Enhanced Photothermal Therapy of Breast Cancer via Co-Loading With Autophagy Inhibitor

Photothermal therapy (PTT) usually causes hyperthermia and damages healthy tissues. Developing a PTT platform using mild irradiation, while enhance the theraputical effects attracts increasing attention. Here, this study developed a theranostic poly(D,L-lactic-co-glycolic acid) (PLGA) nanoplatform loaded with a near-infrared (NIR) dye (new indocyanine Green IR820), autophagy inhibitor (chloroquine, CQ) and a uorescence imaging agent (ZnCdSe/ZnS quantum dot, QD) by the double emulsion solvent evaporation technique (W 1 /O/W 2 ). The resulting hybrid PLGA nanoparticles with IR820/ZnCdSe/ZnS/CQ co-loading (termed PIFC NPs) approximated 240 nm in diameter and had excellent monodispersity, uorescence and size stability, and biocompatibility. PIFC NPs displayed photothermal effects, and the released CQ remarkably decreased autolysosome degeneration by lysosomes in cancer cells, thereby enhancing the suppressive effect on autophagy as well as resistance to photothermia. Anticancer effects were enhanced both in cellular and animal experiments attributed to the combined effects exerted by PIFC NPs and mild NIR irradiation. Moreover, PIFC NPs signicantly accumulated in tumors because of enhanced permeability and retention (EPR) effect, enabling high-spatial resolution, real-time uorescence imaging of solid tumors. The present study developed a novel PTT platform showing potentially enhanced therapeutic ecacy. distribution in cells. The above ndings, alongside punctuate uorescence appearing along the cells suggest the cellular internalization of PIFC NPs. treatment temperatures. This showed that regulating autophagy in cancer may be promising in assisting PTT for therapy. the successful preparation of PIFC NPs in one provides a simple approach for manufacturing nanotheranostics that may be useful in both uorescent imaging and PTT in cancer.


