Rationally designed monoamine oxidase A‐activatable AIE molecular photosensitizer for the specific imaging and cellular therapy of tumors

Due to both the requirements of repeated complex organic synthesis and tedious biochemical verification experiments for the screening of new photosensitizers, the development of a satisfactory photosensitizer with excellent photosensitivity and highly specific response ability is still a great challenge for the accurate imaging localization and the precise photodynamic therapy (PDT) of tumors. Herein, under the help of theoretical calculations, a high‐efficient target‐activatable aggregation‐induced emission (AIE) molecular photosensitizer, TPE‐TThPy, is rationally designed and synthesized by the conjugation of tetraphenylethylene (TPE, which is used as the electron donor) and tetrahydropyridine (ThPy, which can be converted to methylpyridine salts as the electron acceptor) using thiophene (T) as the π‐bridge. This TPE‐TThPy molecule exhibits not only good cellular uptake and mitochondrial targeting ability but also ultra‐high monoamine oxidase A (MAO‐A) response specificity and excellent photosensitivity when oxidized under the action of MAO‐A. The specifically imaging ability and cellular PDT performance of the MOA‐A‐activatable AIE photosensitizer of TPE‐TThPy is demonstrated by using different cell lines and mouse tumor models. The successful development of this MOA‐A‐activatable AIE photosensitizer also provides insight for the development of single‐molecule PDT therapeutic drugs with excellent photosensitivity and highly specific targeting‐response ability.


INTRODUCTION
Photodynamic therapy (PDT), as a promising lightmodulated therapy for cancer and other diseases, has attracted extensive attention in recent decades. [1][2][3] PDT requires the combined action of photosensitizers (PSs), light, and oxygen to produce highly toxic reactive oxygen species (ROS), which can cause cell apoptosis or death, destroy blood vessels, and trigger immune responses for the treatment of pathological tissues. [4,5] As an important component of PDT, PSs not only play a pivotal role in ROS generation but can also guide the irradiation position of light through its fluorescence signal to improve the effect of PDT. However, improving both the photosensitization ability of PSs and the image-guided precision of photodynamic therapy is a major challenge for PDT.
Traditional PSs such as porphyrin and rose bengal are planar structures. [6] Due to the strong π-π stacking effect, the aggregation-induced quenching effect (ACQ) occurs when they aggregate. [7,8] This will cause the fluorescence signal to abate and damage the photosensitive properties, greatly weakening the fluorescence imaging and PDT effect. Moreover, traditional PSs is always in the "on" state and lack of targeted activation, which makes it unable to accurately distinguish the target position and easily damages other normal tissues. [9,10] Therefore, the development of novel PSs with higher photosensitivity and specific target response ability is of great significance for the accurate imaging localization and PDT of tumors.
Theoretically, the increase of the highest occupied molecular orbital (HOMO) energy or the decrease of the lowest unoccupied molecular orbital (LUMO) energy, that is, the decrease of the HOMO-LUMO energy gap (ΔE) of the PS S C H E M E 1 Schematic illustration of (A) the synthesis of the MAO-A-activatable AIE molecular photosensitizer (TPE-TThPy) and (B) its working principle for the highly specific imaging and cellular level therapy of tumors molecules, will cause a red-shift emission of the fluorescent molecules and also reduce the energy level difference (ΔE ST ) between the singlet state (S 1 ) and triplet state (T 1 ) of the molecules, promoting the intersystem crossing (ISC) process and enhancing their photosensitive efficiency. [11] Generally, the construction of a donor-π-acceptor (D-π-A) system can reduce the HOMO-LUMO energy gap of an organic PS molecule. [12,13] In addition, in contrast to ACQ fluorophore, the aggregation induced emission (AIE) fluorophore with a rotor structure can emit intense fluorescence when they are aggregated, which is due to restriction of the molecular rotation and activation of the radiation attenuation pathway upon aggregation of the AIE type fluorophore. [14][15][16][17] Moreover, the AIE type photosensitizers can also exhibit greatly enhanced photosensitivity in the aggregated state. [18,19] Monoamine oxidase (MAOs) is a kind of flavinase that can oxidize amino groups to corresponding imines and then hydrolyze them to aldehydes. It also acts on methylated secondary and tertiary amines, mainly distributed in the outer membrane of mitochondria of various cells. [20] Currently, there are two known human MAOs with approximately 70% homology (MAO-A and MAO-B), among which MAO-A is overexpressed in depression, [21] liver disease, [22] kidney disease, [23] and cancer. [24] Nonetheless, the function of MAO-A outside the brain has not been explored to a large extent, especially its role in anti-tumor. [25] Therefore, MAO-A has the potential to become a biomarker for the identification of tumor cells and a worthy target for fluorescence probes and antitumor drug studies, but there is no research on MAO-A activation of photosensitizer.
Therefore, as illustrated in Scheme 1, we herein designed a novel target-activated aggregation-induced emission (AIE) molecular probe of TPE-TThPy by conjugating tetraphenylethylene (TPE) with tetrahydropyridine (ThPy) using thiophene (T) as the π-bridge. Where, TPE is used as the electron donor, [26] and ThPy can be converted to methylpyridine salts (Pys) as the electron acceptor under the specific catalysis of MAO-A. [27][28][29] When the TPE-TThPy molecules enter tumor cells with a high expression level of MAO-A, [25,30,31] the ThPy group of the TPE-TThPy molecule will oxidize and become a TPys group. The TPE-TThPy molecule will then be converted to TPE-TPys with perfect D-π-A structure and highly positive charges under the action of MAO-A, reducing the HOMO-LUMO energy gap and promoting the ISC process. This results in both red-shifted fluorescence emission and excellent photosensitivity in the tumor cells. Moreover, due to the highly positive charges of the TPE-TPys molecules, they will stay on the mitochondrial membrane of the tumor cells to generate AIE. [32][33][34][35] This leads to the efficient and high production of singlet oxygen ( 1 O 2 ) from the oxygen-rich mitochondria, [36,37] damaging the tumor cells and realizing cellular therapy. [38,39] 2 RESULTS AND DISCUSSION

