Abstract

   Inducing cuproptosis in tumor cells is significantly impeded by the
   challenges of arduous copper ion delivery in vivo and the unbreakable
   intracellular copper homeostasis, which leads to insufficient
   mitochondrial copper accumulation. Here, a carrier-free
   metal-polyphenolic (CF-MPs) based nanoplatform (T-T@Cu) that features
   tumor-mitochondria cascade-targeting, glutathione (GSH) depletion and
   near-infrared Ⅱ photothermal performance is designed to induce
   mitochondria copper-overload and exacerbate cuproptosis in tumor cells.
   By leveraging the enhanced permeability and retention (EPR) effects and
   the mitochondria-targeting capabilities of tannic acid, T-T@Cu
   effectively increases mitochondrial copper accumulation in tumor cells.
   Upon exposure to a 1064 nm laser, T-T@Cu triggers mild
   photothermal-boosted ferroptosis, which down-regulates intracellular
   ATP levels. This reduction dramatically impacts the expression of
   copper-ion efflux proteins ATP7A/7B, ultimately inhibiting copper ion
   efflux. Additionally, T-T@Cu exhibits robust GSH consumption and
   dual-responsive degradation in tumor microenvironments characterized by
   overexpressed cysteine (Cys) and GSH. This results in alleviated
   GSH-induced inactivation of copper ions and specific copper release
   within the tumor microenvironment. In vitro and in vivo therapeutic
   evaluations demonstrate the outstanding tumor inhibition of T-T@Cu in
   4T1-breast-cancer models, with no significant systemic toxicity
   observed. This novel mild photothermal-boosted ferroptosis strategy for
   exacerbating tumor cell cuproptosis holds great promise for future
   clinical applications in oncotherapy.

Graphical abstract

   [52]graphic file with name 12951_2025_3722_Figa_HTML.jpg

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12951-025-03722-2.

   Keywords: Mild-photothermal, Cuproptosis, Ferroptosis, Copper
   homeostasis, GSH depletion

Introduction

   Cuproptosis was recently identified by Tsvetkov et al. as a novel
   programmed cell death (PCD) modality characterized by copper-induced
   lipoylated protein aggregation and the loss of iron-sulfur (Fe-S)
   cluster proteins in the mitochondrial tricarboxylic acid (TCA) cycle
   [[53]1]. This process arises from mitochondrial copper overload and
   culminates in cell death due to proteotoxic stress. Despite being in
   its infancy, cuproptosis has aroused significant interest in
   oncotherapy, as multiple refractory malignancies, including
   drug-resistant tumors, are susceptible to this form of cell death
   [[54]2–[55]4]. However, before cuproptosis can be widely applied in
   clinical settings, several challenges must be addressed [[56]5].
   Notably, the poor targeting of copper ions to tumor mitochondria and
   the difficulty in overcoming intracellular copper homeostasis present
   significant obstacles to achieving mitochondrial copper overload, which
   is crucial for inducing cuproptosis in tumor cells [[57]6–[58]9].
   Additionally, the low selectivity of cuproptosis between normal and
   tumor cells raises potential biosafety concerns [[59]10, [60]11].
   Therefore, it is imperative to explore strategies for efficiently
   inducing cuproptosis in tumor cells while minimizing damage to normal
   cells.

   Due to the short blood circulation half-life and non-specific targeting
   of copper ions, the induction of cuproptosis in tumor cells heavily
   relies on copper ion delivery systems that can selectively transport
   copper ions into the mitochondria of tumor cells. While numerous copper
   ion delivery systems, including copper ionophores, inorganic
   nanoparticles (NPs), and biomacromolecules, have been successfully
   developed, most focus primarily on enhancing the tumor enrichment of
   copper ions [[61]12–[62]16]. The mitochondrial targeting efficiency of
   these copper ions, which is crucial for effectively inducing
   cuproptosis in tumor cells, is often overlooked. Although some
   nanosystems based on elesclomol and triphenylphosphine have been
   designed to achieve tumor-mitochondria cascade targeted delivery of
   copper ions, their clinical applications are hindered by high costs and
   unknown long-term biosafety due to complex preparation processes and
   non-degradability in vivo [[63]17, [64]18]. Carrier-free
   metal-polyphenolic nanocomplexes (CF-MPs), formed by the self-assembly
   of metal ions and polyphenols through coordination interactions, have
   garnered significant attention in the field of oncotherapy due to their
   easy preparation, tumor-specific targeting, and high metal ion loading
   capacity [[65]19–[66]21]. Tannic acid (TA), a plant-derived natural
   polyphenol, is particularly notable as a building block for CF-MPs,
   owing to its excellent biological properties, including
   biodegradability and strong protein affinity [[67]22–[68]25]. Given
   TA’s robust noncovalent affinity for mitochondrial outer membrane
   proteins, we speculate that it may serve as an ideal candidate for
   developing tumor-mitochondria cascade targeted copper ion delivery
   systems.

   Furthermore, to prevent intracellular copper overload, tumor cells have
   evolved a sophisticated copper homeostasis regulation system [[69]26].
   When excess copper ions are internalized by tumor cells, they face the
   risk of being either expelled by copper-ion efflux proteins ATP7A and
   ATP7B (ATP7A/7B) or inactivated by overexpressed intracellular
   glutathione (GSH) [[70]27–[71]30]. This process can lead to
   insufficient mitochondrial copper accumulation, which is necessary to
   trigger effective cuproptosis in tumor cells. Therefore, in addition to
   exploring ideal tumor-mitochondria cascade targeted copper ion delivery
   systems, breaking intracellular copper homeostasis is another critical
   factor for enhancing cuproptosis in tumor cells. Ferroptosis, a
   non-apoptotic programmed cell death modality characterized by oxidative
   damage to plasma membranes and mitochondria, has been demonstrated to
   disrupt cellular energy metabolism through mitochondrial adenosine
   triphosphate (ATP) depletion [[72]31]. Given that copper-transporting
   ATPases (ATP7A/7B) depend on ATP hydrolysis to facilitate copper ion
   efflux [[73]32], the ferroptosis-associated decline in ATP levels may
   compromise their function, resulting in intracellular copper
   accumulation. Furthermore, ferroptosis generates excessive reactive
   oxygen species (ROS) and lipid peroxides (LPO), which can directly
   oxidize and impair ATP7A/7B transporters, exacerbating copper
   homeostasis dysregulation [[74]33]. Based on these findings, we propose
   that ferroptosis induction in tumor cells may suppress
   ATP7A/7B-mediated copper efflux, thereby disrupting intracellular
   copper homeostasis. However, the poor Fenton reaction activity between
   Fe^2+ and H[2]O[2] in acidic tumor microenvironments (TME) results in
   insufficient ROS generation for effective ferroptosis induction
   [[75]34]. Given that copper ions exhibit higher Fenton-like catalytic
   activity than Fe^2+ in the TME, which can be further enhanced by
   mild-photothermal effects, we propose that mild photothermal-boosted
   ferroptosis triggered by copper ion-based nanosystems could effectively
   downregulate the expression of ATP7A/7B [[76]35]. This downregulation
   may alleviate copper ion efflux and exacerbate cuproptosis in tumor
   cells.

   Herein, we developed a versatile carrier-free metal-polyphenolic
   nanocomplex-based nanoplatform (T-T@Cu) to induce robust cuproptosis in
   tumor cells through efficient tumor-mitochondria cascade-targeted
   delivery of copper ions and regulation of intracellular copper
   homeostasis. As illustrated in Scheme [77]1, T-T@Cu was constructed by
   the simple self-assembly of tannic acid (TA),
   7,7,8,8-tetracyanoquinodimethane (TCNQ), and copper ions via
   coordination and charge transfer (CT) interactions. As a natural
   polyphenol, TA not only specifically binds to mitochondrial outer
   membrane proteins but also forms CT nanocomplexes with excellent
   near-infrared (NIR) II (1000–1700 nm) photothermal performance in
   conjunction with copper ions and TCNQ. To prevent the inactivation of
   copper ions by overexpressed glutathione (GSH) in tumor cells, TCNQ is
   employed to consume cysteine (Cys), an essential amino acid for GSH
   biosynthesis. Following intravenous injection of T-T@Cu in 4T1 cell
   line-derived allograft (CDA^4T1) mouse models, T-T@Cu effectively
   accumulates at tumor sites via enhanced permeability and retention
   (EPR) effects and specifically targets the mitochondria of tumor cells.
   Upon exposure to a 1064 nm NIR II laser, T-T@Cu triggers
   mild-photothermal boosted ferroptosis, impacting the activity and
   expression of copper-ion efflux proteins (ATP7A/7B). This process leads
   to enhanced mitochondrial copper overload and exacerbates cuproptosis
   in tumor cells. Notably, due to the low sensitivity of normal cells to
   ferroptosis, this novel cuproptosis strategy mediated by T-T@Cu
   demonstrates extraordinary tumor cell-specific killing effects without
   damaging normal cells, highlighting its promising potential in
   oncotherapy.

