Engineering biocompatible TeSex nano-alloys as a versatile theranostic nanoplatform

Abstract Photothermal nanotheranostics, especially in the near infrared II (NIR-II) region, exhibits a great potential in precision and personalized medicine, owing to high tissue penetration of NIR-II light. NIR-II-photothermal nanoplatforms with high biocompatibility as well as high photothermal effect are urgently needed but rarely reported so far. Te nanomaterials possess high absorbance to NIR-II light but also exhibit high cytotoxicity, impeding their biomedical applications. In this work, the controllable incorporation of biocompatible Se into the lattice of Te nanostructures is proposed to intrinsically tune their inherent cytotoxicity and enhance their biocompatibility, developing TeSex nano-alloys as a new kind of theranostic nanoplatform. We have uncovered that the cytotoxicity of Te nanomaterials primarily derives from irreversible oxidation stress and intracellular imbalance of organization and energy, and can be eliminated by incorporating a moderate proportion of Se (x = 0.43). We have also discovered that the as-prepared TeSex nano-alloys have extraordinarily high NIR-II-photothermal conversion efficiency (77.2%), 64Cu coordination and computed tomography contrast capabilities, enabling high-efficacy multimodal photothermal/photoacoustic/positron emission tomography/computed tomography imaging-guided NIR-II-photothermal therapy of cancer. The proposed nano-alloying strategy provides a new route to improve the biocompatibility of biomedical nanoplatforms and endow them with versatile theranostic functions.


INTRODUCTION
Nanotheranostics makes use of nanotechnology to integrate diagnostics and therapeutics, exhibiting a great potential in precision and personalized medicine [1][2][3]. The emergence of diverse multifunctional nanomaterials and advanced nanotechnologies unprecedentedly simulates the evolution of nanotheranostics, and enables the integration of multimodal imaging and therapeutic functions in a single theranostic nanoplatform for high-efficacy theranostics of diseases [4][5][6]. In engineering of theranostic nanoplatforms, biocompatibility and multifunction are two of the most important factors which need to be considered. Among various nanotheranostics, multimodal imaging-guided photothermal therapy has attracted intensive attention owing to the fact that it is less invasive and has fewer side effects compared with conventional radiotherapy and chemotherapy [7][8][9][10][11]. In recent years, a number of photothermal theranostic nanoagents, including noble metal nanoparticles [12][13][14][15][16], two-dimensional (2D) nanosheets [17][18][19][20][21][22][23][24][25][26] and organic polymer nanomaterials [27][28][29][30], have been explored for cancer treatment, but most of them only work in the NIR-I region and the candidates of theranostic nanomedicine for NIR-II-thermal imaging and therapy are quite rare. There are many photothermal theranostic nanoagents with strong absorbance ranging from NIR-I to NIR-II. However, photothermal performance of nanoagents C The Author(s) 2020. Published by Oxford University Press on behalf of China Science Publishing & Media Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. is determined by three factors: molar extinction coefficient, photothermal conversion efficiency and allowable laser power density. Even though some nanoagents have relatively high molar extinction coefficient or considerable photothermal conversion efficiency in the NIR-II window, their photothermal performances in the NIR-II window are not as good as those in the NIR-I window because of lower allowable laser power density [31]. However, it is true that compared with NIR-I light, NIR-II light possesses some intrinsic advantages in lesser photo-scattering and higher maximum permissible exposure (MPE), consequently exhibiting higher tissue penetration depth with less background interference and higher spatial resolution, and allowing tissue illustration at a relatively higher power density of laser (1.0 W cm −2 for NIR-II versus 0.3 W cm −2 for NIR-I) [29]. Therefore, to develop biocompatible NIR-II-photothermal nanoplatforms with versatile imaging functions is significant to precision cancer theranostics but challenging.
