Abstract

   Cuproptosis, a recently defined copper‐dependent cell death pathway,
   remains largely unexplored in tumor therapies, particularly in breast
   cancer. This study demonstrates that triple‐negative breast cancer
   (TNBC) bears a relatively elevated copper levels and exhibits
   resistance to cuproptosis. Mechanistically, copper activates the AKT
   signaling pathway, which inhibits ferredoxin‐1 (FDX1), a key regulator
   of cuproptosis. AKT1‐mediated FDX1 phosphorylation not only abrogates
   FDX1‐induced cuproptosis and aerobic respiration but also promotes
   glycolysis. Consequently, the combination of AKT1 inhibitors and the
   copper ionophores synergistically alleviate TNBC tumorigenesis both in
   vitro and in vivo. In summary, the findings reveal a crucial mechanism
   underlying TNBC resistance to cuproptosis and suggest a potential
   therapeutic approach for TNBC.

   Keywords: AKT1, breast cancer, cuproptosis, FDX1, metabolic
   reprogramming
     __________________________________________________________________

   This study demonstrates that copper activates the AKT signaling
   pathway, which inhibits ferredoxin‐1 (FDX1), a key regulator of
   cuproptosis. AKT1‐mediated FDX1 phosphorylation not only abrogates
   FDX1‐induced cuproptosis and aerobic respiration but also promotes
   glycolysis. Consequently, the combination of AKT1 inhibitors and the
   copper ionophores synergistically alleviates TNBC tumorigenesis both in
   vitro and in vivo.

   graphic file with name ADVS-12-2408106-g001.jpg

1. Introduction

   Copper, an essential trace element, plays a crucial role in various
   physiological processes, including mitochondrial respiration,
   antioxidant defense, redox signaling, and kinase signaling.^[ [44]^1 ,
   [45]^2 , [46]^3 ^] Copper concentrations are typically elevated in
   tumor tissues and blood, which indicates a potential risk factor and
   prognostic indicator.^[ [47]^4 , [48]^5 , [49]^6 , [50]^7 , [51]^8 ,
   [52]^9 , [53]^10 , [54]^11 ^] Copper metabolism plays dual roles in
   tumorigenesis and cancer therapy. Cuproplasia refers to the role of
   copper in promoting cell proliferation and growth. Copper directly
   activates oncogenic signaling pathways, leading to the expression of
   key proteins such as HIF1α, GPER, and VEGF in breast cancer, and
   promoting tumorigenesis and angiogenesis.^[ [55]^4 , [56]^12 ^] Recent
   studies have highlighted the role of copper in the immune regulation of
   breast cancer, including the upregulation of PD‐L1 expression via
   activation of the STAT3 and EGFR signaling pathways, which enable
   immune evasion.^[ [57]^4 , [58]^13 ^]

   Paradoxically, copper also facilitates cell death through various
   mechanisms, including apoptosis, pyroptosis, ferroptosis, and more
   recently, cuproptosis.^[ [59]^4 , [60]^5 ^] Copper‐induced reactive
   oxygen species (ROS) can activate NLRP3 inflammasomes, leading to
   pyroptosis.^[ [61]^5 , [62]^14 , [63]^15 , [64]^16 ^] Increasing
   evidence suggests that copper affects ferroptosis.^[ [65]^17 ^] Copper
   chelators or complexes, such as disulfiram and elesclomol, disrupt
   mitochondrial homeostasis, resulting in oxidative stress and
   ferroptosis.^[ [66]^18 , [67]^19 , [68]^20 ^] Cuproptosis, a newly
   identified form of cell death, is characterized by the binding of
   copper to lipoylated proteins in the tricarboxylic acid (TCA) cycle,
   causing abnormal protein aggregation and subsequent cell death.^[
   [69]^21 ^] Despite the elevated copper levels, TNBC cells exhibit
   resistance to cuproptosis. Therefore, the regulation and therapeutic
   targeting of this novel death mechanism in tumorigenesis warrant
   further investigation. FDX1 is a crucial regulator of cuproptosis and
   induces the conversion of Cu(II) to Cu(I), which promotes mitochondrial
   proteotoxic stress via protein lipoylation and subsequent
   dihydrolipoamide S‐acetyltransferase (DLAT) aggregation.^[ [70]^21 ,
   [71]^22 , [72]^23 ^] Thus, targeting FDX1 profoundly affects cellular
   sensitivity to cuproptosis. Lactylation of METTL16 at residue K229
   promotes FDX1 accumulation through m^6A modification of FDX1 mRNA,
   followed by cuproptosis.^[ [73]^24 ^] Conversely, p38‐MAPK activation
   leads to the phosphorylation and nuclear translocation of FOXO3, which
   inhibits FDX1 expression, thereby suppressing cuproptosis.^[ [74]^25 ^]

   In our study, we discovered that excessive copper in TNBC activates
   AKT1, which correlates with cuproptosis resistance. Specifically, AKT1
   inhibited the protein lipoylation by phosphorylating FDX1 while
   simultaneously regulating tumor metabolic reprogramming, ultimately
   suppressing cuproptosis. These findings revealed a crucial mechanism
   underlying TNBC resistance to cuproptosis and suggested a potential
   therapeutic approach for TNBC.

2. Results

2.1. Triple Negative Breast Cancer is Resistant to Cuproptosis

   Copper levels are elevated in tumors and play a potential role in
   tumorigenesis, including breast cancer.^[ [75]^26 , [76]^27 ^] As a
   major copper transporter, copper transporter 1 (CTR1) is tightly
   regulated at the transcriptional level by HIF1α and C‐myc and at the
   post‐translational modification level by Nedd4‐1 and AMPK, leading to
   copper elevation in tumors.^[ [77]^28 , [78]^29 ^] Consistent with
   these findings, copper levels were significantly elevated in TNBC
   tissues compared to adjacent tissues (Figure [79]1A) and were
   positively correlated with advanced stages and poor prognosis in
   patients (Figure [80]1B,C). TNBC cells exhibited higher copper levels
   than that in normal breast and LuminalA breast cancer cells; similar
   findings were observed in breast cell lines (Figure [81]1D). Based on
   previous findings that elesclomol (ES) and disulfiram (DSF) deliver
   copper into cells and subsequently induce cuproptosis, we hypothesized
   that elevated copper levels in TNBC enhance the efficacy of
   drug‐induced cuproptosis.^[ [82]^27 , [83]^30 ^] However, cancer cells
   with higher copper levels exhibited resistance to cuproptosis, as
   indicated by the cell viability and colony formation assays
   (Figure [84]1E–H). To confirm that the reduction in cell viability was
   due to cuproptosis rather than other cell death pathways, we assessed
   the insoluble lipo‐DLAT protein levels and oligomerization. The results
   demonstrated that elesclomol‐induced cuproptosis occurred more robustly
   in normal cells with lower copper levels, however, not in TNBC cells
   (Figure [85]1I,J). Collectively, these findings indicated that TNBC
   cells elevated copper levels and exhibited resistance to
   elesclomol‐induced cuproptosis.

Figure 1.

   Figure 1
   [86]Open in a new tab

   Triple negative breast cancer exhibits resistance to cuproptosis. A)
   Paired line scatter plot showing Cu concentration in TNBC tissues
   (right, n = 30) and adjacent normal tissues (left, n = 30). Copper
   concentrations were significantly elevated in TNBC tissues compared to
   adjacent normal tissues. The means were indicated in purple font for
   clarity. B) Box plot illustrating Cu concentrations in patients with
   Stage I + II (n = 15) and Stage III + IV (n = 15) TNBC. C) Kaplan–Meier
   analysis depicting the association between Cu concentrations and
   overall survival in TNBC patients. D) Box plot showing Cu
   concentrations in normal breast cells and TNBC cells, n = 3. E, F) The
   IC[50] values of elesclomol‐Cu or disulfiram‐Cu were determined for
   normal breast cells and TNBC cells, n = 3. G,H) Colonies were
   visualized by crystal violet staining (G), and the bar graphs (H)
   quantified the colony counts, n = 3. I) Lipoylated DLAT expression was
   analyzed in MCF‐10A and MDA‐MB‐231 cells after a 2 h pulse treatment
   with elesclomol‐Cu. J) The DLAT oligomerization was analyzed in MCF‐10A
   and MDA‐MB‐231 cell treated with elesclomol‐Cu for 2 h pulse. The
   culture medium was supplemented with 1 µm CuCl[2] for experiments shown
   in panels (E–J). All data are presented as the mean±SD (n ≥ 3). The
   p‐value in panels (A) and (B) was calculated using a paired t‐test. The
   p‐value in panel (D) was calculated using one‐way ANOVA. The p‐value in
   panel (H) was determined using two‐way ANOVA. ^** p < 0.01; ^*** p <
   0.001; ^**** p < 0.0001.