Scheme 1.

   [78]Scheme 1
   [79]Open in a new tab

   Preparation of T-T@Cu for exacerbating tumor cell cuproptosis. (A)
   Schematic illustration showing the preparation and Cys/GSH-responsive
   degradation mechanism of T-T@Cu. (B) The biological mechanism of
   exacerbated tumor cell cuproptosis induced by T-T@Cu. T-T@Cu can
   effectively accumulates at tumor sites of CDA^4T1 mouse models via
   intravenous injection and specifically targets the mitochondria of
   tumor cells. Once irradiated by a 1064 nm NIR II laser, T-T@Cu can
   down-regulate intracellular ATP levels via mild-photothermal boosted
   ferroptosis, leading to disruption of the activity and expression of
   copper-ion efflux proteins (ATP7A/7B). Meanwhile, T-T@Cu can prevent
   the GSH-induced inactivation of copper ions via depleting overexpressed
   Cys and GSH in tumor cells. These processes significantly enhanced
   mitochondrial copper overload, which facilitates the proteotoxic
   aggregation of DLAT and destabilized Fe-S cluster proteins, hence
   exacerbating tumor cells cuproptosis

Experimental section

Preparation of T-T@Cu

   10 mg of TCNQ, 26 mg of CuCl[2] and 82 mg of TA were co-dissolved in N,
   N-dimethylformamide (DMF). 15 mL of pure water was added dropwise to
   the above mixed solution under ultrasonic treatment at room
   temperature. After continued 5 min of ultrasonic treatment, the
   as-prepared T-T@Cu was washed with pure water for several time via
   centrifugation (13000 rpm, 10 min) and re-dispersed in PBS 7.4 for
   usage.

In vitro biodegradability test of T-T@Cu

   T-T@Cu was dispersed in PBS (pH 4.0) and PBS (pH 7.4) with different
   concentrations of Cys (15 µM, 200 µM and 400 µM), GSH (20 µM, 1 mM and
   10 mM) or Cys + GSH (15 µM + 20 µM, 200 µM + 1 mM and 400 µM + 10 mM),
   respectively. At predesigned time intervals, the digital photos,
   absorbance at 808 nm and size distribution of T-T@Cu were recorded to
   investigate the biodegradation process.

In vitro Cys and GSH consumption assay

   0.5 mL of Cys or GSH PBS 7.4 solution was mixed with 0.5 mL of T-T@Cu
   dispersion with various concentrations. After incubation for 3 h, the
   mixed solution was subsequently centrifuged (13000 rpm, 10 min). The
   supernatant was collected for determining the Cys or GSH level via
   measure the residual -SH groups of Cys and GSH using
   5,5’-Dithiobis-(2-nitrobenzoic acid) (DTNB) as a -SH indicator.

In vitro Cu release assay

   0.5 mL of T-T@Cu dispersion sealed in dialysis bag (MWCO ≈ 1000 Da) was
   soaked in PBS (pH 7.4) with different concentrations of Cys (15 µM, 200
   µM, 400 µM), GSH (20 µM, 1 mM, 10 mM) or Cys + GSH (15 µM + 20 µM, 200
   µM + 1 mM and 400 µM + 10 mM), followed by incubating at 37 ℃. At
   predesigned time intervals, 1 mL of dialysate was collected before
   replenishing with 1 mL corresponding fresh PBS. The Cu content in
   collected dialysate was detected via ICP-MS and the cumulative release
   curve of Cu was plotted.

In vitro detection of hydroxyl radical (•OH)

   2 mL of T-T@Cu dispersion containing H[2]O[2] and TMB were incubated at
   room temperature. During this process, T-T@Cu + H[2]O[2] + TMB + L
   treatment group received an extra irradiation by a 1064 nm laser
   (1W/cm^2). H[2]O[2] + TMB and H[2]O[2] + TMB + L treatment groups were
   served as controlled groups. At different time intervals, the
   absorption at the range of 500 nm to 900 nm was recorded for evaluation
   of the •OH generation.

   1 mL of T-T@Cu dispersion containing H[2]O[2] was incubated at room
   temperature or oil bath (45℃) for 12 min. The ESR signals of different
   treatment groups were recorded by an ESR spectrometer (EMXplus-6/1,
   Bruker) at pre-designed time intervals. 5,5-dimethyl-1-pyrroline
   N-oxide (DMPO) was employed to trap •OH indicator. H[2]O[2],
   H[2]O[2] + 45℃ and T-T@Cu treatment groups were set as control groups.

Cell culture and cytotoxicity study

   Human umbilical vein endothelial cells (HUVEC), Human hepatocellular
   carcinomas (HepG2) and Mouse breast cancer cells (4T1) were purchased
   from the cell bank of the Chinese Academy of Sciences (Shanghai,
   China). All the cells were cultured in the corresponding cell culture
   medium supplemented with 10% (v/v) fetal bovine serum at 37 ℃ under a
   5% CO[2] humid atmosphere.

   HUVEC, HepG2 and 4T1 cells were seeded in 96-well plates and cultured
   overnight before cytotoxicity assay. The cells were then incubated with
   CuCl[2] and T-T@Cu of various copper concentrations for another
   different hours (8 h and 24 h). In addition, cells in laser treatment
   group (T-T@Cu + laser and T-T@Cu + ammonium tetrathiomolybdate
   (ttm) + laser) were irradiated by a 1064 nm laser (1 W/cm^2). PBS,
   CuCl[2] and T-T@Cu plus ttm treated cells were regarded as control
   groups. The cell viability was evaluated by a standard Cell Counting
   Kit-8 (CCK-8) Assay.

Mitochondrial targeting validation

   4T1 cells were incubated in 1640 medium with RhB@T-T@Cu for 6 h. Before
   confocal laser scanning microscopy (CLSM) observation, the cells were
   stained by DAPI and Mito-tracker Green.

   The mitochondria from 4T1 cells treated with PBS, CuCl[2] and T-T@Cu
   for 8 h were isolated by the Cell Mitochondria Isolation Kit. Firstly,
   the 4T1 cells in various treatment groups were collected and well
   homogenized. Subsequently, mitochondria were separated from the cell
   homogenates via two steps of centrifugation at 4 ℃. The isolated
   mitochondria were digested and used for copper quantification by
   ICP-MS.

Determination of intracellular GSH and Cys

   4T1 cells were seeded in 6-well plates and cultured overnight. The
   cells were incubated with CuCl[2] and T-T@Cu at various mass
   concentrations for 8 h. After thoroughly washed with ice-cold PBS, the
   intracellular GSH and Cys levels were measured by a Reduced GSH Content
   Assay Kit and a Cys Colorimetric Assay Kit, respectively.

Validation of DLAT aggregation

   4T1 cells were seeded onto a glass coverslip in 24-well plate and
   cultured overnight. The cells were divided into seven groups (PBS,
   PBS + L, CuCl[2], T-T@Cu, T-T@Cu + ttm, T-T@Cu + ttm + L, T-T@Cu + L)
   and incubated with various drugs at fixed Cu (3 µg/mL) and ttm
   (5 µg/mL) concentrations for 8 h. The cells in the irradiation groups
   received an extra 1064 nm laser (1 W/cm^2) irradiation. Afterward, the
   cells were washed thoroughly with ice-cold PBS and fixed with 4%
   paraformaldehyde at 4℃. Before CLSM observation, the cells were
   incubated sequentially with corresponding primary and secondary
   antibodies. The cell nuclei were stained by DAPI.

Evaluation of intracellular ROS and LPO

   4T1 cells were seeded in 24-well plates and cultured till monolayer
   growth. Subsequently, the cells were further cultured with T-T@Cu
   (25 µg/mL) for 9 h. During this process, the cells in T-T@Cu + L
   treatment group received an extra irradiation by a 1064 nm laser
   (1 W/cm^2). PBS treated cells were served as blank control group.
   Before the flow cytometry analysis, the cells in different treatment
   groups were collected, washed by ice-cold PBS and stained by DCFH-DA or
   BODIPY™ 581/591 C11.

