Self-Chemiluminescence-Triggered Ir(III) Complex Photosensitizer for Photodynamic Therapy against Hypoxic Tumor

The limited optical penetration depth and hypoxic tumor microenvironment (TME) are key factors that hinder the practical applications of conventional photodynamic therapy (PDT). To fundamentally address these issues, self-luminescent photosensitizers (PSs) can achieve efficient PDT. Herein, a self-chemiluminescence (CL)-triggered Ir complex PS, namely, IrL2, with low-O2-dependence type I photochemical processes is reported for efficient PDT. The rational design achieves efficient chemiluminescence resonance energy transfer (CRET) from covalently bonded luminol units to the Ir complex in IrL2 under the catalysis of H2O2 and hemoglobin (Hb) to generate O2•– and 1O2. Liposome IrL2H nanoparticles (NPs) are constructed by loading IrL2 and Hb. The intracellular H2O2 and loaded Hb catalyze the luminol part of IrL2H, and the Ir2 part is then excited to produce types I and II reactive oxygen species (ROS) through CRET, inducing cell death, even under hypoxic conditions, and promoting cell apoptosis. IrL2H is used for tumor imaging and inhibits tumor growth in 4T1-bearing mouse models through intratumoral injection without external light sources. This work provides new designs for transition metal complex PSs that conquer the limitations of external light sources and the hypoxic TME in PDT.


■ INTRODUCTION
−5 In a typical PDT process, a specific wavelength of external light from a laser or a light emitting diode (LED) activates a photosensitizer (PS) from its singlet ground state (S 0 ) to the first excited singlet state (S 1 ), which is converted to the excited triplet state (T 1 ) through intersystem crossing (ISC).−12 Therefore, PDT is not usually a firstline therapeutic method, and there is an urgent need to develop new PSs that do not require external light excitation for PDT, especially in deep tissues.
Chemiluminescence (CL) is a process whereby a molecule is activated by chemical oxidation and then transfers its energy to an adjacent fluorophore, the excited state of which returns to its ground state, accompanied by luminescence.−24 Luminol, as a classical chemiluminescent reagent, is converted in a reaction sequence that is catalyzed by H 2 O 2 and metal ions to yield excited-state aminophthalate, which emits blue light. 13,25One of the features of the tumor microenvironment (TME) is overexpression of H 2 O 2 , which is obtained from superoxide ions generated by mitochondria in a process that is catalyzed by the enzyme superoxide dismutase. 26Therefore, H 2 O 2 can be an ideal trigger to serve as a catalyst for luminol CL, so that the CL can be generated in situ within the diseased tissue without the need for an external light source to drive the PDT. 26In addition, the ROS production of CL-triggered PDT is dependent on the CRET efficiency, which complies with the Forster resonance theory, that is, the efficiency of energy transfer is inversely proportional to the sixth power of the donor−acceptor distance. 27,28−31 However, there are only a few reports of covalently linking luminol with PSs, namely, chlorin (Ce6) 32−34 and borondipyrromethene (BODIPY) 9,35 for CRET-mediated PDT without needing an external light source.These PSs are confined to the traditional type II PDT process, which has high O 2 dependence, and their use is restricted by the hypoxic TME, which arises due to rapid tumor growth and insufficient oxygen supply. 36,37lternative strategies for PDT have been explored including O 2 delivery and blood circulation enhancement to address the limitation of hypoxia; however, their efficacy is not ideal due to fast O 2 efflux and complicated operation. 38,39The type I process has low O 2 dependence, which is superior for PDT with the PSs reacting with intracellular substrates to generate cytotoxic ROS, such as O 2 •− and • OH. 40,41 CL-triggered PSs with type I photochemical properties urgently need further research.Metal-complex PSs with versatile electronic excited state properties are known to achieve type I photochemical processes via their promoted electron transfer pathways and high rates of ISC. 42,43For examples, coumarin derivatives have been covalently linked to the ligands of cyclometalated Ru(II) 44 and Ir(III) complexes. 45However, these conjugates required external light irradiation to effect type I PDT.−50 Most importantly for the present work, Ir complexes can absorb in the blue region, 43,51,52 which overlaps well with the fluorescent donor luminol.We, therefore, postulated that Ir complexes could produce ROS through a luminol-mediated CRET process.Applying Ir complex PSs with self-CL properties for PDT is novel and, to our knowledge, has not been reported until now.
