Mitochondria-targeting Cu3VS4 nanostructure with high copper ionic mobility for photothermoelectric therapy

Thermoelectric therapy has emerged as a promising treatment strategy for oncology, but it is still limited by the low thermoelectric catalytic efficiency at human body temperature and the inevitable tumor thermotolerance. We present a photothermoelectric therapy (PTET) strategy based on triphenylphosphine-functionalized Cu3VS4 nanoparticles (CVS NPs) with high copper ionic mobility at room temperature. Under near-infrared laser irradiation, CVS NPs not only generate hyperthermia to ablate tumor cells but also catalytically yield superoxide radicals and induce endogenous NADH oxidation through the Seebeck effect. Notably, CVS NPs can accumulate inside mitochondria and deplete NADH, reducing ATP synthesis by competitively inhibiting the function of complex I, thereby down-regulating the expression of heat shock proteins to relieve tumor thermotolerance. Both in vitro and in vivo results show notable tumor suppression efficacy, indicating that the concept of integrating PTET and mitochondrial metabolism modulation is highly feasible and offers a translational promise for realizing precise and efficient cancer treatment.


Photothermal effect of CVS NPs
CVS NPs solutions with gradient concentrations (0, 50, 100, 200, and 400 μg mL -1 ) were irradiated under 808 nm laser (1.5 W cm −2 ) for 5 min, and the temperature variations of CVS NPs solution were recorded by an IR thermographic camera.Moreover, the influence of power density was also investigated by recording the temperature of CVS NPs aqueous solution (200 μg mL -1 ) at varied power densities (0.5, 1.0, 1.5, and 2.0 W cm −2 ) of the 808 nm laser for 10 min.To further evaluate the photothermal stability of CVS NPs, CVS NPs solution (100 μg mL −1 ) was irradiated by an 808 nm laser at 1.5 W cm −2 during the heating and cooling processes for three cycles.To study the photothermal conversion effect, the CVS NPs aqueous solution (100 μg mL -1 ) was irradiated under an 808 nm laser (1.5 W cm −2 ) for 10 min, followed by shutting off the NIR laser.The photothermal conversion efficiency (η) of CVS NPs can be calculated according to the following equation: where h (mW m -2 °C-1 ) indicates the thermal conversion efficiency of the system, S (m 2 ) is the surface area of the container, Tmax (°C) represents the equilibrium temperature of the sample solution, Tsurr represents the surrounding ambient temperature, I is the power density of the 808 nm laser (mW), and A808 is the absorbance of CVS NP solution at λ = 808 nm.Besides, Qdis represents the heat loss due to the light absorption of the container itself, and it was determined as Qdis = (5.4 × 10 -4 ) I (mW).To calculate hS, another equation was introduced: where m is the mass of the sample, Cwater is the heat capacity of water (4.2 J g -1 K -1 ), and the time constant for heat transfer from the system is determined to be τs = 559.50s.Therefore, the hS is calculated to be 7.51 mW °C-1 .In addition, the (TMax-TSurr) is 35.1 °C, I808 is 1500 mW, and A808 is 0.798.By substituting these values into these equations, the 808 nm laser photothermal conversion efficiency (η) of CVS NPs was calculated to be 32.78%.

FDTD simulations
The finite element simulations were performed by using the finite difference-time-domain (FDTD) method with Lumerical FDTD solutions, a software package.The nanostructure is simulated in xy planes with periodic boundary conditions while the broadband plane waves are incident from z directions.Along the propagation directions (z directions), perfect matched layers (PML) are used to absorb all the light coming out to the boundaries.The complex refractive indexes were adopted from the reference (67).For structures dispersed in aqueous solutions, the refractive index of the medium (H2O) was set to 1.33.The geometric parameters for simulations were consistent with the average actual size of the as-prepared samples shown in the TEM image.In the FDTD simulations, the side length of the CVS nanocube was set as 17.4 nm.For the detailed FDTD parameter setting, a unit cell of 300 nm × 300 nm × 300 nm in 3D was employed for the nanostructure.Total-field scattered-field light source with a wavelength range of 400-900 nm was injected into the unit cell along the positive z-axis.The frequency-domain field profile monitor with a fixed wavelength of 808 nm was introduced to acquire field profile data, localized at Z = 0 nm in the x-y plane, Y = 0 nm in the x-z plane, and X = 0 nm in the y-z plane, respectively.

Thermoelectric property measurements
To estimate the thermoelectric properties, the as-prepared powders were first pressed into bulk

Cell culture
The 4T1 murine mammary carcinoma cell line was acquired from FDCC (Ruilu in Shanghai, China) and cultured in RPMI 1640 medium with 10% FBS and 1% Penicillin-Streptomycin. Raw 264.7 murine mononuclear macrophage cell line was obtained from the American Type Culture Collection (ATCC) and cultured in Dulbecco's Modified Eagle Medium supplemented with 10% FBS and 1% Penicillin-Streptomycin.The cell lines were incubated in the incubator.All the cultures were conducted in the incubator at 37 °C and 5% CO2 atmosphere.

In vivo biodistribution of CVS NPs
Female BALB/c mice (n = 3) were i.v.injected with CVS NPs (100 μL, 15 mg kg −1 ).At 1, 3, 6, 12, 24, and 48 h post-injection, these mice were sacrificed to collect their major organs including liver, spleen, kidney, heart, lung, and tumor for biodistribution analysis via an inductively coupled plasma−optical emission spectrometry (ICP-OES).The biodistributions of CVS NPs in different organs and tumors were calculated as the Cu percentage of the injected dose per gram of tissue.

