NIR Photoregulated Theranostic System Based on Hexagonal-Phase Upconverting Nanoparticles for Tumor-Targeted Photodynamic Therapy and Fluorescence Imaging

Although photodynamic therapy (PDT) is an effective, minimally invasive therapeutic modality with advantages in highly localized and specific tumor treatments, large and deep-seated cancers within the body cannot be successfully treated due to low transparency to visible light. To improve the therapeutic efficiency of tumor treatment in deep tissue and reduce the side effects in normal tissue, this study developed a near-infrared (NIR)-triggered upconversion nanoparticle (UCNP)-based photosensitizer (PS) carrier as a new theranostics system. The NaYF4:Yb/Er UCNPs were synthesized by a hydrothermal method, producing nanoparticles of a uniformly small size (≈20 nm) and crystalline morphology of the hexagonal phase. These UCNPs were modified with folic acid-conjugated biocompatible block copolymers through a bidentate dihydrolipoic acid linker. The polymer modified hexagonal phase UCNPs (FA-PEAH-UCNPs) showed an improved dispersibility in the aqueous solution and strong NIR-to-vis upconversion fluorescence. The hydrophobic PS, pheophorbide a (Pha), was then conjugated to the stable vectors. Moreover, these UCNP-based Pha carriers containing tumor targeting folic acid ligands exhibited the significantly enhanced cellular uptake efficiency as well as PDT treatment efficiency. These results suggested that this system could extend the excitation wavelength of PDT to the NIR region and effectively improve therapeutic efficiency of PSs.


Introduction
Photodynamic therapy (PDT) has been recognized as a promising treatment method for a variety of cancers due to its minimal normal tissue toxicity, little invasiveness, highly localized and specific tumor treatments, fewer adverse effects as compared with chemotherapy and radiation therapy [1][2][3][4]. PDT is a powerful noninvasive therapeutic technique for a range of diseases including cancers, based on the photochemical reactions mediated by the interaction of photosensitizers (PSs) with a particular type of light and molecular oxygen. When PSs are exposed to a specific wavelength of light, the PS

Surface Modification of UCNPs
FA-conjugated block copolymer (FA-PEAH) composed of PEG, FA, poly(aspartate), and a dihydrolipoic acid end group, for surface modification of UCNPs was prepared as we reported previously [40]. The surface modification of UCNPs was performed by a ligand exchange method using the synthesized FA-PEAH block copolymers. The UCNPs (80 mg) were dispersed in 10 mL aqueous solution. The dispersion was stirred for 2 h while maintaining the pH at 4 by adding 0.1 M HCl solution to remove the oleate ligands. To remove the oleic acid by extraction with hexane, the solution was mixed with hexane, and repeated three times. The combined hexane layers were re-extracted with water. In addition, the water layers were combined and re-extracted with hexane. After precipitation with cold acetone, the ligand-free UCNPs in the water dispersible fraction were collected by centrifugation. The product was re-dispersed in acetone and then the particles were collected by centrifugation. Finally, the particles were dispersed in water (20 mL) for future use. FA-PEAH (160 mg) was dissolved in 10 mL water, and the solution was added into ligand-free UCNP aqueous solution. The mixture was stirred at room temperature for 24 h, and then purified by dialysis against deionized water for 6 h. The resulting product was freeze-dried for further study [41,42].

Preparation of Pha-Conjugated UCNP Nanocarriers (FA-PEAH-UCNPs-Pha)
To introduce the ketone groups to Pha molecules, Pha (0.176 mol) was dissolved in 30 mL methanol. 4-Hydroxy-2-butanone (1.39 mmol), EDC (0.53 mmol), and DMAP (0.53 mmol) were added to the Pha solution. The mixture was stirred at 400 rpm for 24 h in a dark place at room temperature. Next, the solvent from the resulting mixture was removed in a vacuum oven, and the residue was washed with deionized water. The ketone group-modified Pha (Pha-HB) product was collected by centrifugation, and then freeze-dried.
Next, FA-PEAH-modified UCNPs (120 mg) were dissolved in DMSO (10 mL). The modified Pha (40 mg) dissolved in 4 mL DCM was added into the FA-PEAH-UCNPs solution. Subsequently, four drops of acetic acid were added into the mixture. The reaction mixture was stirred for 24 h without light interference under room temperature. After the DCM was removed under vacuum, DMSO was added into the residue (3 mg/mL). The solution was dialyzed against NaHCO 3 (pH 8.0) solution for 1 day, and then against deionized water for another 12 h. The resulting products were collected by centrifugation, and then the products were freeze-dried. In addition, an FA-unconjugated UCNP carrier sample (CH 3 -PEAH-UCNPs-Pha) was prepared using a similar method as a control group.