Herein, by rational choice of ligands, the blue-lightabsorbing Ir complex, Ir2, with carboxyl groups on the N^N auxiliary ligand, was combined with luminol through amide condensation reactions to obtain self-CL triggered PSs, IrL2 (Scheme 1).IrL2 achieves CL under H 2 O 2 and hemoglobin (Hb) catalysis, producing O   Photophysical Properties and ROS Generation Ability of Ir1 and Ir2.The photostability of Ir1 and Ir2 was tested to avoid the possible impact of photobleaching during the PDT process. 54,55The CL of luminol is mainly in the blue region, about 425 nm.Therefore, a 425 nm LED was used as the light source to monitor the photostability of Ir1 and Ir2 during 15 min of irradiation.The results showed no significant change in the absorption spectra of Ir1 and Ir2 (Figure S12), indicating their excellent photostability.
To determine the potential of the Ir complexes to connect with luminol to construct PSs, their photophysical properties and ROS generation abilities were explored.As shown in Figures 1a,b, and Table S1, the absorption of Ir1 and Ir2 is mainly in the ultraviolet and blue-violet regions, and they emit red light with λ max of 615 and 632 nm, respectively, representing a large Stokes shift, which is beneficial for reducing the signal-to-noise ratio in practical therapeutic applications.The absorption of Ir1 has a limited overlap with the emission of luminol, which suggests only limited energy transfer efficiency in the CRET process.On the contrary, Ir2 shows a significant red shift in absorption, which overlaps well with the emission of luminol at 400−500 nm (Figure 1a,b), indicating its great potential for combining with luminol to construct an efficient CL material.
Density functional theory (DFT) calculations showed that the highest occupied molecular orbital (HOMO) of Ir1 and Ir2 is distributed in the cyclometalated C^N ligands and the metal Ir center.In contrast, the lowest unoccupied molecular orbital (LUMO) is distributed on the auxiliary N^N ligand (Figure 1c), indicating that these two complexes have excellent intramolecular charge transfer (ICT) characteristics, which can enhance the charge transfer state energy and further promote ROS production. 56Compared to Ir1, the π-extended C^N ligands of Ir2 increase the electron conjugation within the complex, reduce the HOMO−LUMO gap, increase the electronic transition ability, and cause a red shift in the absorption spectra (Figure S15), which is consistent with the experimental results.
To further verify the ROS generation from Ir1 and Ir2 under 425 nm irradiation, 2,7-dichlorofluorescencein (DCFH) was used as an indicator.As shown in Figures 1d and S16, both Ir1 and Ir2 generate ROS within 30 s of irradiation, and Ir2 has an excellent ability to produce ROS.Dihydrorhodamine 123 (DHR-123) was used as an indicator to detect the type I ROS produced by Ir1 and Ir2 (Figures 1e and S17).Both complexes produce type I ROS upon 425 nm irradiation, with Ir2 showing a superior ability.Similar results were observed from the electron paramagnetic resonance (EPR) detection using 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) as an indicator (Figure S19).After the irradiation of Ir1 and Ir2, DMPO was added and the DMPO−OOH • adduct was registered, giving a spectrum consisting of a quadruplet, which indicates the generation of O 2 •− in the solution. 57,58ndocyanine green (ICG) was used to assess type II ROS production (Figures 1f and S20−S23).Compared with the control group, Ir2 reduced the absorption of ICG at 790 nm to below 40% within 30 s under 425 nm irradiation.The quantum efficiency of singlet oxygen generation was calculated to be 0.69 for Ir1 and 0.79 for Ir2, demonstrating the excellent capacity of Ir2 to produce 1 O 2 .Calculations analyzed the effect of the functional groups of the two Ir complexes on their excited states.At the level of TD-B3LYP/6-31G(d) SMD (solvent = H 2 O), the singlet−triplet energy difference of the two complexes is relatively small (Figure S24), and the degeneracy of the Ir2 excited states significantly increases at lower energy levels, which enhances its ISC channel and further optimizes its ISC ability, promoting the production of triplet excitons and thus improving its ROS generation ability.Therefore, Ir1 and Ir2 are type I and II PSs, respectively; especially, Ir2 has enhanced ROS production ability under 425 nm irradiation compared to Ir1, which provides experimental support for the application of Ir2 in PDT.
Chemiluminescence Behavior of IrL1 and IrL2.Given the ROS production ability of Ir1 and Ir2 and the overlap in their absorption spectra with luminol emission, chemilumines-cent molecules IrL1 and IrL2 were synthesized.Their photophysical properties are shown in Figure 1a and Table S1.The absorption spectra of IrL1 and IrL2 contain the expected features of both luminol and Ir complexes.Luminol is mainly converted into excited 3-aminophthalic acid in the presence of H 2 O 2 catalyzed by reagents such as hemoglobin (containing Fe 2+ ) or catalase enzyme. 59,60The excited state relaxes to the ground state with the emission of light.Therefore, H 2 O 2 and Hb were used to activate the CL of IrL1 and IrL2.As shown in the CL spectra in Figure 1a, the blue emission peak of IrL1 is from luminol, while the red emission peak is from the luminescence of luminol transferred to Ir1 through CRET.The emission of IrL2 at 635 nm is mainly from the Ir2 unit.This derives from the excellent overlap between the absorption of Ir2 and the emission of luminol, resulting in a more efficient CRET process (compared to IrL1), and the emission in the blue region is negligible.The design strategy has successfully achieved efficient CRET for realizing self-CL.