RNA-sequencing analysis

   4T1 cells were seeded in 6-well plate at a density of 1 × 10^6 per
   well. After 24 h of incubation, the cells received various treatments
   including PBS, PBS + L, CuCl[2], T-T@Cu and T-T@Cu + L. The cells in
   the irradiation groups were irradiated by a 1064 nm laser (1 W/cm^2).
   TRIzol reagent (Invitrogen, CA, USA) was used to extract total RNA.
   VAHTS Universal V6 RNA-seq Library Prep Kit was used to construct the
   libraries. Illumina Novaseq 6000 was applied in RNA sequencing. Fastp^1
   was used to quantify the transcription levels. R (v 3.2.0) was employed
   to generate the volcano graphs and the heat maps. R (v 3.2.0) also
   utilized to complete the Kyoto Encyclopedia of Genes and Genomes (KEGG)
   pathway and Gene Ontology (GO) pathway enrichment analysis. GSEA
   software was used to perform the Gene Set Enrichment Analysis (GSEA).
   Cytoscape was utilized to generate the protein-protein interaction
   (PPI) network.

Photothermal induced in vivo infrared thermal imaging

   4T1 allograft tumor-bearing mice were administered intravenously with
   PBS and PBS suspension of T-T@Cu (4.4 µg NPs/g body weight). After 3 h
   of administration, the tumors of the mice were irradiated by a 1064 nm
   laser (1 W/cm^2, 80 s). Meanwhile, the in vivo infrared thermal images
   and temperature variation of tumors were recorded by an infrared
   thermal camera (FLIR A300).

In vivo bio-distribution study

   36 of 4T1 allograft tumor-bearing mice were randomly divided into two
   groups and received intravenous injection of PBS and PBS suspension of
   RhB@T-T@Cu (5 µg NPs/g body weight). Thereafter, at pre-designed time
   intervals (1 h, 3 h, 6 h, 10 h, 13 h and 24 h), three mice were
   randomly picked up from each group and sacrificed for harvesting their
   main organs. The ex vivo fluorescence imaging of the dissected organs
   was acquired an IVIS Lumina XRMS Series III small animal imaging
   system. The PBS treated mice were served as blank control group.

In vivo metabolism of T-T@Cu

   4T1 allograft tumor-bearing mice were randomly grouped and injected
   intravenously with PBS and PBS suspension of T-T@Cu (5 µg NPs/g body
   weight). At pre-designed time intervals (6 h, 12 h, 36 h, 48 h and
   72 h), the urine and feces of mice were collected by using metabolic
   cages. The copper content in the urine and feces was quantified by
   ICP-MS.

Evaluation of in vivo therapeutic efficacy

   When the tumor volume reached about 100 mm^3, 35 of 4T1 allograft
   tumor-bearing mice were randomly grouped and treated with PBS, PBS + L,
   CuCl[2], T-T@Cu, T-T@Cu + ttm, T-T@Cu + ttm + L and T-T@Cu + L,
   respectively. The administrated drug concentration of copper and ttm in
   various treatment groups were fix at 0.8 µg/g body weight and 5 µg/g
   body weight, respectively. During the treatment duration, the mice in
   each group received three doses of different drugs, and the tumors of
   the mice were irradiated by a 1064 nm laser (1 W/cm^2) every three day
   within the first 9 days of treatment. The tumor volume and body weight
   variations of mice were recorded every day for evaluating the
   therapeutic efficacy. On the 3th day of treatment, one mouse was
   randomly selected from each group and its tumor was dissected for FDX1,
   LIAS, DLAT, GPX4, TUNEL and H&E staining observation.

Results and discussion

Preparation and characterization of T-T@Cu

   Equimolar amounts of TA and TCNQ, serving as donor (D) and acceptor (A)
   molecules, respectively, have been shown to form greenish CT complexes
   with near-infrared (NIR) absorption in dimethyl formamide (DMF)
   (Fig. [80]1A&S1). Additionally, TA, as a natural polyphenolic compound,
   can form metal-polyphenol complexes with copper ions (Cu^2+) through
   oxidative-mediated self-assembly. This interaction is evidenced by the
   bathochromic shift of the absorption peak of TA from 275 nm to 310 nm
   in the UV-Vis-NIR absorption spectrum of the TA@Cu complexes (Fig. S2).
   Given these intermolecular interactions, we hypothesize that the
   combination of TA, TCNQ, and Cu^2+ can form CT nanocomplexes with
   potential NIR absorption properties.

Fig. 1.

   [81]Fig. 1
   [82]Open in a new tab

   Structural and performance characterization of T-T@Cu. A)
   Ultraviolet-visible-near-infrared (UV-vis-NIR) absorption spectra of
   TA, CuCl[2], TCNQ, TA-TCNQ and T-T@Cu. B) Copper loading efficiency
   (L.E.) and encapsulation efficiency (E.E.) of T-T@Cu. C) Representative
   TEM image of T-T@Cu. The bar value is 200 nm. D) High-angle annular
   dark-field (HAADF) TEM image and elemental mapping images of C, N, O
   and Cu atoms in T-T@Cu. The bar value is 100 nm. E) FTIR spectra of TA,
   TCNQ and T-T@Cu. High-resolution XPS spectra of copper in T-T@Cu F)
   before and G) after GSH reduction. H) DLS analysis of the hydrodynamic
   sizes of T-T@Cu in PBS (pH 7.4) during 7 days. I) PDI and Zeta
   potential variation of T-T@Cu in PBS (pH 7.4) during 7 days

   Herein, a carrier-free metal-polyphenol nanoplatform (T-T@Cu) was
   prepared via a nanoprecipitation method by adding pure water to the
   mixed DMF solution of CuCl[2], TA and TCNQ under ultrasonic conditions.
   To optimize the feeding ratio of TA, TCNQ, and Cu^2+ for the
   preparation of T-T@Cu, we systematically studied the morphology, size
   distribution, and UV-Vis-NIR absorption spectra of the as-prepared
   T-T@Cu. Transmission electron microscopy (TEM) images revealed that the
   as-prepared T-T@Cu exhibited various morphologies and sizes. By fixing
   the molar ratio of TA and TCNQ at 1:1, we observed that the size of
   T-T@Cu increased with the feeding ratio of Cu^2+ (Fig. S3&S4). This
   increase in size can be attributed to the higher feeding ratio of
   Cu^2+, which provides sufficient copper ions to crosslink the TA-TCNQ
   CT complexes, resulting in larger T-T@Cu particles. Meanwhile, the
   copper loading efficiency (L.E.) and encapsulation efficiency (E.E.)
   were measured, with maximum values determined to be approximately 22%
   and 51%, respectively, at feeding ratios of TA, TCNQ, and Cu^2+ of
   1:1:5 and 1:1:0.8 (Fig. [83]1B). Additionally, the absorption peak of
   T-T@Cu exhibited a bathochromic shift into the NIR II bio-window
   (1000–1700 nm) and increased slightly with the feeding ratio of Cu^2+
   (Fig. [84]1A&S5). We speculate that the presence of Cu^2+ facilitates
   the delocalization of p-electrons from the highest occupied molecular
   orbital of TA (D) to the lowest unoccupied molecular orbital of TCNQ
   (A), leading to a significant reduction in the energy gap and a
   corresponding redshift in absorption [[85]36, [86]37]. When the feeding
   ratio of TA, TCNQ, and Cu^2+ was 1:1:4, we obtained cauliflower-shaped
   T-T@Cu with concentration-dependent NIR II absorption (Fig. S6) and a
   suitable size (~ 120 nm) for in vivo administration (Fig. [87]1C).
   Elemental mapping images confirmed the uniform distribution of C, O, N
   and Cu, validating the successful preparation of T-T@Cu (Fig. [88]1D).
   Fourier transform infrared (FTIR) spectroscopy further verified the
   composition of T-T@Cu, showing characteristic peaks at 2200 cm^−1 and
   1600 cm^−1 corresponding to TCNQ, and peaks at 1300 cm^−1 and 1200
   cm^−1 associated with TA. Notably, the disappearance of the hydroxyl
   stretching vibration peak at 3400 cm^−1 in T-T@Cu indicates the
   oxidation of TA by coordinated copper ions. Moreover, the intensity
   decay of peaks at 1315 cm^−1 and 1189 cm^−1 suggests distortion of TA
   due to coordination with copper ions (Fig. [89]1E). High-resolution
   X-ray photoelectron spectroscopy (HR-XPS) was conducted to investigate
   the oxidation state of copper in T-T@Cu. Both Cu^2+ and Cu^+ binding
   energy peaks, along with strong satellite peaks, were detected,
   indicating that approximately 40.56% of Cu^2+ was reduced to Cu^+ by
   the phenolic hydroxyl groups of TA (Fig. [90]1F). Importantly, we found
   that Cu^2+ in T-T@Cu can be further reduced to Cu^+ by GSH (Fig.
   [91]1G), suggesting that T-T@Cu may release Cu^+ in response to the
   tumor microenvironment, which is characterized by overexpressed GSH,
   thereby facilitating cuproptosis in tumor cells. Dynamic light
   scattering (DLS) analysis showed that the hydrodynamic diameter of
   as-prepared T-T@Cu in PBS (pH 7.4) was approximately 122 nm and
   remained nearly constant over seven days (Fig. [92]1H). The
   polydispersity index (PDI) and zeta potential were maintained around ~
   0.15 and − 37 mV, respectively, throughout the week, indicating
   excellent stability of T-T@Cu in PBS (pH 7.4) (Fig. [93]1I).

