Enhancing Intersystem Crossing by Intermolecular Dimer-Stacking of Cyanine as Photosensitizer for Cancer Therapy

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE7 Nov 2022Enhancing Intersystem Crossing by Intermolecular Dimer-Stacking of Cyanine as Photosensitizer for Cancer Therapy Haiqiao Huang, Dandan Ma, Qiang Liu, Daipeng Huang, Xueze Zhao, Qichao Yao, Tao Xiong, Saran Long, Jianjun Du, Jiangli Fan and Xiaojun Peng Haiqiao Huang State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Google Scholar More articles by this author , Dandan Ma State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Google Scholar More articles by this author , Qiang Liu State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Google Scholar More articles by this author , Daipeng Huang State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Google Scholar More articles by this author , Xueze Zhao State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Google Scholar More articles by this author , Qichao Yao State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Google Scholar More articles by this author , Tao Xiong State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Google Scholar More articles by this author , Saran Long *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Ningbo Institute of Dalian University of Technology, Dalian University of Technology, Ningbo 315016 Google Scholar More articles by this author , Jianjun Du State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Ningbo Institute of Dalian University of Technology, Dalian University of Technology, Ningbo 315016 Google Scholar More articles by this author , Jiangli Fan State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Ningbo Institute of Dalian University of Technology, Dalian University of Technology, Ningbo 315016 Google Scholar More articles by this author and Xiaojun Peng *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024 Ningbo Institute of Dalian University of Technology, Dalian University of Technology, Ningbo 315016 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202101479 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Development of new photosensitizers (PSs) with high singlet oxygen quantum yield and minimal side effects is of great interest in photodynamic therapy (PDT). Herein, a facile strategy to significantly improve photosensitization has been demonstrated for the first time with two pentamethine dyes connected by a varying alkyl chain resulting in a series of cyanine dimers. The photophysical properties of the dimers were studied with steady-state optical spectroscopies, a time-correlated single photon counting technique, and laser flash photolysis spectrometry. X-ray crystallography confirmed that the molecular packing modes of Cy-Bu-D and Cy-He-D were dominated by H-aggregation. The H-aggregation by dimerization suppresses the radiative singlet decay, which helps to stabilize the highly efficient triplet excitation state. Moreover, the dimers show more intense wavelength absorption in the near-infrared (ε 1.5–2.0 times more than monomer Cy-H at 650 nm), better singlet oxygen quantum yield, and a longer triplet-state lifetime than monomer Cy-H, providing excellent performance as a triplet PS. In vivo experiments demonstrated that Cy-He-D successfully suppresses tumor growth after PDT treatment. This work is beneficial to the design of novel heavy atom free PSs for PDT-based theranostic systems. Download figure Download PowerPoint Introduction With the increase of the mortality rate caused by cancer, considerable effort has been devoted to seeking effective treatment.1–6 Photodynamic therapy (PDT) is emerging as an effective modality with precise spatiotemporal control for cancer treatment due to its minimal invasion character, negligible side effects, inappreciable drug resistance, and low systemic toxicity.7–12 PDT relies on photosensitizers (PSs) and light to generate cytotoxic reactive oxygen species (ROS), particularly singlet oxygen (1O2), to destroy cancer cells.13–16 PSs with high 1O2 quantum yield (ΦΔ) induce cancer cell apoptosis or death most effectively, and deep light penetration of near-infrared (NIR, 650–800 nm) irradiation is beneficial for cancer ablation.17,18 Hence, the design of PSs that have NIR excitation wavelengths, high molar extinction coefficients, long triplet state lifetime, and a high ΦΔ, is essential for highly efficient PDT.19,20 To satisfy the above criteria, many recent candidates for PSs use heavy atoms to improve the spin–orbital coupling thus elevating the intersystem crossing (ISC) rate