Subsequently, more detailed studies were conducted on the CL behaviors of IrL1 and IrL2.The effects of different materials, reaction time, pH, Hb concentration, H 2 O 2 concentration, and IrL2 concentration were investigated in black 96-well plates.As shown in Figure S25, the CL of IrL2 is significantly brighter than that of IrL1 under the same catalytic conditions.IrL2 CL is time-dependent, and its emission intensity decreases over time (Figure 2a).This result is similar to previous reports on luminol. 33,34The effect of pH on the CL of IrL2 was investigated by adding the same concentration of IrL2, H 2 O 2 , and Hb to phosphate-buffered saline (PBS) solutions with different pH values (Figure 2b), indicating that the CL of IrL2 almost disappeared at pH 5.0 and gradually enhanced with increasing pH.IrL2 emitted CL at pH = 6.5 (about the typical pH in the TME), 61,62 which favors its application in cancer cells.The effect of catalysts on IrL2 CL was explored by changing the Hb concentration; with increasing Hb concentration, the CL gradually increased, which means that the Hb concentration has a significant impact on the CL of IrL2 (Figure 2c).Hence, in the subsequent preparation of NPs, Hb was added as the catalyst in the system to convert luminol to excited-state aminophthalate in situ.In addition, Figure 2d indicates that the CL of IrL2 gradually enhances with the increase in the H 2 O 2 concentration.Usually, there is about 100 μM H 2 O 2 in cancer cells, 26,63 and according to our experimental results, this concentration is sufficient to activate the CL of IrL2 for application in cell experiments.The effect of the IrL2 concentration on its CL is not very significant within the range tested (Figure 2e).Therefore, cell experiments can be conducted within this concentration range.Moreover, the comparison of CL ability between covalently bonding the compounds and simply mixing the two components is shown in Figure S26.It is obvious that the covalently linked IrL2 exhibits stronger luminescence than simple (noncovalent) mixtures of Ir2 and luminol, indicating the higher CRET efficiency of the covalently bonded compounds  electron delocalization compared to the pyridine ring in IrL1.The band gap of IrL2 decreased, its absorption spectrum redshifted (Figure S31), and its exciton transition ability significantly increased.Meanwhile, combined with the calculated singlet and triplet excitation energies of the Ir complexes at the TD-B3LYP/6-31G (d) SMD (solvent = H 2 O) level (Figure 2i), it was found that the singlet triplet energy level degeneracy of IrL2 was significantly enhanced, especially in the lower triplet states, where the increase in degenerate orbitals contributes to the ISC ability of S 1 and triplet orbitals.This is more conducive to the generation of ROS in IrL2, consistent with the experimental results.In IrL2, luminol is activated to the excited state with H 2 O 2 and Hb, then, energy is transferred to the Ir2 unit through CRET, and finally, the type I or II ROS are generated through electron transfer or energy transfer of the excited state of the PSs (Scheme 1).The above results demonstrate that IrL2 has properties for type I and II ROS generation without an external light source.
Preparation and Characterization of IrL2H NPs.Based on the above discussion, catalysis by Hb is needed for the outstanding CL-induced ROS production ability of IrL2.To improve the water solubility of IrL2 while incorporating Hb into the NPs to release them into cancer cells together, liposome IrL2H NPs were prepared by combining IrL2 with Hb (see the Experimental Section below; IrL2 in a hydrophobic layer and Hb in a hydrophilic layer).The absorption, emission, and Fourier transform infrared spectra demonstrate the successful preparation of IrL2H (Figures 2j  and S32).The morphology of IrL2H NPs was evaluated by transmission electron microscopy (TEM) to be spherical in shape (Figure 2k).The hydrodynamic size and polydispersity index (PDI) of IrL2H NPs were measured by dynamic light scattering (DLS), as shown in Figures 2k and 2l.The size of IrL2H was 153.8 nm, and there was no significant change in size within 5 days, implying that its stability in water basically satisfied the prerequisite for in vitro and in vivo applications.