Theory calculation of the experimental design
Before the synthesis of the TPE-TThPy and TPE-TPys molecules, we first calculated the HOMO-LUMO energy levels (ΔE) and the ΔE ST of both molecules using Gaussian software and density functional theory (DFT) to verify the feasibility of our experimental design. The TPE-TPys molecule was found to have a longer absorption wavelength due to its smaller HOMO-LUMO gap than that of the TPE-TThPy molecule ( Figure 1A). This can be attributed to the strong electron acceptor (pyridine salts) in TPE-TPys, leading to stronger electron donating/accepting interactions in TPE-TPys. DFT calculations also show that the delocalization of the HOMO of TPE-TThPy is mainly in the TPE region, while the delocalization of the LUMO is mainly in the thiophene region. Conversely, for the TPE-TPys molecule, HOMO electron density delocalization is mainly distributed in the TPE region, while the LUMO electron density delocalization is mainly in the pyridine salt group. Similarly, according to DFT calculations, S 1 of the TPE-TThPy and TPE-TPys molecules are 2.5070 and 1.3919 eV, respectively, indicating that the TPE-TPys molecule had more red-shifted fluorescence emission than TPE-TThPy. Furthermore, the calculated ΔE ST of the TPE-TThPy and TPE-TPys molecules were 1.57 and 0.54 eV, respectively, revealing the excellent photosensitivity of TPE-TPys. [13] The above results suggest that both ΔE and ΔE ST significantly decrease when the TPE-TThPy molecule is converted to TPE-TPys with perfect D-π-A structure and highly positive charges, which would result in the redshifted fluorescence emission and significantly enhanced photosensitivity.
In addition, one can also conclude from the experimental design illustrated in Scheme 1 that whether the TPE-TThPy molecules can be specifically oxidized to TPE-TPys molecules under the action of MOA-A is another very important fundament for the feasibility of this experimental design. To explore whether MAO-A can specifically recognize TPE-TThPy molecule and convert it to TPE-TPys, we then studied the docking effect of the TPE-TThPy molecule with the crystal structure of MAO-A (PDB: 2Z5X) and MAO-B (PDB: 1S3E) through docking studies ( Figure 1B). As shown in Figure 1B, in the docked TPE-TThPy/MAO-A complex, the ThPy group of the TPE-TThPy molecule was closer to flavin adenine dinucleotide (FAD). The calculated distance between the tetrahydropyridine nitrogen and C5 in FAD was 3.7 Å, while the binding energy of TPE-TThPy and MAO-A was −8.17 kcal mol −1 , suggesting that MAO-A showed good selectivity and high efficient catalytic transformation ability toward the TPE-TThPy molecule. This is derived from the shorter and wider substrate cavity of MAO-A, allowing the TPE part of TPE-TThPy to be well embedded in the substrate cavity. [40,41] However, in the docked TPE-TThPy/MAO-B complex, the ThPy group remained separated from FAD. The measured distance between the tetrahydropyridine nitrogen and C5 in FAD was 6.2 Å, while the binding energy of TPE-TThPy and MAO-B was −3.87 kcal mol −1 . This indicated that the TPE-TThPy molecule likely cannot combine with MAO-B well to efficiently catalyze its transformation of TPE-TThPy. Together, the above results theoretically indicate the feasibility of our experimental design. Therefore, both the TPE-TThPy and TPE-TPys molecules were subsequently synthesized (Scheme S1, Supporting Information) and all the products in the synthesis were characterized and confirmed by NMR and MS (Figures S1-S15, Supporting Information).