Both selenium (Se) and tellurium (Te) belong to the chalcogen elements, and their nanomaterials exhibit some unique semi-conductive features [32][33][34][35][36]. Te nanoneedles and nanosheets have an extremely narrow band gap (about 0.35 eV) and a strong absorbance in the NIR-II region in support of NIR-II-photothermal therapy and imaging, but also demonstrate high cytotoxicity and poor biocompatibility owing to their strong reducibility, restricting their biomedical applications [37][38][39][40][41]. By comparison, Se is an essential element for human beings and the selenizing can eliminate the cytotoxicity of many metals such as Cd and Cu [42,43]. Therefore, we hypothesize that controllable incorporation of biocompatible Se into the lattice of Te nanostructures for construction of TeSe x nano-alloys could intrinsically tune the inherent cytotoxicity of Te nanomaterials, enhance the biocompatibility of Te nanomaterials and extend their functions for biomedical applications. In this work, we synthesize a series of TeSe x nano-alloys with different Se incorporating proportions, and investigate their biocompatibility and develop their theranostic functions. We have discovered that the toxicity of Te nanomaterials mainly comes from irreversible oxidation stress and intracellular imbalance of organization and energy, which is exterminated by the nano-alloying by incorporating a moderate proportion of Se (x = 0.43) (Scheme 1). The synthesized TeSe x nano-alloy (x = 0.43) exhibits extraordinarily high NIR-II-photothermal conversion efficiency (77.2%), 64 Cu coordination and computed tomography (CT) contrast capabilities, enabling high-efficacy photothermal therapy of cancer under the guidance of multimodal photothermal (PT)/photoacoustic (PA)/positron emission tomography (PET)/CT imaging (Scheme 1).

Synthesis and characterization of TeSe x nano-alloys
A series of rod-like TeSe x nano-alloys with various ratios of Se/Te and length/diameter were synthesized by a facile co-precipitation method. Tellurite and selenite were reduced simultaneously by hydrazine to form TeSe x nano-alloys, and the Se contents were adjusted by tuning the molar ratio of tellurite to selenite (Supplementary Table S1). TeSe x nano-alloys with Se/Te precursor molar ratios of 1 : 3, 2 : 3, 1 : 1 and 3 : 2 were named as TS1, TS2, TS3 and TS4, respectively. To ascertain the crystal structure of the as-prepared TeSe x nanomaterials, X-ray diffraction (XRD) characterization was conducted ( Fig. 1a and Supplementary Fig. S1). XRD patterns were further refined using the total pattern solution (TOPAS) Rietveld crystal-structure refinement software ( Fig. 1a and Supplementary Fig. S2). The refinement results suggested the formation of Te-Se alloys, which were crystallized in a rhombohedral structure with P3 1 21 space group. The refined structure of typical TS3 nano-alloy was further investigated by high-resolution transmission electron microscope (TEM). In a high-angle annular dark field (HAADF) image and the corresponding elemental mapping in Fig. 1b-e, Te and Se were dispersed throughout the whole rod-like TeSe x nano-alloys, and no core-shell structure can be obviously observed. To study the radial elemental distribution of the as-prepared TeSe x nano-alloys, depth profiling X-ray photoelectron spectroscopy (XPS) analysis was conducted on sample TS3 which was exposed to Ar + for 0, 1 and 2 min. Te 3d and Se 3d XPS spectra of TS3 in Fig. 1f revealed that there were only Te (0) and Se (0) in the TeSe x nano-alloy, and the binding energy of Te 3d and Se 3d gradually decreased with the increase of Ar + etching time. Accordingly, the atomic ratio of Se to Te obtained from XPS (Supplementary Table  S2) decreased gradually, suggesting the gradient increase of Se content from inside to outside in support of the formation of TeSe x nano-alloy [44]. To confirm the atomic structure of TeSe x nano-alloy, atomically resolved HAADF-scanning transmission electron microscopy (STEM) was conducted. Figure 1g-i and Fig. 1j-l showed the simultaneously recorded HAADF and bright-field (BF) STEM images acquired along a-axis and c-axis, respectively. The observed atomic crystal structure from STEM was in high accordance with the simulated one from XRD refinements (yellow spheres in Fig. 1i and l). Besides, no megascopic difference between Te and Se can be observed, pinpointing that Se and Te were thoroughly miscible in each other and formed homogenous trigonal-system TeSe x nano-alloy.