2.2. Phosphoproteomic Profiling Identified AKT1 as a Key Gene for Cuproptosis
Sensitivity

   To explore the key factors modulating the sensitivity to cuproptosis,
   we initially generated cuproptosis‐resistant TNBC cells. As a result,
   the cuproptosis‐resistant MDA‐MB‐231 cells exhibited a significantly
   increased IC[50] compared to WT cells (55.58 vs. 9.34 µm) under chronic
   stimulation of copper‐elesclomol (Figure [87]2A,B; Figure [88]S1A,B,
   Supporting Information). Consistently, cuproptosis‐resistant cells
   showed lower levels of lipoylated DLAT, and reduced DLAT
   oligomerization, and no increase in DLAT foci upon elesclomol treatment
   (Figure [89]S1C–F, Supporting Information).

Figure 2.

   Figure 2
   [90]Open in a new tab

   AKT1 contributes to resistance to cuproptosis. A) Schematic diagram
   depicting the workflow for establishing cuprotosis‐resistant MDA‐MB‐231
   cells. B) The IC[50] value of elesclomol‐Cu was determined in wild‐type
   (WT) and resistant MDA‐MB‐231 cells, n = 3. C) The key differential
   proteins were mapped onto a PPI network. D, E) MDA‐MB‐468 or MDA‐MB‐231
   cells transduced with lentiviral shRNAs targeting AKT1 were treated
   with DMSO or elesclomol‐Cu and subjected to CCK‐8 assays to evaluate
   cell viability, n = 3. F) Cells generated in (D, E) were treated with
   DMSO or elesclomol‐Cu and subjected to colony formation assays. G) Bar
   graphs provided a statistical evaluation of colony counts for
   MDA‐MB‐231 cells (top panel) and MDA‐MB‐468 cells (bottom panel), n =
   3. H) The lipoylated DLAT expression was analyzed in cells generated in
   (D, E) following a 2 h pulse treatment with elesclomol‐Cu. I) DLAT foci
   in cells generated in (E) were visualized through immunofluorescence
   staining (DLAT: green, Mitotracker: red, DAPI: blue). J) DLAT foci were
   measured and quantified by ImageJ, n = 3. K) DLAT oligomerization was
   analyzed in cells generated in (D, E) following a 2 h pulse treatment
   with elesclomol‐Cu. The culture medium was supplemented with 1 µm
   CuCl[2] for experiments shown in all panels. All data are presented as
   the mean ± SD (n ≥ 3). The p‐value in panels (D), (E), and (G) was
   calculated using two‐way ANOVA. The p‐value in panel (J) was calculated
   using one‐way ANOVA. ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001.

   Given the crucial role of copper in promoting protein phosphorylation
   to maintain copper homeostasis,^[ [91]^7 , [92]^28 ^] we speculated
   that abnormal phosphorylation might influence cuproptosis sensitivity.
   Further, we performed phosphoproteomic profiling to compare resistant
   and intact MDA‐MB‐231 cells. In total, 7372 proteins and 13831
   phosphosites were quantified, of which 2170 proteins were upregulated,
   and 1344 proteins were downregulated in the resistant cells compared to
   the intact cells, using an FDR cutoff of 1.2. KEGG enrichment analysis
   of upregulated differentially phosphorylated proteins in the
   cuproptosis‐resistant cells revealed significant enrichment in several
   pathways, including the insulin signaling pathway (Figure [93]S1G,
   Supporting Information). Furthermore, domain enrichment analysis
   revealed that the upregulated phosphosites were enriched in AGC kinase
   motifs (RxxS/T) (Figure [94]S1H, Supporting Information). As multiple
   AGC kinases, such as AKT, PDK1, [95]S6K, and SGK, are key factors in
   the insulin signaling pathway, we subsequently conducted a
   protein‐protein interaction analysis, which revealed that AKT1 was the
   most pivotal protein among them (Figure [96]2C). Based on the close
   connection between copper and AKT,^[ [97]^29 ^] we observed that copper
   ions combined with elesclomol promoted AKT1 phosphorylation (Figure
   [98]S1I, Supporting Information), possibly due to enhanced PDK1
   interaction with AKT1 (Figure [99]S1J, Supporting Information).
   Elesclomol and copper ions failed to activate AKT1 in PDK1‐deficient
   TNBC cells (Figure [100]S1K, Supporting Information). In addition, the
   activation of AKT1 oncogenic signaling in cuproptosis‐resistant TNBC
   cells has been reported (Figure [101]S1L, Supporting Information).
   Therefore, we hypothesized that AKT1 was a potential mediator of
   cuproptosis resistance.

   To further explore the relationship between AKT1 and cuproptosis
   resistance, we analyzed the effect of AKT1 depletion on cell viability
   following treatment with elesclomol or disulfiram combined with copper
   ions. The results demonstrated that AKT1 knockdown significantly
   sensitized TNBC cells to cuproptosis, as assessed by cell viability and
   colony formation assays (Figure [102]2D–G; Figure [103]S2A–C,
   Supporting Information). Cuproptosis is dependent on the FDX1‐mediated
   lipoylation pathway. In other words, copper directly binds to
   lipoylated DLAT, inducing DLAT oligomer formation and insoluble DLAT,
   ultimately leading to cell death. Consistently, both pharmacological
   and genetic repression of AKT1 led to a marked increase in insoluble
   lipoylated DLAT expression, decreased soluble lipoylated DLAT
   expression (Figure [104]2H; Figure [105]S2D,E, Supporting Information),
   and increased DLAT oligomerization and DLAT foci formation
   (Figure [106]2I–K; Figure [107]S2F–H, Supporting Information).
   Consistent with this, the AKT1 inhibitor, MK2206, promoted cuproptosis
   in vivo, as evidenced by increased insoluble lipoylated DLAT expression
   (Figure [108]6I). Conversely, activation of AKT1 by insulin
   significantly reduced insoluble lipoylated expression and
   oligomerization of DLAT (Figure [109]S2I–K, Supporting Information).
   Collectively, these findings suggest that AKT1 plays a pivotal role in
   the negative regulation of cuproptosis sensitivity.

Figure 6.

   Figure 6
   [110]Open in a new tab

   AKT1 inhibitors synergize with elesclomol in TNBC therapies. A)
   MDA‐MB‐231 cells were treated with DMSO, MK2206, elesclomol, or a
   combination of MK2206 and elesclomol‐Cu as indicated for 2 days and
   then subjected to a cell viability assay (top panel). The combination
   index (CI) was calculated (bottom panel), n = 3. B) MDA‐MB‐468 cells
   were treated with DMSO, MK2206, elesclomol, or a combination of MK2206
   and elesclomol as indicated for 2 days and then subjected to a cell
   viability assay (top panel), n = 3. The combination index (CI) was
   calculated (bottom panel). C) MDA‐MB‐231 and MDA‐MB‐468 cells treated
   with MK2206, elesclomol, or a combination of MK2206 and elesclomol as
   indicated for 10 days were subjected to colony formation assays. D, E)
   Bar graphs provided a statistical evaluation of colony counts for
   MDA‐MB‐231 cells (D) and MDA‐MB‐468 cells (E), n = 3. F, G, H)
   MDA‐MB‐231 cells were subjected to mouse xenograft assays. Mice were
   administered vehicle, MK2206 (20 mg kg^−1), elesclomol (20 mg kg^−1),
   or a combination of MK2206 (20 mg kg^−1) and elesclomol (20 mg kg^−1)
   three times a week for 24 days (G). Dissected tumors were weighed (F),
   and tumor sizes were monitored (H), n = 8. I) IHC staining was
   performed on tissues derived from dissected tumors in (G), as
   indicated. J) The real‐time ATP production rates of cells extracted
   from tumor tissues generated in (G) were measured following sequential
   injections of oligomycin and Rot/AA using the Seahorse XFe96 Analyzer
   (Agilent), n = 3. K) Pyruvate production in cells extracted from tumor
   tissues generated in (G) was measured, n = 6. L) Lactate production in
   cells extracted from tumor tissues generated in (G) was measured, n =
   6. All data are presented as the mean ± SD (n ≥ 3). The p‐value in
   panels (G), (J), (K) and (L) was calculated using one‐way ANOVA. The
   p‐values in panels (D), (E) and (F) were calculated using two‐way
   ANOVA. ^* p < 0.05; ^**< 0.01; ^*** p < 0.001; ^**** p < 0.0001; ns,
   not significance.

