Reconstructing Tumor Microenvironment Using Photoresponsive Cyanobacteria to Reversal Chemoresistance for Robust Chemotherapy

The solid tumor microenvironment (TME) plays a crucial role in tumor biological behavior, development, and chemoresistance. Herein, a promising strategy is reported to remodel the TME and combat chemoresistance by employing the photoresponsive cyanobacteria (Synechococcus 7942, Syne). Syne exhibits inherent motility and enhanced permeability and retention effects to penetrate deep into the tumor. Under a 660 nm laser irradiation, Syne keeps a controllable, continuous, and robust O2 production ability through photosynthesis to alleviate tumor hypoxia and reduce monocarboxylate transporter 4 (MCT4) expression and exerts a gentle photodynamic effect by generating reactive oxidative species (ROS) in situ. In addition, adequate O2 supplement and ROS can not only facilitate intracellular doxorubicin (DOX) accumulation but also increase the drug sensitivity of tumor cells by downregulating the expression of chemoresistance‐related genes (e.g., heat shock factor‐1, mutant P53, and P‐glycoprotein). Compared with free DOX treatment, photoresponsive Syne with laser irradiation facilitates the deep penetration and accumulation of DOX in the tumor. Importantly, this Syne‐boosted chemotherapy achieves 100% survival in mice and complete tumor ablation with no evident systemic toxicity over a span of 90 days. Overall, this study presents a new insight and strategy to overcome chemotherapeutic resistance and eliminate tumors.

The solid tumor microenvironment (TME) plays a crucial role in tumor biological behavior, development, and chemoresistance. Herein, a promising strategy is reported to remodel the TME and combat chemoresistance by employing the photoresponsive cyanobacteria (Synechococcus 7942, Syne). Syne exhibits inherent motility and enhanced permeability and retention effects to penetrate deep into the tumor. Under a 660 nm laser irradiation, Syne keeps a controllable, continuous, and robust O 2 production ability through photosynthesis to alleviate tumor hypoxia and reduce monocarboxylate transporter 4 (MCT4) expression and exerts a gentle photodynamic effect by generating reactive oxidative species (ROS) in situ. In addition, adequate O 2 supplement and ROS can not only facilitate intracellular doxorubicin (DOX) accumulation but also increase the drug sensitivity of tumor cells by downregulating the expression of chemoresistance-related genes (e.g., heat shock factor-1, mutant P53, and P-glycoprotein). Compared with free DOX treatment, photoresponsive Syne with laser irradiation facilitates the deep penetration and accumulation of DOX in the tumor. Importantly, this Syne-boosted chemotherapy achieves 100% survival in mice and complete tumor ablation with no evident systemic toxicity over a span of 90 days. Overall, this study presents a new insight and strategy to overcome chemotherapeutic resistance and eliminate tumors. mutation of P53, rendering tumor cells less sensitive to drugs. [33][34][35][36] Moreover, monocarboxylate transporter 4 (MCT4) is expressed at high levels in hypoxic regions in many fast-growing tumors, including breast cancer. Indeed, MCT4 promotes tumor cell proliferation and migration, and its expression is correlated with a shorter overall survival. [37,38] Thus, various strategies, such as O 2 carriers, catalystbased O 2 generators, and the inhibition of hypoxia-related protein expression, were developed to improve the TME and enhance the efficacy of chemotherapy. [26,27,[39][40][41][42] However, these strategies have limitations, such as continuous and controllable oxygen supply, and the efficiency to attenuate hypoxia in the TME is still unsatisfactory.
In addition, cancer cells reduce sensitivity to antitumor drugs by decreasing drug uptake or enhancing drug efflux, ultimately leading to the failure of clinical chemotherapy of tumors. [43,44] Reactive oxidative species (ROS) are important physiological regulators of intracellular signaling pathways. Different intracellular ROS levels determine whether to induce chemoresistance or enhance chemosensitivity. Previous studies indicated that the upregulation of ROS in mitochondria to over the cytotoxic threshold of malignant cells can not only promote apoptosis in drugresistant tumor cells [45] but also reduce drug efflux and overcome chemotherapy resistance. [46][47][48][49][50] Li et al. found that the use of drugloaded nanoparticles under near-infrared (NIR) laser irradiation to generate ROS can overcome cancer drug resistance by reducing the synthesis of ATP. [47] Some studies have also found that the upregulation of intracellular ROS can reverse chemotherapeutic resistance in drug-resistant cells by inhibiting P-gp gene expression. [49,50] Thus, increasing intracellular ROS could be an effective approach to make cancer cells lose their drug resistance capability.