Introduction
Multifunctional nanodrugs combining diagnostic and therapeutic functions currently constitute a novel approach in cancer treatment. [1,2] Ideally, theranostic platforms should both exert e cient therapeutic effects and low side effects.
Therapeutic agents and imaging compounds are essential for the fabrication of nanotheranostics for cancer therapy. [3,4] On one hand, photothermal therapy (PTT), an important light-dependent therapeutic approach, is considered as a novel, e cient and noninvasive method for cancer treatment, with reduced tumor recurrence and elevated selectivity. In PTT, light energy is absorbed to yield heat energy by photothermal molecules to kill cancer cells. [5,6] Albumin nanoparticles carrying PTT compounds exerted excellent anticancer effects. IR820, a photosensitizer, after NIR light irradiation, e ciently converts light to heat, acting as PTT agent and inhibiting malignant cells. [7,8] Photothermal therapy shows broad application prospects, but the problems arising in the actual treatment process need to be addressed urgently. At present, photothermal therapy is mostly applied in super cial tumors. The treatment temperature needs to reach 50 ℃ or higher to effectively inhibit tumor cells. [9,10] Meanwhile, excessive temperatures can cause severe burns to the skin. The burned skin further induces a series of self-defense reactions in the body, such as the release of in ammatory factors, increasing the risk of tumor metastasis and recurrence. [11] Therefore, photothermal treatment of tumors at lower temperatures is more in line with practical clinical application. However, photothermal treatment at lower temperatures cannot effectively inhibit tumor growth, and its tumor treatment effect is not signi cant. [12,13] According to previous reports, tumor cells actually have varying degrees of anti-photothermal effects, leading to ineffective tumor ablation.
Such resistance mainly originates from an "autophagy" mechanism in the cell. [14,15] Autophagy represents an important cellular pathway that breaks down impaired organelles, aged proteins and other constituents, reusing them as cellular nutrients for self-renewal. [16,17] Additionally, autophagy has a vital function in resisting cell stress as well as various treatment methods. [18,19] Indeed, autophagy can repair and reverse cell damage caused by heating, resulting in incomplete cell necrosis. [20][21][22] In order to overcome this treatment resistance, severer conditions (e.g., high temperature and extended irradiation time) are applied in PTT, which may cause accidental damage to nearby non-cancerous tissues. [23,24] Therefore, developing an advanced PTT platform that can effectively inhibit autophagy is very attractive, and could help achieve more targeted and speci c inhibition of tumors.
On the other hand, in comparison with alternative imaging techniques, e.g., computed tomography (CT) and magnetic resonance imaging (MRI), [25,26] uorescence imaging has been assessed for bioimaging and medical application thanks to elevated sensitivity, simple operation and cost effectiveness. [27][28][29][30] Consequently, albumin nanoparticles carrying PTT compounds and uorescent dyes were designed to effectively diagnose and treat cancer. Recently, due to synergistic effects exerted by both polymer nanospheres and inorganic nanomaterials, polymer-inorganic hybrid nanospheres attract increasing attention. Speci cally, biocompatible compounds that can emit uorescence may help monitor the nanospheres in cells and body, revealing tumor locations and re ecting treatment responses. In comparison with organic uorescent dyes, inorganic uorescent emitters, including quantum dots (QDs), have stronger uorescence intensity, higher photostability, wider excitation wavelength range and smaller emission spectrum. [31] Based on the above properties, extensive research has focused on the use of quantum dots in multicolor cell, tissue and animal imaging protocols. [27,32] However, low biocompatibility and stability under a variety of physiological conditions limit the biomedical/clinical application of QDs. Multiple approaches are under development for overcoming the above limitations. The most successful strategy has involved QD coating with biocompatible hydrosoluble polymers, which improves the colloidal stability and biocompatibility of QDs in aqueous solutions and biological environments. [28,33] However, such a thin polymer layer on QDs hardly meets the requirement of carrying su cient active molecules. Using biocompatible polymer nanospheres to wrap QDs instead of thin polymer layers would provide ample space to load other activators (e.g., drugs and inorganic nanoparticles), which can spread their value for disease treatment and diagnosis, making them multifunctional nanomaterials. In addition, quantum dots in the polymer matrix may reduce the shell shedding of QDs, which usually leads to photodegradation. Meanwhile, most of the previously reported constructions of QD-related polymer nanocarriers require complex synthetic processes and could cause environmental toxicity because of the use of surfactants and/or organic solvents. [34,35] Herein, a PLGA-based multifunctional treatment system, with both uorescence imaging and autophagy inhibition, was examined for PTT effectiveness with mild irradiation. The system uses PLGA as the nanocarrier of the composite, with chloroquine (CQ) loaded in the PLGA core to form an internal water phase and ZnCdSe/ZnS QD and IR820 used as the oil phase ( Fig. 1). Interestingly, e cient inhibition of autophagy was accomplished by interfering lysosomes and reducing autolysosome degradation by CQ released in malignant cells. [36] Integrating the strong photothermal features and autophagy inhibition of malignant cells, the PIFC NPs system exerted anticancer effects in vivo and in vitro under mild near-infrared irradiation. Therefore, this study provides a novel potential platform for PTT-based tumor therapy with greater e cacy.

Synthesis of PIFC NPs
PIFC NPs were synthesized by the double emulsion (water/oil/water: W 1 /O/W 2 ) technique, with some modi cations. In brief, PLGA (20 mg) and IR820 (1 mg) were added to dichloromethane (1 mL), followed by ZnCdSe/ZnS QD solution (0.1 mL, 3 mg/mL) and CQ (0.2 mL, 2.5 mg/mL). The above mixture was then emulsi ed in an ice water bath for 30 seconds with an ultrasound probe (Sonics & Materials, USA). The resulting mixture was further mixed with cold PVA solution (5 mL, 4% w/v). The resulting emulsion was re-emulsi ed in an ice water bath for 3 minutes using the above ultrasound probe, with subsequent addition of water (15 mL) and stirring until dichloromethane volatilization. Finally, PIFC NPs were obtained by centrifugation (12000 rpm, 20 minutes) and washed thrice with DI water.

Characterizations.
Transmission electron microscopy (TEM) was performed on a JEM 2001F (JEOL, Japan) for image acquisition. An energy dispersive spectrometer (EDS) was also utilized for material characterization. UV-Vis spectroscopy was carried out on a UVmini-1240 (Shimadzu, Japan). Dynamic light scattering (DLS)-based zeta potential and size distribution were determined with a nano-ZS90 Zetasizer (Malvern Panalytical, UK).