Characterization
The modification of copolymers was determined by 600 MHz 1 H NMR (AVANCE III 600, Bruker, Rheinstetten, Germany) using D 2 O, and DMSO-d 6 as the solvent. Size distributions and sizes of UCNPs were measured by dynamic light scattering (ELS-Z, OTSUKA, Hirakata, Japan) at 25 • C using a He-Ne laser (633 nm) as a light source. The surface modification of UCNPs was measured by UV-visible spectrophotometry (UVmini-1240, Shimadzu, Japan), and energy dispersive spectroscopy (EDS) (Tecnai G 2 F30 TEM system). The morphologies of the UCNPs were observed by field-emission transmission electron microscopy (FE-TEM) (Tecnai G 2 F30, FEI, Amsterdam, The Netherlands). The crystalline morphology of nanoparticles was also investigated by selected area electron diffraction (SAED) and X-ray Diffraction (XRD) (D8 DISCOVER, Bruker, Rheinstetten, Germany).

Detection of Singlet Oxygen Generation
The generation of singlet oxygen from FA-PEAH-UCNPs-Pha was determined under 980 nm laser irradiation by monitoring the photobleaching of 1,3-diphenylisobenzofuran (DPBF), that is bleached by singlet oxygen. Experiments were carried out in a quartz cuvette containing 3 mL of FA-PEAH-UCNPs-Pha solution (12 µM) in acetonitrile, and 150 µL of DPBF solution (2 mM). The optical Nanomaterials 2020, 10, 2332 5 of 15 density at 411 nm (λ max of DPBF) was monitored every 2 min using UV-visible spectrometer as a function of time.

Cellular Uptake of FA-PEAH-UCNPs-Pha
The MCF7 breast cancer cell is a FA receptor-overexpressing cell line [43,44], used here for cellular uptake and phototoxicity tests. MCF7 cells were seeded onto 6-well culture plates (1 × 10 5 cells/well) and cultured in RPMI 1640 supplemented with 1% penicillin-streptomycin and 10% FBS at 37 • C under a humidified atmosphere with 5% CO 2 . After incubation for 24 h, the medium was replaced with 1.5 mL of fresh medium containing FA-PEAH-UCNPs-Pha (10 µg/mL Pha equiv.) and free Pha (10 µg/mL), and then incubated for 4 h. The cell nucleus and F-actin were stained by using 4 ,6-Diamidine-2-phenylindol (DAPI) and Alexa Fluor 488 phalloidin. All experiments were performed without light interference under room temperature to prevent photodegradation of the probes. The stained cells were observed using a confocal laser scanning microscope (LSM800, Carl-Zeiss, Jena, Germany).

In Vitro Phototoxicity Assay of FA-PEAH-UCNPs-Pha
MCF7 cells were seeded onto 96-well culture plates and cultured in RPMI 1640. The cell density was about 1 × 10 4 cells per well. After 24 h, the medium was replaced with RPMI 1640 medium containing free Pha and FA/CH 3 -PEAH-UCNPs-Pha under a series of concentrations (0, 5, 10, 20, and 30 µg/mL, Pha equiv.) with 980 nm laser treatment at 0.1 mW/cm 2 for 5 min. After incubation for 24 h, the cell viability was evaluated using a cell viability assay kit (CCK-8, DoGenBio, Korea). Untreated cells served as 100% viable cells. To determine the effect of laser exposure time on the phototoxicity, we also evaluated the in vitro phototoxicity of free Pha, FA-PEAH-UCNPs-Pha, and CH 3 -PEAH-UCNPs-Pha samples after laser (980 nm) radiation for 0, 0.5, 1, and 5 min at 0.2 mW/cm 2 , the concentration of Pha was selected at 10 µg/mL (Pha equiv.). Dark-toxicity of the FA-PEAH-UCNPs-Pha was also investigated by incubating for 4 h under 10 µg/mL (Pha equiv.) without laser irradiation.