Intracellular CL and Uptake of IrL2H.The in vitro application of IrL2H was investigated with breast cancer 4T1 mice cells as a standard and readily available model cell line.The CL of IrL2H within cells was studied by adding different concentrations of IrL2H into 96-well plates with incubating 4T1 cells (Figure S34).At a concentration of 30 μg/mL, weak CL was observed, while at 60 μg/mL, the CL of IrL2H was more pronounced, indicating its good CL ability.The cellular uptake of IrL2H was tested by confocal laser scanning microscopy (CLSM) (Figure S35).After coincubation with 4T1 cells for 10 min, red luminescence of IrL2H was observed, indicating excellent cellular uptake of IrL2H.Subsequently, as the incubation time increased, red luminescence could be observed within 3−12 h, and the intensity was not weakened, indicating that the material could accumulate in cells.These results are prerequisite for its application in subsequent experiments.
In Vitro CL-Excited PDT.The viability of 4T1 cells was evaluated by using an MTT assay under various treatments in vitro.Figure 3a shows that the viability of the 4T1 cells is <20% in 60 μM IrL2H, while >80% of the 4T1 cells were viable in IrL2 and >60% of 4T1 cells are viable in IrLH NPs.These results clearly demonstrate the superior PDT effect of IrL2H.To assess the effect of IrL2H on normal tissue cells, the viability of mouse fibroblasts L929 cells was studied by an MTT assay at different concentrations of IrL2H (Figure 3b).
The results indicate that the toxicity of IrL2H to L929 cells is negligible, demonstrating its good biocompatibility.Furthermore, to assess the effect of different components in IrL2H on 4T1 cells, Hb and luminol + Hb were tested; almost no toxicity was observed (Figure S37).The cytotoxicity of Ir2 under dark and light conditions showed that Ir2 had almost no cytotoxicity under dark conditions, whereas a high level of phototoxicity was observed under 425 nm light, indicating that the cytotoxicity of IrL2H most likely originates from the excited Ir2 unit (Figure 3c).The above results illustrate that IrL2H exhibits excellent cytotoxicity to 4T1 cells at 60 μM, which is attributed to the PDT effect of the Ir2 unit irradiated by CRET, and IrL2H has almost no toxicity to normal cells, demonstrating its good biosafety.
The cytotoxicity of IrL2H in hypoxic cancer cells was assessed.As shown in Figure 3d, the viability of 4T1 cells decreases to <50% even in 1% O 2 conditions with 60 μM IrL2H treatment.The IC 50 values of IrL2H in 4T1 cells under normoxic and hypoxic conditions were calculated to be 36.89and 80.91 μM, respectively.Live/dead cell staining (Figure 3e) shows that only red fluorescence could be observed in the normoxic environment compared with the control group, indicating the death of 4T1 cells after being treated with IrL2H.Both green and red fluorescence were observed after 4T1 cells were treated with IrL2H, suggesting that IrL2H can still partially kill cells under hypoxic conditions, which is consistent with the results of the MTT assay.Therefore, IrL2H is phototoxic through CL excitation in the hypoxic TME.
To better understand the PDT mechanism, the ROS production of IrL2H under different conditions was tested by using DCFH-DA as an indicator after different incubation times.As shown in Figure 4a, the intracellular green fluorescence increased within 1−3 h of incubation with IrL2H and then weakened by 6 h, indicating an initial increase in the ROS level and then a decrease, maybe due to the decay of the CL. Figure 4b shows that the green fluorescence increased as the IrL2H concentration increased, implying that the intracellular ROS level increased with the enhancement of the IrL2H concentration.The type of ROS generated by IrL2H in normoxic and hypoxic 4T1 cells was investigated by using DHR-123 and Singlet Oxygen Sensor Green (SOSG) as indicators for type I and II ROS, respectively.As shown in Figure 4c, the green fluorescence of DHR-123 and SOSG was observed under normal oxygen conditions, demonstrating the good ability of IrL2H to produce type I and type II ROS.However, under hypoxic conditions, only the fluorescence of DHR-123 was observed, indicating that IrL2H can generate type I ROS under hypoxic conditions but not type II ROS due to insufficient O 2 to provide energy transfer substrates.In summary, the intracellular ROS production of IrL2H increases with increasing IrL2H concentration, and IrL2H could still perform PDT in hypoxic cells through a type I process.
