Tumor-Selective Cascade-Amplified Dual-Prodrugs Activation for Synergistic Oxidation-Chemotherapy

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE7 Dec 2022Tumor-Selective Cascade-Amplified Dual-Prodrugs Activation for Synergistic Oxidation-Chemotherapy Xuan Xiao†, Qingyu Zong†, Jisi Li and Youyong Yuan Xuan Xiao† School of Biomedical Sciences and Engineering, South China University of Technology, Guangzhou 511442 National Engineering Research Center for Tissue Restoration and Reconstruction, South China University of Technology, Guangzhou 510006 †X. Xiao and Q. Zong contributed equally to this work.Google Scholar More articles by this author , Qingyu Zong† National Engineering Research Center for Tissue Restoration and Reconstruction, South China University of Technology, Guangzhou 510006 School of Medicine, South China University of Technology, Guangzhou 510006 †X. Xiao and Q. Zong contributed equally to this work.Google Scholar More articles by this author , Jisi Li National Engineering Research Center for Tissue Restoration and Reconstruction, South China University of Technology, Guangzhou 510006 School of Medicine, South China University of Technology, Guangzhou 510006 Google Scholar More articles by this author and Youyong Yuan *Corresponding author: E-mail Address: [email protected] School of Biomedical Sciences and Engineering, South China University of Technology, Guangzhou 511442 National Engineering Research Center for Tissue Restoration and Reconstruction, South China University of Technology, Guangzhou 510006 Key Laboratory of Biomedical Materials and Engineering of the Ministry of Education, South China University of Technology, Guangzhou 510006 Guangdong Provincial Key Laboratory of Biomedical Engineering, South China University of Technology, Guangzhou 510006 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.022.202101564 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Efficacy of prodrugs in cancer therapy requires selective and efficient drug activation in cancer cells. Here, we report a novel dual-prodrug delivery system with tumor-selective cascade-amplified prodrug activation for synergistic oxidation-chemotherapy. Cancer cells overexpressing cathepsin B-activatable near-infrared (NIR) hemicyanine prodrug (CyNH-Citval) were encapsulated by the reactive oxygen species (ROS)-responsive polyprodrug of doxorubicin (DOX) (PTKDOX) to obtain PTKDOX/Cy. Upon uptake of PTKDOX/Cy by cancer cells and subsequent prodrug CyNH-Citval activation, NIR fluorescence was turned on and toxicity toward mitochondria was restored, thereby elevating intracellular ROS levels, which subsequently activated the polyprodrug PTKDOX to initiate the cascade and amplify DOX. Overall, these results indicate that enzyme-mediated initiation of drug activation and amplification of cascade ROS ultimately causes selective and efficient prodrug activation in tumors with synergistic oxidation and chemotherapy. These findings provide new insights to inform precise cooperative cancer therapy. Download figure Download PowerPoint Introduction Although chemotherapy is a major clinical approach for tumor therapy, its application remains constrained by poor selectivity and serious side effects.1–3 To improve the selectivity and therapeutic efficacy of chemotherapy, various stimulus-responsive drug delivery systems (DDSs) have been developed in the past decade.4–6 For example, numerous research has focused on prodrugs that are specifically activated by tumor associated stimuli to release the potent naïve drug which allows them to improve selectivity of chemotherapy.6–8 Various stimuli, including pH,9,10 glutathione,11,12 reactive oxygen species (ROS),13–15 and enzymes,16,17 are present in tumor microenvironments. Thus, tumor-associated enzyme-activated prodrugs have received numerous attention due to the high selectivity of enzymes overexpressed in cancer cells.18–23 However, use of enzyme-activated prodrugs is limited by ineffective drug activation owing to the paucity of tumor-associated enzymes that represent an essential step for the functioning of the prodrug.24,25 For example, Chen and co-workers26 amplified prodrug activation by sequentially delivering combretastatin A4 to upregulate metalloproteinase 9 (MMP9) and MMP9-activated doxorubicin (DOX) prodrug and promoted activation of tumor-selective prodrug for cancer therapy. In addition, Yin and co-workers27 reported that a pro-protein therapy was activated by self-amplified hypoxia associate enzymes. Consequently, development of an enzyme-responsive prodrug, which can simultaneously retain selectivity of enzyme-responsiveness for tumor targeting and amplification of the enzyme response signal for enhanced therapeutic efficiency, remains a great challenge. Recently, numerous