Synthesis and Characterization of the UCNP-Based Nanocarrier
The synthetic scheme of the FA-PEAH-UCNPs-Pha nanocarrier is illustrated in Figure 2. Characterization of the FA-PEAH block copolymer used for UCNPs modification was described in detail in our previous report [40].
As shown in Figure 3A, the synthesis of Pha-conjugated UCNP nanocarrier (FA-PEAH-UCNPs-Pha) was confirmed by the presence of the characteristic peaks of Pha at 8.9 ppm, 9.4 ppm, and 9.8 ppm, as well as characteristic peaks of FA at 7.2 ppm, PEG at 3.5 ppm, P(Asp) at 8.2 ppm, and DHLA at 1.2 ppm. In addition, the conjugation content of Pha to FA-PEAH-UCNPs was determined by 1 H NMR using the relative intensity ratio of the methylene protons of the PEG chain (-OCH 2 CH 2 -, 3.5 ppm) to the methylene protons of Pha (8.9 ppm, 9.4 ppm, and 9.8 ppm). The conjugation content of photosensitizer Pha in FA-PEAH-UCNPs-Pha carrier was determined as about 14.7%.
Furthermore, the formation and surface modification of lanthanide-doped NaYF4:Yb/Er UCNPs were characterized by UV-visible spectroscopy and EDS measurements. Figure 3B showed the UV-visible absorption spectra of FA-PEAH-UCNPs and FA-PEAH-UCNPs-Pha. As shown in Figure 3B, the peak at about 279 nm was assigned to FA, while the peaks at about 401 nm, and 690 nm were attributed to Pha. This observation indicated that the Pha was introduced successfully to FA-PEAH polymer chain.
EDS was also employed to evaluate the elemental composition of UCNPs before and after surface modification. As shown in Figure 3C, the characteristic peaks of F, Na, Yb, Y, Er, and C were observed in the free UCNPs sample. After FA-PEAH modification on the surface of UCNPs by a ligand cap exchange method, new S and N characteristic peaks belonging to the FA-PEAH copolymer appeared, and the relative intensity of the C peak increased significantly ( Figure 3D). These results indicated that the NaYF 4 :Yb/Er UCNPs were formed and the FA-PEAH layer was successfully immobilized onto the surface of UCNPs. After surface modification, the solubility of UCNPs improved significantly at the macroscopic level.
The morphologies of UCNPs, FA-PEAH-UCNPs and FA-PEAH-UCNPs-Pha were observed by FE-TEM. As shown in Figure 4A-C, these NaYF 4 :Yb/Er nanocrystals were uniform submicron in size and monodisperse size distribution. The particle size was about 20 nm in dimeter with a hexagonal plate-like shape. The sizes of UCNPs were almost the same before and after modification in TEM images as shown in Figure 4A-C, while the sizes before and after modification were quite different in the DLS data ( Figure 4D). Since TEM measurement is sensitive only to the electron dense metal particles, the size of all samples in the TEM images were almost the same and the polymers used for surface modification were not clearly observed. However, DLS measurement is sensitive to the hydrodynamic diameter of the whole nanocomposite. Thus, the DLS results of surface-modified UCNPs, FA-PEAH-UCNPs and FA-PEAH-UCNPs-Pha samples, exhibit a larger size than the TEM results.