   [Correction added on 26 February 2025, after first online publication:
   figure 6 is updated].

2.3. AKT1 Directly Phosphorylates FDX1

   To investigate the underlying mechanism by which AKT1 inhibits
   cuproptosis, we performed MS‐based proteomic analyses and found that
   AKT1 interacted with FDX1 (Figure [111]3A), a key regulator of
   cuproptosis. This interaction was further confirmed in both the ectopic
   expression system and the endogenous setting in TNBC cells and tissues
   (Figure [112]3M; Figure [113]S3A,B, Supporting Information).
   Consistently, in vitro pulldown assays demonstrated direct binding
   between FDX1 and AKT1 (Figure [114]S3C, Supporting Information), while
   confocal microscopy confirmed their co‐localization (Figure [115]S3F,
   Supporting Information). Moreover, we identified the N‐terminal domain
   of FDX1 as a crucial component mediating its interaction with AKT1
   (Figure [116]S3D,E, Supporting Information).

Figure 3.

   Figure 3
   [117]Open in a new tab

   AKT1 phosphorylates FDX1 at Ser63. A) GST‐pulldown products from
   HEK‐293T cells transfected with GST‐FDX1 and HA‐myr‐AKT1 were analyzed
   by mass spectrometry. B) Schematic representation of the evolutionarily
   conserved putative AKT1 kinase phosphorylation consensus sequence in
   FDX1. C) IB analysis of WCL and GST‐pulldown products derived from
   HEK‐293T cells transfected with GST‐FDX1 WT or variants and
   HA‐myr‐AKT1. D) IB analysis of the specified synthetic peptides
   immobilized on a nitrocellulose membrane. E) In vitro kinase assays
   were conducted with the purified AKT1 proteins from HEK‐293T cells
   transfected with HA‐myr‐AKT1 as the kinase source, and bacterially
   purified GST‐FDX1 as substrate. F) IB analysis of WCL derived from
   MDA‐MB‐231 cells infected with the lentiviral shRNA targeting AKT1. G)
   IB analysis of WCL derived from MDA‐MB‐231 cells. Cells were
   serum‐starved for 24 h and treated with or without AKT inhibitors
   (MK2206, 10 µm; GSK690693, 10 µm) for 1 h, and then subjected to
   insulin (100 nM) stimulation for 30 min before harvesting. H) IB
   analysis of WCL derived from MDA‐MB‐231 cells stably overexpressing
   HA‐AKT1 WT or variants. I, J) MDA‐MB‐231 and MDA‐MB‐468 cells stably
   overexpressing FDX1 WT or variants were subjected to CCK‐8 assays to
   assess cell viability, n = 3. K) Cells generated in (I) were subjected
   to colony formation assays, and the relative colony numbers were
   quantified (bottom panel), n = 3. L) Cells generated in (J) were
   subject to colony formation assays, and the relative colony numbers
   were quantified (bottom panel), n = 3. M) DLAT, FDX1, and AKT1 staining
   were visualized through immunofluorescence in TNBC and adjacent normal
   tissues. All data are presented as the mean ± SD (n ≥ 3). The p‐value
   in panels (I) and (J) was calculated using two‐way ANOVA. The p‐value
   in panel (K) and (L) was calculated using one‐way ANOVA. ^*** p <
   0.001; ^**** p < 0.0001; ns, not significance.

   Given that AKT1 functions as a kinase that phosphorylates its
   substrates, we investigated whether it could directly phosphorylate
   FDX1. MS analysis identified several potential phosphorylation sites on
   FDX1, with serine 63 (S63) drawing particular attention due to its
   location within the conserved AGC kinase phosphorylation motif
   (RxRxxS/T) (Figure [118]3A,B). Screening of a panel of AGC kinases,
   including constitutively active forms of AKT1 (myr‐AKT1), AKT2
   (myr‐AKT2), S6K1 (S6K1‐R3A), and SGK1 (SGK1‐Δ60), revealed that only
   AKT1 significantly promoted FDX1 phosphorylation, as detected by a
   pan‐AKT substrate phosphorylation antibody (Figure [119]S4A, Supporting
   Information).

   Further experiments confirmed that AKT1‐induced FDX1 phosphorylation of
   FDX1 predominantly occurred at its N‐terminal region in vitro and in
   vivo (Figure [120]S4B,C, Supporting Information). Notably, mimicking
   the non‐phosphorylated mutant FDX1‐S63A abolished AKT1‐mediated
   phosphorylation (Figure [121]3C). Notably, we successfully generated
   and validated phosphorylation‐specific antibodies that recognized
   FDX1‐S63 (Figure [122]3D). Consistent with this finding, in vitro
   kinase assays coupled with MS analysis confirmed that the S63A mutation
   abolished AKT1‐mediated FDX1 phosphorylation (Figure [123]3A,E).
   Additionally, insulin stimulation enhanced FDX1‐S63 phosphorylation in
   TNBC cells (Figure [124]3G), whereas the patient‐derived
   AKT1‐activating mutation (E17K) significantly accelerated this
   phosphorylation (Figure [125]3H). Conversely, AKT1 knockdown or
   treatment with AKT1 inhibitors significantly reduced FDX1‐S63
   phosphorylation (Figure [126]3F,G). The phosphorylation level of
   FDX1‐S63 highly correlated with AKT1 activity in both breast cancer
   cell lines and tumor tissues (Figure [127]S5C–E, Supporting
   Information). Taken together, these findings suggested that AKT1 can
   directly phosphorylate FDX1 at Ser63.

2.4. AKT1‐Mediated FDX1 Phosphorylation Promotes TNBC Tumor Growth by
Inhibiting Cuproptosis

   Although FDX1 plays a crucial role in mediating cuproptosis, its
   biological function in breast cancer remains largely unelucidated. Our
   initial analysis indicated that TNBC patients with low FDX1 expression
   had worse prognosis (Figure [128]S5A, Supporting Information). Further
   examination of FDX1 expression in the TNBC patients showed
   significantly higher expression levels in normal tissues than those in
   tumor tissues, both in the TCGA database and the database of our center
   (Figure [129]S5B,C,E, Supporting Information). Consistent with these
   findings, TNBC cells displayed reduced FDX1 expression compared with
   normal breast cells (Figure [130]S5D, Supporting Information). Knocking
   down FDX1 enhanced the proliferative and colony‐forming abilities of
   triple‐negative breast cancer cells (Figure [131]S5F–K, Supporting
   Information). To further dissect the biological function of
   AKT1‐mediated FDX1 phosphorylation, we introduced various FDX1 mutants
   (WT, S63A, and S63D) into TNBC cells (Figure [132]S4D,E, Supporting
   Information). While the WT and S63A mutant FDX1 suppressed tumor
   progression, the S63D mutant substantially enhanced tumor growth
   (Figure [133]3I–L). Similar observations were obtained in xenograft
   mouse models, where the S63D mutant accelerated tumor growth and
   elevated the levels of the proliferation marker Ki67 (Figure
   [134]4F–I). These results suggested that AKT1‐mediated FDX1
   phosphorylation contributes to TNBC tumor growth.

Figure 4.