Histochemical analysis
At the end of therapy, tumor tissues of mice in each treatment group were collected followed by fixed in 37% formalin and embedded in paraffin for one day.Then, paraffin-embedded tumor sections (4 μm) were cut by a microtome (Leica RM2235, Germany).For the pathological investigation, the obtained samples were stained with hematoxylin and eosin (H&E) and TdTmediated dUTP nick-end labeling (TUNEL) by following the standard protocol, and the slides were visualized using a CLSM.For immunofluorescence staining, the sections were placed into xylene two times for 15 min to remove paraffin, hydrated through an ethanol gradient (100%, 95%, 90%, 80%, and 70%), and deionized water for 5 min in each solution.Then, the antigen retrieval was conducted by sodium citrate buffer solution (0.01 M) at 95 °C for 15 min.Subsequently, the obtained samples were incubated with primary antibodies of Anti-HSP70 antibody (1:200 dilution), at 4 °C overnight, followed by incubation with Alexa Fluor488-conjugated anti-rabbit IgG antibody (1:1000 dilution) for 1.5 h at 37 °C.After washing with PBS for three times, the samples were stained with DAPI solution for 5 min at room temperature.At last, the stained tissues were photographed by using a CLSM.

1, 2 -
photoelectron spectroscopy (XPS) spectra were carried out using an ESCALAB 250Xi.Fourier transform infrared (FTIR) spectrum was conducted on a Perkin-Elmer 580b spectrophotometer.The dynamic light scattering and zeta potential measurements for different samples were performed on a Malvern Zetasizer Nan Nano ZS90.A UV-1601 spectrophotometer was used for obtaining the specimens at 623 K for 15 min under 45 MPa pressure in a vacuum atmosphere, with a heating rate of 100 °C min −1 .The resulting cylindrical tablets were sectioned into bars of 11.00 mm × 2.00 mm× 2.00 mm for simultaneous measurements of the temperature-dependent electrical conductivity (σ) and Seebeck coefficient (S) over 300-923 K using a ZEM-3 instrument (Ulvac-Riko ZEM-3).The thermal conductivity (κ) was obtained from κ = D × CP × d, where D, CP, and d mean the thermal diffusivity, heat capacity, and mass density, respectively.The thermal diffusivity (D) was detected using an LFA 457 (NETZSCH) under high-purity helium atmosphere.The heat capacity (CP) was estimated by using the Dulong-Petit rule.The Archimedes method was performed to calculate the density (d).

Fig. S2 .
Fig. S2.FT-IR spectra and TGA curves.(A) FT-IR spectra of Cu3VS4, TPP, and CVS.Compared to the Cu3VS4 sample, the additional peaks of benzene from 1600 to 1400 cm −1 in the CVS sample indicate the successful modification of DSPE-PEG-TPP.(B) Thermogravimetric analysis (TGA) curves of Cu3VS4 and CVS.

Fig. S3 .
Fig. S3.TEM image and size distribution.(A) TEM image and (B) particle-size distribution of CVS NPs.

Fig. S6 .
Fig. S6.UV-vis spectra and size distribution.(A) UV-vis absorbance spectra and (B) particlesize distributions of CVS NPs before and after laser radiation for 30 min.

Fig. S7 .
Fig. S7.Linear fit of PA intensity of CVS NPs as a function of sample concentration at 808 nm.

Fig. S12 .
Fig. S12.Band structure of Cu3VS4 with the optical transitions for absorption peak.

Fig. S13 .
Fig. S13.Schematic illustration for the redox reaction of NBT with •O2 -, which can be used for monitoring intracellular •O2 -content.

Fig. S14 .
Fig. S14.Valence band XPS spectrum and schematic illustration.(A) Valence band XPS spectrum of Cu3VS4 NPs, representing the valence band maximum position.(B) Schematic illustration of the energy bands of Cu3VS4 NPs, showing that the formation of •O2 − free radical is possible.

Fig. S17 .
Fig. S17.Cell viability of RAW 264.7 cells after incubation with CVS NPs for 24 h.

Fig. S20 .
Fig. S20.In vitro T1-MR images and relaxation rate r1 versus Cu concentration.(A) In vitro T1-MR images of CVS NPs with different concentrations (obtained under 9.4-T magnetic resonance scanner).(B) Relaxation rate r1 versus Cu concentration of CVS NPs with or without the addition of H2O2.

Fig. S21 .
Fig. S21.PA images and quantification of PA intensities.(A) PA images of living mice bearing xenograft 4T1 tumors at 0, 3, 6, 12, 24, and 48 h after i.v.injection of CVS NPs.The PA images were acquired at 808 nm.(B) Quantification of PA intensities at the tumor regions after various durations.(n = 3).

Fig. S22 .
Fig. S22.Blood circulation curve and accumulated Cu amount.(A) Blood circulation curve of intravenously injected CVS NPs.Data are expressed as mean ± S.D. (n = 3).(B) Accumulated Cu in feces and urine excreted out of the mouse body within 14 days after i.v.injection of CVS NPs (n = 3).

Fig. S24 .
Fig. S24.NADH/NAD + redox ratios in the tumor regions of mice after different treatments.Statistical analysis was performed via one-way ANOVA with Tukey's multiple comparisons post hoc test.*P < 0.05; **P < 0.01; n.s., not significance.Data were represented as mean ± S.D. from three independent replicates.