Apoptosis is the most common PDT cell death pathway. 68itochondrial membrane potential (MMP) is a parameter that reflects the level of cellular health. 68The loss of MMP may stimulate apoptosis.A JC-1 dye assay was applied in 4T1 cells to evaluate the damage of the MMP (Figure 4d).Strong red fluorescence from aggregated-JC-1 was observed in the control group, while the group treated with IrL2H showed a green fluorescence from free-JC-1, indicating that IrL2H induced a loss of MMP.Annexin V-FITC/PI dual fluorescence staining experiments were performed using CLSM to detect the effect of IrL2H on 4T1 cells.Annexin V-FITC emits green fluorescence, while PI emits red fluorescence.As shown in Figure 4e  undertaken in mice bearing 4T1 tumors.The CL images of IrL2H in the tumors were obtained.As shown in Figure 5b,c, a tumor model was constructed by injecting 4T1 cells into the right thigh of mice, and IrL2H and H 2 O 2 were intratumorally injected to detect their CL image.The CL intensity of IrL2H was bright after injection, then slightly decreased between 10 and 30 min, then weakened more rapidly after 1.5 h, and finally disappeared after 3 h.The results indicate that IrL2H has a promising in vivo CL imaging performance.
We further applied IrL2H to in vivo therapy.As shown in Figure 5a, a mouse model was established by subcutaneous injection of 4T1 cells into the left thigh.Then, tumor-bearing mice were randomly divided into two groups, each consisting of 4 mice: (i) PBS and (ii) IrL2H.Intratumoral injection therapy was performed on days 0, 2, 4, and 6, and then, each group of mice was observed for 14 days.The changes in tumor volume and weight of each mouse were recorded (Figure 5d− f).The relative tumor volume of the PBS group increased about 10 times after 14 days.In contrast, there was no significant change in tumor volume in the IrL2H group.After 14 days of treatment, the mice were sacrificed, and the tumors, as well as the subject organs (heart, liver, spleen, lung, and kidneys), were excised for histological analysis.As shown from hematoxylin and eosin (H&E) staining (Figure 5g), there were many stromal tight cells and intact nuclei and cytoplasm in the control group, indicating the good condition of the cells.The tissue space presented a hiatus, and the tumor cells were destroyed in the IrL2H group, indicating that IrL2H has a good inhibitory effect on tumor growth.Within 14 days, there was no significant change in the weight of the mice, and histological analysis of the main organs after 14 days of treatment showed almost no pathological changes (Figure 5g), indicating that IrL2H imparts almost no damage to normal tissues, demonstrating its good biocompatibility.The above results indicate that IrL2H has an excellent ability to inhibit tumor growth through self-CL-induced PDT, verifying its potential application in cancer treatment.

■ CONCLUSIONS
In conclusion, Ir complex Ir2 was rationally synthesized by extending the π-conjugation of the N ligand to give a good overlap between the absorption spectrum of the complex and the emission of luminol.Luminol was covalently attached to the auxiliary ligand of Ir2 through amide condensation reactions to construct a chemiluminescent molecule IrL2.The experimental results show that the CL of IrL2 increases with an increase of H 2 O 2 and Hb concentrations, and IrL2 still exhibits excellent CL properties at pH = 6.5.IrL2 generates types I and II ROS excited by CL under the catalysis of H 2 O 2 and Hb.Liposome IrL2H NPs, constructed with IrL2 and Hb, exhibited good stability in aqueous solutions.In vitro experiments showed that IrL2H is effectively internalized by cells, producing CL in cells, and is nontoxic to normal cells.The CRET from luminol to the Ir2 unit excited the Ir complex to generate type I and type II ROS, leading to apoptosis of 4T1 cells.In vivo experiments demonstrate that IrL2H images tumors by self-CL and inhibits tumor growth following intratumoral injection.This work provides a versatile precedent for using functionalized transition metal complexes to self-generate CL for PDT, thereby extending the molecular platform for tackling the enduring problems of an external light source and the hypoxic tumor environment in clinical applications.
NMR spectra were recorded at 25 °C on a Varian 500 MHz spectrometer.HPLC used a Shimadzu LC-16A liquid chromatograph.UV−vis absorption spectra were recorded on a Shimadzu UV-3100 spectrophotometer.The photoluminescence spectra, excited state lifetimes (τ), and photoluminescence quantum yields (Φ) were recorded on an Edinburgh FLS920 spectrofluorometer under air at room temperature.TEM images of the samples were taken with a TECNAI F20 microscope.Diameter and diameter distributions of the NPs were determined by a Malvern Zetasizer Nano instrument for DLS.CLSM images were taken using a ZeissLSM 700 instrument (Zurich, Switzerland).The hypoxic conditions were achieved using a biological hypoxic incubator (MIC-101 from Billups-Rothenberg Company).