   Figure 4
   [135]Open in a new tab

   AKT1‐mediated FDX1 phosphorylation inhibits cuproptosis. A) MDA‐MB‐231
   and MDA‐MB‐468 cells stably overexpressing FDX1 WT or its variants were
   subjected to IB analysis to assess effect on Fe‐S cluster biogenesis.
   B) Cells generated in (A) were subjected to IB analysis to assess
   effect on the lipoylation pathway. C) DLAT foci were measured and
   quantified using ImageJ, n = 3. D) DLAT foci in cells generated in (A)
   were visualized through immunofluorescence staining (DLAT: green,
   Mitotracker: red, DAPI: blue). E) DLAT oligomerization was analyzed in
   cells generated in (A) following a 2 h pulse treatment with elesclomol.
   F,G,H) Cells generated in (A) were subjected to mouse xenograft assays
   with vehicle or elesclomol (20 mg kg^−1), administered three times a
   week for 21 days. Dissected tumors were weighed (G), and tumor sizes
   were monitored (H), n = 8. I) IHC staining was performed on tissues
   derived from dissected tumors in (F), as indicated. All data are
   presented as the mean ± SD (n ≥ 3). The p‐value in panels (C) and (G)
   were calculated using one‐way ANOVA. The p‐value in panels and (H) was
   calculated using two‐way ANOVA. ^*** p < 0.001; ^**** p < 0.0001.

   Based on these findings, we hypothesized that AKT1 negatively regulated
   cuproptosis by phosphorylating FDX1. Cuproptosis relies on the
   lipoylation pathway and disruption of Fe‐S cluster biogenesis. To
   explore the relationship between phosphorylated FDX1 and Fe‐S cluster
   biogenesis, we examined the expression of Fe‐S cluster‐containing
   proteins. The FDX1‐S63D mutation did not affect the protein levels of
   POLD1, ACO2, SDHB, or DPYD, and a similar pattern was observed in cells
   overexpressing FDX1‐WT or FDX1‐S63A mutants (Figure [136]4A).
   Therefore, neither FDX1‐WT nor its phosphorylated form affected the
   synthesis of Fe‐S clusters, which was consistent with previous reports
   (Figure [137]4A).^[ [138]^22 ^] FDX1 acts as an upstream regulator of
   protein lipoylation, specifically affecting TCA cycle enzymes,
   including DLAT. Our observations revealed that the FDX1‐S63D mutant did
   not interact with DLAT (Figure [139]S6G, Supporting Information). The
   FDX1‐S63D mutant showed a significant lipoylation defect in both DLAT
   and DLST, in contrast to FDX1‐WT (Figure [140]4B). Notably, compared to
   the WT or S63A mutant, the FDX1‐S63D mutant impaired cuproptosis in
   TNBC, as evidenced by reduced lipoylation and oligomerization of DLAT
   (Figure [141]4C–E; Figure [142]S6H, Supporting Information). Next, we
   assessed the effect of FDX1‐WT and FDX1‐S63D overexpression cuproptosis
   sensitivity. FDX1‐WT increased TNBC cell sensitivity to cuproptosis,
   whereas the S63D mutation conferred resistance to copper ions (Figure
   [143]S6A–F, Supporting Information). Consistently, in vivo studies
   demonstrated that the ectopic expression of the S63D mutant conferred
   resistance to elesclomol (Figure [144]4F–H). IHC staining confirmed
   that the S63D mutation impaired copper‐mediated DLAT lipoylation
   (Figure [145]4I). In summary, these findings suggested that
   AKT1‐mediated FDX1 phosphorylation contributed to cuproptosis
   resistance in TNBC cells.

2.5. AKT1‐Mediated FDX1 Phosphorylation Promotes TNBC Glycolysis

   Cuproptosis interacts with components of the TCA cycle in the
   mitochondria, whereas FDX1 acts upstream of the lipoylation pathway,
   which is crucial for respiratory chain function.^[ [146]^22 ^]
   Therefore, AKT1‐mediated FDX1 phosphorylation may also influence tumor
   metabolic reprogramming. Metabolomic analysis revealed that FDX1‐WT and
   S63A mutants induced similar changes in metabolites, whereas the S63D
   mutation exhibited notable differences, indicating distinct metabolic
   profiles (Figure [147]5A). Further analysis of metabolite differences
   between the FDX1‐S63A and S63D mutants showed that the FDX1‐S63D mutant
   significantly promoted glucose, amino acid, and energy metabolism,
   favoring tumor malignant tumor proliferation and adaptation to adverse
   survival environments (Figure [148]5B,C). Moreover, the total ATP
   production rate in cells with the FDX1‐S63D mutation was significantly
   increased, primarily due to the activation of glycolysis, whereas
   mitochondrial ATP production was suppressed (Figure [149]5D). The
   oxygen consumption rate (OCR) assay results further indicated that
   phosphorylated FDX1 inhibited OCR, whereas FDX1‐WT and the FDX1‐S63A
   mutant enhanced OCR (Figure [150]5E). These findings suggested that
   phosphorylated FDX1 impaired mitochondrial function. To test this
   hypothesis, we assessed mitochondrial membrane potential using the JC‐1
   dye assay and measured the red‐to‐green fluorescence ratio. The results
   showed that MDA‐MB‐231 cells transfected with FDX1‐S63A mutant
   exhibited strong red fluorescence, indicating increased mitochondrial
   membrane potential due to JC‐1 aggregation. Conversely, cells
   transfected with FDX1‐S63D mutant showed an elevated green‐to‐red
   fluorescence ratio (Figure [151]5F,G), suggesting mitochondrial
   dysfunction.

Figure 5.

   Figure 5
   [152]Open in a new tab

   AKT1‐mediated FDX1 phosphorylation promotes glycolysis. A) MDA‐MB‐231
   cells stably overexpressing FDX1 WT or variants were extracted for
   metabolomics analysis. Metabolites were visualized using principal
   component analysis. B) Differential metabolites between FDX1 S63A and
   S63D mutants were visualized using a volcano plot. C) Differential
   metabolites were subjected to KEGG pathway enrichment analysis. D) The
   real‐time ATP production rates of MDA‐MB‐231 cells overexpressing FDX1
   S63A and S63D were measured following sequential injections of
   oligomycin and Rot/AA using the Seahorse XFe96 Analyzer (Agilent), n =
   3. E) OCR responses of MDA‐MB‐231 cells stably overexpressing FDX1 WT
   or its variants were determined following sequential injections of
   oligomycin, trifluoromethoxy carbonylcyanide phenylhydrazone (FCCP),
   and rotenone, n = 8. F) Changes in mitochondrial membrane potential
   (ΔΨm) in MDA‐MB‐231 cells stably overexpressing FDX1 WT or its variants
   were evaluated using JC‐1 fluorescence. G) Fluorescence intensity was
   measured and quantified using ImageJ, and the red/green fluorescence
   ratio was calculated, n = 3. H) ECAR responses of MDA‐MB‐231 cells
   stably overexpressing FDX1 WT or its variants were measured following
   sequential injections of glucose, oligomycin, and 2‐deoxy‐D‐glucose
   (2DG), n = 8. I) IB analysis of MDA‐MB‐231 cells stably overexpressing
   FDX1 WT or its variants targeting key enzymes in the glycolytic
   pathway. J) Pyruvate production in MDA‐MB‐231 cells overexpressing FDX1
   WT or its variants was measured, n = 3. K) Lactate production in
   MDA‐MB‐231 cells overexpressing FDX1 WT or its variants was measured, n
   = 3. All data are presented as the mean ± SD (n ≥ 3). The p‐values in
   panels (D), (G), (J) and (K) were calculated using one‐way ANOVA. ^***
   p < 0.001; ^**** p < 0.0001.

   Tumor cells shift their energy metabolism toward glycolysis because of
   impaired aerobic respiration.^[ [153]^31 , [154]^32 , [155]^33 ^] Our
   findings demonstrated that phosphorylated FDX1 increased the
   extracellular acidification rate (ECAR), whereas FDX1‐WT or the
   FDX1‐S63A mutations resulted in a lower ECAR (Figure [156]5H).
   Phosphorylated FDX1 also upregulated several glycolysis‐related genes,
   including glucose transporter 1 (GLUT1), hexokinase 2 (HK2), pyruvate
   kinase M2 (PKM2), and lactate dehydrogenase A (LDHA), which ultimately
   promoted the glycolytic pathway (Figure [157]5I). Increased lactate and
   pyruvate production were observed in the FDX1‐S63D mutant, further
   indicating the pivotal role of the FDX1‐S63D mutation in promoting TNBC
   glycolysis (Figure [158]5J,K). Taken together, these data showed that
   phosphorylated FDX1 promoted the transition from mitochondrial
   respiration to glycolysis, allowing tumor cells to adapt to rapid
   growth.