Optical properties of TPE-TThPy and TPE-TPys
The optical properties of the two molecules (TPE-TThPy and TPE-TPys) were first investigated by both UV-vis and fluorescence spectroscopy ( Figure 2B) to 660 nm (TPE-TPys; excitation 450 nm; Figure 2C). Moreover, TPE-TThPy had no significant absorption for light with a wavelength greater than 430 nm (Figure 2A), and its significant fluorescence emission could not be observed in the TPE-TPys emission range of 550-800 nm when both molecules were excited using 450-nm light ( Figure S16, Supporting Information). In addition, although both the two molecules showed enhanced fluorescence emission in aqueous solution, it was much weaker than the fluorescence emission when they were dispersed in glycerol or in a poor solvent ( Figure 2B,C; Figure S17, Supporting Information). It was likely that their excessive aggregation in the aqueous solution reduced the effective absorption of the excitation light, preventing strong fluorescence emission. However, when these molecules are uniformly dispersed in glycerol or in a poor solvent, they can absorb the excitation light efficiently while their molecular rotation is restricted and produce strong AIEs.

Photosensitivity investigation of TPE-TThPy and TPE-TPys
To investigate the 1 O 2 -generation ability of TPE-TThPy and TPE-TPys, we used 9,10-anthracenylbis(methylene)dimalonic acid (ABDA), which can be selectively oxidized by 1 O 2 to provide peroxide with a reduction in absorbance, as the ROS probe (Figure 2D-F; Figure S18, Supporting Information). [42] Under white light irradiation (50 mW cm −2 ), a significant absorbance change of ABDA at 380 nm could not be observed for the solution containing TPE-TThPy ( Figure 2D), while the absorbance of ABDA at 380 nm decreased rapidly for the solution containing TPE-TPys ( Figure 2E). Moreover, the rate of decrease was significantly faster than that for commercial PSs of the solution containing Rose Bengal ( Figure 2F; Figure S18, Supporting Information), indicating that TPE-TThPy has no obvious photosensitive property but TPE-TPys has excellent 1 O 2 -generation ability. Together, the above results indicate the feasibility of both accurate imaging localization and precise PDT at the cellular level through the specific target response conversion from TPE-TThPy to TPE-TPys and the insertion of the resulted molecular probes into phospholipid membrane to generate highly efficient AIEs. It also suggests that the use of lasers with a wavelength greater than 450 nm for excitation would further avoid the background fluorescence signal and damage to normal cells or tissues of the hydrophobic AIE PSs in a physiological environment.

