Disruption of NF-κB-Mediated Copper Homeostasis Sensitizes Breast Cancer to Cuproptosis

Abstract

Copper plays a key role in inflammation and recent tumorigenesis. However, copper homeostasis and its role in cuproplasia and cuproptosis for cancer intervention remain incompletely explored. Here, it is unveiled that copper enhances the NF-κB pathway by directly binding to transforming growth factor β-activated kinase 1 (TAK1), thereby promoting TRAF2 interaction with and mediation of TAK1 ubiquitination and activation, leading to IκB kinase β (IKKβ) activation and mediating copper's inflammatory and oncogenic functions. Notably, copper is indispensable for TNFα/LPS-induced NF-κB activation and subsequent PD-L1 promotion. Thus, copper chelators offer protection against acute infection in murine models. Meanwhile, NF-κB represses copper uptake by negatively controlling the expression of copper transporter 1 (CTR1) transcriptionally, providing a negative feedback regulation for maintaining copper homeostasis. As a result, targeting NF-κB appears to elevate CTR1 expression, leading to excessive copper uptake and downstream MAPK and AKT activation, in turn, conferring resistance to anti-NF-κB therapies. Therefore, disruption of NF-κB not only synergizes with copper chelators to overcome drug resistance and cuproplasia, but also combines with copper ionophores to facilitate cuproptosis, providing a dual approach for combating chronic inflammation-driven cancers.

Keywords: Copper/CTR1; NF‐κB; PD‐L1; TAK1; breast cancer therapy.

Figure 1

Copper activates NF‐κB pathway to mediate its oncogenic functions. A) MDA‐MB‐231 cells were infected with lentiviruses encoding shCTR1 and then selected with puromycin (1 µg mL⁻¹). The resulting cells (MDA‐MB‐231 shCTR1‐tet on) were treated with or without doxycycline (1 µg mL⁻¹) for 72 h and were collected for RNA extracting and sequencing (accession code: PRJCA027211). B) T‐47D cells were disposed with serum‐free medium and then treated with different concentrations of CuSO₄ for indicated time point. The cells were collected and lysed for immunoblot (IB) analysis (left panel). The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (right panel). C) HEK293T cells were transfected with NF‐κB luciferase reporter plasmid and then disposed with CuSO₄ (50 µm) or/and TTM (50 µm) for 16 h. The cells were collected and lysed for the detection of luciferase activity. The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001). D) MDA‐MB‐231 cells were cultured in serum‐free medium for 16 h. Then the cells were pretreated with TTM (100 µm) for 1 h and disposed with different concentrations of CuSO₄ for 1 h. The resulting cells were subjected to IB analysis (left panel). The results were normalized and analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (right panel). E) MDA‐MB‐231 cells were cultured in serum‐free medium and treated with distinct concentrations of CuSO₄ (25 µm, 50 µm) and TTM (50 µm) for 18 h and then subjected to IB analysis (left panel). The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (right panel). F) MDA‐MB‐231 shCTR1‐tet on cells were treated with or without diverse concentrations of doxycycline for 72 h. Then the resulting cells were collected and lysed for IB analysis (left panel). The results were normalized and analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (right panel). G,H) T‐47D cells were disposed in serum‐free medium for 18 h and pretreated with TTM for 1 h, and then diverse concentrations of CuSO₄ were added to the cells for another 2 h. The cells were subjected for immunofluorescence (IF) analysis (G). Bar indicates 10 µm. The cells with nuclear p65 were counted. The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (H). I) MDA‐MB‐231 cells were treated in the presence or absence of CuSO₄ (50 µm) or/and TTM (50 µm) for 16 h. The cells were collected for RNA extraction and then subjected to qRT‐PCR analysis. The results were analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). J,K) MDA‐MB‐231 shCTR1‐tet on cells were infected with lentiviruses encoding IKKβ and then selected with hygromycin (100 µg mL⁻¹). The resulting cells were treated with or without doxycycline (1 µg mL⁻¹) for 72 h and then subjected to IB analysis (J, left panel), colony formation (K, top panel) and soft agar assay (K, bottom panel). Relative protein expression (J, right panel) and relative colony numbers (K, right panel) were normalized and plotted. The results were analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001).