2.6. Combination of AKT1 Inhibitor with Copper ion Carriers Effectively
Inhibits TNBC Progression

   Our previous study demonstrated that AKT1‐mediated FDX1 phosphorylation
   inhibited cuproptosis and promoted TNBC progression. We observed a
   positive correlation between AKT1 activation, FDX1 phosphorylation at
   residue S63, and copper concentration, whereas lipoic DLAT levels were
   negatively correlated, as measured by IHC and IF staining
   (Figure [159]3M; Figure [160]S5C, Supporting Information). Next, we
   evaluated whether combining AKT1 inhibition with copper ion carriers
   could offer therapeutic benefits for TNBC. The combination of MK2206,
   an AKT1 inhibitor, with copper ion carriers, elesclomol, and copper
   ions showed a strong synergistic effect with a combination index CI<1
   in TNBC cells (Figure [161]6A,B) and significantly suppressed colony
   formation (Figure [162]6C–E).

   The combination treatment markedly inhibited tumor growth compared to
   treatment with either agent alone in vivo (Figure [163]6F–H). IHC
   staining further confirmed enhanced cuproptosis in TNBC tissues treated
   with MK2206 and elesclomol (Figure [164]6I). To explore the metabolic
   effects of this combination therapy, we assessed the ATP production
   pathways in tumor tissues. Combination treatment significantly reduced
   glycolysis‐driven ATP production (glycoATP) without affecting
   mitochondrial ATP (mitoATP) levels (Figure [165]6J). Moreover, lactate
   and pyruvate production notably decreased under the combination
   treatment (Figure [166]6K,L), indicating inhibition of glycolysis. In
   summary, these findings indicated that the combination of MK2206 and
   elesclomol effectively targeted this pathway and represented a
   promising therapeutic strategy for TNBC (Figure [167]7 ).

Figure 7.

   Figure 7
   [168]Open in a new tab

   Model of AKT1 regulating cuproptosis in TNBC tumorigenesis. Schematic
   illustration of AKT1 role in regulating cuproptosis. Copper activates
   AKT signaling pathway, which inhibits cuproptosis by directly
   phosphorylating ferredoxin‐1 (FDX1), a key regulator of cuproptosis.
   AKT‐mediated FDX1 phosphorylation not only abrogates FDX1‐induced
   cuproptosis and aerobic respiration, but also promotes glycolysis.
   Consequently, combination of the AKT inhibitor MK2206 and the copper
   ionophore elesclomol synergistically induces cuproptosis and inhibits
   glycolysis, suggesting a feasible and effective therapeutic strategy
   for TNBC.

3. Discussion

   Cuproptosis is a form of mitochondrial cell death induced by
   proteotoxic stress. However, the regulatory mechanism in TNBC remains
   still unclear. In our study, we observed that copper overload resulted
   in abnormal activation of AKT1 in TNBC. AKT activation has been widely
   exploited in response to extracellular stimuli, including insulin,
   hypoxia, and high‐fat diet,^[ [169]^34 , [170]^35 ^] contributing to
   tumorigenesis and anti‐apoptosis effects. However, the roles of AKT in
   regulating cuproptosis remain undefined. Here, AKT1 has been
   characterized to directly phosphorylate FDX1 and inhibit aerobic
   respiration and cuproptosis. Furthermore, phosphorylated FDX1 shifts
   the cellular metabolism toward glycolysis, facilitating a rapid energy
   supply. Notably, the combination of the AKT1 inhibitor MK2206 and the
   copper ionophore elesclomol (ES) synergistically inhibit TNBC
   tumorigenesis both in vitro and in vivo.

   Copper is an essential trace element that plays a pivotal role as a
   cofactor in biological redox reactions. It is extensively involved in
   biosynthesis within the human body and is tightly regulated by the
   coordinated activities of copper uptake by CTR1 and efflux by ATPases
   under normal physiological conditions.^[ [171]^26 , [172]^36 , [173]^37
   , [174]^38 ^] The disruption of copper homeostasis can lead to disease
   development. In several solid tumors, including TNBC, copper levels in
   cancer tissues are higher than those in adjacent normal tissues, which
   correlates with advanced TNM stages, poor prognosis, and increased
   recurrence rates, making it a potential biomarker for cancer
   progression.^[ [175]^10 , [176]^39 , [177]^40 ^] Dysregulated copper
   metabolism plays a dual role in tumorigenesis and cancer therapy.^[
   [178]^41 ^] Cuproplasia refers to copper‐dependent cell growth and
   proliferation, encompassing processes such as hyperplasia, metaplasia,
   and neoplasia through signaling pathways modulated by both enzymatic
   and non‐enzymatic copper.^[ [179]^2 , [180]^41 ^] More recently, copper
   treatment of cell cultures increased the phosphorylation of tyrosine
   kinases such as TRKB, EGFR, and MET, which are upstream regulators of
   the MAPK signaling pathway.^[ [181]^42 ^] Copper can also directly
   activate MEK, enhancing ERK phosphorylation and leading to MAPK pathway
   activation, which is a key driver of tumor cell growth and division.^[
   [182]^43 ^] Additionally, copper promotes angiogenesis by transiently
   upregulating VEGF expression and activity, presumably through the
   generation of ROS.^[ [183]^44 , [184]^45 ^] Conversely, excess copper
   exerts cytotoxic effects on tumor cells by triggering cell death,
   particularly apoptosis, paraptosis, pyroptosis, ferroptosis, and
   cuproptosis. Copper overloads generate ROS, causing oxidative stress
   and damage to cellular components.^[ [185]^46 , [186]^47 ^] It disrupts
   mitochondrial function, impairs energy production and promotes
   apoptosis.^[ [187]^48 ^] Furthermore, copper induces ferroptosis by
   perturbing lipid peroxidation pathways.^[ [188]^17 , [189]^49 ^] It was
   recently reported that copper triggers cuproptosis by binding to
   lipoylated DLAT, a recently discovered cell death mechanism that
   disrupts the TCA cycle and leads to DLAT aggregation.^[ [190]^21 ^]

   The recent discovery of cuproptosis has promoted the targeting of
   copper metabolism in cancer cells. Copper promotes METTL16 lactylation
   at site K229 regulated by SIRT2 and subsequently mediates m6A
   modification of FDX1, followed by cuproptosis. Thus, combining
   elesclomol with the SIRT2‐specific inhibitor AGK2 induces cuproptosis
   in gastric tumors both in vitro and in vivo.^[ [191]^24 ^] MELK
   enhances the expression of the cuproptosis‐related signature gene DLAT,
   particularly by increasing the DLAT monomer portion through the
   activation of the PI3K/mTOR pathway, thereby facilitating the
   progression of HCC.^[ [192]^50 ^] In our study, we found that the
   copper ion content was significantly elevated in TNBC tissues and was
   closely associated with prognosis. However, TNBC cells are insensitive
   to cuproptosis. To investigate the mechanism contributing to TNBC
   resistance to cuproptosis, we performed a phosphoproteomic analysis and
   identified AKT1 as a key factor.

   Copper is closely related to the AKT1 signaling pathway. Our findings
   demonstrated that copper/ES directly induced AKT1 phosphorylation in
   TNBC cells. Specifically, elesclomol and copper ions enhanced the
   interaction between AKT1 and PDK1, and knocking out PDK1 abolished the
   ability of elesclomol and copper ions to activate AKT1, which is
   consistent with the results of previous studies.^[ [193]^29 , [194]^51
   ^] The AKT1 signaling pathway is a crucial oncogenic pathway that is
   abnormally activated in TNBC. Our previous study revealed that AKT1
   phosphorylated SIK1, thereby disrupting its tumor‐suppressive
   function.^[ [195]^52 ^] Moreover, AKT1 is also closely associated with
   tumor tolerance to cell death.^[ [196]^53 ^] AKT1 directly
   phosphorylates key apoptotic regulators such as BAD, Caspase‐9, and
   BIM, thereby inhibiting tumor apoptosis.^[ [197]^54 ^] In addition,
   AKT1 directly phosphorylates TRPML1 at Ser343, and prevents its
   degradation of TRPML1, which leads to ferroptosis inhibition.^[
   [198]^55 ^] In this study, we found that AKT1 inhibited cuproptosis in
   TNBC by phosphorylating FDX1. Thus, when copper ions are overloaded in
   TNBC, the excess copper paradoxically activates AKT1, ultimately
   suppressing cuproptosis.