Specific recognition of TPE-TThPy by MAO-A
The specific catalysis ability of MAO-A for the transformation of TPE-TThPy to TPE-TPys was carefully investigated  Figure S19A). This indicates that the TPE-TThPy molecule can be indeed converted to TPE-TPys with ideal D-π-A structure and highly positive charges under the action of MAO-A and demonstrates its excellent specificity for MAO-A. The catalytic oxidation performance of MAO-A for the transformation of TPE-TThPy to TPE-TPys was also investigated. As shown in Figure 2H, during the incubation process, the fluorescence emission at 650 nm attributed to TPE-TPys increased gradually for the 10 μM of TPE-TThPy in PBS (pH 7.4) upon the addition of MAO-A (20 μg mL −1 ) and reached a maximum that was nine times stronger after 48 min. In contrast, no obvious fluorescence emission at 650 nm attributed to TPE-TPys was observed for the 10 μM of TPE-TThPy in PBS (pH 7.4) upon addition of MAO-B (20 μg mL −1 ) during the incubation period. The effect of temperature on the MAO-A-catalyzed reaction was also investigated and 37 • C was found to be an optimum temperature ( Figure S19B, Supporting Information). Furthermore, a significant TPE-TPys fluorescence emission can be observed in the presence of MAO-A at very low concentrations (1.0 μg mL −1 , Figure 2I) but not for either 10 μM of  Figure S19D, Supporting Information). Finally, it was confirmed by HPLC that TPE-TThPy can be catalyzed by MAO-A to TPE-TPys ( Figure  S19E, Supporting Information). These results demonstrate that MAO-A can convert TPE-TThPy to TPE-TPys quickly and efficiently, further suggesting its highly specific catalytic property.
To demonstrate the efficient photosensitivity of TPE-TPys, TEMP (2,2,6,6-Tetramethylpiperidine) as the singlet oxygen trapping agent was first added to the solution of TPE-TThPy (10 μM) in PBS (pH 7.4) pre-incubated with and without MAO-A (20 μg mL −1 ) and mixed thoroughly. The resulting mixture solution was subsequently subjected to ESR (electron spin resonance) measurement after irradiation with a white light (50 mW cm −2 ) for 1 min ( Figure S20, Supporting Information). Strong ESR signals attributed to 1 O 2 were observed for the TPE-TThPy solution pre-incubated with MAO-A, while no significant ESR signal of 1 O 2 could be observed for the TPE-TThPy solution pre-incubated without MAO-A. This further supported the conversion of the TPE-TThPy to TPE-TPys under the action of MAO-A and the excellent 1 O 2 -generation ability of the resulting product.
The above results agree well with the theoretical calculations, further demonstrating the feasibility of our experimental design.

Cellular uptake and cellular imaging of the TPE-TThPy molecule
After the successful demonstration of the experimental design in vitro, the applicability of the TPE-TThPy molecular probes in real biomedical systems was first investigated by using the SH-SY5Y, HepG2, and NIH-3T3 cell lines as models. Human SH-SY5Y and HepG2 cells are two known tumor cell lines that express elevated levels of MAO-A and MAO-B activity, respectively, while NIH-3T3 cells do not express MAO. [43][44][45] We also confirmed that MAO-A was overexpressed in only SH-SY5Y cells among the three cell lines by using ELISA method ( Figure S21, Supporting Information). To investigate the cellular uptake and cellular imaging of the TPE-TThPy probe, SH-SY5Y, HepG2, and NIH-3T3 cells were first incubated with 5 μM of TPE-TThPy and then washed with PBS for fluorescence imaging ( Figure 3A-D). The fluorescence emission in the blue channel (450-500 nm) attributed to TPE-TThPy rapidly increased with the incubation time for all three cell lines ( Figure 3A,B). However,  Figure S22, Supporting Information). Furthermore, the co-localization images of the three cell lines, illustrated in Figure 4A,B and Figure S23, Supporting Information, also indicated that only the resulting TPE-TPys product with positive charges in SH-SY5Y cells showed high targeting ability towards mitochondria (P = 0.92) ( Figure 4A), while the non-charged TPE-TThPy molecules in all three cell lines neither show this specific mitochondria-targeting property (P = 0.72, 0.65, and 0.53 for HepG2, NIH-3T3, and SH-SY5Y cells, respectively) ( Figure 4B) nor show specific lysosome-targeting property (P = 0.39, 0.24, and 0.45 for HepG2, NIH-3T3, and SH-SY5Y cells, respectively) ( Figure S23, Supporting Information). This is perhaps being due to how positively-charged TPE-TPys can bind to the membrane of mitochondria through electrostatic action, [46][47][48] limiting its intramolecular rotation and enhancing the red emission. These above results indicate that this TPE-TThPy molecular probe has no selectivity for cell type or suborganelles and can effectively enter tumor cells or normal cells, but the red-emitted TPE-TPys resulting from the TPE-TThPy probes can be generated only in the cells with high expression levels of MAO-A and then target mitochondria to generate AIEs. This demonstrates the applicability of the good cellular uptake and excellent specific response activated cellular imaging and therapy of the proposed TPE-TThPy molecular probe.