and improving the ΦΔ of PSs, which is called the “heavy atom effect.” Nevertheless, heavy atoms incorporated into the structure often lead to increased “dark toxicity.”21–23 To overcome this challenge, the rational design of heavy-atom-free PSs with high efficiency and enhanced production of ROS is significant.11,24–26 Cyanine derivatives are recognized as potential candidates for bioimaging and PDT therapeutic agents because of their excellent optimal biosafety and high molar extinction coefficient in NIR.27 However, they usually suffer from short triplet state lifetime and low ΦΔ, which are key factors to determine the performance of PSs. Herein, a series of cyanine dimers with high NIR absorption ability, long-triplet lifetime, and high ΦΔ, were built based on pentamethine dyes (Cy-H monomer) intraconnected by various alkyl chains (Figure 1). For the dimers, the exciton model of molecular aggregates was assumed to cause the split of the excited S1-state level into two levels including a higher energy (S1H) and a lower energy (S1L, Figure 1) level, leading to highly efficient triplet excited state.27,28 The ΦΔ of the dimer was far greater than 20 times that of the monomer, and the triplet lifetime of Cy-He-D reached 37.5 μs. In addition, Cy-He-D demonstrated an excellent photodamage effect on cancer cells and exhibited low dark cytotoxicity. This study demonstrates that Cy-He-D can be a powerful candidate for PDT and provides a protocol to develop highly efficient PSs. Figure 1 | Illustration of energy levels of the monomer and dimer. Chemical structures of Cy-H and Cy-X-Ds dimers. Download figure Download PowerPoint Experimental Methods The general chemicals used in the report were purchased from Energy Chemical Co. (Shanghai, China) and J&K Scientific Ltd. (Beijing, China), and all the solvents were of analytic grade. 1,3-Diphenylisobenzofuran (DPBF) were obtained from Energy Chemical Co. ROS assay kit [2,7-dichlorodihydrofluorescein diacetate (DCFH-DA)], mitochondrial membrane potential assay kit (JC-1), and AnnexinV-FITC (fluorescein isothiocyanate, AV-FITC)/propidium iodide (PI) apoptosis detection kit were purchased from KeyGEN BioTECH (Nanjing, China). Hoechst 33324 and MitoTracker Green FM were purchased from Thermo Fisher Scientific Inc. (Shanghai, China). 1H NMR and 13C NMR spectra were detected by a Varian DLG400 spectrometer (Varian, USA) and Bruker Avance III 500 spectrometer (Bruker, Germany). Mass spectrometry (MS) was carried out using time-of-flight liquid chromatography (TOF LC)/MS instruments (Agilent, USA). Absorption spectra were measured on a CARY 60 UV-Vis spectrophotometer (Agilent, USA). Fluorescence spectra were obtained with a Agilent Cary Eclipse fluorescence spectrophotometer (Agilent, USA). Absolute photoluminescence quantum yields (Φf) was obtained from the Quantaurus-QY Absolute PL quantum yields spectrometer (Hamamatsu, Japan). The triplet excited-state lifetime of dyes were obtained from a Flash Photolysis spectrophotometer (Edinburgh, UK). Confocal laser scanning microscope (CLSM) images were performed on an Olympus FV3000 CLSM (Olympus, Japan). Small animals’ fluorescence imaging was carried out by the NightOWL II LB983 living imaging system (Berthold Technologies, Germany). Other experimental details for compound synthesis, characterization, crystal diffraction data, photophysical properties, and the cell imaging and anticancer study are available in the Supporting Information. All animal studies involved in this work were performed in compliance with the Guidelines for the Care and Use of Laboratory Animals, published by the National Institutes of Health. The animal protocol was approved by the local research ethics review board of the Animal Ethics Committee of Dalian University of Technology (Certificate number//Ethics approval no. is 2020-043). Results and Discussion Molecular design, synthesis, and basic photophysical properties General synthetic routes of the Cy-H monomer and Cy-X-Ds dimers are shown in Supporting Information Scheme S1. Cy-X-Ds with double Cy-H salts were designed and synthesized by varying the alkyl chain length (X) and were characterized by 1H NMR, 13C NMR, and high-resolution mass spectrum ( Supporting Information Figures S1–S12). The steady-state absorption spectra for the Cy-X-Ds homodimers and the reference monomeric dye Cy-H in different solution are shown in Supporting Information Figure S13. The photophysical properties of these novel dimeric dyes in ethanol (EtOH) and water were characterized (Figures 2a–2d and Table 1). The spectral data obtained for