescent sensor CF4 (2 μM) for 5 min at 37°C were imaged by confocal microscopy (n=3). D. ICP-MS measurement for Cu, Fe and Zn contents in HAEC lysates after DF or LF for 24 hrs (n=4). E. Partial carotid ligation (PCL) surgery on left carotid artery (LCA) and schematic diagram of experimental design for (F). Male and female Ctr1^WT/Apoe^−/− and Ctr1^iECKO/Apoe^−/− mice were treated with tamoxifen, followed by PCL surgery on LCA and high fat diet (HFD) for 21 days. F. Representative images of cross sections of right carotid artery (RCA) or LCA stained with hematoxylin and eosin (H&E), Oil Red O, or Mac-3 (n=6–7) Scale bar: 100 μm. Right panel: Quantification of lesion size, lipid deposition, and macrophage infiltration. G. Schematic diagram of experimental design for (H). Apoe^−/− mice were treated with Cu chelator tetrathiomolybdate (TTM) for 14 days, followed by PCL surgery on LCA and HFD for 21 days. H. Representative images of cross sections of RCA or LCA stained with H&E, Oil Red O, or Mac-3 (n=6–7) Scale bar: 100 μm. Right panel: Quantification of lesion size, lipid deposition, and macrophage infiltration.

Figure 2.. D-flow promotes mitochondrial dysfunction and cuproptosis via CTR1- and Cu-dependent mechanisms in ECs.

A. Ex vivo Seahorse assay to measure mitochondrial O₂ consumption rate (OCR) in aortic arch (D-flow, DF) and thoracic aorta (L-flow, LF) with and without endothelial removal by denudation, isolated from Apoe^−/− mice fed with HFD for 30 days (n=7). B. OCR in HAECs transfected with control siRNA or CTR1 siRNA, or pretreated with or without Cu chelator, TTM (20nM, 24 hrs) exposed to LF or DF for 48 hrs. (n=5–6). C. HAECs exposed to DF or LF for 4 hr or 24 hrs were used to measure Cu content in cytosolic and mitochondrial fraction using ICP-MS (n=3–4). D. Left: Representative images of en face immunofluorescence staining of DLAT aggregation (green). The endothelium was visualized by VE-Cadherin (VE-Cad)(Red) staining, and nuclei were counterstained with TO-Pro (blue). Right: Quantification of DLAT aggregation in the endothelial layer of aorta exposed to LF or DF (n=5). E. Left: Representative images of en face immunofluorescence staining with DLAT aggregation (red) in the endothelium of greater curvature (LF) and lesser curvature (DF) of aortic arch of human atherosclerotic aorta. The endothelium was visualized by CD31 (green) and nuclei were counterstained with DAPI (blue) (n=3). Right: Quantification of DLAT aggregation. Details of donor tissues are described in previous reports (41). Additionally, patient #3’s information is: 69/F/Caucasian/female, Respiratory failure (cause of death), and Coronary Artery disease/hypertension/kidney disease (Medial history). F. HAECs transfected with CTR1 siRNA or control siRNA were exposed to DF for 48 hrs. Representative images of en face immunofluorescence staining DLAT aggregation (green, DLAT; red, Mitotracker; blue, DAPI). G and H. HAECs were transfected with CTR1 siRNA or control siRNA or pretreated with 20 nM TTM for 24 hrs and exposed to DF for 48 hrs. Cell lysates were immunoblotted with the indicated antibodies (n=3–4). I. Schematic model showing that DF promotes Cu accumulation in cytosol and mitochondria, as well as mitochondrial dysfunction and cuproptosis in a CTR1/Cu-dependent manner in ECs.

Figure 3.. D-flow-induced mitochondrial Cu elevation contributes to cuproptosis and atherosclerosis.