In nature, active microalgae with photoautotrophic ability can grow under light and stop growing without light. More importantly, microalgae can keep producing abundant oxygen in the presence of light, and the photosynthetic O 2 production capacity of microalgae is light-controllable. Over the last 2 years, scientific researchers have increasingly investigated the O 2 production capacity of microalgae and used microalgae in related antitumor research. [42,[51][52][53][54][55][56][57][58] Our group found that cyanobacteria under laser irradiation can continuously generate O 2 by photosynthesis, and produce ROS through a certain photodynamic effect. [42] Hence, applying versatile photosynthetic cyanobacteria with controlled O 2 and ROS production to improve the TME and enhance the chemotherapy effect is a promising approach for tumor treatment.
Here, we propose using Synechococcus 7942 (Syne), a natural photosynthetic cyanobacteria capable of intrinsic tumor-targeting ability, combined with doxorubicin (DOX) to achieve high-efficiency chemotherapy for breast cancer (Scheme 1). Syne produced O 2 Scheme 1. Schematic illustration of Syne as an in situ photocatalyzed O 2 and ROS generator combined with DOX for high-efficient antitumor therapy. (1) Syne capable of intrinsic tumor-targeting ability were effectively enriched at the tumor site. (2) Administering 660 nm laser irradiation at the tumor site, Syne provided abundant O 2 through photosynthesis to overcome tumor hypoxia and reduce the expression of MCT4 and generated ROS through the PDT effect, which synergistically inhibited tumor cell proliferation and migration. In addition, the as-produced O 2 and ROS played a key role in reducing the exocytosis of DOX and increasing cell sensitivity to DOX by downregulating chemoresistance-related proteins (e.g., HSF-1, P53, HIF-1α, and P-gp), thereby synergistically increasing the intracellular accumulation of DOX and attenuating the drug resistance. (3) Consequently, Syne combined with DOX effectively inhibited tumor growth and recurrence in an MCF-7 murine model. by photosynthesis in situ to overcome tumor hypoxia, reduced the expression of MCT4, and generated ROS through a photodynamic therapy (PDT) effect as photosensitizers upon 660 nm laser irradiation, which synergistically inhibited tumor cell proliferation and migration. In addition, adequate O 2 and ROS produced by Syne could not only reduce the exocytosis of DOX, but also increase cell sensitivity by downregulating of chemoresistance-related genes (e.g., HIF-1α, heat shock factor-1 [HSF-1], mutant P53, and Pgp), thereby synergistically overcome chemoresistance and increased the intracellular accumulation of DOX. As a result, this Syne-boosted chemotherapy achieved 100% of mice survival and completed tumor ablation with no evident systemic toxicity over a span of 90 days. Thus, this chemotherapeutic strategy with the help of photoresponsive Syne can be a promising candidate for malignant tumor treatments.

Characterization of Syne
Confocal microscopy imaging showed that autofluorescent Syne in the medium maintained a rod shape ( Figure 1A). A representative scanning electron micrograph (SEM) ( Figure 1B) and transmission electron micrograph (TEM) ( Figure S1, Supporting Information) showed that the length of Syne was approximately 2-5 μm. Syne grew quickly under light but not in the dark ( Figure 1C), which is important for the mass production and light-controllable distribution of cyanobacteria in vivo. Next, we evaluated whether Syne can generate O 2 under 660 nm laser by a portable oxygen meter. Large number of bubbles were observed in the Syne culturing solution after laser irradiation ( Figure S2, Supporting Information). As shown in Figure 1D-E, Syne produced a high amount of O 2 under 660 nm laser but not without light, suggesting that Syne can carry out efficient photosynthesis and control O 2 production by controlling laser irradiation. Subsequently, we measured whether Syne can produce singlet oxygen ( 1 O 2 ) in BG11 medium under 660 nm laser with singlet oxygen sensor green (SOSG). The results showed that compared with the groups of BG11 irradiated with a 660 nm laser (here after medium (þ)) and free Syne, Syne irradiated with a 660 nm laser (here after Syne (þ)) produced more 1 O 2 , which increased with the time of laser exposure ( Figure 1F), indicating that Syne under laser irradiation can produce a PDT effect. These above results indicate that active Syne continuously produces O 2 by photosynthesis and exert a PDT effect as a photosensitizer.