Figure 2

Copper is essential for TNFα/LPS‐induced NF‐κB activation. A) MDA‐MB‐231 shCTR1‐tet on cells were treated with or without doxycycline (1 µg mL⁻¹) for 72 h and then the cells were disposed with TNFα for diverse time points. The resulting cells were subjected to IB analysis (left panel). The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (right panel). B) MDA‐MB‐231 cells were pretreated with TTM (100 µM) for 1 h and disposed with distinct concentrations of TNFα for 1 h and then subjected to IB analysis (left panel). The results were normalized and analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (right panel). C) HEK293T cells were transfected with NF‐κB luciferase reporter plasmid and then disposed with TNFα (10 ng mL⁻¹) or LPS (0.5 µg mL⁻¹) in combination with or without TTM (50 µM) for 16 h. The cells were collected and lysed for the detection of luciferase activity. The results were normalized and analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). D) MDA‐MB‐231 cells were pretreated with TTM (100 µm) for 1 h and then disposed with TNFα (20 ng mL⁻¹) and LPS (10 µg mL⁻¹) for 1 h, respectively. The cells were fixed, permeated and subjected for IF staining (left panel). Bar indicates 10 µm. The cells with nuclear p65 were counted. The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (right panel). E) MDA‐MB‐231 shCTR1‐tet on cells were treated with or without doxycycline (1 µg mL⁻¹) for 72 h and then treated with TNFα (20 ng mL⁻¹) and LPS (10 µg mL⁻¹) for 1 h, respectively. The cells were fixed, permeated and subjected for IF staining (top panel). Bar indicates 10 µm. The cells with nuclear p65 were counted. The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (bottom panel). F) Ctr1^‐/‐ and counterpart MEFs cells were treated with TNFα (50 ng mL⁻¹) and LPS (10 µg mL⁻¹) for 1 h, respectively. The resulting cells were fixed, permeated, and subjected for IF staining (left panel). Bar indicates 50 µm. The cells with nuclear p65 were counted. The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (right panel). G) MDA‐MB‐231 cells were treated with TNFα (10 ng mL⁻¹) in the presence or absence of TTM (50 µM) for 15 h. The cells were collected for RNA extraction and then subjected to qRT‐PCR analyses. The results were analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001). H) The time schematic of the female C57BL/6 mice disposed with LPS or/and TTM. I–K) The female C57BL/6 mice (6/each group) were pretreated with indicated concentration of TTM and then disposed with LPS (10 mg kg⁻¹) or/and TTM. After 6 h, the lung tissues were collected and utilized for H&E staining (J) and IB analyses (K, left panel). Relative protein expression was normalized and plotted (K, right panel). Survivals of the mice were recorded every day until no mice lived in the group of treated with LPS alone. Survival curve was drawn by GraphPad Prism version 6.0 (I) and the results were analyzed using Log‐rank (Mantel‐Cox) test (**p < 0.01). Control group (blue), LPS group (red), TTM group (green), LPS/TTM group (violet). Bar indicates 100 µm. The results in (K) were analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001). L) A proposed model for TNFα or LPS‐induced activation of NF‐κB pathway mediated by copper. Copper has a key role in TNFα or LPS‐induced activation of NF‐κB pathway, whereas, copper chelator TTM suppresses activation of NF‐κB pathway induce by TNFα or LPS.