Cy-H, Cy-Bu-D, Cy-He-D, Cy-Oct-D, and Cy-Dec-D at 1 μM were normalized for direct comparison. The homodimer dyes showed obvious enhanced absorption spectra of short-wavelength bands (H-aggregates). The intensity of the short-wavelength bands (shoulder peak, 590–600 nm) regularly decreases with increasing chain length (increasing n) of the polymethylene bridge, which indicates the degree of the chromophores’ interaction of dimer cyanines Cy-X-Ds correspondingly decreases (in EtOH). The long-wavelength bands of absorption of homodimer dyes (main peak, 642–655 nm) had a bathochromic-shift compared with Cy-H (640 nm). Figure 2 | Normalized absorption (Abs) and fluorescence (Fl) emission spectra of Cy-H and Cy-X-Ds in ethanol (a and b) and in water (c and d). Download figure Download PowerPoint Table 1 | Photophysical Parameters of the Cy-H and Cy-X-Ds Dimers Dyes Solvent λabsa λemb Φfc εd BTe Cy-Bu-D EtOH 655 672 0.010 3.31 0.033 Water 651 666 0.001 1.65 0.002 Cy-He-D EtOH 646 670 0.060 2.77 0.166 Water 639 661 0.014 1.51 0.021 Cy-Oct-D EtOH 644 670 0.106 4.14 0.439 Water 636 660 0.010 1.65 0.017 Cy-Dec-D EtOH 642 670 0.140 4.02 0.563 Water 632 661 0.008 1.23 0.010 Cy-H EtOH 640 660 0.150 1.94 0.291 Water 637 657 0.096 1.69 0.162 The entries were measured at 25 °C; the dye concentration was 1 μM. aThe max absorption peaks of dyes (nm). bThe max fluorescence peaks of dyes (nm). cThe fluorescence Φf. dMolar extinction coefficient (105 mol−1cm−1L). eBrightness is defined as ε × Φf (105 mol−1cm−1L). Attractively, Cy-He-D showed different characteristics in the intensity of the longer-wavelengths bands in water compared with the other three dimers. The larger extinction coefficient of Cy-He-D in long-wavelength bands (639 nm, in water) indicates that Cy-He-D has a greater probability of long-wavelength electronic transition and high solubility in water. Cy-Oct-D and Cy-Dec-D exhibited longer wavelength bands of absorption than Cy-H, and the extinction coefficients of the two at the maximum absorption wavelength was approximately twice as large as that for Cy-H in dichloromethane (DCM) or EtOH. These results suggest that the two remote moieties of Cy-Oct-D and Cy-Dec-D do not interact in organic solvent. Thus, we speculate that Cy-Oct-D and Cy-Dec-D in organic solvent completely opens from its dimeric form, resulting in the aforementioned spectral behavior. In the fluorescence spectra of Cy-X-Ds dimers, only one band, independent of the excitation in the short- or long-wavelength band of absorption, was observed. The fluorescence maximum lies in the long wavelength in comparison with the monomer dye Cy-H. This data demonstrates that molecular fluorescence is generated by the S1L state returning to the ground state. Moreover, all dimer cyanines have a Φf less than the monomer Cy-H. A regular increase of Φf accompanied the abating of the chromophores’ interactions in the Cy-Bu-D, Cy-He-D, Cy-Oct-D, and Cy-Dec-D sequence (Table 1). In terms of fluorescence brightness (BT = ε × Φf) of Cy-X-Ds, Cy-Bu-D was the least bright, Cy-He-D was approximately half the brightness of Cy-H, and Cy-Oct-D and Cy-Dec-D were approximate to Cy-H. Finally, the fluorescence emission of these dimers in the far-red region (630–750 nm), remained compatible with the classical microscopy setups for cellular imaging. We also determined the lipophilicity of all PSs using a PION μ-Diss Profiler (Pion, USA). This were done by measuring PS partitioning between n-octanol and phosphate buffer (pH 7.4) and calculating the resulting log Poct values, which were 1.46, 2.75, 2.23, 2.56, and 2.63 for Cy-H, Cy-Bu-D, Cy-He-D, Cy-Oct-D, and Cy-Dec-D, respectively. The logP of dimers were moderately higher than that of monomer Cy-H. Considering that the cell membranes are composed of lipid bilayers, it has been predicted that the lipophilic sensitizers should be easily taken up into cells. Structural characterization by X-ray crystallography Further evidence for the formation of dimeric aggregates was provided by solid-state X-ray crystallography. Single crystals of Cy-Bu-D and Cy-He-D were readily obtained by vapor diffusion from a hexane/methanol solution at room temperature. Unfortunately, Cy-Oct-D and Cy-Dec-D failed to grow single crystals, which may be due to their longer connecting alkyl chain. Furthermore, the angle between transition dipoles and the interconnected axis (θ) is larger than the critical value of 54.7° ( Supporting Information Figure S14), manifesting the existence of H-aggregates.28,30 As shown in Figures 3a–3c, Cy-Bu-D shows both intra- and inter- molecular monomers parallel. Along the