In Vitro Mechanisms Overcoming Chemoresistance
To investigate whether Syne (þ) can overcome hypoxia-induced chemoresistance to improve the therapeutic efficiency in vitro, the experiment was tested in MCF-7 cells that were maintained at hypoxia and compared to cells at normoxia. First, MCF-7 cells were cultured under hypoxic conditions, incubated with or Bottom right picture: bubbles generated in the Syne suspension. E) Light-dependent oxygen production capacity of Syne (1 Â 10 8 CFUs in 10 mL BG11 medium) when irradiated with or without a 660 nm laser at different time intervals. The black arrow represents the laser off, and the red arrow represents the laser on. F) Singlet oxygen generation of Syne in BG11 medium, irradiated with or without laser for different durations (n = 3). "(þ)" represents exposure to a 660 nm laser (0.1 W cm À2 ). Data are expressed as mean AE SD. The asterisks indicate that the difference between medium (þ) and the other treatment groups is statistically significant. ***p < 0.001. One-way ANOVA with Tukey's post hoc test was employed to analyze the results.
without Syne, irradiated with a 660 nm laser, and then treated with DOX for 1 h. As shown in Figure 2A, DOX fluorescence was significantly enhanced in the MCF-7 cells, especially in the nucleus, after the hypoxia þ Syne (þ) or normoxia treatment. The DOX fluorescence intensity in the hypoxia þ Syne (þ) treated cells was 3-4 times higher than that in the hypoxia or hypoxia þ Syne group, but was almost similar to that in the normoxic group ( Figure 2B), indicating that Syne (þ) could facilitate effective intracellular DOX accumulation. Then, the in vitro antitumor effect of Syne and DOX was detected in MCF-7 cells cultured under hypoxic conditions. The results showed that the DOX treatment alone induced only approximately 30% cell death, but the DOX treatment after Syne (þ) in advance induced nearly 85% cell death ( Figure 2C). The above results suggest that Syne (þ) could reverse chemoresistance under the hypoxic condition.
Previous studies have shown that chemoresistance in hypoxic tumors is related to a high expression of HIF-1α. [26,[30][31][32] An increase in HIF-1α expression causes an increase in the expression of MDR1 and P-gp, leading to the export of anticancer drugs and consequent drug ineffectiveness. It has also been shown that hypoxic conditions are often correlated with the overexpression of mutant P53 protein, which is related to the sensitivity of cells to chemotherapy drugs. [33][34][35][36] Moreover, a high expression of the HSF-1 gene is also closely related to poor chemotherapy results. [43,44,59] The mRNA levels of drug-resistant genes were first measured in cells in different treatment groups via quantitative real-time polymerase chain reaction (qRT-PCR). As shown in Figure 2D-E, there was no apparent accumulation of mutant P53 and MDR1 in MCF-7 cells under normoxia conditions, while corresponding high expression levels were observed under hypoxia conditions. However, when Syne was added to the culture medium of hypoxic cells and irradiated with a 660 nm laser, the expression levels of mutant P53 and MDR1 in the cells were downregulated to levels similar to those in normoxic cells. Meanwhile, no apparent accumulation of HIF-1α was detected in MCF-7 cells under normoxic condition, while high expression of HIF-1α occurred under hypoxic condition. It was noteworthy that HIF-1α downregulated significantly in hypoxic MCF-7 cells treated with Syne (þ) The asterisks indicate statistically significant differences between the hypoxia and other treatment groups. **p < 0.01, ***p < 0.001; # indicates a statistically significant difference between the two groups: #p < 0.05. One-way ANOVA with Tukey's post hoc test was employed to analyze the results. ( Figure S3, Supporting Information). The protein expression of HSF-1, mutant P53, and P-gp was further assessed with confocal fluorescence microscopy ( Figure 2F-H). The results demonstrated that the red fluorescence signals (recognizing HSF-1, mutant P53, or P-gp) accumulated in MCF-7 cells after the treatment in the hypoxia or hypoxia þ Syne groups but decreased in the normoxia or hypoxia þ Syne (þ) group.