Figure 3

Copper activates NF‐κB pathway through binding with and activating TAK1‐mediated IKKβ phosphorylation. A,B) HEK293T cells were transfected with indicated construct and then were treated with CuSO₄ (50 µm) and TTM (100 µm) for 1 h after disposed in serum‐free medium for 16 h. The resulting cells were collected and subjected to IB analysis (A). The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (B). C,D) MDA‐MB‐231 were infected with lentiviruses encoding shTAK1 and then selected with puromycin (1 µg mL⁻¹). The resulting cells and control cells were disposed in serum‐free medium for 16 h and then treated with CuSO₄ (100 µM) for 1 h. The cells were collected and subjected to IB analysis (C). Relative protein expression was normalized and plotted (D). The results were analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001). E,F) MDA‐MB‐231 cells were disposed in serum‐free medium and treated with CuSO₄ (50 µm) or/and TTM (50 µm) for 14 h. The resulting cells were collected and lysed. The lysate was incubated with anti‐TAK1 antibody and protein A/G agarose for 12 h a 4 °C. The agaroses were washed with NETN buffer (20 mm Tris‐HCl (PH 8.0), 150 mm NaCl, 0.5% NP‐40, 1 mm EDTA) for 4 times and then subjected to IB analysis (E). Relative protein expression was normalized and plotted (F). The results were analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001). G) The amino acid sequence of human MEK1 and TAK1 was aligned using Clustal W. Blue letters: amino acids. Blue letters in yellow background represented copper‐binding conserved sites between MEK1 and TAK1. Spatial structure and the intervening space (Å) of interaction of TAK1 containing amino acids H154 and M196 with copper were performed by Schrodinger software package. H) HEK293T cells were transfected with indicated constructs, resulting cells were harvested for copper pull‐down and then subjected to IB analysis (top panel). The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (bottom panel). I) The fragment TAK1‐N containing 300 amino acids derived from N terminal of human TAK1 and the mutant fragment TAK1‐N‐2A containing H154A and M196A sites were inserted into the vector pGEX‐4T1. The proteins TAK1‐N and TAK1‐N‐2A were purified from bacteria (E. coli strain BL21). The GST tag was cleaved using Thrombin. The resulting protein was incubated with NTA beads with or without copper for 0.5 h at 4 °C. The beads were washed using NETN buffer (20 mm Tris‐HCl (PH 8.0), 300 mm NaCl, 0.5% NP‐40, 1 mm EDTA) for 5 times and then subjected to IB analysis (top panel). Relative protein expression was normalized and plotted (bottom panel). The results were analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001). J) HEK293T cells were transfected with indicated constructs and then disposed with CuSO₄ (100 µm) individually or in combination with TTM (100 µm) for 40 mins after cultivated in serum‐free medium for 16 h, following the kinase TAK1 was immuno‐precipitated from cells above. The substrate IKKβ was immuno‐precipitated and purified from HEK293T transfected with indicated constructs. The indicated purified TAK1 and IKKβ were used for kinase reaction and then subjected to IB analysis (top panel). The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (bottom panel). K,L) HEK293T cells were cotransfected with indicated constructs. Following, the cells were disposed with CuSO₄ (50 µm) in the serum‐free medium (K) or treated with TNFα (10 ng mL⁻¹) (L) for 10 h. The cells were collected and lysed. The lysate was incubated with Ni‐NTA beads for 3 h at room temperature. The beads were washed and then subjected to IB analysis. M) HEK293T cells were transfected with the plasmid Flag‐TRAF2. The resulting cells were harvested and lysed. The cell lysate was incubated with anti‐Flag beads for 3 h at 4 °C and then the beads were washed using NETN buffer (20 mm Tris‐HCl (PH 8.0), 150 mm NaCl, 0.5% NP‐40, 1 mm EDTA) for 4 times. Following, the protein Flag‐TRAF2 was replaced from the beads using 3 x Flag peptide. The purified Flag‐TRAF2 and commercialized TAK1 were co‐incubated with anti‐TAK1 antibody and protein A/G agarose with or without CuSO₄ (50 µM) for 10 h at 4 °C and then the beads were washed using NETN buffer for 4 times and subjected to IB analysis (top panel). The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (bottom panel). N) HEK293T cells were cotransfected with indicated constructs. The resulting cells were harvested and subjected to immunoprecipitation (IP) with anti‐HA beads and IB analysis (top panel). The results were normalized and analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (bottom panel). O) HEK293T cells were transfected with NF‐κB luciferase reporter plasmid and indicated constructs. Following, the cells above were disposed with CuSO₄ (50 µm) for 16 h. The cells were collected and lysed for the detection of luciferase activity. The results were analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, *** p < 0.001, ****p < 0.0001). P) T‐47D cells stably expressing wild‐type or mutant TAK1 were harvested and subject to IB analysis (left panel). The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001) (right panel). Q–S) T‐47D cells expressing wild‐type or mutant TAK1 were subjected to xenograft assays. The tumor size was monitored (mean+/‐SD, n = 6) (****p < 0.0001, ANOVA test) (Q). The tumors were dissected and weighed (R) (mean+/‐SD, n = 6). (***p < 0.001, student t test). The tumors were subjected to IHC assay with indicated antibodies (S, left panel). Bar indicates 100 µm. The staining intensity was normalized, plotted (S, right panel), and analyzed (mean+/‐SD, n = 3) (**p < 0.01, ***p < 0.001, student t test). T) A proposed model for the activation of NF‐κB pathway by copper. Under CTR1 amplification, excess copper directly binds to TAK1 and strengthens the E3 ubiquitin ligase TRAF2 to bind with TAK1 to facilitate K63‐linked ubiquitination of TAK1, resulting in activation of the kinases TAK1 and IKKβ and subsequent activation of NF‐κB pathway to facilitate cytokines (TNFα, IL‐6) and PD‐L1 expressions.