Cell-selective killing property of the TPE-TThPy molecule
The toxicity of TPE-TThPy to SH-SY5Y, HepG2, and NIH-3T3 cells was then investigated using the CCK-8 method to evaluate the cell survival rate of these cells treated with TPE-TThPy and irradiation ( Figure 4C). The experimental results illustrated in Figure 4C indicate that, under an extremely low radiation dose of white light (5 mW cm −2 , 15 min), over half of the SH-SY5Y cells were killed when a lower concentration of TPE-TThPy (e.g., 5 μM) was used, while over 95% cell viability was still obtained for HepG2 and NIH-3T3 cells even after treatment with a higher concentration of TPE-TThPy (e.g., 10 μM). This suggests the excellent selective killing ability of TPE-TThPy toward the target cells with high expression levels of MAO-A, and also agreed well with the cellular uptake and imaging results, indicating the conversion of TPE-TThPy to TPE-TPys and insertion into the mitochondrial membrane of only cells with high expression of levels MAO-A (SH-SY5Y). Meanwhile, TPE-TThPy showed no obvious toxicity to all three cell lines treated without irradiation, even when incubated with very high concentrations of TPE-TThPy (e.g., 50 μM, survival rate > 80%), suggesting the potential good biocompatibility of TPE-TThPy ( Figure  S24, Supporting Information). Then, an 1 O 2 -indicator of 2,7dichlorodihydrofluorescein diacetate (DCFH-DA), which can be converted to dichlorofluorescein (DCF) in the presence of 1 O 2 and provide a strong green fluorescence signal, [49,50] Figure 4D, no significant green fluorescence emission of DCF was detected for either HepG2 or NIH-3T3 cells treated with TPE-TThPy/DCFH-DA and irradiation, CL-incubated SH-SY5Y cells treated with TPE-TThPy/DCFH-DA and irradiation, or SH-SY5Y cells treated with TPE-TThPy/DCFH-DA but without irradiation, indicating the absence of 1 O 2 in these cells due to the absence of TPE-TPys photosensitizer or the presence of TPE-TPys but without irradiation. In contrast, the strong green fluorescence emission of DCF can be observed for TPE-TThPy/DCFH-DA-treated SH-SY5Y cells with irradiation, indicating the highly efficient generation of 1 O 2 in living SH-SY5Y cells due to the production of TPE-TPys photosensitizer in the presence of MAO-A, which agreed well with the cellular imaging results and cytotoxicity measurements of the TPE-TThPy molecular probes.

Cell death mechanism of the TPE-TThPy molecule
To further prove that the generation of 1 O 2 using TPE-TPys as the photosensitizer led to the damage of SH-SY5Y cells, calcein-AM and propidium iodide (PI) fluorescence imaging were adopted to measure the TPE-TThPy-treated SH-SY5Y, HepG2, and NIH-3T3 cells before and after irradiation ( Figure S25, Supporting Information). Calcein-AM can be degraded by esterase in living cells to calcitonin, which emits a strong green fluorescence, whereas PI can cross the membrane of dead cells to reach the nucleus, and emitting a red fluorescence signal. [51,52] Only a strong green fluorescence emission was detected for HepG2 and NIH-3T3 cells treated with TPE-TThPy and irradiation as well as SH-SY5Y cells treated with TPE-TThPy without irradiation. In contrast, only a strong red fluorescence emission was observed for SH-SY5Y cells treated with both TPE-TThPy and irradiation, further indicating that the TPE-TThPy can be converted to TPE-TPys only in the cells with high expression levels of MAO-A to efficiently produce 1 O 2 under irradiation, resulting in the selective death of these cells.
Annexin V-FITC and PI fluorescence imaging were also used to measure the apoptosis of SH-SY5Y cells and further investigate the cell-killing mechanism of TPE-TThPy ( Figure 4E). Annexin V-FITC is a Ca 2+ -dependent phospholipid-binding protein labeled with FITC, which can specifically bind to the high nucleophilicity of phosphatidylserine that flips to the outside of the membrane in the early stages of apoptosis, and shows as a strong green fluorescence on the cell membrane. The progression of apoptosis causes the permeability of the cell membrane to increase, allowing for PI to enter the nucleus and thus generate a strong red fluorescence. [53] Even after 5 h of incubation, neither the green fluorescence emission of Annexin V-FITC nor the red fluorescence emission of PI can be observed for the TPE-TThPy-treated SH-SY5Y cells without irradiation ( Figure 4E), indicating that the cell apoptosis was not induced  Figure S26, Supporting Information, also revealed that the generation of 1 O 2 using the produced TPE-TPys as the photosensitizer resulted in the apoptosis of the SH-SY5Y cells.