intrastack direction, the intermolecular monomer of Cy-Bu-D in the next neighboring molecules forms a slip-stacked molecular arrangement (θ = 79.03°) and tightly packed chromophores with a π–π distance only 6.85 Å (Figure 3d). We also calculated energies of intramolecular and other intermolecular exciton splitting energies (Δε) for general comparison ( Supporting Information Figure S16a and Table S1). According to the calculation results, the intermolecular arrangement of Cy-Bu-D exhibited a larger positive Δε, which was dominant among four selected molecular pairs due to its shorter intermolecular distance (d), suggesting powerful H-type coupling in Cy-Bu-D. In addition, due to its relatively long flexible connecting arm, Cy-He-D exhibits a more complex single-crystal structure. Analysis of three adjacent Cy-He-D dimers in its unit cell, as illustrated in Figure 3e and Supporting Information Figure S15, showed all the angles between transition dipoles and the interconnected axis (θ) were larger than the critical value of 54.7°. Meanwhile, Δε was dominantly higher than zero, as calculated from the angles between the transition moments in the three intermolecular chromophores ( Supporting Information Figure S16b, Table S2). These results were also consistent with the absorption of dimers, further confirming the formation of H-aggregates. Figure 3 | (a–c) Single-crystal structures of Cy-Bu-D from three perspectives; (d) packing mode of Cy-Bu-D with a slippage angle (θ) and center-to-center distances; (e) packing mode of Cy-He-D with a slippage angle (θ) and center-to-center distances. Download figure Download PowerPoint The fluorescence lifetime and triplet state lifetime To confirm the interaction between the two chromophores of dimer Cy-X-Ds, the fluorescence lifetime of Cy-X-Ds and Cy-H were recorded using the time-correlated single photon counting technique excited at 400 nm in DCM ( Supporting Information Figure S17 and Table S3). The monomer Cy-H presented a single-exponential lifetime (∼1.37 ns), and the Cy-X-Ds dimers showed at least double-exponent lifetime. Specifically, the time constant of the longer fluorescent decay component (τf1) ( Supporting Information Table S3) of Cy-X-Ds corresponds with the fluorescence lifetime (about 1 ns) of Cy-H, attributed to the decay of the monomeric chromophores; the time constant of the shorter fluorescent decay component (τf2) of Cy-X-Ds is attributed to the decay of the dimer. However, Cy-Bu-D has a three-exponential lifetime, which may be due to the strong interaction of Cy-Bu-D with the shortest polymethylene bridge. Meanwhile, the average fluorescence lifetime (τavf) of Cy-X-Ds regularly increased with the lengthening of the methylene chains and approached the average fluorescence lifetime of Cy-H. This data confirmed that the ability of the dimer to inhibit the singlet decay of radiation decreases with the lengthening of the connected alkyl chain. In addition, to better comprehend the energy decay process of the dimers, the triplet-excited state of Cy-X-Ds and Cy-H were investigated by an LP980 laser flash photolysis spectrometer (Edinburgh Instruments, Edinburgh, UK). By monitoring the decay of the ground-state bleaching signal at the maximum absorption of the compounds, the triplet-excited state was found to be longer time (dozens of microseconds) in deaerated solution (Figures 4a–4e). Cy-He-D, Cy-Oct-D, and Cy-Dec-D contained two decay component. The shorter time component corresponds well with the intrinsic triplet-excited-state lifetime of Cy-H (2.33 μs). This longer time component could be assigned to the dimer. Only one long-lived (dozens of microseconds) component was measured from Cy-Bu-D, which may be attributed to its strong interaction between the two chromophores of the dimer. All dimer PSs have a longer triplet-state lifetime than monomer dye Cy-H, which corresponds well with the exciton model of dimerization leading to a highly efficient triplet excited state from Kasha and McRae.28,29 Figure 4 | (a–e) Nanosecond time-resolved transient dynamics decay traces of Cy-X-Ds and Cy-H at 650 nm in oxygen-free DCM. (f) The ΦΔ of Cy-H and Cy-X-Ds. Download figure Download PowerPoint Generation of 1O2 The 1O2 generation capability of Cy-X-Ds and Cy-H under red light irradiation was investigated by using DPBF as an indicator because the absorbance attenuation at 415 nm is positively correlated with 1O2 production. As displayed in Supporting Information Figure S18, under the same irradiation, the absorbance of DPBF at 415 nm remarkably decreased in the presence of Cy-X-Ds dimers, whereas the