Figure 4

NF‐κB transcriptionally represses CTR1 expression. A) MDA‐MB‐231 cells were treated with different concentrations of QNZ for 14 h. The resulting cells were collected and subjected to IB analysis (left panel). The results were normalized and analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01) (right panel). B) MDA‐MB‐231 cells were disposed with different concentrations of QNZ for 14 h and then collected for measuring the copper content. The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, ** p < 0.01). C) MDA‐MB‐231 cells were treated with different concentrations of NF‐κB inhibitor JSH‐23 for 14 h. The resulting cells were collected and subjected to IB analysis (left panel). The results were normalized and analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01) (right panel). D) MDA‐MB‐231 cells were disposed with NF‐κB inhibitor JSH‐23 (20 µm) or PDTC (1 µM) for 14 h and then collected for measuring the copper content. The results were normalized and analyzed using student t‐test (mean+/‐SD, n = 3) (****p < 0.0001). E) MDA‐MB‐231 cells were infected with shRNA lentiviruses targeting IKKβ and then selected with puromycin (1 µg mL⁻¹). The resulting cells were collected and subjected to IB analysis (left panel). The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ****p < 0.0001) (right panel). F) The cells from (E) were collected for measuring the copper content. The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01). G) MDA‐MB‐231 cells were treated with different concentrations of TNFα for 14 h. The resulting cells were collected and subjected to IB analysis (top panel). The results were normalized and analyzed using student t‐test (mean+/‐SD, n = 3) (**p < 0.01) (bottom panel). (H‐I) HEK293T cells were transfected with indicated construct or disposed with LPS (1 µg mL⁻¹) or/and QNZ (10 µM) for 14 h. The resulting cells were fixed, permeated and stained (H). Bar indicates 10 µm. The fluorescent intensity of CTR1 was measured and plotted (I). The results were analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01). J) MDA‐MB‐231 cells were transfected with small interfering RNAs targeting p65 for 72 h. The resulting cells were collected and subjected to IB analysis (top panel). The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, N.S., no significance) (bottom panel). K) MDA‐MB‐231 cells were transfected with small interfering RNAs targeting p65 for 72 h. Then the cells were collected for measuring the copper content. The results were normalized and analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, N.S., no significance). L) MDA‐MB‐231 cells were transfected with small interfering RNAs targeting p65 for 48 h. The resulting cells were collected for RNA extraction and then subjected to qRT‐PCR analysis. The results were analyzed using student t test (mean+/‐SD, n = 3) (**p < 0.01, ***p < 0.001). M,N) The binding affinity of p65‐SLC31A1 promoter with or without IL‐1β stimulation was acquired from GEO database using the IGV (version 2.17.0) genome browser. Green indicated SLC31A1 gene tracks and blue indicated annotation tracks (M). ChIP‐qPCR analyses of the enrichment of SLC31A1 promoter and IL‐1β promoter by p65 (N, left panel) and the products were identified by DNA gel electrophoresis (N, right panel). The results were analyzed using student t test (mean+/‐SD, n = 3) (****p < 0.0001). O) Construct of the luciferase reporter plasmid containing human SLC31A1 promoter. The promoter sequences were located at between the upstream 2438 locus and the downstream 563 locus from transcription start site (TSS) (top panel). HEK293T cells were transfected with small interfering RNAs targeting p65 for 24 h and then transfected with the luciferase reporter plasmid containing human SLC31A1 promoter or/and the indicated constructs. After 36 h of transfection, the cells were treated with or without TNFα (5 ng mL⁻¹) for 16 h. The resulting cells were collected and lysed for the detection of luciferase activity (bottom panel). The results were analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, ***p < 0.001). P) IHC analyses of human TNBC tissues and normal breast tissues with indicated antibodies (left panel). Bar indicates 100 µm. The staining intensity was normalized and plotted (right panel). The results were analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, ***p < 0.001). Q–S) IKKβ knock‐down MDA‐MB‐231 and control cells were disposed with different concentrations of TTM (Q‐R, 60 µm; S, 20 µm or 30 µm) for 72 h (apoptosis assay) or 14 days (colony formation) and then subjected to apoptosis assay (Q), IB analysis (R, left panel), and colony formation (S). Apoptotic cells, relative colony numbers and relative protein expression (R, right panel) were normalized and plotted. The results were analyzed using student t test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001). T) A proposed model for copper‐induced NF‐κB activation to suppress CTR1 transcriptional expression and subsequently reducing cellular copper level.