Tumor-Targeted Oxygen Generation of Syne and Biodistribution of DOX In Vivo
Based on the results of reversing chemoresistance by Syne in vitro, we investigated whether Syne could modulate the tumor microenvironment and enhance the retention and penetration of DOX in solid tumors in vivo. First, Syne was injected intravenously (i.v.) into MCF-7 tumor-bearing mice, and the FL signals monitored by an in vivo imaging system showed that Syne largely accumulated in tumors within 24 h ( Figure S4, Supporting Information). Moreover, the distribution of Syne in major organs and tumors was quantitatively analyzed by bacterial colony counting at different times. The results demonstrated that the Syne colony numbers (normalized by weight; CFUs g À1 ) were higher in the tumors than in other organs at 24 h ( Figure 3A,B), mainly due to the inherent motility and enhanced permeability and retention effects of Syne, which is consistent with our previous results and proves that Syne had good tumor targeting ability. [43]  The asterisks indicate statistically significant differences between the groups. *p < 0.05. The differences between two groups were analyzed using Student's t-test analysis. To study the oxygen production capacity of Syne in vivo, photoacoustic (PA) imaging was used to collect the signals of oxygenated hemoglobin (HbO 2 ) before and after injection of Syne. The results showed that the PA signal in the Syne-treated tumor was significantly enhanced after 660 nm laser irradiation ( Figure 3C and S5, Supporting Information), suggesting that Syne-mediated photosynthesis enhanced the intratumoral O 2 levels. However, the PA signal in the phosphate balanced solution (PBS)-treated tumor was almost unchanged before and after laser irradiation. To further visualize the hypoxia variation, tumors were collected after different treatments and then immunochemically stained with hypoxyprobe. The confocal microscopy images of the tumor slices showed that compared with the PBS (þ) group, the hypoxic areas (green signals) in the Syne (þ) group were significantly reduced ( Figure 3D). By quantitatively analyzing the region of interest (ROI) in these images, the hypoxia distribution in the Syne (þ) group dramatically decreased ( Figure 3E). Meanwhile, notably, the expression of HIF-1α was greatly downregulated in the Syne (þ) group ( Figure 3F), indicating that Syne could generate O 2 by photosynthesis under the 660 nm laser to relieve tumor hypoxia. MCT4 plays an important role in intracellular pH homeostasis and tumor aggressiveness and has been shown to be highly expressed in hypoxic regions of rapidly growing tumor masses, as its expression has been shown to be HIF-1dependent. Double immunofluorescence staining of tumor sections showed that both a hypoxia-related marker (PIMO) and MCT4 were highly expressed and colocalized in the tumors.
In contrast, PIMO and MCT4 signals were rarely observed in the Syne (þ) tumors ( Figure 3G). Then, a SOSG probe was selected to investigate the PDT effect in vivo. Strong green fluorescence was observed in the Syne (þ) treated group ( Figure 3H), indicating that Syne generated abundant 1 O 2 through the PDT effect. These data demonstrate that after laser irradiation, photoresponsive Syne improved the tumor hypoxic microenvironment, reduced MCT4 expression, and generated 1 O 2 , which could play important roles in inhibiting tumor cell growth and metastasis.
Hypoxia has been proven to promote tolerance to chemotherapy. Hence, 24 h after the DOX injection, the distribution of DOX in major organs and tumor tissues in the mice was determined. As demonstrated in Figure 4A, the DOX fluorescence intensity in the tumor tissue in the DOX/Syne (þ) treatment group was significantly higher than that in the free DOX treatment group. The average fluorescence intensity of DOX at the tumor site in the DOX/Syne (þ) treatment group was approximately 1.45 times that in the free DOX group ( Figure 4B), suggesting that photoresponsive Syne with laser irradiation facilitated DOX penetration and accumulation in the tumor.
Previous studies reported that only when the drug penetrates a human tumor up to a distance of 200 μm it can reach all surviving cells in the tumor. [14] Hence, the localization of DOX in tumors was further investigated. Compared to the DOX group, the green fluorescence signals (DOX) in the DOX/Syne (þ) group were significantly increased and distributed in the deep part of the tumor, even as deep as 300 μm from the edge of the tumor ( Figure 4C,D). . "(þ)" represents exposure to a 660 nm laser (0.1 W cm À2 , 30 min). Data are expressed as mean AE SD. The asterisks indicate statistically significant differences between the groups. ***p < 0.001. The differences between two groups were analyzed using Student's t-test analysis. These results indicate that Syne promoted intratumoral DOX accumulation through adequate O 2 supplement and ROS production in situ, which could be superior to conventional chemotherapy.