   Cuproptosis depends on the lipoylation pathway and the destabilization
   of Fe‐S cluster proteins. Neither WT nor phosphorylated FDX1 affected
   Fe‐S cluster biosynthesis, consistent with the results of previous
   studies showing that FDX2, rather than FDX1, is essential for Fe‐S
   cluster biosynthesis.^[ [199]^23 ^] FDX1 plays a pivotal role in
   cuproptosis by reducing Cu (II) to Cu (I) in the mitochondria, thereby
   promoting excessive protein lipoylation and subsequent aggregation,
   which are closely tied to mitochondrial respiration.^[ [200]^56 ^] In
   our study, we observed that the FDX1‐S63D mutant caused a significant
   DLAT lipoylation defect compared to FDX1‐WT. FDX1‐mediated lipoylation
   pathway is closely associated with aerobic respiration and cellular
   metabolism.^[ [201]^41 ^] FDX1‐knockout cells exhibit a failure of
   oxidative phosphorylation and accumulation of Kennedy pathway
   intermediates. We found that phosphorylated FDX1 induced mitochondrial
   damage, as evidenced by a decrease in mitochondrial membrane potential
   and OCR. Additionally, phosphorylated FDX1 promoted the expression of
   glycolysis‐related genes, thereby facilitating glycolysis.
   Phosphorylated FDX1 stimulated lactate and pyruvate production, thereby
   promoting rapid energy metabolism. Glycolysis status determines the
   susceptibility to cuproptosis. Clinical studies have revealed that
   patients with low lactate levels exhibit an increased sensitivity to
   elesclomol.^[ [202]^30 , [203]^57 ^] Therefore, inhibiting glycolysis
   enhances cuproptosis‐mediated suppression of tumor growth. AMPK
   increases the sensitivity of human pancreatic cancer cells to
   cuproptosis, owing to AMPK‐induced suppression of glycolysis.^[
   [204]^58 ^] Similarly, ACOD1 enhances the susceptibility of CRC cells
   to cuproptosis by inhibiting aerobic glycolysis.^[ [205]^59 ^] Our
   findings revealed that the AKT1 inhibitor, MK2206, inhibited glycolysis
   in TNBC cells, thus, the combination of elesclomol with the AKT1
   inhibitor has been shown to synergistically reduce cell viability.

   Collectively, we reported that copper ion levels are elevated levels in
   TNBC cells, leading to AKT1 activation via PDK1. Nevertheless, AKT1
   suppresses cuproptosis by phosphorylating FDX1, which explains the
   resistance of TNBC cells to cuproptosis. Phosphorylated FDX1 impairs
   mitochondrial function, inhibits aerobic respiration, and promotes
   glycolysis. Therefore, we propose a novel therapeutic strategy
   involving the synergistic combination of AKT1 inhibitors and copper
   ionophores to treat TNBC.

4. Experimental Section

Cell Lines, Transfection, and INFECTION

   MDA‐MB‐231, MDA‐MB‐468, and HEK293T cell lines were cultured in DMEM
   medium supplemented with 10% FBS. All cell lines were maintained at
   37 °C in a humidified atmosphere with 5% CO[2]. Cell transfection was
   performed using Lipofectamine or PEI (Polysciences) according to the
   manufacturer's instructions. Packaging of lentiviral shRNA or
   over‐expression viruses, as well as subsequent infection of various
   cell lines were performed according to the methods described
   previously. After viral infection, cells maintain continuous screening
   with different concentrations of hygromycin or puromycin based on the
   cell line and the viral vector used to infect the cells.

Antibodies and Reagents

   Anti‐AKT Substrate (RxRxxpS/T) antibody (9614), anti‐phospho‐Ser473‐AKT
   antibody (4060), anti‐phospho‐Thr308‐AKT antibody (2965), anti‐AKT
   total antibody (4691), anti‐phospho‐Ser9‐GSK3β antibody (5558),
   anti‐GSK3β antibody (12456), anti‐PRAS40 antibody (2691),
   anti‐phospho‐PRAS40 (Thr246) antibody (2997), anti‐DLST antibody
   (11954), anti‐GST antibody (2622) and anti‐HA antibody (2367) were
   obtained from Cell Signaling Technology. Anti‐FDX1 antibody
   (12592‐1‐AP), anti‐DLAT antibody (13426‐1‐AP), anti‐GAPDH antibody
   (10494‐1‐AP) and anti‐AKT antibody (60203‐2‐Ig) were obtained from
   Proteintech. Anti‐LA antibody (ab58724) was obtained from Abcam.
   Anti‐Aconitase 2 antibody (A4524), anti‐SDHB antibody (A23832),
   anti‐POLD1 (A4218), anti‐DPYD (A1620), anti‐PDK1 (A8930) were obtained
   from ABclonal. AKT inhibitor MK2206 (HY‐108232), capivasertib
   (HY‐15431), copper ionophore elesclomol (HY‐12040), disulfiram
   (HY‐B0240), and insulin (HY‐[206]P73243) were obtained from MCE.

DNA Constructs and Mutant

   pcDNA3‐HA‐myr‐AKT1, pcDNA3‐HA‐myr‐AKT2, pcDNA3‐HA‐myr‐AKT3,
   pcDNA3‐HA‐S6K1‐R3A, pcDNA3‐HA‐Δ60‐SGK1, pcDNA3‐HA‐AKT1 were cloned into
   mammalian expression pCDNA3‐HA vectors. pCMV‐GST‐FDX1 was cloned into
   mammalian expression pCMV‐GST vectors. pGEX‐4T1‐FDX1 were cloned into
   mammalian expression pGEX‐4T1 vectors. pLenti‐hygro‐FDX1 was cloned
   into mammalian expression pLenti‐hygro vectors. FDX1‐S63A, FDX1‐S63D,
   and AKT1‐E17K were obtained via Q5 Site‐Directed Mutagenesis Kit (New
   England Biolabs) according to the manufacturer's instructions and the
   sequences were verified by DNA sequencing. Specific shRNA primers were
   cloned into pLKO.1‐puro lentiviral vectors, human AKT1 target sequence
   #1 5′‐CGTGCCATGATCTGTATTTAA, human AKT1 target sequence #2
   5′‐GGACAAGGACGGGCACATTAA, human AKT1 target sequence #3
   5′‐CGCGTGACCATGAACGAGTTT, human FDX1 target sequence #1
   5′‐GATGCCAGACAATCCATTGAT, human FDX1 target sequence #2
   5′‐GCAATCACTGATGAGGAGAAT, human FDX1 target sequence #3
   5′‐GCTGCCAAATCTGTTTGACAA.

Breast Cancer Tissues and Immunohistochemical Staining

   All breast cancer and paired normal specimens were obtained from the
   Women and Children's Medical Center of Guangzhou Medical University and
   usage of these specimens was approved by the institutional ethics
   committee (2023, No. 306B00). Informed consent was obtained from each
   patient. Fix tumor xenografts or surgical specimen tissue sections in
   10% neutral buffered formalin, embedded in paraffin, and cut into 4 µm
   sections. For IHC staining, the sections were incubated with 3%
   H[2]O[2]for 15min at room temperature to eliminate the endogenous
   peroxidase activity and blocked with 5% normal goat serum for 1 hr at
   room temperature. Sections were incubated with primary FDX1, FDX1‐pS63,
   AKT1‐ pS473, LA, Ki67 antibody at 4 °C overnight, followed by
   incubation with monoclonal mouse anti‐rabbit immunoglobulins
   (Dako#M0737) for 1 h at room temperature. Afterward, sections were
   incubated with Envision+ System‐HRP Labeled Polymer Anti‐Rabbit (Dako
   #K4003) for 30 min Specific detection was developed using the DAB
   chromogen kit (Dako#K3468) and lightly counterstained with hematoxylin.

Copper Microplate Assay

   The Copper (Cu) Colorimetric Assay Kit was purchased from Elabscience
   (E‐BC‐K300‐M). It measures the absorption or transmission of light by a
   copper‐containing solution. By analyzing the resulting spectra, the
   concentration of copper ions can be determined.