Figure 5

Combination of NF‐κB inhibitor and copper chelator for cancer intervention in breast cancer cells and in vivo. A,B) MDA‐MB‐231 cells were treated with indicated concentrations of QNZ (A, 8 µm) or IKK‐16 (B, 0.5 µm) and TTM (A, 30 µm; B, 40 µm) individually or in combination for 24 h, then the cell viabilities were detected and analyzed using student t‐test (mean+/‐SD, n = 3) (****p < 0.0001). Coefficient of drug in interaction (CDI) was calculated. CDI<1 means synergistic effect, CDI<0.7 means significantly synergistic effect. (C) MDA‐MB‐231 cells were treated with QNZ (10 µm) and TTM (30 µm) individually or in combination for 12 h, and the resulting cells were fixed and labeled with EdU (top panel), relative EdU‐labeled cell numbers were normalized and plotted (bottom panel). The results were analyzed using student t‐test (mean+/‐SD, n = 3) (****p < 0.0001). Bar indicates 100 µm. D,E) MDA‐MB‐231 cells were treated with indicated concentrations of QNZ and TTM individually or in combination for colony formation assay. The resulting cells were fixed and stained with crystal violet solution (D) and heat map of quantified colony numbers was plotted (E). F,G) MDA‐MB‐231 cells were treated with QNZ (8 µm) and TTM (35 µm) individually or in combination for 24 h, the resulting cells were subjected to Annexin V‐PE/7‐AAD‐labeled apoptosis assays (F, left panel) and IB analysis (G). Apoptotic cells were quantified and analyzed (F, right panel) using student t test (mean+/‐SD, n = 3) (****p < 0.0001). H–L) MDA‐MB‐231 cells were subjected to xenograft assay. The mice bearing MDA‐MB‐231 xenografts were treated with QNZ and TTM individually or in combination. The tumor size was monitored (H) (mean+/‐SD, n = 7) (***p < 0.001, ANOVA test). The tumors were dissected and weighed (I‐J) (mean+/‐SD, n = 7) (**p < 0.01, ***p < 0.001, student t‐test). The tumors were subjected to IHC assay (K, top panel) and IB analysis (L, top panel) with indicated antibodies. Red bar indicates 100 µm. Black bar indicates 50 µm. The IHC staining intensity (K, bottom panel) and relative protein expression (L, bottom panel) were measured and plotted. The results were analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001).

Figure 6

NF‐κB inhibitor synergizes with copper chelator for treating breast cancer organoids and in MMTV‐PyMT mice. A,B) Breast cancer organoids were treated with QNZ (5 µm) and TTM (20 µm) individually or in combination for 96 h, then the live and dead organoids were stained (A), the viabilities of organoids were detected and analyzed (B) using student t‐test (mean+/‐SD, n = 3) (***p < 0.001, ****p < 0.0001). Bar indicates 50 µm. C) The time schematic of female MMTV‐PyMT mice treated with QNZ and TTM individually or in combination. D–H) Female MMTV‐PyMT mice were treated with indicated concentrations of QNZ and TTM individually or in combination for indicated time point. The mice were euthanized, and then the tumors and lungs were dissected (D and G). The tumor weights (E) (n = 6) and lung lesions (G‐H) (n = 3) of all mice were recorded and analyzed using student t‐test (mean+/‐SD) (*p < 0.05, **p < 0.01, ****p < 0.0001). The mammary tumors of mice were subjected to IHC staining assay (F) with indicated antibodies. The whole lungs of all mice were sliced and disposed with H&E staining (G). Red bar indicates 2 mm. Black bar indicates 100 µm. Green bar indicates 50 µm. I) A proposed model for the potential roles of copper‐NF‐κB‐CTR1 axis in breast cancer. Under physiological conditions, copper activates NF‐κB pathway to increase PD‐L1 expression via facilitating the binding of activated TAK1 and IKKβ and further promoting IKKβ phosphorylation, whereas activated NF‐κB signaling suppresses CTR1 expression (left panel). Under NF‐κB active conditions, NF‐κB inhibitor QNZ restrains NF‐κB activation stimulated by cytokines (TNFα), LPS or viruses to reduce PD‐L1 expression, but increase CTR1 expression, further facilitating copper uptake and activating oncogenic AKT and ERK pathways, which possibly antagonizes to QNZ tumor suppressor roles (middle panel). Thus, this finding provides a strategy to combine NF‐κB inhibitor QNZ with copper chelator TTM for breast cancer therapy (right panel). The figure was generated with Biorender.com.