The compositions of a series of TeSe x nanoalloys were measured by the inductively coupled plasma optical emission spectrometry (ICP-OES). As shown in Supplementary Table S1, the chemical structures of TS1, TS2, TS3 and TS4 were Te 0.82 Se 0.18 , Te 0.75 Se 0.25 , Te 0.7 Se 0.3 and Te 0.67 Se 0.33 , respectively. By varying the amounts of Se and Te precursors, the molar fraction of Se in TeSe x nanoalloys could be controllably tailored. XRD patterns of TS1, TS2, TS3 and TS4 in Supplementary Fig. S1 showed that the diffraction peaks were well matched with the standard ones of Te (JCPDS card number  in the absence of impurity. In contrast to the reported core−shell Te@Se nanowires [45] and Se-coated Te nanoheterojunctions [46], no characteristic XRD peaks of Se were observed in TeSe x nano-alloys in this work. It is worth carefully noticing that the finely identified TeSe x nanoalloying structure is easily mistaken for core-shell structure and heterojunction, possibly attributed to improper sampling, characterization and analysis. As the content of Se in TeSe x nano-alloys increased, all the diffraction peaks shifted slightly towards high-angle direction, suggesting the decrease of the interlayer distances which agreed with the variation of c/a obtained from the Rietveld refinement results (Supplementary Table S3). XPS analysis (Supplementary Fig. S3) displayed that with the increase of the amount of Se, Te 3d and Se 3d peaks of TeSe x nano-alloys shifted towards lower binding energy, which was also induced by the alloying formation between Te and Se. TEM images in Supplementary  Fig. S4 indicated that the diameter and length of rod-like TeSe x nano-alloys decreased with the increase of Se incorporation amount, while their morphologies remained nearly unchanged, suggesting that the incorporation of Se inhibited the growth of Te nano-rods. The particle size of synthesized TeSe x nano-alloys was less than 100 nm in favor of passive targeting accumulation in tumor by the enhanced permeability and retention (EPR) effect.

Evolution of biocompatibility and photothermal properties of TeSe x nano-alloys
The biocompatibility of nanomedicines is vitally important to their biomedical application. Here we evaluated the biocompatibility of TeSe x nano-alloys and checked the effect of the Se incorporation amount. Two cell lines (breast 4T1 cells and liver L-O2 cells) were employed for in vitro cytotoxicity assay. In Fig. 2a   b c d e a Figure 1. The structure, morphology and chemical composition of TeSe x nano-alloys (TS3). (a) Measured (red line) and simulated (blue line) XRD patterns, and differentiated profiles (green line) between them obtained from the Rietveld refinement of TS3 using P3 1 21 space group with hexagonal unit cell, where the inset is the perspective view of the simulated crystal structure of TeSe x nano-alloys (the atoms represent either Te or Se). HAADF-STEM (b) and energy dispersive spectrum (EDS) elemental-mapping images (c−e) of TeSe x nano-alloys. (f) Depth profiling XPS spectra of TeSe x nano-alloys exposed to Ar + for 0, 1 and 2 min. Atomically resolved HAADF-STEM images acquired along the a-axis direction (g) and c-axis direction (j), with more detailed views (h, k) and corresponding BF-STEM images (i, l). The scale bars in (h), (i), (k) and (l) correspond to 2 nm, 1 nm, 2 nm and 1 nm, respectively.