In Vivo Synergistic Antitumor Therapy
The synergistic antitumor effect in vivo was further investigated in the MCF-7 tumor-bearing mouse model. A schematic diagram of the drug treatment program is shown in Figure 5A. As shown in Figure 5B, Syne (þ) did not significantly inhibit tumor growth. Even though the DOX treatment was able to inhibit tumor growth to a certain extent before day 20, all tumors in this group still grew to approximately 1800 mm 3 on day 57. However, the tumors in DOX/Syne (þ) group displayed shrinkage from day 10 and were completely eliminated without recurrence within a period of 60 days, indicating significant tumor-growth inhibition efficiency. Furthermore, the survival curve revealed that all mice treated with DOX/Syne (þ) had an extended life span, and 100% of the mice survived for at least 90 days, while all mice in the other three groups died within 82 days ( Figure 5C). Moreover, the apoptotic level of the tumor cells was further evaluated by a terminal deoxynucleotidy1 transferase-mediated dUTP-biotin nick end labeling (TUNEL) analysis ( Figure 5D). The TUNEL staining assay showed that the mice treated with DOX/Syne (þ) presented much larger areas of tumor apoptosis and necrosis in the slice, whereas moderate apoptosis was observed in the Syne (þ) and free DOX groups, further proving the superior therapeutic efficacy of DOX/Syne (þ). Therefore, the combination of Syne and DOX can potently suppress tumor growth and recurrence in vivo. The asterisks indicate statistically significant differences between the PBS and other treatment groups. ***p < 0.001; # indicates a statistically significant difference between the two groups: #p < 0.05. One-way ANOVA with Tukey's post hoc test was employed to analyze the results.

In Vivo Drug Resistance Reversal Mechanisms
To further demonstrate the possible molecular mechanism of overcoming chemoresistance, the expression of chemoresistancerelated genes in vivo was measured. As shown in Figure 6A,B, the qRT-PCR analysis of the tumor tissues showed that the expression levels of MDR1 and mutant P53 after the DOX/Syne (þ) treatment were reduced by 71.43% and 75.21%, respectively, compared with those in the PBS group. Moreover, the P-gp and mutant P53 protein levels were also notably downregulated in the tumors after the Syne (þ) or DOX/Syne (þ) treatments ( Figure 6C), indicating that Syne with 660 nm laser irradiation could overcome tumor hypoxia-induced chemoresistance. In addition, the tumor sections were stained with hypoxyprobe or P-gp or mutant P53 antibodies for immunofluorescence imaging to explore the relationship between tissue hypoxia and resistance-related proteins. As illustrated in Figure 6D,E, the tumor slices from the mice with i.v. injection of Syne (þ) or DOX/Syne (þ) exhibited weak green and red fluorescence signals, suggesting that Syne potently ameliorated tumor hypoxia through photocatalytic O 2 generation and further reduced the expression of P-gp and mutant P53 in the tumors. These results indicate that the oxygen-producing and ROS-generating effects of Syne can effectively alleviate chemotherapy resistance in vivo, further revealing the effective antitumor mechanism of cyanobacteria-mediated chemotherapy.

Toxicity Evaluation
The biosafety of the Syne and DOX-based therapy on day 30 after treatment was also evaluated. The body weights of the mice were recorded throughout the treatment. The free DOX and DOX/ Syne (þ) groups exhibited a slight weight loss compared with the other groups after the last injection. At the end of the experiment, the weight of the mice was recovered and was not significantly different from that in the PBS group ( Figure S6, Supporting Information), illustrating that the dosage of DOX used in the experimental treatment was well tolerated. The histology analysis of the main organs in each treatment group showed no abnormal tissue structure compared with the PBS group ( Figure 7A), suggesting the good biocompatibility of the treatment with DOX and Syne. To evaluate long-term toxicity in the heart, liver, spleen, lung, and kidney, the cardiac troponin I (cTnI), creatine kinase MB fraction (CK-MB), blood urea nitrogen (BUN), serum creatinine (CRE), aspartate aminotransferase (AST), and alanine aminotransferase (ALT) concentrations in serum were collected on the 30th day postinjection. The blood tests showed no difference in the concentrations of all these markers among the four groups ( Figure 7B-D). Taken together, these observations suggest that the treatment with DOX/Syne (þ) at a DOX dose of 5 mg kg À1 and a Syne concentration of 2 Â 10 7 CFUs within 30th days poses no obvious signal of toxic The asterisks indicate statistically significant differences between the PBS and other treatment groups. *p < 0.05, ***p < 0.001; # indicates a statistically significant difference between the two groups: # p < 0.05. One-way ANOVA with Tukey's post hoc test was employed to analyze the results. side effects in mice. These results demonstrate that DOX/Syne (þ) is a highly effective combination therapy with minimal toxicity.