Figure 7

NF‐κB inhibitor synergizes with cuproptosis inducer for treating breast cancer. A) IKKβ knock‐down MDA‐MB‐231 and control cells were disposed with elesclomol (ES, 3.2 µm) for 48 h. The cell viabilities were detected and analyzed using student t test (mean+/‐SD, n = 3) (***p < 0.001, ****p < 0.0001). B) IKKβ knock‐down MDA‐MB‐231 and control cells were treated with elesclomol (ES, 250 nm) for 48 h. The resulting cells were harvested and then subjected to IB analysis. C) IKKβ knock‐down MDA‐MB‐231 and control cells were disposed with elesclomol (ES, 500 nm) for 48 h. After adding the mitochondrial dye PK Mito Deep Red (250 nm) for 2 h. The resulting cells were fixed, permeated and stained (top panel). Bar indicates 10 µm. The cell numbers with DLAT aggregation were counted and plotted (bottom panel). The results were analyzed using student t‐test (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01). D) BT‐549 cells were treated with different concentrations of QNZ and elesclomol (ES) for 48 h. The heatmap was plotted and indicated the synergistic effect of QNZ and ES. E) BT‐549 cells were treated with QNZ (10 µm) and ES (20 nM) for 48 h. The cell viabilities were detected and analyzed using student t‐test (mean+/‐SD, n = 3) (***p < 0.001, ****p < 0.0001). Coefficient of drug in interaction (CDI) was calculated. CDI<1 means synergistic effect, CDI<0.7 means significantly synergistic effect. F) BT‐549 cells were treated with QNZ (1 µm) and ES (5 nm) for colony formation assay. The resulting cells were fixed and stained with crystal violet solution (top panel) and the colony numbers were counted and plotted (bottom panel). The results were analyzed using student t‐test (mean+/‐SD, n = 3) (***p < 0.001, ****p < 0.0001). G) MDA‐MB‐231 (left panel) and BT‐549 (right panel) cells were disposed with QNZ (5 µM) and ES (20 nM) for 48 h, respectively. The resulting cells were collected and then subjected to IB analysis. H) MDA‐MB‐231 and BT‐549 cells were disposed with QNZ (5 µm) and ES (MDA‐MB‐231, 20 nm; BT‐549, 80 nm) for 48 h, respectively. After adding the mitochondrial dye PK Mito Deep Red (250 nm) for 2 h. The resulting cells were fixed, permeated and stained (left panel). Bar indicates 10 µm. The cell numbers with DLAT aggregation were counted and plotted (right panel). The results were analyzed using student t‐test (mean+/‐SD, n = 3) (***p < 0.001, ****p < 0.0001). I–N) BT‐549 cells were subjected to xenograft assay. The mice bearing BT‐549 xenografts were treated with QNZ and ES individually or in combination. The tumor size (I) and body weight of mouse (L) were monitored (mean+/‐SD, n = 5) (****p < 0.0001, ANOVA test). The tumors were dissected and weighed (J‐K) (mean+/‐SD, n = 5) (*p < 0.05, **p < 0.01, student t‐test). The tumors were subjected to IF assay (M) with indicated antibodies. The mitochondria were labelled using anti‐TOM20 antibody. Bar indicates 50 µm. The fluorescent intensity was measured and plotted (N) (mean+/‐SD, n = 3) (*p < 0.05, **p < 0.01, ***p < 0.001, student t test). O) NF‐κB inhibitor QNZ represses NF‐κB activation to increase CTR1 expression, further facilitating copper uptake. Meanwhile, cuproptosis inducer elesclomol (ES) delivers copper to mitochondria to cause cuproptosis. Combination of QNZ and ES further intensifies cuproptosis to inhibit tumor growth.

Figure 8

Model of targeting NF‐κB with copper chelator or cuproptosis inducer for cancer therapy. Copper activates NF‐κB signaling to accelerate tumor progression, and yet activated NF‐κB suppresses CTR1 expression to restrain copper uptake, whereas, NF‐κB inhibitor QNZ increases CTR1 expression to promote copper uptake to activate oncogenic AKT and ERK pathways, which antagonizes to QNZ tumor suppressor roles (top panel). Thus, while NF‐κB inhibitor QNZ have no good outcome in cancer therapy, combination of copper chelator TTM or cuproptosis inducer elesclomol (ES) provides a novel strategy for overcoming QNZ drug resistance in cancer therapy (bottom panel). The figure was generated with Biorender.com.