Typical average particle size distributions measured by DLS for free UCNPs, FA-PEAH-UCNPs, and FA-PEAH-UCNPs-Pha are shown in Figure 4D as 996.0, 68.6, and 90.3 nm, respectively. Since free UCNPs were quite hydrophobic due to the hydrophobic oleate capping ligand before surface modification, the macroscopic aggregations were observed. Thus, DLS data of free UCNPs exhibited a much larger size compared with the surfaced-modified UCNP samples (FA-PEAH-UCNPs and FA-PEAH-UCNPs-Pha). However, after surface modification of the hydrophilic FA-PEAH polymer instead of the hydrophobic oleate ligand, the dispersity and solubility in aqueous solution significantly improved. As shown in Figure 4D, the size of the nanoparticles significantly decreased (about 68.6 nm for FA-PEAH-UCNPs and 90.3 nm for FA-PEAH-UCNPs-Pha), and the size distribution maintained a narrow monodisperse unimodal pattern. In order to evaluate the deep-penetration PDT application, hexagonal-phase UCNPs are the best choice, because hexagonal-phase NaYF 4 :Yb/Er UCNPs usually produce a bright green emission around 550 nm along with a weak dark red emission around 660 nm under 980 nm NIR irradiation [45]. It has been also reported that NaYF 4 crystals with hexagonal-phase (β-NaYF4) are the most efficient host materials for upconverting lanthanide ions due to the low phonon energy of the crystal lattice. The crystalline morphology of the synthesized NaYF 4 :Yb/Er UCNPs was investigated by the SAED pattern, as shown in Figure 4E. The SAED pattern of the UCNPs was shown as spotty polycrystalline diffraction rings, which can be indexed to the (100), (110), (101), (110), (200), (111), (201), (210), (002), (300), (211), and (321) planes of hexagonal NaYF 4 lattice. We also employed XRD to further confirm the crystalline morphology of NaYF 4 :Yb/Er UCNPs. In Figure 5, the peak positions and intensities of the free UCNPs agree well with the standard pattern of hexagonal phase NaYF 4 crystal ( Figure 5; , JCPDS 16-0334). These results indicated that the synthesized NaYF 4 :Yb/Er UCNPs have the hexagonal β-phase. Additionally, in Figure 5, the peak positions between 15 and 30 degrees were well matched with the standard pattern of PEG ( , JCPDS . It also could be evidence that the FA-PEAH layer was successfully immobilized onto the surface of UCNPs.
Nanomaterials 2020, 10, x FOR PEER REVIEW 8 of 14 [45]. It has been also reported that NaYF4 crystals with hexagonal-phase (β-NaYF4) are the most efficient host materials for upconverting lanthanide ions due to the low phonon energy of the crystal lattice. The crystalline morphology of the synthesized NaYF4:Yb/Er UCNPs was investigated by the SAED pattern, as shown in Figure 4E. The SAED pattern of the UCNPs was shown as spotty polycrystalline diffraction rings, which can be indexed to the (100) We also employed XRD to further confirm the crystalline morphology of NaYF4:Yb/Er UCNPs. In Figure 5, the peak positions and intensities of the free UCNPs agree well with the standard pattern of hexagonal phase NaYF4 crystal ( Figure 5; ■, JCPDS 16-0334). These results indicated that the synthesized NaYF4:Yb/Er UCNPs have the hexagonal β-phase. Additionally, in Figure 5, the peak positions between 15 and 30 degrees were well matched with the standard pattern of PEG (▼, JCPDS . It also could be evidence that the FA-PEAH layer was successfully immobilized onto the surface of UCNPs.