Conclusion
In summary, we reported an active photosynthetic cyanobacteria (Syne) for in situ photocatalyzed O 2 and ROS generation, which reconstructed the TME and enhanced the ability of chemotherapy to inhibit tumor growth and relapse. The in vitro investigations demonstrated that Syne could provide controllable, continuous, and robust O 2 through photosynthesis and generate ROS through the PDT effect under a 660 nm laser irradiation. In breast cancer MCF-7 subcutaneous xenograft models, the findings indicated that Syne could not only target the tumor but also generate O 2 to overcome tumor hypoxia, reduce the expression of MCT4, and generate ROS in situ under a 660 nm laser irradiation, which could synergistically inhibit tumor cell proliferation and migration. In addition, adequate O 2 and ROS produced by Syne decreased chemoresistance-related genes (e.g., mutant P53, HSF-1, and P-gp) to synergistically overcome the hypoxia-induced chemoresistance and facilitate the deep penetration and accumulation of DOX in the tumor. More importantly, this Syne-boosted chemotherapy achieved 100% survival in the mice and completed tumor ablation with no evident systemic toxicity over a span of 90 days. Therefore, this study provides new insight for achieving an improved therapeutic efficacy in overcoming chemotherapeutic resistance via photosynthetic cyanobacteria-boosted chemotherapy.

Experimental Section
Materials and Bacterial Strains: Hoechst 33 358 and SOSG were provided by Invitrogen (USA). Hematoxylin and eosin (H&E), BCA protein assay kit, and TUNEL Apoptosis Assay Kit were provided by Beyotime Institute of Biotechnology (CHN). Doxorubicin (DOX) was obtained from J&K Scientific Ltd. (CHN). Anti-MCT4 antibodies were purchased from Santa Cruz Biotechnology (USA). Aspartate aminotransferase (AST), alanine aminotransferase (ALT), blood urea nitrogen (BUN), creatinine (CRE), creatine kinase MB fraction (CK-MB), and cardiac troponin (cTn-1) were purchased from Jian Cheng Biotech (CHN). Human MCF-7 breast cancer cell line was purchased from the Cell bank of Chinese Academy of Sciences (CHN). Syne and BG11 medium were purchased from Freshwater algae seed bank at the institute of hydrobiology (CHN). Syne was cultured with BG11 medium at room temperature under light.
Characterization of Syne: The morphology of Syne was obtained using a confocal laser scanning microscope (CLSM, TCS SP5II, Leica, Germany). SEM (VEGA 3 SBH, Czech Republic) and TEM (TecnaiG2 F20 S-Twin, FEI, 200 kV, USA) were applied to further identify the morphology of Syne. To determine the O 2 production of Syne, a dissolved O 2 meter (Mettler Toledo, Switzerland) was applied to detect the O 2 release profiles of Syne (1 Â 10 8 CFUs in 10 mL BG11 medium) irradiated with or without a 660 nm laser. The detector was injected into the solution and the value of the dissolved O 2 was recorded at the predetermined times. In vitro singlet oxygen ( 1 O 2 ) generated by Syne under 660 nm laser irradiation was monitored by the SOSG indicator. Syne was dispersed in BG11 medium and SOSG (1 μM) was added, followed by irradiation of the medium with or without a 660 nm laser (0.1 W cm À2 ) for different time.
qRT-PCR and Western Blot Assays: MCF-7 cancer cells (1 Â 10 5 per well) were seeded in 24-well plates and cultured in normoxic (21% O 2 ) or hypoxic (2.5% O 2 ) conduction at 37°C for 24 h. Hypoxic cells were stochastically separated into three groups and treated with or without Syne. One . "(þ)" represents exposure to a 660 nm laser (0.1 W cm À2 , 30 min). Data are expressed as mean AE SD. One-way ANOVA with Tukey's post hoc test was employed to analyze the results. No statistically significant difference was found among various groups.