incorporation amounts still can inhibit cell growth to a certain extent, while TS3 and TS4 did not (Fig. 2b). In vivo toxicity of TeSe x nano-alloys was further investigated. After intravenous injection with TeSe x nano-alloys at the higher dose of 50 mg kg −1 for one week, all the treated mice were alive and well, and their blood samples were taken from the orbital sinus to investigate the toxicity of TeSe x nano-alloys. From the standard blood biochemical indexes in Fig. 2c and d, the concentrations of aspartate transaminase (AST) and creatinine (CREA) in the Te-treated group were remarkably higher and lower than that of the blank control group, respectively, suggesting that Te nano-rods caused distinct damage to liver and kidney functions. By comparison, all the investigated TeSe x nano-alloys did not demonstrate visible toxicity to liver and kidney. These in vitro and in vivo toxicity results therefore suggested that the incorporation of Se into Te nano-rods at a relatively high amount can effectively reduce their toxicity. In addition, different from TeSe nanoheterojunctions [46], Te nano-rods synthesized in this work can degrade by only 1.4% in water after immersion for 8 months, and nano-alloying of TeSe x inhibited the degradation of Te nano-rods remarkably (Supplementary Table S4) in support of depressed toxicity.
To understand the mechanisms of the toxicity of Te nano-rods and alloying detoxification, gene expression studies were performed by RNA sequencing (RNA-Seq) which allowed quantitative measurement of expression levels of genes in 4T1 cells from six groups with different treatments (blank control, Te, TS1, TS2, TS3 and TS4 at 200 μg mL −1 ). We first screened the differentially expressed genes (DEGs) between TeSe x nano-alloys and blank control to characterize the functional consequences of gene expression changes induced by TeSe x nano-alloys. As shown in Supplementary Figs S6   , several genes, involving subunit organization and positive regulation of reactive oxygen species (ROS) metabolic process, were remarkably down-regulated by Te nano-rods, but very slightly affected by TeSe x nano-alloys, especially TS3 and TS4 with higher Se incorporation amounts. Furthermore, defense response was provoked by Te nano-rods and can also be avoided to a certain extent by TS1, TS2 and TS3 with relatively lower Se incorporation amounts. However, excessive Se incorporation for TS4 caused the strongest defense response. These GO results indicated that Te nano-rods disturbed the normal metabolic process of ROS and thus caused oxidative stress and damage to subunit organization. The moderate incorporation of Se into Te nano-rods (such as TS3) can recover the normal metabolic process of ROS and avoid damage to subunits, thereby greatly reducing the toxicity of Te nano-rods. But the defense response from overhigh incorporation of Se (such as TS4) would possibly cause toxicity. To further identify the pathways of Te toxification and TeSe x detoxification, we performed the pathway analysis of DEGs based on the KEGG (Kyoto Encyclopedia of Genes and Genomes) database. As summarized in Fig. 2f, and Supplementary Tables S5 and S6, Te nano-rods positively stimulated the drug metabolism pathway, which was similar to the response of cells to many toxic substances ( Supplementary Fig. S9a). Furthermore, Natl Sci Rev, 2021, Vol. 8, nwaa156 Te nano-rods significantly promoted the metabolism of glutathione (GSH) by up-regulating glutathione S-transferase (GSTs), resulting in the decrease of intracellular GSH level and thus oxidative stress to impair subunit organization, as illustrated in Fig. 2f. Moreover, Te nano-rods also significantly caused ribosome disorders and inhibited glycolysis by suppressing RPL7A and GAPD, consequentially causing subunit organization dysfunction and reduced energy production, as illustrated in Fig. 2f. The increased levels of GSTs protein expression and decreased levels of RPL7A and GAPD protein expression in Te evidently confirmed the associated GSH metabolism and energy production. In contrast, the TS3 got rid of the negative effects on GSTs, RPL7A and GAPD ( Supplementary Fig. S10). Therefore, the Se incorporation into Te nano-rods to form TeSe x nano-alloys got rid of the negative effects on GSTs, RPL7A and GAPD (Supplementary Tables S5 and  S6 and Fig. S10), suppressing the cytotoxicity of Te (Fig. 2f). The GSH depletion of Te nanomaterials by surface coordination between Te and hydrosulfide group is generally thought to be the main reason for their toxicity [47]. Indeed, this work also found that Te nano-rods could adsorb GSH but TeSe x nano-alloys almost not ( Supplementary Fig. S11). Additionally, this work discovered that Te can also reduce GSH by up-regulating GSTs, and also uncovered other pathways involving ribosome and glycolysis for the first time. The identified mechanisms for Te toxification and TeSe x detoxification in this work would greatly favor deep understanding of the origination of Te nanomaterials toxicity and also provide a strategy for developing biocompatible Tebased nanomaterials for biomedical applications. After detoxification, the NIR-II-photothermal effect of TeSe x nano-alloys was evaluated. As shown by UV-VIS-NIR spectra in Fig. 2g, all the investigated TeSe x nano-alloys (TS1-TS4) had distinct NIR-II light absorption. The Se incorporation led to the blue shift of absorption spectra and the reduction in the NIR-II absorbance of Te nano-rods, but remarkably enhanced their NIR-II-photothermal conversion efficiencies. As shown in Fig. 2h and Supplementary Fig. S12, the NIR-II-photothermal conversion efficiency of TeSe x nano-alloys gradually increased and then decreased with the increase of Se incorporation amount. TS3 exhibited the highest NIR-II-photothermal conversion efficiency (η) of 77.2% under the irradiation of a 1060 nm laser, which is much higher than that of Te nano-rods (53.3%) and other reported NIR-II-photothermal nanomaterials such as Au nanostar@MOF (48.5%) [48], Nb 2 C nanosheet (46.7%) [19] and Pt spiral (52.5%) [49]. Moreover, the aqueous solution of TS3 nano-alloy was exposed to the 1060 nm laser at varied laser power densities (0.2, 0.5 and 1.0 W cm -2 ) and at different particle concentrations (50, 100 and 200 μg mL -1 ) to investigate its NIR-II-photothermal effect. As shown in Fig. 2i and j, the NIR-II-photothermal effect of TeSe x nano-alloy was positively related to both the power density of the laser and the concentration of TeSe x nano-alloy. Typically, the temperature of the TeSe x solution containing 200 μg mL -1 TS3 rose by 36.2 • C after 7 min of 1060 nm laser irradiation at 1.0 W cm -2 in great support of thermal therapy of cancer. The NIR-II-photothermal stability of TS3 was further investigated for five laser on/off cycles. As shown in Supplementary Fig. S13, a temperature change of 36.6 • C was achieved and did not show significant deterioration during five-cycle irradiation, suggesting that TeSe x nano-alloy had high NIR-II-photothermal stability. From the above results, TeSe x nano-alloys with moderate Se incorporation demonstrated highest comprehensive performances including good biocompatibility and high NIR-IIphotothermal efficiency, and was therefore chosen as a theranostic platform to execute the following evaluation of theranostic performances. In addition, we further measured photothermal performances of TeSe x nano-alloys using a 808 nm laser and compared them with the use of a 1060 nm laser. As in Fig. 2 and Supplementary Fig. S14, it could be found that NIR-II-photothermal conversion efficiency of TS3 nano-alloys (77.2% for 1060 nm) was higher than that in the NIR-I window (62.3% for 808 nm, Supplementary Fig. S14b). Although TeSe x nanoalloys had higher extinction coefficient at 808 nm than at 1060 nm ( Supplementary Fig. S15), they still exhibited higher photothermal performance at 1060 nm (Fig. 2, Supplementary Figs S14 and S16) owing to higher photothermal efficiencies. Therefore, at the same laser power density, a 1060 nm laser should have a higher tissue penetration depth than a 808 nm laser, and PAI performance of TS3 nano-alloys in the NIR-II window could be better than that in the NIR-I window [27], implying the possibility of using TS3 nano-alloys for photoacoustic imaging in both NIR-I and NIR-II windows. In addition, we have executed the measurement of singlet oxygen yield under NIR irradiation, and found that no distinct singlet oxygen was generated by TeSe x nano-alloys ( Supplementary Fig. S17), possibly because most of the photo energy had been converted to heat.