Luminescence Properties of UCNPs
The upconversion fluorescence spectra of free UCNPs and surface modified UCNPs (FA-PEAH-UCNPs) in aqueous solution excited with a 980 nm laser at room temperature are shown in Figure 6. As shown in Figure 6, the luminescence efficiency of FA-PEAH-UCNPs increased significantly compared to free UCNPs after surface modification. The free UCNPs stabilized by oleic acid had very weak luminescence intensity because they were rarely dispersed in aqueous solution, whereas the dispersion properties of FA-PEAH-UCNPs increased and the luminescence intensity increased significantly after modifying the water-soluble FA-PEAH.

Luminescence Properties of UCNPs
The upconversion fluorescence spectra of free UCNPs and surface modified UCNPs (FA-PEAH-UCNPs) in aqueous solution excited with a 980 nm laser at room temperature are shown in Figure 6. As shown in Figure 6, the luminescence efficiency of FA-PEAH-UCNPs increased significantly compared to free UCNPs after surface modification. The free UCNPs stabilized by oleic acid had very weak luminescence intensity because they were rarely dispersed in aqueous solution, whereas the dispersion properties of FA-PEAH-UCNPs increased and the luminescence intensity increased significantly after modifying the water-soluble FA-PEAH.
The upconversion fluorescence spectra of free UCNPs and surface modified UCNPs (FA-PEAH-UCNPs) in aqueous solution excited with a 980 nm laser at room temperature are shown in Figure 6. As shown in Figure 6, the luminescence efficiency of FA-PEAH-UCNPs increased significantly compared to free UCNPs after surface modification. The free UCNPs stabilized by oleic acid had very weak luminescence intensity because they were rarely dispersed in aqueous solution, whereas the dispersion properties of FA-PEAH-UCNPs increased and the luminescence intensity increased significantly after modifying the water-soluble FA-PEAH.

Detection of Singlet Oxygen Generation
The generation of singlet oxygen from FA-PEAH-UCNPs-Pha was determined under 980 nm laser irradiation by monitoring the photobleaching of 1,3-diphenylisobenzofuran (DPBF), that is bleached by singlet oxygen. The decrease in absorbance peak intensity of DPBF at 411 nm was monitored every 2 min using UV-visible spectrometer as a function of time ( Figure 7A,B). FA-PEAH-UCNPs-Pha sample showed a sharp decline in optical density at 411 nm (λ max of DPBA), indicating rapid generation of singlet oxygen as a function of time of 980 nm laser (2 W/cm 2 ) exposure. On the other hand, the DPBF absorbance of free Pha as a function of time has not changed significantly since free Pha was not activated by 980 nm NIR laser. In the case of FA-PEAH-UCNPs-Pha, the absorbance of DPBF decreased by about 6% within 30 min. These results indicate that the luminescence intensity of FA-PEAH-UCNPs-Pha by NIR 980 nm laser irradiation can activate the Pha molecules and produces singlet oxygen leading to cytotoxicity. rapid generation of singlet oxygen as a function of time of 980 nm laser (2 W/cm 2 ) exposure. On the other hand, the DPBF absorbance of free Pha as a function of time has not changed significantly since free Pha was not activated by 980 nm NIR laser. In the case of FA-PEAH-UCNPs-Pha, the absorbance of DPBF decreased by about 6% within 30 min. These results indicate that the luminescence intensity of FA-PEAH-UCNPs-Pha by NIR 980 nm laser irradiation can activate the Pha molecules and produces singlet oxygen leading to cytotoxicity.

Cellular Localization of UCNP-Based Nanocarrier in Tumor Cells
The cellular uptake and accumulation in malignant tissues of PS greatly affect the efficiency of PDT treatment [46]. The fate of the UCNP-based carriers in tumor cells was observed using a confocal