www.advancedsciencenews.com www.small-structures.com group treated with Syne was exposed to a 660 nm laser irradiation (8 min, 0.1 W cm À2 ) and cultured for another 24 h. For detecting the mRNA level of MDR1 and mutant P53 in vitro, total RNA was purified from MCF-7 cells using Trizol reagent (Life Technologies, USA). The complementary DNA (cDNA) strand was synthesized with an inverse transcription kit (Toyobo, Japan). QRT-PCR was performed using SYBR qPCR mix (Toyobo, Japan). All reactions were performed in a final volume of 20 μL with Roche light cycler 480II (USA). For western blot assay, tumor cells were harvested and treated with 1 Â RIPA lysis buffer containing 1 mM phenylmethanesulfonyl fluoride (PMSF) to extract the proteins. The lysates were centrifuged with 10 000 rpm for 20 min at 4°C. Then, protein concentration in cell lysates was detected with a BCA protein assay kit. Equal amounts of protein were separated by 8% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred onto a polyvinylidene fluoride (PVDF) membrane. After blocking for 1 h by blocking buffer, the membranes dip in 2% BSA buffer with anti-P-gp antibody (Abcam, USA), anti-HIF-1α antibody (Cell Signal Technology, USA), anti-mutant P53 antibody (Abcam, USA), or anti-GAPDH antibody (Beyotime Biotechnology, CHN) at 4°C overnight, followed by subsequent imaging tests.
In Vitro Immunofluorescence Staining: 1 Â 10 4 cells per well of MCF-7 cells were inoculated onto 8-well chambered slides (Thermo Scientific, USA), cultured overnight under normoxic or hypoxic conditions, followed by treating with or without Syne. The cells were then irradiated with or without 660 nm laser at 0.1 W cm À2 for 8 min and cultured for an additional 24 h. Cells were rinsed with PBS and handled with 4% paraformaldehyde (PFA) at 4°C, followed by incubating with 3% BSA and 0.1% TritonX-100 for 1 h at room temperature. The cells were incubated with P-gp antibody, anti-HSF-1 antibody (Abcam, USA), and mutant P53 antibody at 37°C for 2 h and then washed with PBS 3 times. Afterward the cells stained with secondary antibody for 1 h. Then, the cells were incubated with Hoechst 33 358 for nuclei staining. After 5 min, CLSM was selected to detect the fluorescence signals.
Cellular Uptake Analysis: CLSM was selected to investigate the cell uptake of DOX in MCF-7 cells. 1 Â 10 4 per well of MCF-7 cells were cultured in eight-well chambered slides under normoxic or hypoxic condition for 24 h. Subsequently, the cells treated with or without Syne were exposed to or not 660 nm laser at 0.1 W cm À2 for 8 min. The next day, the medium of MCF-7 was replaced with new medium containing DOX (6 μg mL À1 ). Cells were washed thrice with PBS to remove DOX not taken up by cells after 1 h incubation at 37°C. Then, the cell nucleus was stained with Hoechst 33 358 solution for 5 min, and rinsed with PBS 3 times. Finally, the cells were imaged under a CLSM, and the total fluorescence intensity of DOX was quantified via Image J software.
Tumor Model: Female Balb/c nude mice (6-8 weeks old) were provided by Beijing Vital River Laboratory Animal Technology Co., Ltd. (CHN) and the animal analysis was allowed by the Animal Care and Use Committee (Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences). The right flank of each mouse was inoculated with MCF-7 cells (1 Â 10 6 cells per mouse) suspended in free Dulbecco's modification of Eagle's medium (DMEM) medium. The tumor-bearing mice were randomized to receive corresponding treatments when the volume of tumor closed to 70 mm 3 .
In Vivo Biodistribution of Syne and DOX: To assess the distribution of Syne in mice, MCF-7 tumor-bearing mice were sacrificed at 1, 12, and 24 h after treatment with Syne, and their main organs and tumors were collected. The tissues were homogenized in BG11 solution, and then placed on solid BG11 agar plates, and counted after 72 h to detect Syne CFUs in organs and tumors.
To evaluate the distribution of DOX, Syne (2 Â 10 7 CFUs per mouse) were injected into MCF-7 tumor-bearing mice via tail vein. After 1 day, the tumors were irradiated with a 660 nm laser for 30 min (0.1 W cm À2 ), followed by intravenous injection of DOX (5 mg kg À1 ) into the mice, with free DOX as a control. After 24 h, the main organs and tumor were removed and placed on a plate. An IVIS spectrum imaging system (CRI Maestro, USA) was taken to capture the fluorescence signals of DOX. Fluorescence intensity values were calculated for the major organs and tumor ROI.