Cys and GSH dual-response degradation of T-T@Cu

   The tumor microenvironment (TME) is characterized by distinct metabolic
   features that create unique conditions for drug delivery and efficacy.
   The development of nanodrugs with TME-responsive degradability is
   crucial for achieving high tumor-specific killing efficacy while
   minimizing systemic toxicity [[94]38]. T-T@Cu leverages the specific
   interactions between Cys and aryl nitriles, as well as the redox
   reactions involving transition metal ions (i.e. Cu^2+) and the thiol
   (-SH) groups in GSH [[95]39, [96]40]. Therefore, the overexpressed Cys
   and GSH in tumor tissues as endogenous stimulus can trigger the
   degradation of T-T@Cu. As illustrated in Fig. [97]2A, when TCNQ was
   mixed with Cys in dimethyl sulfoxide (DMSO), a bathochromic shift of
   the absorption peak from 405 nm to 420 nm was observed. This shift
   confirmed the formation of mono-thiazolidine derivatives resulting from
   the reaction between TCNQ and Cys. Additionally, a new absorption peak
   around 485 nm emerged, likely due to the formation of bi-substituted
   and multi-substituted compounds (Fig. S7). The redox reaction between
   Cu^2+ ions and the -SH groups of GSH was investigated using
   5,5’-Dithiobis-(2-nitrobenzoic acid) (DTNB) as a -SH indicator. As
   shown in Fig. [98]2B, the intensity of the absorption peak at 412 nm,
   which corresponds to the reaction products of -SH and DTNB,
   progressively decreased with increasing concentrations of Cu^2+. This
   result indicated that Cu^2+ ions effectively consume the -SH groups of
   GSH through a redox reaction. Furthermore, in vitro studies were
   conducted to measure the residual -SH groups of both Cys and GSH after
   treatment with T-T@Cu. The results demonstrated that T-T@Cu can react
   with Cys and GSH simultaneously, and the consumption of Cys and GSH by
   T-T@Cu occurred in a dose-dependent manner (Fig. [99]2C).

Fig. 2.

   [100]Fig. 2
   [101]Open in a new tab

   In vitro evaluation of Cys and GSH dual-responsive degradation of
   T-T@Cu. A) UV-vis absorption spectra of TCNQ obtained at different
   reaction time with Cys. B) UV-vis absorption spectra of DTNB in GSH
   solution with Cu^2+ at different concentrations. C) Concentration
   dependent Cys and GSH consumption by T-T@Cu. D-F) Degradation kinetics
   of T-T@Cu in various mimic in vivo microenvironments including plasma,
   intracellular microenvironments of normal cells and tumor cells. G-I)
   DLS analysis of the hydrodynamic size variations of T-T@Cu in various
   mimic in vivo microenvironments. J) TEM images of T-T@Cu in various
   mimic in vivo microenvironments for 6 h

   To evaluate the stability of T-T@Cu in vivo, photographs of T-T@Cu in
   various media, including mimic plasma (PBS (pH 7.4) with Cys (15 µM),
   GSH (20 µM) or Cys (15 µM) + GSH (20 µM)), mimic intracellular
   lysosomal microenvironment (PBS (pH 4.0)), mimic intracellular
   microenvironments of normal cells (PBS (pH 7.4) with Cys (200 µM), GSH
   (1 mM) or Cys (200 µM) + GSH (1 mM)) and tumor cells (PBS (pH 7.4) with
   Cys (400 µM), GSH (10 mM) or Cys (400 µM) + GSH (10 mM)), were recorded
   at predesigned time intervals (Fig. S8). Negligible color change was
   observed in the mimic plasma and intracellular lysosomal
   microenvironment dispersion of T-T@Cu within 12 h, suggesting
   satisfactory stability during the blood circulation and lysosomal
   internalization process. In contrast, T-T@Cu gradually faded in the
   mimic intracellular microenvironments of both normal and tumor cells,
   with a faster fading rate noted in the latter. The results indicated
   that T-T@Cu might undergo degradation after being internalized by both
   normal and tumor cells. Subsequently, the degradation ratios of T-T@Cu
   in various media were analyzed by recording its absorption variation at
   808 nm. As shown in Fig. [102]2D and F, T-T@Cu exhibited the highest
   degradation ratios of approximately 40%, 62%, and 75% in the mimic
   intracellular microenvironment of tumor cells (PBS (pH 7.4) with Cys
   (400 mM), GSH (10 mM), or Cys (400 mM) + GSH (10 mM)) within 12 h
   [[103]41, [104]42]. This indicated that T-T@Cu could degrade more
   thoroughly in tumor cells, especially in the intracellular
   microenvironment containing both Cys and GSH. In comparison, T-T@Cu
   exhibited lower degradation ratios in mimic plasma, intracellular
   lysosomal microenvironment and the intracellular microenvironments of
   normal cells. Notably, the degradation ratio of T-T@Cu in mimic plasma
   and intracellular lysosomal microenvironment within 12 h was determined
   to be ~ 19% and 9.3%, validating that T-T@Cu can remain relatively
   intact during the blood circulation and lysosomal internalization
   process (Fig. [105]2F&S9). DLS and TEM analysis further demonstrated
   that T-T@Cu can undergo degradation in the mimic intracellular
   microenvironments of tumor cells, dissociating into ultra-small
   nanoparticles (NPs) with diameters of less than 5 nm (Fig. [106]2G and
   J). Since NPs smaller than 5 nm are reported to be capable of urinary
   metabolism, we speculate that the degradation products of T-T@Cu can
   eventually be metabolized.

In vitro copper ion release and •OH generation

   Then, the release of copper ions from T-T@Cu in various media was
   systematically studied using inductively coupled plasma mass
   spectrometry (ICP-MS). The final released copper ions in the mimic
   intracellular microenvironments of tumor cells reached ~ 93% within
   96 h, which was significantly higher than the release observed in mimic
   plasma (~ 38%) and the intracellular microenvironment of normal cells
   (~ 73%) (Fig. [107]3A). This specific copper ion-responsive release in
   tumor microenvironments may be beneficial for triggering ferroptosis
   and cuproptosis in tumor cells. To validate the Fenton-like reaction
   between T-T@Cu and H[2]O[2], the catalytic activity of T-T@Cu for •OH
   generation was investigated. The oxidation of
   3,3’,5,5’-tetramethylbenzidine (TMB) by •OH results in a measurable
   change in absorption intensity at 652 nm, which indicates •OH
   generation. In T-T@Cu + H[2]O[2] + TMB and T-T@Cu + H[2]O[2] + TMB + L
   treatment groups, the absorption intensity of TMB at 652 nm
   progressively increased over time (Fig. S10A&3B), while negligible
   changes were detected in the absorption curves of TMB + H[2]O[2] and
   TMB + H[2]O[2] + L treatment groups (Fig. S10B&S10C). Additionally, the
   electron spin resonance (ESR) spectra of different treatment groups
   were measured using 5,5-dimethyl-1-pyrroline N-oxide (DMPO) as a
   trapping agent for •OH. The characteristic quartets (1:2:2:1) of the
   DMPO/•OH adduct were observed in CuCl[2] + H[2]O[2], T-T@Cu + H[2]O[2]
   and T-T@Cu + H[2]O[2] + 45℃ treatment groups, but not in T-T@Cu,
   CuCl[2]and H[2]O[2] treatment groups. Notably, the characteristic
   signals of •OH were significantly enhanced with a mild-photothermal
   (45℃) treatment (Fig. [108]3C). The above results verified the
   Fenton-like catalytic activity of T-T@Cu for •OH generation, which
   could be dramatically enhanced by mild-photothermal effects.

Fig. 3.