Synthesis.The synthesis of Ir1 and Ir2 followed standard procedures, as reported in the Supporting Information.The synthetic route to [Ir(C^N) 2 (N^N)] + complexes leads to a racemic mixture of Δ and Λ stereoisomers, which (as in the present case) are generally not separated. 69omplexes IrL1 and IrL2.Ir1 (74.5 mg, 0.1 mmol) [or Ir2 84.5 mg (0.1 mmol)] and luminol (42.5 mg, 0.24 mmol) were dissolved into dry DMF (3 mL) and stirred for 30 min at 0 °C.Then, N,Ndiispropylethylamine (DIPEA) (38.78 mg, 0.3 mmol) was added into the mixture.Subsequently, 1-hydroxybenzotriazole (HOBt) (37 mg, 0.3 mmol) and 1-ethyl-3-(3-(dimethylamino)propyl)carbodiimide (EDCl) (57.5 mg, 0.3 mmol) were added to the mixture and stirred for 1 h in 0 °C.After that, the mixture was stirred under N 2 at room temperature for 3 days.Cold water was added to terminate the reaction, and the product was extracted with dichloromethane; the organic layer was collected, and the solvent was evaporated to obtain the crude product as a red solid.The product was purified by silica gel column chromatography with CH 2 Cl 2 /MeOH (100:1−30:1 v/v) as the eluent.
1 O 2 Quantum Yield Measurements.The 1 O 2 quantum yield of aggregation-induced emission NPs in water (Φ) upon LED irradiation (425 nm, 20 mW cm −2 ) was determined using ICG as an indicator and using Rose Bengal (RB) as the standard; ref 72 The absorbance decrease of ICG at 790 nm was recorded for different durations of light irradiation to obtain the decay rate of the photosensitizing process.The ROS yield was calculated by using the following equation: where K PS and K RB are the decomposition rate constants of the photosensitizing process determined by the plot ln(A 0 /A) versus irradiation time (Figures S21 and S22).A 0 is the initial absorbance of ICG, while A is the absorbance of ICG after different irradiation times.A PS and A RB represent the light absorbed by Ir complex PSs and RB at 425 nm, respectively.Φ RB is the 1 O 2 quantum yield of RB, which is 0.75 in water.Calculation Methods.All calculations were completed in the Gaussian 16 software package.The Hay Wadt effective nuclear potential (ECP) and double ξ optimize the structure of the base set LANL2DZ to obtain a stable structure without imaginary frequencies.FMO orbital information on Ir complex molecules, including energy levels and distributions, was obtained at the same theoretical level.Excited state calculations were conducted in water at the levels of TD-B3LYP/6-31Gd, SMD, solvent = H 2 O, and the results were used to analyze the optoelectronic properties of the material.
Cell Culture Methods.4T1 cells were cultured in this experiment.The culture medium was prepared with RPMI medium 1640 containing 10% (v/v) FBS.The cell culture flask was placed in an incubator under the following conditions: normoxic conditions: cells were maintained in a humidified atmosphere containing 5% CO 2 of 37 °C; hypoxic conditions: cells were maintained at 37 °C in a humidified atmosphere containing 1% O 2 , 5% CO 2 , and 94% N 2 .L929 cells were cultured in this experiment.The culture medium was prepared with Dulbecco's modified Eagle medium (DMEM) containing 10% (v/v) FBS.The cell culture flask was placed in an incubator in an atmosphere containing 5% CO 2 at 37 °C.
Cytotoxicity Test Methods.4T1 cells were seeded in 96-well plates at a density of 10,000 cells per well and cultured in the incubator under different conditions (normoxia or hypoxia) for 24 h.After aspirating the old culture medium, 100 μL of RPMI medium 1640 containing different concentration gradient PSs (0−60 μM) were added to each well, and the cells were placed in the incubator for 24 h.After incubation, the cells were placed in an incubator for 24 h.Then, the viability of the cells was detected by an MTT assay.
Intracellular Uptake.4T1 cells were seeded in confocal culture dishes at a density of 50,000 cells per well.RPMI medium 1640 containing NPs was incubated for different times.After incubation, the cells were fixed and then stained with 4′,6-diamidino-2phenylindole (DAPI).The uptake of IrL2H by cells was observed by CLSM.
Evaluation of Intracellular ROS Production Capacity.4T1 cells were seeded in confocal dishes at a density of 50,000 cells per well.Different formulations (control, IrL2H) were added to the medium, and the cells were incubated for another 24 h.For the ROS level test, DCFH-DA (1 μL) [or SOSG (1 μL) or DHR-123 (3 μL)] was dissolved in blank RPMI medium 1640 without FBS and added to the confocal culture dish.After treatment in the dark for 20 min, the medium containing indicator was aspirated, washed twice with PBS, and 1 mL of PBS was added, followed by CLSM analysis.To observe the green fluorescence intensity in the cells, λ ex = 465−495 nm and λ em = 415−555 nm for indicators.