research groups have exploited the ability of cancer cells to overproduce ROS to develop ROS-responsive DDSs containing oxidation-labile groups, such as thioketal, boronic ester, and proline, for cancer treatment.28 However, intracellular concentration of ROS is still not high enough for efficient drug activation, which represents an intrinsic limitation for the ROS-responsive systems despite their great potential.29–31 For example, Mokhir and co-workers29 reported an ROS-dependent aminoferrocene-based prodrug that amplified intracellular ROS level for efficient cancer therapy. Therefore, development of new strategies for enzyme-activated ROS generation is imperative to improvement of tumor selectivity. The generated ROS can be further utilized for efficient prodrug activation. In this study, we developed a tumor-selective cascade-amplified dual-prodrug activation system (denoted PTKDOX/Cy) consisting of cancer cells overexpressing cathepsin B (CTB), CTB-activated hemicyanine (CyNH2) prodrug (CyNH-Citval), and ROS-responsive polyprodrug of DOX (PTKDOX) conjugated on the side chain of poly(thioketal) (PTK; Schemes 1a and 1b). Previous studies have shown that amino containing near-infrared (NIR) CyNH2 can selectively accumulate in mitochondria and efficiently lower their membrane potential, thereby increasing intracellular ROS levels to cause oxidation-induced cell death.32 Prior to activation, the prodrug CyNH-Citval shows a weak fluorescence due to intramolecular charge transfer (ICT) and low toxicity, which can reduce its toxicity to normal cells.33,34 Upon interacting with CTB overexpressed in cancer cells, CyNH2 is activated to restore toxicity, and NIR fluorescence is triggered for drug activation monitoring. Activated CyNH2 leads to mitochondrial dysfunction in cancer cells, thereby elevating levels of intracellular ROS. Consequently, high ROS levels mediate activation of the polyprodrug PTKDOX, thereby initiating a cascade and amplifying the DOX prodrug. More importantly, CyNH2 and DOX showed synergistic oxidation and chemotherapy. Moreover, CyNH-Citval exhibited low toxicity and insignificant elevation of ROS levels in normal cells due to the low CTB expression. This phenomenon disrupted initiation of cascade DOX activation and resulted in low cytotoxicity of PTKDOX/Cy to normal cells. Integrating dual prodrugs into a single PTKDOX/Cy with cascade and amplified drug activation may increase tumor selectivity and efficiency of drug activation for synergistic oxidation-chemotherapy of cancer. Scheme 1 | Profile of the tumor-selective cascade-amplified dual-prodrug activation system (PTKDOX/Cy) developed in this research. (a) Chemical structure. (b) Schematic representation of the tumor-selective cascade-amplified dual-prodrugs activation for synergistic oxidation-chemotherapy. Download figure Download PowerPoint Experimental Methods Details of the materials and instruments utilized are provided in the Supporting Information. Preparation of PTKDOX and PTKDOX/Cy CyNH-Citval was synthesized by conjugating CyNH2 with CTB-specific citrulline-valine (Cit-Val) peptide linker. PTK was obtained by fast polycondensation of 1,3-dimercapto-2-propanol and acetone, with a molar ratio of 1∶1.05, in the presence of concentrated hydrogen chloride (HCl). Then PTK pyridine (PTK-SS) was obtained by the disulfide-thiol exchange reaction of PTK with 2,2′-dithiodipyridine at a molar ratio of 1∶3. Finally, PEG-TK-DOX (TK = polythioketone) was first synthesized via conjugating DOX and PEG with a fixed ratio of 10∶1 to the hydroxyl groups of PTK-SS. Then PEG-PTK-DOX (30 mg) was dissolved in 1 mL of dimethyl sulfoxide (DMSO), and then gradually added into 9 mL of ultrapure water under stirring. After additional stirring for 2 h, the solution was transferred into a dialysis bag (MWCO 3500) to remove DMSO against ultrapure water for 24 h, and then the solution was filtered through a 0.45 μm filter to obtain PTKDOX. The preparation of PTKDOX/Cy was similar to that for PTKDOX, but the polymer PEG-PTK-DOX was replaced with PEG-PTK-DOX and CyNH-Citval (1.0 mg). Cell culture and tumor model Mouse breast cancer cell line 4T1 cells were cultured in Roswell Park Memorial Institute 1640 medium with 10% fetal bovine serum and 1% penicillin-streptomycin. Cell cultures were incubated in a 5% CO2 and 21% O2 incubator at 37 °C. Female BALB/c mice and BALB/c nude mice (20 ± 2 g, 6–8 weeks old) were purchased from Hunan SJA Laboratory Animal Co. Ltd (Hunan, China). 