   [109]Fig. 3
   [110]Open in a new tab

   In vitro •OH generation and photothermal performance study. A)
   Releasing profiles of copper ions from T-T@Cu under various mimic in
   vivo microenvironments. B) UV-vis absorption spectra of TMB incubated
   with T-T@Cu and H[2]O[2] under a 1064 nm laser irradiation for
   different time intervals. C) ESR spectra of various treatment groups
   for the detection of •OH generation. D) Heating curves of T-T@Cu at
   different mass concentrations under a 1064 nm (1 w/cm^2) laser for
   6 min. E) Heating curves of T-T@Cu (80 µg/mL) under a 1064 nm laser
   with different power density for 6 min. F) Infrared thermal images of
   T-T@Cu at different mass concentrations after irradiated by a 1064 nm
   laser with different power density for 6 min. G) Temperature variation
   curve of T-T@Cu during a 1064 nm laser ON/OFF cycle. H) Plot of linear
   fitting curve of cooling time versus – ln (q) of T-T@Cu. I)
   Photothermal stability of T-T@Cu and ICG over four cycles of a 1064 nm
   laser ON/OFF irradiation

Photothermal performance

   Based on the outstanding absorption in NIR II bio-window, the NIR II
   photothermal performance of T-T@Cu was systematically studied. When
   exposed to a 1064 nm NIR II laser, the photothermal performance of
   T-T@Cu was found to be dependent on both concentration and laser power
   density. After 6 min of irradiation with a 1064 nm laser (1 W/cm^2),
   the temperature of T-T@Cu aqueous solutions at various concentrations
   (20 µg/mL, 30 µg/mL, 40 µg/mL, 80 µg/mL and 200 µg/mL) rose from 28℃ to
   44℃, 48℃, 52℃, 61℃ and 71℃, respectively (Fig. [111]3D). Moreover, for
   80 µg/mL T-T@Cu aqueous solution, the ultimate temperatures after 6 min
   of irradiation with different power densities (0.5 W/cm^2[,] 1 W/cm^2
   and 1.5 W/cm^2) were measured to be 43℃, 61℃ and 68℃, respectively
   (Fig. [112]3E). Corresponding infrared thermal imaging photos of T-T@Cu
   aqueous solutions subjected to different doses of irradiation further
   demonstrated its excellent photothermal performance (Fig. [113]3F). The
   photothermal conversion efficiency (h) of T-T@Cu at 1064 nm was
   calculated using a standard detection method. As shown in Fig. [114]3G
   and H, the h value of T-T@Cu was measured to be ~ 48.5%, which is
   higher than that of many traditional photothermal agents, including
   polydopamine NPs (~ 40%), polypyrrole NPs (~ 45%), and Au NPs (~ 22%)
   [[115]43, [116]44].The photostability analysis indicated that there was
   no impairment of the photothermal performance of T-T@Cu after four
   ON/OFF cycles of laser irradiation. In contrast, the photothermal
   performance of indocyanine green (ICG), a food and drug administration
   (FDA) approved photothermal agent, was found to gradually deteriorate
   under the same irradiation conditions (Fig. [117]3I). These results
   demonstrate the outstanding NIR II photothermal performance of T-T@Cu,
   suggesting its potential as a promising photothermal nano-agent for
   cancer therapy.

Efficient cellular internalization of T-T@Cu for scavenging intracellular Cys
and GSH

   Cys and GSH are well-known thiol-containing metal ion chelators and
   intracellular antioxidants that are often overexpressed in tumor
   microenvironments [[118]42, [119]45]. Their elevated levels facilitate
   the strong elimination of copper ions and reactive oxygen species
   (ROS), which can dramatically suppress the processes of cuprotosis and
   ferroptosis [[120]46]. Therefore, scavenging intracellular Cys and GSH
   is considered a promising strategy to enhance cellular cuprotosis and
   ferroptosis for cancer therapy. Moreover, Cys is a necessary amino acid
   for the biosynthesis of GSH and has been reported to undergo selective
   reactions with the cyanide groups of TCNQ [[121]47]. In this context,
   we expected that the designed T-T@Cu could effectively deplete
   intracellular Cys and GSH simultaneously.

   Prior to evaluating the Cys and GSH scavenging activity of T-T@Cu, the
   cellular internalization of T-T@Cu was confirmed using flow cytometry.
   The results showed that the detected fluorescence signals of RhB@T-T@Cu
   in 4T1 cells gradually increased over a 4 h incubation period,
   indicating successful cellular internalization of RhB@T-T@Cu. This
   uptake process was found to be time-dependent (Fig. S11). Following the
   incubation with T-T@Cu and CuCl[2] for 8 h, the intracellular Cys and
   GSH content was measured using the Cys Colorimetric Assay Kit
   (Elabscience) and the Reduced Glutathione (GSH) Content Assay Kit
   (Solarbio), respectively. As shown in Fig. S12A&12B, intracellular Cys
   and GSH content decreased drastically with increasing concentrations of
   T-T@Cu. Although the GSH content in the CuCl[2] treated group showed a
   slight reduction, the magnitude of this reduction was significantly
   lower (~ 2.3-folds) compared to that observed in the T-T@Cu treated
   cells (Fig. S12B). These results revealed the bioactivity of T-T@Cu in
   Cys and GSH depletion, which is conductive for amplifying cuproptosis
   and ferroptosis.

Evaluation of mitochondria targeting of T-T@Cu

   Considering that mitochondria are the primary targets of ferroptosis
   and cuproptosis, the mitochondrial targeting of T-T@Cu was analyzed
   using CLSM and ICP-MS. The mitochondrial colocalization CLSM images
   showed that the red fluorescence of RhB@T-T@Cu was highly overlapped
   with the green fluorescence of MitoTracker Green-labeled mitochondria,
   confirming the excellent mitochondrial targeting capability of T-T@Cu
   (Fig. [122]4A and B). Furthermore, mitochondria from 4T1 cells treated
   with PBS, CuCl[2], T-T@Cu and T-T@Cu + L were extracted using a cell
   mitochondria isolation kit (Beyotime) to determine the copper ion
   content via ICP-MS. The results indicated that the copper ion
   concentration in the mitochondria of T-T@Cu-treated cells was
   approximately 4.6-fold and 22-fold higher than that in the CuCl[2] and
   PBS-treated cells, respectively. Notably, after receiving an extra
   1064 nm laser irradiation, the copper ion concentration in mitochondria
   of T-T@Cu treated cells further increased about 0.2-folds
   (Fig. [123]4C). The significant enrichment of copper ions in the
   mitochondria of T-T@Cu-treated cells suggests a potential for inducing
   serious cytotoxicity in 4T1 cells.

Fig. 4.

   [124]Fig. 4
   [125]Open in a new tab

   T-T@Cu enable high mitochondrial enrichment and photothermal promoted
   cell cuproptosis and ferroptosis in 4T1 cells. A) Intracellular
   location of RhB@T-T@Cu (Red) in Mito-tracker Green labeled 4T1 cells.
   Nuclei were labeled with DAPI (blue). The bar value is 20 μm. B)
   Mitochondria colocalization analysis of T-T@Cu in 4T1 cells. C)
   Quantification of copper ions concentration (Con.) of isolated
   mitochondria from various treatment group after 8 h of incubation by
   ICP-MS. D) Relative cell viability of tumor cells (4T1 and HePG2) and
   normal cells (HUVEC) after incubation with T-T@Cu for 24 h. E) Relative
   cell viability of 4T1 cells in various treatment groups after 8 h of
   incubation. F) CLSM images of 4T1 MCTs upon different treatments and
   dead (red) and live (green) cell staining by Calcein-AM/PI. The bar
   value is 100 μm. G) Immunofluorescence images of DLAT protein in 4T1
   cells of various treatment groups. The bar value is 20 μm. H) WB
   analysis on the expression levels of DLAT, FDX1 and LIAS proteins in
   4T1 cells of various treatment groups. I) Representative TEM images of
   4T1 cells after different treatments. The red dash line highlighted the
   cellular mitochondria. J) Intracellular ATP levels of 4T1 cells after
   different treatments. K) WB analysis on the expression levels of
   copper-ion-efflux proteins ATP7A and ATP7B in various treatment groups.
   Data are denoted as mean ± SD. p-values were calculated using Student’s
   t-test. *p < 0.05, **p < 0.01, ***p < 0.001, NS represents no
   significance