Loading of IR820 and CQ.
Subsequently, PIFC NPs were centrifuged at 12000 rpm for 20 minutes, and underwent three deionized water washes for removing unloaded CQ and IR820. The drug loading rates were examined using the resulting supernatant. Based on CQ and IR820 absorbance at approximately 344 nm and 680 nm, respectively, calibration curves for the concentrations of CQ and IR820 were generated.
In order to assess drug release under acidic pH, two dialysis bags (Biosharp USA, molecular weight cutoff approximating 3500 kDa) containing 1 mL of PIFC NPs were immersed in 8 mL of PBS (pH 7.4 or pH 5.0) under constant temperature and stirring. At various times, 0.2 mL dialysis solution was obtained for examination, maintaining the dialysate volume by addition of PBS (0.2 mL). Absorption was read at 344 nm on a UV-Vis spectrometer, and the amount of released CQ was determined in the collected sample based on the abovementioned calibration curve.
In order to evaluate the photostability of PIFC NPs, 1 mL PIFC aqueous solution (PIFC NPs at 600 µg/mL and IR820 at 80 µg/mL) and free IR820 (80 µg/mL) were irradiated with a 808-nm laser till temperature stabilization. After turning off the laser, solution cooling occurred, and temperature recording was carried out at an interval of 1 min. This on/off cycle was repeated four times.

Cellular Uptake Assays
To evaluate the cellular uptake behavior of PIFC NPs, breast cancer MDA-MB-231 cells underwent seeding in 6-well plates and 24 h incubation at 37°C under 5% CO 2 . Then, medium with PIFC NPs was used for further culture for 0.5, 1 and 2 h, respectively. After PBS washes, the nucleus and lysosomes underwent staining with DAPI and LysoTracker Green, respectively. Finally, a confocal laser scanning microscope (CLSM, Olympus, Japan) was utilized for data analysis.

Western Blot.
MDA-MB-231 cells underwent seeding in 6-well plates at 4×10 5 cells per well, and were incubated with control medium, CQ, PIF NPs and PIFC NPs for 4 h. This was followed by 808-nm irradiation (1.0 W/cm 2 , 5 min). The harvested cells (trypsinization) were fully lysed with the RIPA buffer containing protease inhibitors (Beyotime, China). After adding SDS buffer, the sample was boiled for 8 min, and protein separation was carried out by 12% SDS-PAGE. This was followed by electro-transfer of the separated proteins to the NC membrane (Beyotime) at low temperature for 45  37°C. Then, cell washes were performed, followed by analysis by confocal laser scan microscopy (CLSM, Olympus, Japan).

TEM Assessment of Autophagy Inhibition.
MDA-MB-231 cells were administered control medium, CQ, PIF NPs and PIFC NPs in culture medium. After laser irradiation, cell xation was carried out with 2.5% glutaraldehyde in PBS overnight. Next, the cells underwent washing, further xation with 1% osmium, dehydration (graded ethanol), epoxy resin embedding, sectioning (70 nm), staining and TEM analysis.

In Vivo Toxicity Analysis
Mice intravenously administered 200 µL of PBS and PIFC NPs (20 mg/kg), respectively, underwent euthanasia at 0, 1, 7 and 21 days, respectively. Then, blood collection was carried out by cardiac puncture for blood biochemistry and complete blood count by the A liated Hospital of Xuzhou Medical University. Next, heart, liver, spleen, lung and kidney specimens underwent 10% formalin xation, para n embedding, sectioning and H&E staining. An inverted microscope was utilized for analysis.