Live/Dead Staining Test Methods.4T1 cells were seeded in confocal dishes at a density of 50,000 cells per well.Different formulations (control, IrL2H) were added to the medium, and the cells were incubated for another 24 h.Detection buffer containing Calcein-AM and PI was added to the confocal dishes.The fluorescence image was observed under an inverted fluorescence microscope to assess the cell survival state: λ ex = 540−580 nm, λ em = 600−660 nm for PI and λ ex = 465−495 nm, λ em = 415−555 nm for Calcein-AM.
CL Response of IrL2H in 4T1 Cells.4T1 cells were incubated with IrL2H at different concentrations in a 96-well plate.Luminescence images (emission filter, long pass 510 nm; exposure time, 10 min) were acquired with a TUCSEN FL 9BW image system.
Detection of Mitochondrial Membrane Potential.The 4T1 cells were incubated in a glass-bottom dish at a density of 50,000 cells per dish for 24 h, and the,n the cells were subjected to different treatments (Control, IrL2H).After 3 h, the culture medium was replaced.Next, the cells were incubated with JC-1 dye (5 μg/mL) at 37 °C for 20 min.Then, the cells were washed three times with PBS.After that, they were imaged by CLSM.λ ex = 540−580 nm, λ em = 600−660 nm for J-A and λ ex = 465−495 nm, λ em = 415−555 nm for J-M.
Annexin V-FTIC/PI Assays.4T1 cells were seeded in confocal dishes at a density of 50,000 cells per well.Different formulations (control, IrL2H) were added to the medium, and the cells were incubated for another 24 h.Finally, the cells were stained with Annexin-FITC/PI for 15 min and then were imaged by CLSM.λ ex = 540−580 nm, λ em = 600−660 nm for PI and λ ex = 465−495 nm, λ em = 415−555 nm for Annexin V-FTIC.
Animals and Tumor Model.The mice experiments were performed in accordance with animal regulations and management protocols.All animal experiments were approved by the Institutional Animal Care and Use Committee of Northeast Normal University.BALB/c mice (female, 6−8 weeks) were purchased from the Vital River Company in Beijing.4T1 cells (1.0 × 10 6 /tumor, 100 μL in PBS) were subcutaneously injected into the flank of each mouse to establish the 4T1 tumor model.
In Vivo CL Imaging.To study the change of CL of IrL2H in vivo, BALB/c mice anesthetized were intratumorally injected with IrL2H solution (100 μg mL −1 ) and 5 μL of H 2 O 2 (1 mM) into the same field.The CL images were captured post injection for different times.The CL images were taken immediately by a TUCSEN FL 9BW image system (emission filter, long pass 510 nm; exposure time 20 min).Following a literature precedent, intratumoral injection can maximize the accumulation of radiosensitizers within a tumor compared to intravenous injection and further enhance efficacy. 73e also attempted in vivo imaging with intravenous injection, and almost no luminescence was observed under the same detection conditions as those under intratumoral injection.Therefore, intratumoral injection was used to accumulate more NPs to improve the therapeutic effect in our system.
Evaluation of Antitumor Efficacy In Vivo.When the tumor volume grew to about 60 mm 3 , the mice were randomly separated into two groups with 4 mice in each group.Then, the mice were injected with PBS (i) and IrL2H (12 mg/kg) (ii).The tumor sizes and body weights of the mice were monitored every 2 days during the 14 days of treatment.The tumor volume was measured as volume = (L × W 2 /2), where L (length) and W (width) are the two tumor dimensions.On day 14, the mice in each group were euthanized; the tumors and major organs were harvested for subsequent analysis.The tissues were fixed with formalin and then paraffin-embedded.Furthermore, the paraffin-embedded tissues were sectioned and stained with H&E to determine tissue damage after observation under a microscope.

■ ASSOCIATED CONTENT
Scheme 1. Schematic Diagram of the IrL2 Structure and Self-CL Triggered PDT Using IrL2H NPs a

Figure 2 .
Figure 2. (a−e) Average fluorescence intensity of IrL2 with different treatments.Inset image: CL image of IrL2 with different treatments.(f) Change of PL intensity of DCFH with the change of time under different conditions.I 0 = initial intensity of 525 nm.I = real-time intensity of 525 nm with various times.EPR signals of TEMP (for type-II ROS detection) (g) and DMPO (for type-I ROS detection) (h) with different treatments in MeOH solution.(i) Singlet and triplet energy levels of IrL1 and IrL2 and their spin−orbit coupling values with S 1 −T 1 and S 0 −T 1 transitions.(j) Solid line: absorption spectra of IrL2, Hb, and IrL2H; dashed line: PL spectra of IrL2 and IrL2H, λ ex = 425 nm.(k) DLS data for the NPs of IrL2H.Inset image: TEM image of IrL2H.(l) Changes in the size (solid line) and PDI (dashed line) of IrL2H in 5 days.