In vivo multimodal imaging performances of TeSe x nano-alloys
Inherent imaging functions of theranostic nanoplatforms are very helpful for precision medicine.
Especially, multimodal imaging with complementary advantages can be used to accurately guide cancer therapy. Based on the unique properties of TeSe x nano-alloys in photothermal conversion, surface incorporation/coordination and high density (high atomic number), we tried to uncover multimodal PT/PA/PET/CT imaging performances of TeSe x nano-alloys with the 4T1 tumor-bearing mice model. As to PT imaging, TeSe x nano-alloys (100 μL of TS3 at 2 mg mL -1 ) were intravenously injected into mice, when tumors grew up to 100 mm 3 , followed by 1060 nm laser irradiation (1 W cm -2 , 5 min) after 8 h post injection. The TeSe x group had a remarkably higher increase of temperature in the irradiated tumor site compared with the phosphate buffer saline (PBS) control group ( Fig. 3a and b). After 1 min NIR-II light irradiation, the increases of temperature in the TeSe x and PBS groups were 17.6 • C and 4.2 • C, respectively, suggesting that the irradiated tissue itself had low NIR-II-photothermal effect but TeSe x nano-alloys effectively accumulated in the tumor in a passive targeting way and exhibited high NIR-II-photothermal effect owing to high NIR-II-photothermal conversion efficiency (Fig. 2h). Based on NIR-II-photothermal effect, the PA imaging (PAI) performance of TeSe x nano-alloys was further evaluated in vitro and in vivo on the 4T1 tumor-bearing mice model. TeSe x nanoalloys exhibited a high photoacoustic coefficient of 0.109 mL mg −1 at 810 nm ( Supplementary Fig.  S18). As shown in Fig. 3c, the tumor itself displayed relatively low PA signal before injection. After intravenous injection of TeSe x nano-alloys (TS3), the intratumoral PA signal intensity gradually augmented over time and reached the maximum value at about 8 h post injection (Fig. 3d), suggesting efficient intratumoral accumulation of TeSe x nano-alloys in accordance with the above-mentioned PT results.
Leveraging the virtue of large X-ray absorption coefficient of high atomic number elements for CT contrast and the strong affinity of chalcogen to transitional metal ions for PET imaging [50,51], we anticipated that TeSe x nano-alloys could impart the quantitative measure of their biodistributions and metabolic processes by PET/CT imaging, beyond the localization photoacoustic/photothermal (PT) imaging. We first investigated CT contrast performances of TeSe x nano-alloys in vitro and in vivo with 4T1 tumor-bearing mice with intravenous injection of TS3 (100 μL, 10 mg mL -1 ). TeSe x nano-alloys exhibited a considerable X-ray absorption coefficient of 2.3 HU/mM equal to that of an aqueous iodine standard (iopamidol) which is popularly used clinically at 140 kV of X-ray tube voltage [52] (Supplementary Fig. S19). In addition, superior to the iodine standard, TeSe x nano-alloys have no visible kidney/heart/lung toxicities and longer circulation time, as well as tumor-targeted ability in favor of tumor-targeted therapy. As in Fig. 3e and f, intratumoral CT signal (yellow circles) gradually increased and achieved the maximum at about 8 h post injection, indicating gradual intratumoral accumulation process of TeSe x nano-alloys in accordance with the above-mentioned PA imaging results. Interestingly, we also observed that CT signals in the kidney (red arrows, Supplementary  Fig. S21) and bladder (red circles, Supplementary  Fig. S21) enhanced with time, indicating that TeSe x nano-alloys could be excreted through the urinary system, possibly owing to their small particle size (about 43 nm, Supplementary Fig. S4). Furthermore, the radionuclide 64 Cu was facilely labeled to TeSe x nano-alloys (TS3) by a surface coordination method for PET imaging. Supplementary  Fig. S22 showed about 90.6% 64 Cu-labeled efficiency of TeSe x nano-alloys which was measured by instant thin layer chromatography (iTLC). Then, we employed PET to evaluate the in vivo delivery and biodistribution of TeSe x nano-alloys. The decay-corrected PET images (Fig. 3g) displayed a high tumor-to-background contrast in the TeSe x -64 Cu treated 4T1 tumor-bearing nude mice. The tumor uptake efficiency of TeSe x nano-alloys was measured by using a quantitative three-dimensional volume-of-interest analysis method. As shown in Fig. 3g-i, the intratumoral accumulation of TeSe x nano-alloys reached the maximum after about 12 h post injection, and TeSe x nano-alloys which were taken up by liver, spleen and kidney were gradually eliminated with time. In Fig. 3g, the metabolic process of TeSe x nano-alloys was also clearly visible (yellow arrows). At 24 h post injection, the mice were sacrificed and major organs were collected for biodistribution study. As shown in Fig. 3j-l, 6.42% ID g −1 tumor uptake of TeSe x nano-alloys was achieved at 24 h post injection, and other particles mainly distributed on liver, spleen, kidney, etc. Even though TeSe x nano-alloys widely distributed in the body, their continuous excretion could reduce the potential risk of toxicity. In addition, the biodistributions of TeSe x nano-alloys in major organs were also determined by ICP-OES at the different time points (2,4,8,12 and 24 h) after injection (n = 3). As shown in Supplementary Figs S20 and S23, the ICP-OES results more accurately reflect the biodistribution of TeSe x nano-alloys in basic accordance with PET results, and the blood circulation half-time of TS3 was calculated to be about 1.21 h. Nevertheless, TeSe x nano-alloys were confirmed to be an excellent theranostic platform with multimodal PT/PA/CT/PET imaging functions in favor of guiding and monitoring cancer treatment.

In vitro and in vivo NIR-II-photothermal therapy performances of TeSe x nano-alloys
Cellular uptake of TeSe x nano-alloys was firstly investigated in vitro. TeSe x nano-alloys were facilely labeled with red fluorescent dye 5,10,15,20-tetra(4pyridyl)-21H,23H-porphine (TPyP) by virtue of its coordination capability. Confocal fluorescence images of 4T1 cells incubated with TPyP-labeled TeSe x nano-alloys for 1 and 2 h showed that red fluorescence gradually increased inside cells (Fig. 4b and Supplementary Fig. S24), indicating that TeSe x nano-alloys were efficiently internalized into 4T1 cells due to their small size of 43 nm. Thereafter, NIR-II-photothermal cytotoxicity of TeSe x nanoalloys against varied cancer cell lines (4T1, B16, HeLa cells) at different particle concentrations and at different laser power densities were investigated using the standard CCK-8 assay. From Fig. 4a and Supplementary Fig. S25, TeSe x nano-alloys without 1060 nm laser irradiation did not show observable   cytotoxicity to all the investigated cancer cells at a concentration up to 200 μg mL −1 . Under 1060 nm laser irradiation, TeSe x nano-alloys exhibited remarkable concentration-and power-dependent cytotoxicity against various cancer cells ( Fig. 4a and Supplementary Fig. S25). Typically, 0.5 W cm −2 1060 nm laser irradiation for 5 min on 100 μg mL −1 TeSe x nano-alloys treated cancer cells killed 83.5% 4T1 cells, 98.2% HeLa cells and 81.6% B16 cells. Additionally, green (calcein-AM) and red (propidium iodide) fluorescence staining results also clearly demonstrated high NIR-II-photothermal cytotoxicity of TeSe x nano-alloys (Fig. 4c). In addition, based on obvious absorption and photothermal conver-sion of TeSe x nano-alloys in the NIR-I window, we also found that TeSe x nano-alloys had