Cy-H group exhibited an extremely slow decline in DPBF absorbance. This data clearly demonstrated that dimerization of Cy-H significantly improved photosensitization efficiency. The ΦΔ were measured to quantify the photosensitization efficiency [methylene blue (MB)] used as a reference, ΦΔMB = 0.57, in DCM). As anticipated, the ΦΔ of all dimers (14–20%) were much higher than that of Cy-H (0.7%). Remarkably, the ΦΔ of Cy-He-D (18.5%) and Cy-Oct-D (20%) were approximately 26- and 28-fold that of Cy-H, respectively. Interestingly, the ΦΔ of dimers showed an obvious trend of first increasing and then decreasing with the lengthening of the alkyl chain. As shown in Figure 4f, the ΦΔ of Cy-Bu-D, Cy-He-D, and Cy-Oct-D has an obvious upward trend. This can be attributed to the enhanced ISC process resulting from the dimer. The ΦΔ of Cy-Dec-D shows a downward trend compared to Cy-He-D and Cy-Oct-D. This may be due to the longest alkyl chain in the dimers along with the recovery of fluorescence, which reduced the interaction of the chromophores in Cy-Dec-D. In vitro application Ideal PSs should possess low dark toxicity and high phototoxicity upon light irradiation for clinical application. First, the dark toxicity of Cy-X-Ds and Cy-H were evaluated by methyl thiazolyltetrazolium assay (Figure 5a). Considering the dimer of Cy-X-Ds contained two identical monomer molecules (Cy-H), we set the concentration of the dimer to half that of the monomer for more accurate comparison. As shown in Figure 5a, for Cy-Bu-D and Cy-He-D at concentrations as high as 5 μM, after incubation for 24 h, negligible reduction in 4T1 cellular viability was found. However, the viability fell to 60% at 10 μM for Cy-H, suggesting significant dark toxicity of monomeric dye towards the cell line. The 4T1 cells suffered severe viability loss with only 10% remaining when exposed to Cy-Oct-D and Cy-Dec-D group at 5 μM. These results confirmed that Cy-Bu-D and Cy-He-D possess low dark toxicity. Figure 5 | In vitro cytotoxicity of Cy-H and Cy-X-Ds against 4T1 cells: dark cytotoxicity (a) and phototoxicity with the irradiation of 660 nm light for 5 min (b); (c) confocal microscopy images of 4T1 cells incubated with 2 μM of Cy-He-D and DCFH-DA (10 μM) under different treatments; (d) apoptosis detection using AV-FITC/PI costaining in 4T1 cells. Download figure Download PowerPoint Next, the photodynamic efficacy of Cy-X-Ds dimers were evaluated. After 2 h incubation with different concentration of Cy-X-Ds, 4T1 cells were subjected to light treatment (660 nm, 20 mW cm−2) for 5 min and then cultured in the dark environment for another 12 h. As illustrated in Figure 5b, cell viability rapidly decreased upon the dosage increase of all dimers. The half-maximal inhibitory concentration (IC50) of Cy-X-Ds are listed in Supporting Information Table S4. The IC50 of Cy-Oct-D and Cy-Dec-D were 0.54 and 0.53 μM respectively due to their high ΦΔ and poor biosafety. Cy-He-D (IC50 0.92 μM) exhibited high photodynamic therapeutic efficiency and low dark toxicity; Cy-Bu-D (IC50 1.63 μM) showed a weak PDT effect. Hence, Cy-He-D was chosen as the PS for subsequent testing. According to previous reports, positively charged dye molecules (cyanines) easily enter mitochondria. Therefore, sub-cellular colocalization in mitochondria was investigated by the commercial probe Mito-Tracker Green FM (green). As shown in Supporting Information Figure S19, the fluorescence distribution of Cy-H and Cy-X-Ds overlapped well with that of Mito-Tracker Green with a high Pearson’s correlation coefficient (>0.8), demonstrating that these dimers exhibited excellent mitochondria localization. In general, mitochondria are a power house for producing adenosine triphosphate to regulate cellular metabolism. Thus, the mitochondria-target ability of dimers will greatly boost the PDT efficacy.31–34 Apoptosis mechanism of Cy-He-D mediated PDT Finally, to further investigate the potential therapeutic mechanism, we conducted a series of studies, including ROS levels, mitochondrial membrane potential, and apoptosis. First, the cellular 1O2 generation capability of Cy-He-D was evaluated using DCFH-DA as an 1O2 indicator, which gives rise to green fluorescence in the presence of ROS (Figure 5c). Obviously, almost no fluorescence was observed for control groups treated with only Cy-He-D in dark, whereas a bright signal appeared in the PDT groups treated with Cy-He-D plus light irradiation. Moreover, JC-1 staining analysis showed that the excess generation of 1O2 by Cy-He-D after PDT led to serious mitochondrial membrane disruption ( Supporting Information