Immunoprecipitation (IP), GST Pull‐Down (PD) Assay and Western Blot

   Cells were lysed in EBC buffer (50 mm Tris pH 7.5, 120 mm NaCl, 0.5%
   NP‐40) supplemented with protease inhibitors (phosphatase inhibitor
   cocktail set I and II, Calbiochem) and phosphatase inhibitors
   (PhosSTOP, Roche). The protein concentrations of whole cell lysates
   were quantified by a bicinchoninic acid (BCA) assay kit (Pierce BCA
   Protein Assay Kit, 23225). An equal amount of whole cell lysates was
   separated by 10% SDS‐PAGE gels and then transferred to an Immobilon‐P
   PVDF membrane (Millipore, Massachusetts, USA). The membrane was
   incubated overnight with a primary antibody at 4 °C overnight and then
   hybridized with a secondary antibody at room temperature for 1h. The
   signals were visualized by enhanced chemiluminescence as previously
   described. For immunoprecipitation, WCL were incubated with the
   indicated antibody (1–2 µg) for 3–4 h at 4 °C followed by Protein A/G
   sepharose beads (GE Healthcare) at 4 °C overnight. For GST pull‐down
   assays, the lysates were incubated with glutathione sepharose 4B (GE
   Healthcare) for 2 hr at 4 °C. The immuno‐complexes were washed four
   times with NETN buffer (20 mm Tris, pH 8.0, 150 mm NaCl, 1 mM EDTA and
   0.5% NP‐40) and then were resolved by SDS‐PAGE and immunoblotted with
   indicated antibodies. Quantification of the immunoblot band intensity
   was performed with ImageJ software.

Soluble and Insoluble Fraction Isolation

   Cells were harvested and lysed in ice‐cold NP40 lysis buffer (20 mm
   Tris (pH 7.4), 150 mm NaCl, 1% NP40) supplemented with protease
   inhibitors for 30 min. Then the cell extracts were centrifuged at
   14 000 rpm at 4 °C for 30 min. Following centrifugation, the
   supernatant and pellet were carefully separated. The supernatant was
   mixed with 5× SDS loading buffer, while the pellet was resuspended in
   1× SDS loading buffer. Two fractions were then boiled and sent for
   western blotting.

Immunostaining (IF)

   Cells grown on confocal dishes were fixed in paraformaldehyde, blocked
   with 10% goat serum, and then incubated with primary FDX1 and AKT
   antibodies, washed thrice with PBS, and then incubated with 488 Alexa
   (green) and 647 Alexa (red)‐labeled secondary antibodies. The DNA dye
   DAPI was used for re‐staining the nuclear DNA.

Peptide Synthesis and Dot Immunoblot Assay

   FDX1‐S63‐WT and FDX1‐S63‐Phospho peptides used for dot blot assays were
   synthesized by HuaBio Technology. The sequences are listed below:

   FDX1‐S63‐WT: C‐SARARSSSEDKITVC

   FDX1‐S63‐Phospho: C‐SARARSS(pS)EDKITVC

   Peptides were diluted into 2 mg mL^−1 for dot blot assays, and spotted
   onto the nitrocellulose membrane at doses of 0.01, 0.03, 0.10, and 0.30
   mg. The membrane was dried and blocked by soaking in TBST buffer with
   5% nonfat milk for immunoblot analysis.

Purification of GST‐Tagged Proteins from Bacteria

   Recombinant GST‐conjugated FDX1 was generated by transforming the BL21
   (DE3) E. coli strain with pGEX‐ 4T1‐FDX1. Starter cultures grown
   overnight at 37 °C were injected (1%) into larger volumes. Cultures
   were grown at 37 °C to an O.D. 0.8, and then the protein expression was
   induced for 16 h using 0.1 mM isopropyl β‐d‐ 1‐thiogalactopyranoside
   (IPTG) at 16 °C with vigorous shaking.

   Recombinant proteins were purified from harvested pellets, which were
   re‐suspended in EBC buffer for sonicated. Insoluble proteins and cell
   debris were discarded, and the supernatant was incubated with 50 µl 50%
   Glutathione‐sepharose slurry (Pierce) for 3 h at 4 °C and then washed 3
   times with PBS buffe. Finnaly, the recombinant protein was stored at 4
   °C in PBS buffer containing 10% glycerol or eluted by elution buffer.
   The purified proteins were analyzed by coomassie blue staining and
   quantified with BSA standards.

In Vitro Kinase Assay

   In vitro kinase assays were adapted from a protocol described
   previously. Briefly,1 µg of the bacterially purified GST‐FDX1 fusion
   proteins were incubated with immunoprecipitated AKT1 from cell lysates
   transfected in the presence of 200 µm adenosine triphosphate (ATP) in
   the kinase reaction buffer (50 mM Tris pH 7.5, 1 µm MnCl[2], 2 mm DTT)
   for 30 min at 30 °C. The reaction was subsequently stopped by adding in
   0.1 mm EDTA. The reaction was resolved by SDS‐PAGE and detected by
   pS63‐SIK1.

Mass Spectrometry

   For mass spectrometry (MS) analysis, GST‐pulldown were performed with
   the WCL derived from three 10 cm dishes of HEK293T cells transfected
   with GST‐FDX1 with/without HA‐myr‐AKT1 as previously described. The
   proteins were resolved by SDS‐PAGE and identified by Coomassie
   staining. The band containing FDX1 was reduced with 10 mm
   Dithiothreitol (DTT) for 30 min, alkylated with 55 mm iodoacetamide for
   45 min, and in‐ gel‐digested with trypsin enzymes. The resulting
   peptides were extracted from the gel and analyzed by microcapillary
   reversed‐phase (C18) liquid chromatography‐tandem mass spectrometry
   (LC‐MS/MS), using a high resolution QExactive HF Orbitrap (Thermo
   Fisher Scientific) in positive ion DDA mode (Top 8) via higher energy
   collisional dissociation (HCD) coupled to a Proxeon EASY‐nLc II
   nano‐high performance liquid chromatography (HPLC). MS/MS data were
   searched against the Uniprot Human protein database (version 20151209
   contain‐ ing 21,024 entries) using Mascot 2.5.1 (Matrix Science) and
   data analysis was performed using the Scaffold 4.4.8 software (Proteome
   Software). Peptides and modified peptides were accepted if they passed
   a 1% false discovery rate (FDR) threshold.

Metabolomics

   MDA‐MB‐231 cell stably overexpressing HA‐FDX1 WT or variants cultured
   for 48 h before washing with PBS. The samples were removed the protein
   by the precooled methanol/acetonitrile (1:1, v/v) and extracted the
   metabolites. The samples were then centrifuged at 14 000 g for 20 min
   at 4 °C to collect the supernatant. The supernatant was dried in a
   vacuum centrifuge. The samples were re‐dissolved in 100 µL
   acetonitrile/water (1:1, v/v) solvent and centrifuged at 14 000 g at
   4 °C for 15 min, then the supernatant was injected for LC‐MS analysis.
   The mass spectrometry was performed using a UHPLC (1290 Infinity LC,
   Agilent Technologies) coupled to a quadrupole time‐of‐flight (AB Sciex
   TripleTOF 6600) and analyzed by NuoBiotech (Guangzhou)  Co., Ltd,
   China. The raw MS data were converted to MzXML files using ProteoWizard
   MSConvert before importing them into freely available XCMS software.
   For peak picking, the following parameters were used: centWave m/z = 10
   ppm, peakwidth = c (10, 60), prefilter = c (10, 100). For peak
   grouping, bw = 5, mzwid = 0.025, minfrac = 0.5 were used. CAMERA
   (Collection of Algorithms of MEtabolite pRofile Annotation) was sued
   for annotation of isotopes and adducts. In the extracted ion features,
   only the variables having more than 50% of the nonzero measurement
   values in at least one group were kept. Compound identification of
   metabolites was performed by comparing of accuracy m/z value (<10 ppm),
   and MS/MS spectra with an in‐house database established with available
   authentic standards. After sum‐normalization, the processed data were
   analyzed by an R package (ropls), where it was subjected to
   multivariate data analysis, including Pareto‐scaled principal component
   analysis (PCA) and orthogonal partial least‐squares discriminant
   analysis (OPLS‐DA). The 7 fold cross‐validation and response
   permutation testing was used to evaluate the robustness of the model.
   The variable importance in the projection (VIP) value of each variable
   in the OPLS‐DA model was calculated to indicate its contribution to the
   classification. Student's t‐test was applied to determine the
   significance of differences between two groups of independent samples.
   VIP > 1 and p‐value < 0.05 were used to screen significant changed
   metabolites. Pearson's correlation analysis was performed to determine
   the correlation between two variables.