Specific imaging and precise therapy of tumor with the TPE-TThPy molecule
After the successful demonstration of the cell system, the applicability of the TPE-TThPy molecular probes in real biomedical systems was further investigated using nude BALB/c mice bearing a subcutaneous SH-SY5Y or subcutaneous HepG2 xenografts, which have higher expression levels of MAO-A and MAO-B, respectively ( Figure 5). To investigate the specific tumor imaging ability of the TPE-TThPy probes, after the injection of the probes into the tumors of the two mouse models, fluorescence images of the red channel (600-650 nm) for TPE-TPys were obtained at various time points by using the Lumina XR ( Figure 5A). A significant fluorescence emission in the tumor area of the mouse model bearing a subcutaneous SH-SY5Y xenograft can be observed only after 10 min of the injection of TPE-TThPy molecular probes and reaches a maximum intensity after 60 min. This likely resulted from the rapid and efficient cellular uptake of TPE-TThPy probes and the high efficiently catalytic capacity of MAO-A for the conversion of TPE-TThPy to TPE-TPys, which is supported by the results shown in Figures 3, 4A,B, and 2H,I, respectively. In contrast, no significant fluorescence emission was detected for the mouse model bearing a subcutaneous HepG2 xenograft throughout the 2 h incubation time due to the absence of MAO-A in the HepG2-tumor leading to no generation of TPE-TPys molecules with red emissions. Together all the above results indicate that the TPE-TThPy molecular probe could be effectively enriched in the tumor sites, but it would only be activated in tumors with high expression levels of MAO-A. TPE-TThPy can then be converted to TPE-TPys with red emission and long retention times, guaranteeing the precise tumor imaging performance and PDT at the cellular level.
Evaluation of PDT efficacy with the TPE-TThPy molecules was also carried out using the above two mouse models (HepG2 and SH-SY5Y). The experimental results illustrated in Figure 5B,C indicate that, for nude BALB/c mice bearing a subcutaneous SH-SY5Y xenograft, treatment with both TPE-TThPy molecules and irradiation showed a strong inhibitory effect on the tumors. For the control groups, however, injection with PBS or with TPE-TThPy molecules but without the light irradiation led to the tumor sizes increasing day by day. Furthermore, for nude BALB/c mice bearing a subcutaneous HepG2 xenograft, a significant inhibitory effect on the tumors was not observed after the group was treated with TPE-TThPy molecules with or without light irradiation. The body weights of all the mice were not affected throughout the treatment ( Figure S27, Supporting Information), and H&E-stained images of other tissues (heart, liver, spleen, lungs, and kidneys) for all groups indicated that no significant physiological or pathological tissue damage or inflammatory lesions were observed in the vital organs ( Figure 5D; Figure S28, Supporting Information). Collectively, our results demonstrate the ability of the TPE-TThPy molecular probe to efficiently retain and precisely treat tumor tissues with higher expression levels of MAO-A in vivo through the PDT mechanism with negligible adverse effects due to the good biocompatibility and the excellent MAO-A-specificity of the TPE-TThPy molecules.

CONCLUSIONS
In summary, we designed and synthesized a high-efficient MAO-A-activatable AIE photosensitizer under the help of theoretical calculations and successfully applied it to selectively kill tumor cells or for the cellular therapy of tumors with a higher expression of MAO-A. This rationally designed TPE-TThPy molecule can be converted to TPE-TPys with perfect D-π-A structure and highly positive charges, showing various excellent properties such as highefficient photosensitivity but good biocompatibility, excellent MAO-A specificity, and mitochondrial targeting property. The successful in vitro and in vivo specific imaging and efficiently PDT applications indicate the potential of our rationally designed and synthesized target-activatable AIE photosensitizer of TPE-TThPy for the precise theragnostic of tumors with higher MAO-A expression and also provide a new strategy for the development of novel photosensitizers with excellent photosensitivity and highly specific targeting ability. In addition, it is worth to note that as an organic small molecule probe, TPE-TThPy still has the problem of poor tumor enrichment ability, and cannot be enriched to ideal concentration in tumors through caudal vein injection, that is another aspect we need to continue to improve on.

A C K N O W L E D G E M E N T S
This work was financially supported by NSFC (21775035) and Natural Science Foundation of Hunan Province (2020SK2096, 2021JJ31137).

C O N F L I C T O F I N T E R E S T
The authors declare no conflict of interest.

E T H I C S S TAT E M E N T
The animal experiment part of the work was approved by the Ethics Committee of the College of Biology of Hunan University (Certificate number//Ethics approval No. is SYXK2018-0006).

D ATA AVA I L A B I L I T Y S TAT E M E N T
All the data supporting the findings of this study are available within the article and its Supporting Information files and from the corresponding author on request.