Study the cytotoxicity of T-T@Cu

   Encouraged by the mitochondrial-targeted delivery of copper ions and
   the superior NIR II photothermal performance of T-T@Cu, the
   cytotoxicity of T-T@Cu toward various cell lines, including normal
   cells (HUVEC) and cancer cells (4T1 and HepG2), was investigated. In
   comparation to normal cells (HUVEC), T-T@Cu exhibited cancer cells (4T1
   and HepG2) specific killing efficacy. Briefly, after 24 h of
   incubation, the IC[50] value of T-T@Cu against 4T1 cells and HepG2
   cells were detected to be ~ 12.87 µg/mL and ~ 11.27 µg/mL, while
   negligible cytotoxicity was detected in HUVEC cells incubated with
   T-T@Cu at the same concentration (Fig. [126]4D). As a control group,
   the IC[50] of CuCl[2] toward 4T1 cells was determined to be about
   7.1-folds higher than that of T-T@Cu (Fig. S13). This distinct
   difference in cytotoxicity can be attributed to the efficient delivery
   of copper ions to the mitochondria by T-T@Cu (Fig. [127]4C).
   Thereafter, the cytotoxicity of T-T@Cu under a 1064 nm laser
   irradiation was further evaluated. After 8 h of incubation, the cell
   viability of 4T1 cells in various treatment groups exhibited a
   concentration-dependent decline of T-T@Cu. The optimal therapeutic
   efficacy was observed in T-T@Cu + L treatment group. Noteworthy, when a
   cuprotosis inhibitor (ammonium tetrathiomolybdate (ttm)) or a
   ferroptosis inhibitor (ferrostatin-1 (Fer-1)) were added, the cell
   viability of 4T1 cells in the corresponding treatment groups increased
   slightly. This observation provides evidence for the occurrence of
   cuprotosis and ferroptosis in 4T1 cells (Fig. [128]4E&S14). To further
   evaluate the in vitro antitumor efficacy, multicellular tumor spheroids
   (MCTs) of 4T1 cells receiving different treatments were stained with
   Calcein-AM/PI to observe live/dead cells via CLSM. The red fluorescence
   from PI-stained dead cells was found to gradually increase in the
   T-T@Cu, T-T@Cu + ttm + L, and T-T@Cu + L treatment groups, further
   confirming the satisfactory in vitro antitumor efficacy of T-T@Cu
   (Fig. [129]4F).

Exacerbate tumor cells cuproptosis via mild-photothermal boosted ferroptosis

   To elucidate whether the mechanisms of cell death were indeed
   ferroptosis and cuproptosis, we further investigated the hallmarks of
   ferroptosis including intracellular ROS, LPO levels and
   ferroptosis-related protein, as well as the expression levels of
   cuproptosis-related protein across various treatment groups. Firstly,
   we employed 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) as a
   ROS probe to quantitatively analyze intracellular ROS levels via flow
   cytometry. As shown in Fig. S15A&S15B, the fluorescence intensity of
   DCFH-DA-labeled intracellular ROS in the T-T@Cu and T-T@Cu + L
   treatment groups was found to be 4.4-folds and 6.0-folds higher,
   respectively, compared to the cells treated with PBS. This indicates
   the superior ROS production capacity of T-T@Cu in 4T1 cells [[130]48].
   Notably, after an additional irradiation with a 1064 nm laser, the
   fluorescence intensity of DCFH-DA-labeled intracellular ROS in the
   T-T@Cu treatment group was dramatically enhanced, indicating the
   promotion of mild-photothermal effects on ROS generation. Subsequently,
   BODIPY™ 581/591 C11 were used to quantitatively analyze intracellular
   LPO level, another hallmark of ferroptosis, via flow cytometry. BODIPY™
   581/591 C11 acts as a LPO sensor that transitions its fluorescence from
   red to green upon oxidation by LPO. As a results, T-T@Cu + L treatment
   group showed the highest ratios of green fluorescence to red
   fluorescence, indicating robust intracellular LPO accumulation for
   triggering mild-photothermal boosted cell ferroptosis (Fig. S16A-S16D).
   Furthermore, Western blot (WB) analysis confirmed that T-T@Cu treatment
   significantly downregulated GPX4, a key ferroptosis marker protein, in
   4T1 cells (Fig. S17). Notably, this suppression was further enhanced
   upon laser irradiation, suggesting that the mild photothermal effect of
   T-T@Cu synergistically amplifies ferroptosis signaling. Subsequently,
   we analyzed the hallmarks of cuproptosis, specifically focusing on
   lipoylated dihydrolipoamide acetyltransferase (DLAT), ferredoxin
   (FDX1), and lipoyl synthase (LIAS) proteins using immunofluorescence
   and WB analysis. The immunofluorescence images of 4T1 cells treated
   with PBS, PBS + L, CuCl[2], T-T@Cu + ttm and T-T@Cu + ttm + L showed a
   uniformly distributed DLAT protein with no obvious foci detected. In
   contrast, 4T1 cells treated with T-T@Cu and T-T@Cu + L exhibited
   pronounced DLAT foci, indicating that T-T@Cu can induce the formation
   of DLAT protein oligomers in these cells (Fig. [131]4G). Besides, the
   WB results demonstrated a significant upregulation of DLAT protein
   oligomers in the T-T@Cu and T-T@Cu + L treatment groups. Conversely,
   the expression levels of FDX1 and LIAS proteins in these groups were
   dramatically downregulated. In particular, after irradiation with a
   1064 nm laser, the variation amplitude of the expression levels of DLAT
   protein oligomers, FDX1 and LIAS in T-T@Cu treatment group were further
   increased, which might be ascribe to the exacerbation of cell
   cuproptosis by the mild-photothermal boosted ferroptosis. Notably, when
   the cuproptosis inhibitor ttm was added to the T-T@Cu and T-T@Cu + L
   treatment groups, the expression levels of DLAT protein oligomers,
   FDX1, and LIAS proteins returned to levels similar to those observed in
   the PBS treatment group (Fig. [132]4H). Moreover, the level of DLAT
   protein oligomers in 4T1 cells was found to be further upregulated with
   increasing concentrations of T-T@Cu, while the expression levels of
   FDX1 and LIAS proteins were detected to be downregulated (Fig. S18).
   These results collectively demonstrated that T-T@Cu + L treatment can
   indeed induce the proteotoxic aggregation of lipoylated DLAT, as well
   as the destabilization of FDX1 and LIAS proteins, indicating the
   occurrence of photothermal-enhanced cell cuproptosis.

   To gain deeper insight into the influence of mild-photothermal-boosted
   ferroptosis on the process of cell cuproptosis, 4T1 cells were
   incubated with CuCl[2] and T-T@Cu at the same copper ion concentration
   for various time intervals, followed by different treatments (CuCl[2],
   T-T@Cu and T-T@Cu + L). After the incubation, the intracellular copper
   ion concentration in 4T1 cells was quantified by ICP-MS. As depicted in
   Fig. S19, the highest intracellular copper ion concentration was
   detected in the T-T@Cu + L treatment group, regardless of whether the
   incubation time was 6, 8, or 10 h. Notably, after 8 h of incubation,
   the intracellular copper ion concentration in the T-T@Cu + L treatment
   group was found to be 5.4-fold higher than that in the CuCl[2]
   treatment group and 1.5-fold higher than that in the T-T@Cu treatment
   group. These results suggested that mild-photothermal treatment
   enhanced the enrichment of intracellular copper ions, thereby leading
   to increased cell cuproptosis. Numerous studies have shown that
   excessive intracellular ROS can cause significant mitochondrial damage,
   leading to reduced levels of adenosine triphosphate (ATP), which is
   crucial for the expression of copper-ion-efflux proteins ATP7A and
   ATP7B [[133]49, [134]50]. Given our previous findings that photothermal
   treatment can elevate intracellular ROS levels, we speculated that the
   high levels of intracellular copper ion enrichment might be attributed
   to the mild-photothermal-induced downregulation of ATP7A and ATP7B,
   resulting in decreased copper ion efflux. To validate this hypothesis,
   we observed mitochondrial morphology using TEM and quantitatively
   analyzed intracellular ATP and ATP7A/7B protein levels across various
   treatment groups. As shown in Fig. [135]4I&S20, ROS induced
   mitochondrial damage was observed, including decreased mitochondrial
   volume, reduced or absent cristae, and increased membrane density in
   the T-T@Cu and T-T@Cu + L treatment groups. Subsequently, to assess
   mitochondrial membrane potential (ΔΨm), JC-1 staining was performed
   followed by confocal laser scanning microscopy (CLSM) analysis.
   Compared to PBS-treated controls, cells treated with T-T@Cu or T-T@Cu +
   L showed progressive ΔΨm loss, evidenced by reduced JC-1 aggregation
   (red fluorescence) in mitochondria. This manifested as a marked
   increase in green fluorescence (JC-1 monomers) concomitant with a
   decrease in red fluorescence (JC-1 aggregates) (Figure S21). These
   results confirm that both T-T@Cu and T-T@Cu + L induce mitochondrial
   depolarization, ultimately causing severe mitochondrial impairment.
   Furthermore, the intracellular ATP levels in the T-T@Cu and T-T@Cu + L
   treatment groups were determined to be 0.16-folds and 0.28-folds lower
   than that in the PBS treatment group, confirming significant
   down-regulation of ATP levels due to these treatments (Fig. [136]4J).
   In addition, the expression of ATP7A and ATP7B proteins in T-T@Cu and
   T-T@Cu + L treatment groups were found to be dramatically inhibited.
   Particularly, after irradiation, the expression levels of ATP7A and
   ATP7B proteins in the T-T@Cu treatment group were further
   down-regulated (Fig. [137]4K). These results demonstrated that
   mild-photothermal treatment leaded to the down-regulation of
   copper-ion-efflux proteins ATP7A and ATP7B, which helped maintain high
   intracellular copper ion concentrations and further exacerbated cell
   cuproptosis.