Synthesis and Characteristics of PIFC Nanoparticles.
The formation process for PIFC NPs is depicted in Fig. 2a. TEM revealed PIFC NPs had an evident core-shell structure and spherical outline (Fig. 2b and c). The remarkably high electron density of QDs potentiates their direct visualization within PLGA nanoparticles, indicating that CdSe/ZnS QD were successfully embedded in PLGA nanoparticles. PIFC NPs were 247 ± 4.1 nm (Fig. 2d); their PDI was 0.127 (Fig. 2f), zeta potential was − 30.4 ± 0.09 mV (Fig. 2e), suggesting it is a stable nanoplatform due to electrostatic repulsion among circulating. [37] The as-prepared PIFC NPs were highly stable and had elevated dispersity in DI water, PBS, FBS, DMEM and L15, with a hydrodynamic average diameter of 240 nm (Fig. 2f). The spheres appeared smaller under TEM compared with DLS because of shrinking after drying.
The absorption properties of PIFC NPs and constituents were assessed by UV-Vis spectrophotometry. Figure 3a shows IR820 had an absorption peak at about 680 nm, and an overt peak was found at 344 nm, which was attributed to CQ and absent in PLGA NPs. PIFC NPs containing IR820 and CQ showed two small peaks at 810 nm and 344 nm, respectively. These ndings con rmed the successful loading of CQ and IR820 into PLGA nanoparticles. Based on respective calibration curves, the loading rates of CQ (Fig. 3b) and IR820 (Fig. 3c) were approximately 39.86% and 41.40%, respectively.
PIFC NPs exhibit remarkable uorescence stability, owing to the powerful protection of QDs by thick polymer matrices. The corresponding Energy Dispersive Spectrometry (EDS) data revealed that the elements sulfur (S), zinc (Zn), selenium (Se) and cadmium (Cd) were homogenously distributed over the entire nanoparticles (Fig. 3d). Figure 3e presents the photoluminescence spectra of ZnCdSe/ZnS QDs and PIFC NPs, alongside those of empty PLGA NPs. It was noticed PIFC NPs have similar shape and peaks as ZnCdSe/ZnS QDs, with the exception of peaks resulting from the scattering of PIFC NPs. After con rming that PLGA was successfully loaded with ZnCdSe/ZnS QDs, we used a uorescence spectrophotometer to characterize its uorescence performance. As shown in Fig. 3g, elevated solution concentration resulted in stronger uorescence intensity of PIFC NPs, indicating that uorescence intensity is related to concentration. In addition, in order to verify the uorescence stability of PIFC NPs, we tested the changes of uorescence intensity in different media. PIFC NPs were stable for several weeks in DI Water, PBS, FBS, DMEM cell culture medium and L-15 cell culture medium (Fig. 3f). Obviously, the protection provided by PLGA NPs is su cient to prevent any photobleaching of the QDs. Therefore, we can consider using PIFC NPs as an optical bioimaging agent and drug carrier.

Release of CQ
The release of CQ was examined based on its calibration curves at pH 7.4 and pH 5.0 (Fig. 3h). Within the measurement time (120 h), under normal physiological conditions (pH 7.4) the release of CQ in PIFC NPs was 15.99%, while under acidic conditions (pH 5.0) it was signi cantly enhanced (66.41%) (Fig. 3i). After the rapid release in the rst 10 hours, CQ was still released slowly in the subsequent time. The above results suggest PIFC NPs constitute an effective drug carrier that can promote the release and accumulation of CQ in the acidic microenvironment of tumors.

Photothermal Features of PIFC Nanoparticles.
To explore the photothermal features of PIFC NPs, aqueous solutions with various particle amounts (0, 150, 300, 450, 600 and 750 µg/mL) were irradiated with different powers using an 808-nm laser (0.5, 1.0 and 1.5 W/cm 2 , respectively, for 5 minutes). Then, heating monitoring and quantitation were performed at 1-min intervals with an infrared thermal imaging camera. Under 0.5 W/cm 2 laser irradiation, the highest concentration of nanoparticles only rose the temperature to ~ 39 ℃ (Fig. 4a). However, under 1.0 W/cm 2 laser irradiation, the PIFC NPs solution (600 µg/mL) reached 45 ℃ after irradiation for 5 min (Fig. 4b and c), while pure water was not signi cantly heated by irradiation, which indicates the PIFC NPs solution can quickly and effectively convert light into heat. Obviously, the heating effect highly depended upon particle levels and irradiation time. As expected, it was also found that a higher laser power density amplitude intensi ed the heating phenomenon ( Fig. 4e and f).
Photostability is considered another key parameter in evaluating potential applications of photothermal agents in cancer treatment. For this purpose, a PIFC aqueous solution (600 µg/mL) was irradiated circularly using an 808-nm laser (1.0 W/cm 2 for 5 min). Interestingly, peak temperature change was found to be smaller (Fig. 4d). Both phenomena indicate that PIFC NPs have good photostability as expected. Overall, these unique photothermal features indicate PIFC NPs constitute a promising platform for PTT.