Cellular Localization of UCNP-Based Nanocarrier in Tumor Cells
The cellular uptake and accumulation in malignant tissues of PS greatly affect the efficiency of PDT treatment [46]. The fate of the UCNP-based carriers in tumor cells was observed using a confocal laser scanning microscopy. Figure 8 exhibited the confocal microscopy images based on the blue fluorescence from DAPI bound to the nucleus, F-actin in the cytoplasm stained by Alexa Fluor 488 phalloidin, and the red fluorescence of Pha.
For MCF7 cells treated with free Pha, the red fluorescence signal from Pha detected in MCF7 cells was relatively weak ( Figure 8B). On the other hand, UCNP-based Pha carriers, FA-PEAH-UCNPs-Pha and CH 3 -PEAH-UCNPs-Pha, showed higher red fluorescence Pha signal compared with the free Pha sample ( Figure 8C,D). Especially, after FA-PEAH-UCNPs-Pha treatment, MCF7 cells exhibited strong red fluorescence of Pha around the nucleus and inside the cells. It means that Pha was effectively internalized into the MCF7 cells ( Figure 8D). To compare the cellular uptake difference between CH 3 -PEAH-UCNP-Pha and FA-PEAH-UCNP-Pha, the fluorescence intensities of Pha in the confocal images was analyzed by dividing the local image into six equal areas and the average fluorescence intensities were plotted. Free Pha and CH 3 -PEAH-UCNP-Pha samples showed similar fluorescence intensities of Pha, while FA-PEAH-UCNP-Pha showed much stronger fluorescence intensity ( Figure 8E). As shown in Figure 8F, flow cytometry assay also showed that the cellular uptake of FA-PEAH-UCNPs-Pha higher than that of Free Pha and CH 3 -PEAH-UCNPs-Pha, in agreement with the confocal microscopy results.
The results indicate that the UCNP-based Pha carrier improved the water solubility of Pha molecules, and FA ligands could enhance the cellular uptake of Pha into MCF7 cells by an active targeting effect. In addition, these strong fluorescence signals from Pha molecules can be used to determine the dynamics of signal transduction in the intracellular networks and in diagnostics.
As the Pha concentration increased, the cell viability gradually decreased. Notably, the viability of MCF7 cells treated with FA-PEAH-UCNPs-Pha sample decreased to nearly 25% after a 5 min treatment at a concentration of 30 µg/mL. In order to determine the effect of laser exposure time on the phototoxicity, we also determined the in vitro phototoxicity of free Pha, FA-PEAH-UCNPs-Pha, and CH 3 -PEAH-UCNPs-Pha after 980 nm laser radiation for 0, 0.5, 1 and 5 min at 0.2 mW/cm 2 (10 µg/mL, Pha equiv.). Under the dark condition, free Pha, FA-PEAH-UCNPs-Pha, and CH 3 -PEAH-UCNPs-Pha exhibited more than 90% cell viability and no significant dark toxicity as shown in Figure 8B (0 min of laser exposure time, 10 µg/mL Pha concentration). However, the cell viability significantly decreased as the laser exposure time increased ( Figure 9B). The PDT efficiency of FA-PEAH-UCNPs-Pha was obviously higher than free Pha and CH 3 -PEAH-UCNPs-Pha. These results are probably due to the increased solubility of hydrophobic Pha molecules in aqueous environments by loading into the block copolymer chain-immobilized UCNP carriers, resulting in an enhanced 1 O 2 quantum yield of Pha.
8B (0 min of laser exposure time, 10 μg/mL Pha concentration). However, the cell viability significantly decreased as the laser exposure time increased ( Figure 9B). The PDT efficiency of FA-PEAH-UCNPs-Pha was obviously higher than free Pha and CH3-PEAH-UCNPs-Pha. These results are probably due to the increased solubility of hydrophobic Pha molecules in aqueous environments by loading into the block copolymer chain-immobilized UCNP carriers, resulting in an enhanced 1 O2 quantum yield of Pha. Several papers have reported that targeting using folic acid receptors does not seem to have a significant effect on increasing phototoxicity [47]. However, many studies have reported that active targeting using folic acid ligands increases phototoxicity [48][49][50]. In this study, FA-conjugated FA-PEAH-UCNPs-Pha showed higher phototoxicity than free Pha and CH3-PEAH-UCNPs-Pha, possibly due to improved cellular uptake by the FA receptor-medicated endocytosis process.