Phosphoproteomic Profiling

   The protein sample was extracted from WT and resistant MDA‐MB‐231 cell.
   The protein concentration was determined with the BCA kit according to
   the manufacturer's instructions. The protein solution was reduced with
   5 mm dithiothreitol for 30 min at 56 °C and alkylated with 11 mm
   iodoacetamide for 15 min at room temperature in darkness. The protein
   sample was then diluted by adding 100 mm TEAB to urea concentration
   less than 2 m. Finally, trypsin was added at 1:50 trypsin‐to‐protein
   mass ratio for the first digestion overnight and 1:100
   trypsin‐to‐protein mass ratio for a second 4 h‐digestion. Finally, the
   peptides were desalted by the C18 SPE column. Peptide mixtures were
   first incubated with IMAC microspheres suspension with vibration in
   loading buffer (50% acetonitrile/0.5% acetic acid). To remove the
   nonspecifically adsorbed peptides, the IMAC microspheres were washed
   with 50% acetonitrile/0.5% acetic acid and 30% acetonitrile/0.1%
   trifluoroacetic acid, sequentially. To elute the enriched
   phosphopeptides, the elution buffer containing 10% NH[4]OH was added
   and the enriched phosphopeptides were eluted with vibration. The
   supernatant containing phosphopeptides was collected and lyophilized
   for LC‐MS/MS analysis. The mass spectrometry was performed and analyzed
   by NuoBiotech (Guangzhou)  Co., Ltd, China. The resulting MS/MS data
   were processed using the MaxQuant search engine (v.1.6.15.0). Tandem
   mass spectra were searched against the
   Rattus_norvegicus_10116_PR_20230103.fasta (47945 entries) concatenated
   with reverse decoy database. Trypsin/P was specified as a cleavage
   enzyme, allowing up to 2 missing cleavages. The mass tolerance for
   precursor ions was set as 20 ppm in the first search and 20 ppm in the
   main search, and the mass tolerance for fragment ions was set as 0.02
   Da. Carbamidomethyl on Cys was specified as fixed modification, and
   acetylation on protein N‐terminal and oxidation on Met and
   Phosphorylation modification were specified as variable modifications.
   FDR was adjusted to < 1%.

Real‑time Cell Metabolism Assay

   Real‐time ATP production, extracellular acidification rates (ECAR), and
   oxygen consumption rates (OCR) were quantified using Seahorse XFe96 and
   XFp Extracellular Flux Analyzers (Agilent, USA) with corresponding
   assay kits following the manufacturer's standardized protocols.
   MDA‐MB‐231 cells or dissected tumor tissues were seeded in appropriate
   Seahorse assay plates (96‐well or XFp) at a density of 1 × 10⁵
   cells/well and prepared for assays by overnight culture or incubation
   to allow attachment. Before measurements, cells were washed, and the
   medium was replaced with Seahorse XF DMEM (pH 7.4) containing 10 mm
   glucose, 2 mm glutamine, and 1 mm sodium pyruvate. ATP production rates
   from mitochondrial respiration (mitoATP) and glycolysis (glycoATP) were
   recorded after sequential injections of 1.5 µm oligomycin and 0.5 µm
   rotenone/antimycin A (Rot/AA), while ECAR was quantified following
   injection of 2‐deoxy‐D‐glucose (2‐DG). Real‐time measurements of OCR
   were recorded under four sequential conditions: 1) basal respiration;
   2) 1 µm oligomycin; 3) 0.3 µM FCCP; and 4) 0.5 µm rotenone and 0.5 µm
   antimycin. All OCR and ECAR values were normalized to the total protein
   content in each well.

JC‐1 Staining

   Treated MDA‐MB‐231 cells were seeded into confocal dishes and stained
   with the JC‐1 dye kit (Beyotime Biotechnology). After PBS washing, the
   JC‐1 staining solution was applied, and the cells were incubated at 37
   °C for 30 min. Subsequently, images were captured with a Leica SP8
   confocal microscope after washing with PBS. In healthy cells with a
   high mitochondrial membrane potential (△ψm), JC‐1 aggregates in the
   mitochondrial matrix, forming polymers that emit red fluorescence (Em =
   590 nm). By contrast, mitochondria with reduced membrane potential
   retain JC‐1 in their monomeric form, emitting green fluorescence (Em =
   529 nm). The green‐to‐red fluorescence intensity ratio was calculated
   as an indicator of △ψm.

The Lactate and Pyruvate Production

   MDA‐MB‐231 cells were seeded into six‐well plates at a density of
   1 × 10^6 cells per well and cultured overnight. The Cells or dissected
   tumor tissues were collected and extracted for further measurement. The
   lactate production and pyruvate production were detected using the
   Lactic Acid Content Assay Kit (Solarbio BC2235) and Pyruvate Content
   Assay Kit (Solarbio BC2205), according to the instruction of the
   manufacturers, respectively. All experiments were performed at least
   three times, and the data were normalized by the cell numbers or
   protein content.

Colony Formation

   Cells were inoculated into a 6‐well plate (300 or 600 cells/well) and
   cultured for 12–20 days until visible colonies were visible. The
   colonies were washed with PBS and fixed with 4% paraformaldehyde for 20
   min and then stained with 0.4% crystal violet in 20% ethanol for 20
   min. After staining, the plate was washed and air‐dried, and the number
   of colonies was numbered and quantified.

Cell Proliferation Assay

   The cells were seeded in the 96‐well plates (5000 cells/well). At the
   indicated time points, the cell proliferation was detected with MTS
   assay (Promega, G3530). The MTS reagent was added to cell culture media
   and incubated for 0.5–4 h in standard culture conditions. After
   incubation, the plate was shaken briefly and measured absorbance at 490
   nm.

Xenograft Models and In Vivo Treatments

   Mouse xenograft assays were performed as described previously. All
   animals were approved by the Institutional Animal Care and Use
   Committee of Guangzhou Women and Children's Medical Center
   (RSDW‐2024‐01240) and purchased from Guangdong GemPharmatech. Briefly,
   2 × 10^6 MDA‐MB‐231 cells stably expressing targeting gene mixed with
   matrigel were injected into the flank of female nude mice (8/group, 4
   weeks of age). Once the tumor size reached 50–100 mm^3, some groups
   were treated with vehicles, MK2206 (20 mg kg^−1), elesclomol (20
   mg kg^−1), or a combination of MK2206(20 mg kg^−1)/elesclomol (20
   mg kg^−1) three times a week. Tumor size was measured every three days
   with a caliper, and the tumor volume was determined with the formula L
   × W^2 × 0.5, where L is the longest diameter and W is the shortest
   diameter. Mice were euthanized when their tumor volume was larger than
   2000 mm^3.

Statistical Analysis

   The key experiments have been repeated at least 3 times, quantitated,
   and subjected to statistical analysis. GraphPad Prism version 10.0 was
   used for statistical analyses. As indicated in the figure legends, all
   quantitative data were presented as the mean & SD of three biologically
   independent experiments or samples; n and p values were indicated in
   every single figure. Two‐way ANOVA was used for multiple‐group
   comparisons, one‐way ANOVA for one‐way comparisons, and unpaired
   two‐tailed t‐tests for two‐group comparisons. Differences between
   groups were considered statistically significant at ^* p < 0.05, ^** p
   < 0.01, ^*** p < 0.001, and ^**** p < 0.0001; “ns” indicated not
   significant.

Conflict of Interest

   The authors declare no conflict of interest.

Author Contributions

   This study was conceived and designed by J.L. Development of
   methodology was performed by Z.S. and H.X. Acquisition of data
   (provided animals, acquired, and managed patients, provided facilities,
   etc.) was performed by Z.S., H.X., G.L., C.Y., X.G., J.Z., X.L., and
   Y.C. The final manuscript was written by Z.S, H.X., J.G., and J.L.
   Administrative, technical, or material support (i.e., reporting or
   organizing data, constructing databases) was provided by Z.S., H.X.,
   K.W., J.G., and J.L. Study supervision was provided by J.L. All authors
   approved the manuscript.

Supporting information

   Supplemental information
   [207]ADVS-12-2408106-s001.docx^ (7.6MB, docx)

Acknowledgements