Transcriptomic analysis of anti-cancer mechanism via RNA sequencing

   RNA sequencing is a powerful technology that enables high-throughput
   sequencing of the transcriptome, identification of differentially
   expressed genes, and facilitates enrichment analysis using the Kyoto
   Encyclopedia of Genes and Genomes (KEGG), Gene Ontology (GO), Reactome,
   and WikiPathways. Herein, transcriptomic analysis was conducted via RNA
   sequencing to elucidate the anti-cancer mechanisms underlying the
   treatment of 4T1 cells across various experimental groups. Among the
   genes analyzed, 3262 and 3754 genes were uniquely expressed in 4T1
   cells treated with T-T@Cu and T-T@Cu + L, respectively, indicating a
   significant impact of these treatments on cellular gene expression
   (Fig. [138]5A). Comparative analysis with the PBS treatment group
   revealed that 50, 61, 2872, and 3132 genes were significantly
   up-regulated, while 48, 69, 2004, and 2082 genes were significantly
   down-regulated in the PBS + L, CuCl[2], T-T@Cu, and T-T@Cu + L
   treatment groups, respectively (Fig. [139]5B-E). Furthermore, in
   comparison to T-T@Cu treatment group, 4T1 cells in T-T@Cu + L treatment
   group showed 242 down-regulated genes and 77 up-regulated genes (Fig.
   [140]5F). KEGG and GO enrichment analyses of the differentially
   expressed genes indicated that multiple signaling pathways were
   significantly affected following T-T@Cu + L treatment. Notably, the
   glutathione metabolism pathway was significantly impacted, suggesting
   that T-T@Cu depletes GSH, thereby enhancing cellular cuproptosis and
   ferroptosis. Additionally, the enrichment of the ferroptosis pathway
   indicated its activation in 4T1 cells post-treatment with T-T@Cu + L,
   while interference with the mitochondrial respiratory chain complex IV
   pathway and ATP binding pathway might have contributed to alterations
   in intracellular ATP levels [[141]51, [142]52]. Moreover, T-T@Cu + L
   treatment greatly influenced genes related to p53 signaling pathway and
   cell cycle pathway, both of which are associated with cuproptosis
   [[143]53, [144]54]. Notably, several immune-related pathways, including
   NF-kappa B signaling pathway, Th17 cell differentiation pathway, and T
   cell-mediated cytotoxicity pathway, were significantly enriched in 4T1
   cells, suggesting that T-T@Cu + L treatment may potentially elicit an
   immune response in vivo (Fig. [145]5G and H).

Fig. 5.

   [146]Fig. 5
   [147]Open in a new tab

   Transcriptomic analysis upon RNA-sequencing reveals the anti-cancer
   mechanism. A) A Venn diagram revealed the intersection of the gene
   expression level of each treatment group. B-F) Volcano plots of
   different treatment groups showed the differentially expressed genes.
   G) KEGG and H) GO pathway enrichment analysis between PBS and
   T-T@Cu + L treatment groups. I) GSEA showed the gene sets of Cellular
   response to reactive oxygen species, Mitochondrion morphogenesis, metal
   ion transmembrane transporter activity and iron-sulfur cluster binding
   pathways. J) Heat map of differentially transcribed genes of interest
   of 4T1 cells in various treatment groups. K) Protein–protein
   interaction network (PPI)

   To further investigate the anti-cancer mechanisms, Gene Set Enrichment
   Analysis (GSEA) was performed to explore specific changes in relevant
   pathways. The results indicated an enhancement of the cellular response
   to reactive oxygen species pathway, alongside the inhibition of
   pathways related to mitochondrion morphogenesis, metal ion
   transmembrane transporter activity, and iron-sulfur cluster binding,
   which aligns with the KEGG results and further supports that T-T@Cu + L
   exacerbates cuproptosis in tumor cells through mild
   photothermal-boosted ferroptosis (Fig. [148]5I). Additionally, the heat
   map generated from the various treatment groups revealed the major
   differentially expressed genes (Fig. [149]5J). Notably, genes
   associated with the iron-sulfur cluster binding pathway (e.g., Iscu,
   Fech, Dph1, and Dph2) and the ferroptosis pathway (e.g., Acsl4, Acsl6,
   Map1lc3b, and Sat1) exhibited significant down-regulation or
   up-regulation in the T-T@Cu and T-T@Cu + L treatment groups, indicating
   successful induction of cellular cuproptosis and ferroptosis.
   Furthermore, genes involved in ATP synthesis, including Cox11, Cox6a2,
   Mfn2, and Mfn1, were down-regulated in the T-T@Cu and T-T@Cu + L
   treatment groups, suggesting that high levels of intracellular
   oxidative stress inhibit ATP synthesis, consistent with previous
   quantitative assessments of intracellular ATP levels. Finally, a
   protein-protein interaction (PPI) network was plotted to illustrate the
   relationships among the corresponding proteins in the relevant pathways
   (Fig. [150]5K).

In vivo biosafety evaluation of T-T@Cu

   Prior to in vivo application, the biosafety of T-T@Cu was thoroughly
   evaluated through several assays, including hemolysis assessment, blood
   biochemical analysis, and histological examination via H&E staining.
   The hemolysis ratio of T-T@Cu at a mass concentration of up to 80 µg/mL
   was determined to be 3.3%. This low hemolysis ratio indicated excellent
   hemocompatibility of T-T@Cu, as well as its feasibility for intravenous
   administration (Fig. S22). Blood biochemical analysis was conducted on
   day fourteen following three intravenous doses of T-T@Cu (4.4 µg/g body
   weight) in mice. Compared to PBS-treated mice, there were no
   significant differences in blood biochemical indices, including ALT,
   AST, BUN, CREA, CK, and CK-MB, in mice treated with T-T@Cu. This
   suggested that there was no loss of cardiac and hepatorenal function in
   the treated mice (Fig. S23A). Besides, H&E staining was performed on
   the main organs, including the heart, liver, spleen, lung, and kidney,
   from T-T@Cu-treated mice. The H&E staining images revealed no serious
   cell damage, such as changes in cell morphology or condensed nuclei,
   further demonstrating the biosafety of T-T@Cu for potential in vivo
   applications (Fig. S23B).

In vivo biodistribution and metabolism of T-T@Cu

   To address complex clinical demands, ideal nanodrugs should exhibit
   multiple functionalities, including prolonged blood circulation,
   specific tumor-targeting, and effective metabolism [[151]55].
   Therefore, we systematically investigated the in vivo biodistribution
   and metabolism of T-T@Cu following intravenous administration in mice
   using ex vivo fluorescence imaging and ICP-MS analysis. Compared to
   CuCl[2] treated mice, the half-life of copper in the blood of T-T@Cu
   treated mice was found significantly prolonged, which guaranteed the
   subsequent ongoing accumulation of T-T@Cu in tumors (Fig. [152]6A).
   ICP-MS analysis revealed that the intra-tumoral copper content in
   T-T@Cu-treated mice was much higher than that in CuCl[2] treated mice,
   confirming the specific tumor-targeting capability of T-T@Cu (Fig.
   [153]6B). Furthermore, in vivo infrared photothermal imaging was
   conducted on PBS and T-T@Cu-treated mice. As shown in Fig. [154]6C and
   D, negligible temperature increase was observed in PBS-treated mice,
   while T-T@Cu-treated mice exhibited significant temperature variation,
   with tumor site temperatures reaching approximately 45℃ after 80 s of
   irradiation. This temperature rise further supported the substantial
   accumulation of T-T@Cu at the tumor site. Additionally, fluorescence
   images of dissected tissues, including heart, liver, spleen, lung,
   kidney, intestine and tumor, were obtained at pre-designed time
   intervals. The fluorescence of RhB@T-T@Cu was primarily detected in the
   liver, kidney, intestine, and tumor of treated mice, with fluorescence
   intensity gradually decreasing over 24 h (Fig. [155]6E and F&S24). The
   gradual attenuation of fluorescence in these tissues might be
   attributed to the high expression of GSH and Cys, which induced the
   degradation of RhB@T-T@Cu in these tissues. To further evaluate the
   metabolism process of T-T@Cu in vivo, feces and urine from mice
   administered with PBS and T-T@Cu were collected for ICP-MS measurement.
   As depicted in Fig. [156]6G and H, significantly increased copper
   concentrations were detected in the feces and urine of T-T@Cu-treated
   mice within 72 h post-administration, suggesting the hepatic and renal
   metabolic pathways of T-T@Cu. Taken together, T-T@Cu demonstrated long
   blood circulation, specific tumor targeting, and effective metabolism
   in vivo, which addressed the multifaceted requirements of nanodrug
   delivery systems.