In Vivo Antitumor Assessment: MCF-7 tumor-bearing mice were separated into four groups randomly. Experimental group: group 1: PBS, group 2: Syne (þ), group 3: DOX, and group 4: DOX/Syne (þ). On days 5, 7, and 9, mice in the Syne (þ) group and DOX/Syne (þ) group were i.v. injected with Syne (2 Â 10 7 CFUs per mouse). After 24 h, the mice were irradiated with a 660 nm laser for 30 min (0.1 W cm À2 ). After the mice were irradiated with or without laser on days 6, 8, and 10, DOX was i.v. injected in the free DOX group or DOX/Syne (þ) group (DOX: 5 mg kg À1 ). The tumor size was monitored every other day with a digital caliper. The tumor volume was assessed as width 2 Â length Â 0.5. The weight of the mice was also monitored with an electronic balance. When the tumor volume exceeded 2000 mm 3 or showed signs of impaired health, the mice were sacrificed by cervical dislocation. The survival status of animals was monitored daily.
In Vivo FL/PA Imaging: ICG-labeled Syne were i.v. injected when the tumor volume in BALB/c mice reached approximately 200 mm 3 to evaluate the accumulation of Syne in tumor. The FL signal of the ICG was recorded with the IVIS spectrum imaging system at the appointed times after injection.
To assess the in vivo photosynthetic oxygen production capacity of Syne by PA imaging, a preclinical PA computerized tomography scanner (Endra Nexus 128, USA) was selected to capture the HbO 2 signal at an excitation wavelength of 850 nm.
Immunohistochemistry: MCF-7 tumor-bearing mice treated with Syne (2 Â 10 7 CFUs) were irradiated with a 660 nm laser for 30 min (0.1 W cm À2 ). After 24 h, the mice injected intraperitoneally pimonidazole hydrochloride (60 mg kg À1 ) according to the protocol of the Hypoxyprobe-1 Plus kit. The animals were sacrificed and tumors were collected at predetermined time points. Subsequently, frozen tumor sections (8 μm) were prepared. After blocking with 3% BSA and 0.1% TritonX-100 in PBS at 37°C for 1 h, the sections were incubated with various antibodies (anti-P-gp, antimutant P53, anti-MCT4, or antipimonidazole antibodies) at the room temperature for 2 h, followed by Alex Fluor secondary antibodies for 1 h. To assess the apoptosis of the tumor, the TUNEL assay was applied according to the protocol provided by the kit. To study the production of 1 O 2 in tumors, mice were administered with Syne through the tail vein. After 24 h, the mice injected intraperitoneally SOSG (6.25 μg mouse À1 ). Tumor then irradiated with a 660 nm laser for 30 min (0.1 W cm À2 ). Tumors were collected and embedded in the Optimum Cutting Temperature Compound. Hoechst 33 358 was used to stain the nucleus of tumor sections. Finally, fluorescence images were obtained with a CLSM.
To investigate the distribution of DOX in tumors, tumors at 24 h after i.v. injected with DOX or DOX/Syne (þ) were harvested. The prepared tumor sections were incubated with Hoechst 33 358 for 5 min and observed by a CLSM. The fluorescence intensity of DOX was quantified by Image J software.
Blood Analysis and Histology Examination: The blood of mice was taken for blood biochemical analysis on the 30th day after the last treatment. Meanwhile, the mice in each group were sacrificed and major organs were harvested and fixed with 4% PFA. Then the tissue sections were stained with H&E and imaged with a CLSM. The serum CK-MB, cTn-1, BUN, CRE, AST, or ALT was assessed in accordance with the manufacturer's instruction.
Statistical Analysis: Fluorescence images were processed and analyzed by Image J. The experimental data of cell viability were normalized using PBS group as control. All experiments were repeated at least three independent samples. All data points are reported as mean AE SD. The statistical analysis was made with GraphPad Prism software. Student's t-test was utilized to evaluate between two groups, and one-way ANOVA with Tukey's post hoc test was applied to estimate the difference between multiple groups. Statistical significance was denoted by asterisks, *p < 0.05, **p < 0.01, and ***p < 0.001; # indicates a statistically significant difference between the two groups, #p < 0.05.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.