Conclusions
In order to demonstrate the NIR-light induced photodynamic therapy, we have designed the NIR light-triggered theranostic system based on hexagonal-phase UCNPs for efficient PDT with enhanced deep tissue penetration ability and fluorescence imaging. Hexagonal-phase NaYF4:Yb/Er UCNPs were prepared by a hydrothermal method and the UCNPs were monodisperse with a uniform size of about 20 nm. The crystalline morphology of the synthesized NaYF4:Yb/Er UCNPs showed a thermodynamically stable hexagonal β-phase. To increase the aqueous solubility and conjugate functional moieties for subsequent biological functionalization, the surface of UCNPs was modified with FA-conjugated copolymers through a bidentate dihydrolipoic acid linker. The FA-PEAH copolymer-modified UCNPs (FA-PEAH-UCNPs) exhibited enhanced solubility in aqueous solution. Then, the hydrophobic Pha molecules were combined to FA-PEAH-UCNPs. These water dispersible FA-PEAH-UCNPs have a much stronger luminescence property compared with hydrophobic UCNPs. The upconversion fluorescence spectra of FA-PEAH-UCNPs excited with a 980 nm laser showed sharp green emissions at 510~530 nm and at 530~570 nm as well as a red emission at 645~680 nm. These FA-PEAH-UCNPs-Pha that produce high energy visible photons from low energy radiation in the NIR region could be promising materials for PDT and bioimaging.
Due to the active tumor targeting FA ligand conjugation, the phototoxicity and cellular uptake against MCF7 cells of FA-PEAH-UCNPs-Pha were significantly enhanced compared with free Pha and FA-ligand unconjugated CH3-PEAH-UCNPs-Pha. These UCNP-based Pha nanocarriers, which Several papers have reported that targeting using folic acid receptors does not seem to have a significant effect on increasing phototoxicity [47]. However, many studies have reported that active targeting using folic acid ligands increases phototoxicity [48][49][50]. In this study, FA-conjugated FA-PEAH-UCNPs-Pha showed higher phototoxicity than free Pha and CH 3 -PEAH-UCNPs-Pha, possibly due to improved cellular uptake by the FA receptor-medicated endocytosis process.

Conclusions
In order to demonstrate the NIR-light induced photodynamic therapy, we have designed the NIR light-triggered theranostic system based on hexagonal-phase UCNPs for efficient PDT with enhanced deep tissue penetration ability and fluorescence imaging. Hexagonal-phase NaYF4:Yb/Er UCNPs were prepared by a hydrothermal method and the UCNPs were monodisperse with a uniform size of about 20 nm. The crystalline morphology of the synthesized NaYF 4 :Yb/Er UCNPs showed a thermodynamically stable hexagonal β-phase. To increase the aqueous solubility and conjugate functional moieties for subsequent biological functionalization, the surface of UCNPs was modified with FA-conjugated copolymers through a bidentate dihydrolipoic acid linker. The FA-PEAH copolymer-modified UCNPs (FA-PEAH-UCNPs) exhibited enhanced solubility in aqueous solution. Then, the hydrophobic Pha molecules were combined to FA-PEAH-UCNPs. These water dispersible FA-PEAH-UCNPs have a much stronger luminescence property compared with hydrophobic UCNPs. The upconversion fluorescence spectra of FA-PEAH-UCNPs excited with a 980 nm laser showed sharp green emissions at 510~530 nm and at 530~570 nm as well as a red emission at 645~680 nm. These FA-PEAH-UCNPs-Pha that produce high energy visible photons from low energy radiation in the NIR region could be promising materials for PDT and bioimaging.
Due to the active tumor targeting FA ligand conjugation, the phototoxicity and cellular uptake against MCF7 cells of FA-PEAH-UCNPs-Pha were significantly enhanced compared with free Pha and FA-ligand unconjugated CH 3 -PEAH-UCNPs-Pha. These UCNP-based Pha nanocarriers, which include FA-conjugated copolymers, could increase the aqueous solubility of hydrophobic Pha molecules and enhance 1 O 2 quantum yield of tumor tissue, as well as improve PDT efficiency. In addition, this UCNP nanocarrier could be triggered by NIR which has low photodamage effects on cells, and high signal-to-noise ratios, and deep tissue penetration. These results indicate that the FA-PEAH-UCNPs-Pha system could have great potential to serve as a targeted PS delivery system for tumor PDT applications in deep tissue.
Author Contributions: Conceptualization, L.Z. and S.Y.K.; analysis, L.Z., Y.L. and J.C.; writing, all the authors; review and editing, J.C. and S.Y.K. All authors have read and agreed to the published version of the manuscript.