remarkable NIR-I-photothermal cytotoxicity against various cancer cells ( Supplementary Fig. S27), which can also be an alternative candidate for photothermal therapy of cancer. Encouraged with the above-confirmed NIR-II-photothermal effects of TeSe x nano-alloys, the NIR-I-/NIR-II-photothermal ablation of solid tumors with TeSe x nano-alloys was further evaluated in vivo. Firstly, mice bearing breast 4T1 tumor were randomly divided into six groups (n = 5 per group) with approximately the same tumor volume (ca. 100 mm 3 ): (i) the PBS group (blank control), (ii) the PBS+NIR808 group, (iii) the PBS+NIR1060 group, (iv) the TeSe x group, (v) the TeSe x +NIR808 group, and (vi) the TeSe x +NIR1060 group. The NIR treatment group referred to the mice intravenously injected with PBS or TeSe x nano-alloys (100 μL TS3, 10 mg kg −1 , three times at Day 1, Day 3 and Day 5) and irradiated with 808 nm or 1060 nm laser irradiation (1.0 W cm −2 for 5 min) at fixed time points (Day 2, Day 4 and Day 6) after 8 h post injection to ensure sufficient thermal damage to tumor cells. From Fig. 4d, tumor growth was not affected by TeSe x nano-alloys, but significantly suppressed by combination of TeSe x nano-alloys with 1060 nm laser irradiation. The 808 nm laser irradiation plus TeSe x nano-alloys also generated remarkable inhibition effect on tumor growth, but the inhibition efficacy was not as efficient as that of the 1060 nm laser irradiation. After 21-day treatment, tumors were dissected to photograph and weigh as shown in Fig. 4e. The results further demonstrated remarkable inhibition effect of NIR-photothermal TeSe x nano-alloys on tumor growth. Although tumors had not been completely eradicated which possibly resulted from short irradiation time of only 5 min, we felt optimistic because in vivo therapeutic efficacy could become better if we extended NIR irradiation time. Furthermore, the hematoxylin-eosin (H&E) staining of major organs and tumor tissues was conducted ( Fig. 4f and Supplementary Fig. S28). It was found that NIR-II-photothermal therapy caused significant tumor cell damage. But there was no obvious damage in all major organs after treatment ( Supplementary Fig. S28) and no loss in body weight during treatment ( Supplementary  Fig. S29), suggesting no obvious systematic toxicity of TeSe x nano-alloys. Even at the high injection dose of 50 mg kg −1 which was five-fold higher than treatment dose, no damage to liver and kidney functions was visible (Fig. 4g and h). These results indicated that TeSe x nano-alloys were a biocompatible and high-efficacy NIR-photothermal platform.

CONCLUSION
In conclusion, a series of TeSe x nano-alloys with different ratios of Se/Te and length/diameter were controllably synthesized by the facile coprecipitation method. Incorporating a moderate content of Se (x = 0.43) into the lattice of Te nanostructure effectively eradicated the toxicity of Te, which mainly originated from GSTs up-regulation, and RPL7A/GAPD down-regulation caused subunit organization dysfunction and energy production loss. TeSe x nano-alloys exhibited high NIR-II-photothermal conversion efficiency (77.2%), and had been proved to be a kind of versatile nanotheranostic platform with multiple functions of NIR-II-photothermal therapy and multimodal PT/PA/PET/CT imaging, enabling multimodal imaging-guided NIR-II-photothermal therapy of cancer with high theranostic performances.

METHODS
The controlled preparation of rod-like TeSe x nano-alloys with different ratios of Te/Se and length/diameter were realized by a facile hydrothermal method, where the Se contents were adjusted by tuning the molar ratio of tellurite to selenite. The comprehensive details, chemicals and characterizations are in the Supplementary data.