A. Molecular components of mitoCDN and illustrated nanoparticle formulation. B-F. HAEC were pretreated with 1 μM mitoCDN for 1 hr and exposed to DF (D-flow) for 48 hrs. Mitochondrial O₂ consumption rate (OCR) measured by a Seahorse analyzer (n=5–6)(B). Cell death was measured by CCK8 assay (n=5) (C) or LDH release assay (n=4)(D). Confocal immunofluorescence imaging of DLAT aggregation (green, DLAT; red, Mitotracker; blue, DAPI)(n=3)(E). Cell lysates were immunoblotted with the indicated antibodies (n=3)(F). G. Apoe^−/− mice were treated with mitoCDN (1.35mg/kg) every 3 days with a total of 10 doses and fed with HFD for 30 days. Representative images of en face staining for DLAT aggregation (green), VE-Cad (endothelium, red) and DAPI (nucleus, blue) in the endothelial layer with LF (L-flow) or DF. (n=5). H. Schematic diagram of experimental design for I to L. DF was induced in the LCA of Apoe^−/− mice using PCL surgery, while the contralateral RCA was used as an internal control. Ligated mice were fed with HFD and treated with mitoCDN (i.v.) (one day prior to surgery and every three days after PCL). I and J. LCA or RCA from 3 mice were pooled and digested at 4 days after PCL. CD31⁺ ECs were isolated and used to analyze cell viability using flow cytometry with viability dye, BV605. Right: Quantification of % dead ECs. (n=3). K and L. Representative images of cross sections of RCA or LCA stained with H&E, Oil Red O, and Mac-3 at 3 weeks after PCL (n=6) Scale bar: 100 μm. Quantification of lesion size, lipid deposition, and macrophage infiltration (Right) (n=6).

Figure 4.. D-flow promotes SLC25A3 translocation to lipid rafts, interaction with CTR1, and mitochondrial Cu accumulation, leading to cuproptosis in ECs.

A and B. HAECs exposed to LF (L-flow) or DF (D-flow) for 24 hrs were used to isolate caveolae/lipid raft (C/LR) using sucrose gradient centrifugation. Fractions from the top (fraction 1) to the bottom (fraction 13) were immunoblotted with antibodies as indicated. C. Equal amounts of C/LR (fraction 5) or non-C/LR (fraction10–13) were immunoblotted with antibodies as indicated (n=5). ATP5A1, a component of ATP synthase, is used as a marker of the mitochondrial inner membrane. D. HAECs transfected with CTR1-flag plasmid were pretreated with or without Methyl-β-cyclodextrin (MβCD, 10 mM, 1 hr) and then exposed to DF for 24 hrs. Cells were immunostained for Flag (green), SLC25A3 (red) and DAPI (blue) (n=3). E. HAECs transfected with control or CTR1 siRNAs were exposed with DF for 24 hrs. Cells were immunostained with SLC25A3 (green) and DAPI (blue). Right: Quantification for % plasma membrane localization of SLC25A3 (n=3). F and G. HAECs were exposed to DF for indicated time (F) or transfected with control or caveolin-1 (Cav1) siRNAs with or without pretreated with MβCD (10 mM, 1 hr) (G). These ECs were exposed to DF for 24 hrs. Lysates were immunoprecipitated with CTR1 antibody or IgG, followed by immunoblotted with SLC25A3 or CTR1 antibody. H, I, and J. HAECs transfected with control or SLC25A3 or CTR1 siRNAs exposed to DF for 24hrs were used to measure Cu contents in cytosol and mitochondrial fractions using ICP-MS (H) or [Cu]i levels using CF4 fluorescence probe (I). (n=3–4). HAECs treated with or without MβCD (10 mM, 1 hr) exposed to DF for 24 hrs were used to measure [Cu]i levels using CF4 fluorescence probe (J)(n=3).

Figure 5.. D-flow-induced cuproptosis is mediated through CTR1-SLC25A3- and lipid rafts-dependent mechanisms.

A, B, and C. HAECs treated with or without MβCD (10 mM, 1 hr) (A) or transfected with control or SLC25A3 siRNA (B) or Cav1 siRNA (C) were exposed to DF (D-flow) or LF (L-flow) for 48 hrs. These ECs were used to measure DLAT aggregation and lipoylation (Left), and protein expression of Fe-S cluster proteins (Right) (n=3). D. Schematic summary of the proposed model. Endothelial CTR1 responds to DF by elevating mitochondrial Cu through forming a complex with the mitochondrial Cu transporter SLC25A3 at lipid rafts. This in turn promotes aggregation of lipoylated mitochondrial proteins, mitochondrial dysfunction, and cuproptosis, thereby driving atherosclerosis. Thus, the endothelial CTR1-SLC25A3-mitochondrial Cu axis is a promising therapeutic target for treatment of Cu-dependent vascular diseases such as atherosclerosis.