Cellular Uptake of PIFC Nanoparticles
Compared to other surface-modi ed QDs, [38][39][40] PIFC NPs show an evident advantage in the photoluminescence of individual PIFC NPs, since many QDs are encapsulated in every nanoparticle, with relatively strong protection by the large polymer matrix without aggregation. Therefore, PIFC NPs could constitute a great probe for real-time optical cell imaging, which may be extremely useful for diagnosing and treating cancers. Using the optical properties inherited from the packaged QDs, the cell internalization and intracellular distribution of PIFC NPs were observed by CLSM. In addition, the majority of the bright spots occupied the cytosol, and uorescent signals showed uneven or random distribution in cells. The above ndings, alongside punctuate uorescence appearing along the cells suggest the cellular internalization of PIFC NPs.

In vitro Anticancer Properties.
Using cell counting kit-8 to evaluate the cyto-compatibility of PIFC NPs to MDA-MB-231 cells, we demonstrated that PIFC NPs solutions at 0 to 750 µg/mL had no cytotoxicity following 24 h of incubation (Fig. 6a). However, after exposing PIF NPs (600 µg/mL) to NIR irradiation (1.0 W/cm 2 for 5 min), cell viability was decreased by 50% within 24 h. Upon CQ loading, PIFC NPs induced remarkable cell death, and under the same NIR irradiation, 80% cell death was observed within 24 h (Fig. 6b). What's more, PIF NPs and PIFC NPs solutions at various levels (0, 150, 300, 450, 600 and 750 µg/mL) were administered to cells submitted to irradiation at different times (0, 1, 2, 3, 4 and 5 min). As expected, cell viability overtly decreased with increasing nanoparticle amounts and irradiation time. Even at low IR820 level (80 µg/mL), cell viability was reduced by 50% after 24 h for PIF NPs (Fig. 6c). Strikingly, PIFC NPs resulted in a more speedy decrease of cell viability compared with PIF NPs under the above conditions ( Fig. 6c and d). Fluorescence microscopy of live and dead cells indicated overtly increased in vitro PTT e ciency of PIFC NPs conferred by the delivered CQ molecules (Fig. 6e).