Fig. 6.

   [157]Fig. 6
   [158]Open in a new tab

   In vivo biodistribution and metabolism of T-T@Cu. A) Pharmacokinetic
   analysis of CuCl[2] and T-T@Cu via ICP-MS quantification of copper ion
   concentration in blood. B) Copper ion concentrations of dissected
   tumors from CuCl[2] and T-T@Cu treated mice at predesigned time
   intervals. C) Temperature variation curves of tumors of mice received
   intravenous injection of PBS and T-T@Cu during 80 s of irradiation. D)
   Time-resolved infrared photothermal images of 4T1 tumor-bearing mice in
   different treatment groups. E) Ex vivo fluorescence distribution images
   of dissected main organs and tumors from RhB@T-T@Cu treatment groups
   (n = 3) at predesigned time intervals after intravenous administration.
   F) Average fluorescence intensity semiquantitative analysis of excised
   main organs and tumors of mice in various treatment groups.
   Time-dependent excretion of copper ions from G) feces and H) urine of
   mice received intravenous administration of PBS and T-T@Cu. Data are
   denoted as mean ± SD. p-values were calculated using Student’s t-test.
   *p < 0.05, **p < 0.01, ***p < 0.001

Exploration of in vivo anti-tumor efficacy and mechanism

   Encouraged by the promising in vitro anti-cancer efficacy and the
   remarkable in vivo biosafety and tumor-targeting properties of T-T@Cu,
   the in vivo anti-tumor effect of various treatments including PBS,
   PBS + L, CuCl[2], T-T@Cu, T-T@Cu + ttm, T-T@Cu + ttm + L and T-T@Cu + L
   was comprehensively investigated. BABL/C mice were selected to
   establish a subcutaneous 4T1 tumor model. As illustrated in
   Fig. [159]7A, mice received four intravenous doses of the different
   drug formulations, along with three sessions of irradiation during the
   therapeutic period. For in vivo therapeutic efficacy evaluation, tumor
   volume, tumor weight, and body weight of the mice were monitored daily
   throughout the treatment. The tumor growth inhibition data revealed a
   dramatic suppression of tumor volume and weight in the T-T@Cu,
   T-T@Cu + ttm, T-T@Cu + ttm + L, and T-T@Cu + L treatment groups
   (Fig. [160]7B and C). Moreover, excised tumors demonstrated that tumors
   in the T-T@Cu + L treatment group were thoroughly cured (Fig. [161]7D).
   Also, H&E staining of tumor tissues from various treatment groups
   showed significant cellular damage, including changes in cell
   morphology and nuclear condensation, particularly in the T-T@Cu + L
   treatment group (Fig. S25). Noteworthy, the introduction of ttm
   significantly impaired the therapeutic efficacy of the corresponding
   treatment group, suggesting that cuproptosis played an important role
   in the tumor inhibition mechanism of T-T@Cu. Finally, no significant
   fluctuations in body weight were observed in the treatment groups,
   indicating excellent biosafety of the formulations (Fig. [162]7E).

Fig. 7.

   [163]Fig. 7
   [164]Open in a new tab

   Mild-photothermal promoted cell cuproptosis and ferroptosis in
   subcutaneous 4T1 breast tumor models. A) Schematic illustration of the
   establishment of subcutaneous 4T1 breast tumor models in mice and the
   therapeutic regimen. B) Tumor growth curves of mice upon various
   treatments. C) Ex vivo tumor weight and D) tumorous digital photos of
   mice upon various treatments. E) Body weight variation of mice in
   different treatment groups during 14 days. F) DLAT, FDX1, LIAS and GPX4
   immunohistochemical staining as well as Tunnel staining of tumor tissue
   sections of each treatment groups. The green fluorescence represents
   DLAT, FDX1, LIAS proteins and apoptotic DNA fragmentation staining,
   while the blue fluorescence indicates nuclear staining. Data are
   denoted as mean ± SD. p-values were calculated using Student’s t-test.
   *p < 0.05, **p < 0.01, ***p < 0.001

   To figure out the potential in vivo anti-tumor mechanism of T-T@Cu, the
   expression of cuproptosis and ferroptosis related proteins including
   DLAT, FDX1, LIAS and GPX4 in tumors were evaluated via
   immunofluorescence and immunohistochemical analyses. Compared to the
   PBS, PBS + L, CuCl[2], T-T@Cu + ttm, and T-T@Cu + ttm + L treatment
   groups, tumors treated with T-T@Cu and T-T@Cu + L exhibited a
   progressive increase in DLAT foci, alongside a marked reduction in
   FDX1, LIAS, and GPX4 expression (Fig. [165]7F). Notably, the
   significantly downregulation of GPX4 in tumor tissues, indicating
   ferroptosis induction in vivo. To further validate this, we assessed
   additional ferroptosis biomarkers, including intratumoral ROS levels
   and malondialdehyde (MDA), a downstream product of lipid peroxidation.
   As shown in Figure S26&S27, T-T@Cu and T-T@Cu + L treated groups showed
   distinct DCFH-DA marked green fluorescence of ROS and up-regulation of
   MDA in tumors, indicating T-T@Cu and T-T@Cu + L treatments effectively
   induced accumulation of intratumoral ROS and lipid peroxides. These
   findings provide evidence for photothermal-promoted cuproptosis and
   ferroptosis in these tumors. Furthermore, the administration of ttm
   resulted in the loss of expression variation for DLAT, FDX1, and LIAS
   in the T-T@Cu and T-T@Cu + L treatment groups, further supporting the
   involvement of cuproptosis in tumor cell death. Additionally, terminal
   deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL)
   staining of tumor tissues revealed that T-T@Cu + L induced the most
   significant apoptosis among the treatment groups (Fig. [166]7F). Taken
   together, T-T@Cu could effectively suppresses tumor growth through the
   activation of cuproptosis, ferroptosis, and apoptosis pathways.

Conclusion

   In summary, we successfully constructed a carrier-free metal-polyphenol
   nanoplatform (T-T@Cu) through the self-assembly of TA, TCNQ, and copper
   ions to facilitate cuproptosis in tumor cells. This nanoplatform
   exhibited excellent tumor-mitochondria cascade targeting, GSH
   depletion, Fenton-like catalytic activity, and NIR II photothermal
   conversion properties (h = 48.5%). Once internalized by tumor cells,
   T-T@Cu not only accumulated specifically in the mitochondria but also
   consumed the overexpressed GSH present in tumor microenvironments,
   alleviating GSH-induced copper inactivation. Furthermore, T-T@Cu
   demonstrated Cys/GSH dual-responsive degradation and controllable
   copper ion release features in the tumor Cys/GSH microenvironment. Upon
   irradiation by a 1064 nm laser, T-T@Cu triggered mild
   photothermal-boosted ferroptosis in tumor cells, impacting cellular ATP
   synthesis. This process further disrupted intracellular copper
   homeostasis by downregulating the expression of copper-ion efflux
   proteins ATP7A/7B, leading to enhanced mitochondrial copper
   accumulation and exacerbated cuproptosis in tumor cells. Based on these
   ideal characteristics, effective tumor targeting, high tumor inhibition
   rate (~ 100%) and no obvious systematic toxicity were achieved. This
   novel strategy for exacerbating tumor cell cuproptosis shows great
   promise for future clinical applications in oncotherapy.

Supplementary Information

   [167]Supplementary Material 1.^ (7.7MB, docx)

Acknowledgements