Figure S20). Furthermore, Annexin V-FITC/PI double staining assay confirmed that the apoptosis mechanism was responsible for the phototriggered cancer cell death. As shown in Figure 5d, the PDT group exhibited the green fluorescence of AV-FITC in the cellular membrane and the simultaneous red fluorescence of PI in the cell nucleus, which indicates apoptosis of the 4T1 cells after PDT. In contrast, negligible AV-FITC green signals and PI red signals were observed in the dark groups (Cy-He-D). Collectively, these observations suggest that excessive 1O2 production after PDT treatment causes severe mitochondrial depolarization, which in turn induces apoptosis and ultimately leads to cancer cell death. In vivo application Subsequently, time-dependent in vivo imaging was conducted to determine the distribution of Cy-H and Cy-X-Ds after intravenous injection into the 4T1 breast tumor model. As depicted in Figure 6a and Supporting Information Figure S21, after injection, intense fluorescence signals were captured at the tumor site in each case. For Cy-H, the fluorescence peaked after 15 min intravenous injection, but, the fluorescence quickly disappeared, which indicates that Cy-H was rapidly excreted from the living body. In addition, Cy-Bu-D had almost no fluorescence detected due to its low fluorescence brightness. For the other three molecules, the fluorescence signal intensity at the tumor site reached a maximum after 24 h injection, which gave the optimal accumulation time point of Cy-He-D for in vivo PDT. This result suggests that the dimer molecule had a longer tumor residence time than the monomer and has maximum enrichment in the tumor location. To accurately determine the biodistribution, imaging of the main organs at 24 h post-injection was performed. As depicted in Supporting Information Figure S22a, Cy-He-D effectively accumulated in the tumor tissue and exhibited strong fluorescence signal. Meanwhile, weaker a fluorescence signal remained in the liver, lung, and kidney as a consequence of the metabolization of Cy-He-D. Furthermore, tumor slices were stained with the 1O2 probe singlet oxygen sensor green (SOSG) to evaluate the ROS production in tumor tissues. The SOSG signal of the phosphate-buffered saline (PBS) group (dark or with irradiation) and Cy-He-D in dark was very weak ( Supporting Information Figure S22b). Conversely, the SOSG signal of Cy-He-D with irradiation was stronger than Cy-He-D in dark. This result demonstrates that Cy-He-D has successfully generated ROS in the tumor after PDT. Figure 6 | (a) In vivo fluorescence imaging of 4T1 tumor-bearing BALB/c mice after intravenous injection of Cy-H and Cy-He-D. (b) Tumor volume change during the treatment. (c) Body weight change of mice after different treatments. (d) Photographs of tumor tissue excised from mice in different groups after treatment. (e) H&E analysis of tumor tissue from different groups after two weeks of treatment (scale bar = 100 μm). Download figure Download PowerPoint Encouraged by the good biocompatibility and high PDT efficiency in vitro, Cy-He-D was applied to in vivo PDT treatment of 4T1 tumor-bearing mice. When the tumor volume reached about 150 mm3, 4T1 tumor-bearing mice were randomly divided into four groups to comprehensively evaluate the in vivo therapeutic efficacy over a 14-day follow-up period. The four groups include: (1) PBS buffer without irradiation; (2) PBS buffer with 650 nm lamp irradiation; (3) Cy-He-D without lamp irradiation; (4) Cy-He-D with 650 nm lamp irradiation. Each group was injected intravenously with PBS, or Cy-He-D, and subsequently, the laser irradiation (0.1 W cm−2, 10 min) was performed at 24 h post injection. As shown in Figures 6b and 6c, the tumor volume significantly increased in the three control groups (1, 2, and 3), proving that Cy-He-D without light irradiation did not show any antitumor effect. The Cy-He-D plus 650 nm lamp irradiation group showed significant inhibition of tumor growth and much better therapeutic efficacy than other groups. The outstanding antitumor performance of Cy-He-D was further demonstrated by the representative tumor images (Figure 6d) as well as hematoxylin & eosin (H&E) staining of tumor slices (Figure 6e). Moreover, body weight was monitored for each group to assess toxicity. As shown in Figure 6c, no significant body weight loss or obvious abnormality was discovered in all groups of mice. In addition, after 14 days of treatment, mice were sacrificed and the main organs were sliced for histological H&E staining. No pathological changes of the main organs (heart, liver, spleen, lun