Mechanisms.
To evaluate the increased cytotoxicity of CQ-loaded PIFC NPs, immunoblot and confocal microscopy were performed. The LC3II protein is considered the gold standard for detecting autophagy. In the process of autophagy, the LC3 precursor molecule is cut at the C-terminal 5 amino-acid peptide by ATG4B and cleaved to form cytosolic LC3 (i.e., LC3I). Subsequently, LC3I is induced by APG7L/ATG7, transferred to ATG3 and combined with fatty acid ethanolamine (PE). This coupling is transformed into the membrane type LC3 (i.e., LC3II), which attaches to the autophagosome membrane to form the structural protein of the autophagosome. [41] In MDA-MB-231 cells incubated with PIF NPs and irradiated with an 808-nm laser, LC3II was overtly upregulated compared with the non-NIR irradiation group ( Fig. 7a and b). The above data demonstrated PTT overtly induced autophagy, corroborating a recent report. [21] Both cells administered CQ and PIFC NPs under NIR irradiation had elevated LC3II protein amounts (Fig. 7c and d). These ndings suggest CQ loaded on PLGA NPs does not prevent LC3 I conversion into LC3 II. Next, LysoTracker Green was utilized for staining acidic vesicles (lysosomes) in cells.
Upon NIR irradiation, PIFC NPs induced remarkably enhanced accumulation of autophagic vesicles (green) in cells in comparison with other groups (Fig. 7g). TEM images indicated that MDA-MB-231 cells administered PIF NPs followed by NIR irradiation generated more autophagic vesicles (both autophagosomes and autolysosomes) in comparison with controls. At the same time, in comparison with MDA-MB-231 cells administered PIF NPs, those administered both PIFC NPs and NIR radiation showed more autophagic vesicles, especially autolysosomes (high density of the content; Fig. 7h). This marked elevation of autophagic vesicles in cells after treatment with PIFC NPs may be attributed to CQ, preventing autolysosome degradation. In order to further assess whether autophagic vesicle elevation results from increased autophagy or its suppressed degradation by CQ, P62 degradation levels were examined. P62 is a ubiquitin-binding protein, which is tightly associated with protein ubiquitination. It regulates various cell signal transduction and autophagic processes. [42] During the process of autophagy, P62 protein interacts with ubiquitinated proteins, forming a complex with LC3-II on the inner membrane of the autophagy body and being degraded together in the autophagic lysosome. [43] Therefore, during autophagy, cytosolic P62 protein is continuously degraded; in case of reduced or defective autophagy, the P62 protein continuously accumulates in the cytosol, with its amounts indirectly re ecting autophagosome clearance. [44] As depicted in Fig. 7e and f, in comparison with control cells, P62 amounts were increased in CQ and PIFC NP-treated cells, indicating that the observed increase in autophagic vesicles was due to suppressed autophagy-lysosomal degradation mediated by CQ.
Generally, the occurrence of autophagy needs to go through the following four stages: (1) the intracellular cargo is engulfed to form a double membrane structure; (2) autophagosome formation; (3) autophagosome and lysosome fusion to produce an autophagolysosome; (4) autophagolysosome degradation. [45] Autophagy removes injured organelles, resists cell stresses and induces resistance to therapeutic agents [36] . As shown in Fig. 7i, CQ molecules delivered by PIFC NPs had no negative effect on LC3 I conversion into LC3 II, likely because the autophagy inhibitor CQ acts on the last stage of autophagy to prevent the degradation of autophagic lysosomes and does not affect LC3I conversion into LC3II. However, it markedly suppressed the degradation of autolysosomes by lysosomes, accumulating autophagic vesicles in cells. The observed in vitro PTT e cacy was therefore highly increased after treatment with PIFC NPs.

In Vivo Fluorescence Imaging with PIFC nanoparticles
Through uorescence signal distribution in internal organs (heart, liver, spleen, lung and kidney), the metabolic pathway of PIFC NPs was explored. As depicted in Fig. 8a, with extended injection time, the uorescence signals of the kidney and liver also increased. At 2 h, the uorescence signal of the kidney was strongest, but the signal intensity was always lower than that of the liver. After that, the uorescence signal of the kidney gradually weakened, while that of the liver continued to increase, reaching a peak at 8 h; the uorescence signal disappeared after 24 h. These results showed that PIFC NPs were mainly metabolized by the liver, and a small amount was metabolized by the kidney.
Based on in vitro cell internalization data, tumor imaging of PIFC NPs was examined in tumor-bearing mice. PIFC NPs were administered via the tail vein to study time-dependent in vivo uorescence (Fig. 8b). With increasing time, uorescence signals were increasingly stronger; within 8 h post-injection, strong uorescence signals were observed in the tumor region, and the tumor was overtly demarcated from the adjacent non-cancerous tissues. The subsequent 16 h witnessed a steady and slow reduction in uorescence intensity at the tumor site. Even after 24 hours, uorescence in the tumor area was weakened, but could still be distinguished from normal tissues. Next, uorescence intensity of total tumor photons was determined as a post-injection time function. There was overtly elevated concentration of PIFC NPs in the tumor that was maintained within approximately 8 h post-injection, and the results showed that the prepared PIFC NPs can remain in the tumor for a long time and could be utilized as an ideal uorescence imaging contrast agent.