. The above results indicate that the CL of IrL2 gradually increases with the increase of pH, Hb content, and H 2 O 2 concentration and could be achieved under appropriate TME pH and H 2 O 2 concentrations.These data provided the basis for subsequent cell experiments.CL-Excited ROS Generation Ability of IrL1 and IrL2.To further verify the possibility of CL Ir complexes for PDT, Inorganic Chemistry the ROS production by CL-excited IrL1 and IrL2 was investigated As shown in Figures 2f and S28, DCFH was used to detect the ROS in the solution under different conditions.Compared with that of the control group, the emission of DCFH gradually increased with time after adding H 2 O 2 to the solutions of IrL1 and IrL2.This may be because H 2 O 2 catalyzes the conversion of the luminol part of the PSs to the excited state and then excites the Ir complexes through CRET to produce ROS.More ROS are generated after adding H 2 O 2 to the solution of IrL2, compared to IrL1, which is due to its the superior CRET with IrL2.Therefore, Hb should also enhance ROS production by IrL1 and IrL2.To verify this hypothesis, the ROS production of IrL1 and IrL2 in the presence of H 2 O 2 and Hb was detected simultaneously.A blank solution with H 2 O 2 and Hb was monitored, and the emission spectra were obtained of DCFH in IrL1 and IrL2 solutions in the presence of the same concentrations of H 2 O 2 and Hb, respectively.Compared with the blank solution (green line in Figure 2f), the emission of DCFH in IrL1 and IrL2 solutions was significantly enhanced, and the curve flattened over time.The data show that especially IrL2 has an excellent ability to produce ROS under catalysis by H 2 O 2 and Hb, again revealing its superiority over IrL1 for self-CL-triggered PSs.2,2,6,6-Tetramethylpiperidine (TEMP) and DMPO were used as indicators to detect the type II and type I ROS production, respectively, by EPR spectroscopy (Figures 2g,h, and S29).Compared with those of the blank control group, the spin signals of the TEMPO • adduct in IrL1 and IrL2 solutions were significantly enhanced to give a spectrum consisting of a triplet, with 1:1:1 relative intensities after the addition of H 2 O 2 and Hb (Figure 2g) which indicate the generation of 1 O 2 . 64,65Significant DMPO−OOH • adduct signals of a quadruplet with 1:1:1:1 relative intensities were also observed for IrL1 and IrL2 after mixing with H 2 O 2 and Hb (Figure 2h), consistent with the production of O 2 •− . 66,67The signals of TEMPO • and DMPO−OOH • for the IrL2 solution were more obvious than for IrL1, as shown in Figure 2g,h, confirming the excellent CLinduced type I and II ROS production ability of IrL2.These results agree with the ROS generation abilities of Ir1 and Ir2 under 425 nm irradiation.The calculated energy level distributions of IrL1 and IrL2 are shown in Figure S30.The isoquinoline group in the C^N ligands in IrL2 increased the

Figure 4 .
Figure 4. Confocal fluorescence images for the detection of ROS after treatment with (a) IrL2H for different times and (b) different concentrations of IrL2H.Scale bar = 10 μm.(c) Confocal fluorescence images for detecting the different types of ROS after treatment with IrL2H.Scale bar = 50 μm.(d) Confocal fluorescence imaging of MMP in 4T1 cells incubated with IrL2H via a subsequent JC-1 dye assay.Scale bar = 50 μm.(e) Death of 4T1 cells induced by IrL2H and staining with dual fluorescence of Annexin V-FITC/PI.Scale bar = 10 μm.
, cells treated with IrL2H show strong green and red fluorescence, indicating the occurrence of apoptosis.The above in vitro results, with excellent cytotoxicity of IrL2H and its CL-triggered PDT-induced cell apoptosis, laid the foundation for in vivo experiments.In Vivo CL Imaging and PDT.Inspired by the excellent in vitro anticancer efficiency of IrL2H, in vivo experiments were

Figure 5 .
Figure 5. IrL2H-mediated inhibition of tumor growth in the 4T1 tumor model.(a) Timeline of the treatment.(b) Imaging of IrL2H in tumors; and (c) curve of the average fluorescence intensity at the tumor site over time.(d) Relative tumor volume in different groups.(e) Photos of tumors after treatment.(f) Changes in the body weights of mice.(g) H&E staining images of tumor tissues from the different groups on day 14; scale bar: 100 μm.