In vivo Anticancer Therapeutic Effects of PIFC Nanoparticles
In vivo anticancer effects in various groups with or without laser therapy were examined in mice harboring MDA-MB-231 cell xenografts. When the tumors approximated 100 mm 3 , the animals were assigned to 6 groups of ve, including the (1) control, (2) NIR irradiation only, (3) PIFC NPs, (4) PIF NPs with NIR irradiation at 45℃, (5) PIF NPs with NIR irradiation at 55℃ and (6) PIFC NPs with NIR irradiation at 45℃ groups. The animals underwent irradiation with an 808-nm laser for 5 min upon administration of NPs for 8 h in groups 2, 4, 5 and 6. An IR thermal camera was utilized to monitor temperature changes at the tumor site under NIR laser irradiation (Fig. 9a). After injection, the tumor site rapidly heated up under nearinfrared irradiation, reaching 45ºC within 5 minutes. Such temperature could e ciently preserve healthy tissues while exerting overt antitumor effects. Mouse body weights had no overt changes 14 days post-treatment, suggesting no obvious biotoxicity for the tested products (Fig. 9b). Tumor volumes were assessed every other day (Fig. 9c). At study end, mouse euthanasia was performed, and tumors were extracted and weighed. Average tumor weights in various groups after 14 days are depicted in Fig. 9d. The imaged tumors follow the order described in Fig. 9e. Tumors in groups 2 and 3 had rapid growth, similar to control animals. After administration of PIF NPs following NIR irradiation (group 4), tumor growth was inhibited signi cantly. In addition, we also assessed a high temperature group (group 5), whose tumor growth inhibition was higher than that of the low temperature group (group 4). However, group 6 (PIFC with NIR irradiation at 45 ℃) had the most signi cant tumor inhibition. Tumor volumes were quickly reduced by this treatment and remained low throughout the study.
As shown in Fig. 9f, the high temperature of 55°C caused the skin tissue on the tumor surface to be severely injured due to heat, and the scab surface was large. During the treatment, the tumor recurred in the unscabbed area, and the temperature was mild at 45°C. Under photothermal conditions, the wound surface is less damaged and the scab is small. During the treatment, the tumor is effectively ablated due to the presence of the autophagy inhibitor CQ. The experimental results showed that mild photothermal therapy combined with autophagy inhibition can effectively prevent excessive damage to the skin tissue, while effectively inhibiting tumor growth. The above ndings con rmed that PIFC NPs have great potential use in PTT, and autophagy inhibition by CQ could evidently improve the antitumor effect in vivo. Indeed, many studies have recently been conducted to provide the PTT method a functional factor that can achieve a certain degree of autophagy inhibition. [46,47] For example, combination of iron oxide nanoparticles with chloroquine e ciently improved treatment effects by inhibiting autophagy suppression. [48] Autophagy inhibitors (CQ and 3-MA) have been used in chemical-PTT synergistic strategies to obtain excellent results in drug-resistant cancers. [49] However, this research mostly examined direct injection of autophagy suppressors, resulting in the inability to control the dose in subsequent clinical trials. For this reason, the PIFC NPs manufactured in this study provide a high potential for the PTT method under mild near-infrared conditions.

In vivo Toxicity Research
To further evaluate PIFC NPs for in vivo safety, routine blood and blood chemistry markers (re ecting renal and hepatic functions) were assessed in mice administered PIFC NPs for 3 weeks, and no marked changes were detected (Fig. 10a and 10b). H&E staining was also carried out, indicating no overt in ammation or damage in the heart, kidney, spleen, liver, lung and intestine within 3 weeks (Fig. 10c). The above ndings indicate PIFC NPs at the administered doses are safe for biological application.

Conclusion
In summary, the prepared PIFC NPs have excellent uorescence performance for imaging, and may render NPs suitable for high-e ciency PTT. The results indicated nanoparticle-mediated PTT enhances tumor cell autophagy, whose suppression obviously increases PTT's e ciency in cancer cell inhibition. In addition, PIFC NPs are biocompatible nanoparticles comprising photothermal NPs and autophagy inhibitors, which can be used for systemic delivery and in vivo cancer therapy. In this study, autophagy suppression in tumor cells signi cantly enhanced PTT's e cacy, thereby completely inhibiting tumors at mild treatment temperatures. This research showed that regulating autophagy in cancer may be promising in assisting PTT for therapy. Finally, the successful preparation of PIFC NPs in one pot provides a simple approach for manufacturing nanotheranostics that may be useful in both uorescent imaging and PTT in cancer.         In vivo anticancer effect. a) Real time infrared thermal images. b) Body weights of mice following different treatments. c) Relative tumor volumes upon different treatments. d) Tumor weights following different treatments. e) Digital photographs of tumors obtained at day 14 in various mouse groups. f) Wound surfaces of the skin in the tumor area with different temperatures in the laser irradiation area. Scale bar, 100 µm. **p<0.01 and ***p<0.001, determined by Student's t-test.