4T1 cells (1 × 106) were injected into the right mammary fat pads to establish an orthotopic 4T1 tumor model. After the tumor volumes reached 100 mm3, the mice were used for subsequent experiments. At the end of experiments, all mice were killed by CO2 inhalation. All animal experiments were approved by the Ethics Committee of the South China University of Technology (Guangzhou, China). All detailed experimental methods are available in the Supporting Information. Results and Discussion Preparation and characterization of PEG-TK-DOX and CyNH-Citval A summary of synthetic routes to the NIR CyNH2 is presented in Supporting Information Scheme S1. The structure and purity of CyNH2 and intermediates were confirmed by 1H NMR spectra ( Supporting Information Figures S1–S6). The synthetic method for preparation of the prodrug CyNH-Citval is displayed in Supporting Information Scheme S2. Briefly, CyNH2 was conjugated with CTB-specific Cit-Val peptide linker to obtain the prodrug CyNH-Citval with a yield of 11.6%. Thereafter, the prodrug and its intermediates were verified via 1H NMR spectra ( Supporting Information Figures S7 and S8). The synthetic route of polyprodrug PTKDOX is shown in Supporting Information Scheme S3. In brief, PTK was obtained by rapid polycondensation of 1,3-dimercapto-2-propanol and acetone, with a molar ratio of 1∶1.05, in the presence of concentrated HCl. PTK appeared as a colorless waxy solid, with a 47% yield and 14 repetitive units, after contrastive analysis of integration intensities of peaks 1 (methylene protons of PTK) and 2 (sulfhydryl protons of 1,3-dimercapto-2-propanol) from the 1H NMR spectra ( Supporting Information Figures S9 and S10). Subsequently, PTK-SS was obtained, as a light-yellow solid, by the disulfide-thiol exchange reaction of PTK with 2,2′-dithiodipyridine, at a molar ratio of 1∶3. Next, PTK-SS was activated with N,N′-carbonyldiimidazole (CDI) then conjugated with DOX and amino-terminated methoxy poly(ethylene glycol) (PEG) to obtain polyprodrug PTKDOX. 1H NMR spectra revealed that the grafting rate for DOX was about 50% ( Supporting Information Figure S13). Moreover, 1H NMR spectra, 13C NMR spectra, MS spectra, and gel permeation chromatography studies were used for characterization analysis of the new compounds, polymers, and their intermediates ( Supporting Information Figures S1–S32 and S34A). The fluorescence change of CyNH-Citval in response to papain Results of analysis of CyNH2 and CyNH-Citval absorption are shown in Supporting Information Figure S34b. Summarily, CyNH2 had a maximum absorption of 710 nm, whereas that of CyNH-Citval blue-shifted to 615 nm, which is attributed to ICT of CyNH-Citval.33,34 In addition, CyNH2 exhibited a strong fluorescence intensity, whereas that of CyNH-Citval was weak, further affirming the ICT of CyNH-Citval ( Supporting Information Figure S34c). Next, we chose papain as a substitute enzyme for analysis of CyNH-Citval’s enzyme-response behavior, due to its similar enzyme activity to CTB,35 and investigated fluorescence changes of CyNH-Citval after treatment with different concentrations of papain over time. CyNH-Citval’s fluorescence intensity increased with prolonged incubation times, reaching saturation after 6 h with a papain concentration of 10 mM ( Supporting Information Figure S34d). In addition, CyNH-Citval’s fluorescence intensity increased upon increased papain concentration, reaching saturation upon addition of 10 μM papain after 6 h (Figure 1a). Notably, this fluorescence intensity increased ∼15-fold and exhibited papain concentration dependence over a wide range (0–10 μM). These results indicated that papain could effectively cleave the amide bond of CyNH-Citval, and release CyNH2 with strong fluorescence activation. Therefore, a high concentration of CTB in cancer cells may cause a release of CyNH2 and turn-on NIR fluorescence. Figure 1 | (a) Fluorescence spectra of CyNH-Citval treated with different concentrations of papain. (b) UV–vis absorbance spectra of DOX, CyNH-Citval, and PTKDOX/Cy. (c) Changes in hydrodynamic diameter of PTKDOX/Cy after treatment with H2O2, ClO−, or ·OH. (d) 1H NMR spectrum for H2O2-responsive degradation of PTK-SS with generation of acetone after treatment with DMSO-d6 and H2O2 (10 mM) at 37 °C. (e) Fluorescence spectra for DOX, PTKDOX, and PTKDOX/Cy. (f) Cumulative release of DOX from PTKDOX/Cy in the presence of different concentrations of H2O2. Download figure Download PowerPoint In vitro ROS-responsive degradation and DOX release Next, we employed a nanoprecipitation method to prepare PTKDOX/Cy by self-assembly from PEG-PTK-DOX via encapsulation of CyNH-Citval. Results showed that PTKDOX/Cy had a hydrodynamic diameter of ∼107 nm in phosphate-buffered saline (PBS) (Figure 1c). In addition, PTKDOX/Cy exhibited an absorbance spectrum with similar absorbance to DOX and CyNH-Citval, with maximum absorption at 480 and 615 nm, respectively (Figure 1b and Supporting Information Figure S34b), indicating that DOX and CyNH-Citval were successfully loaded. DOX and CyNH-Citval had loading capacities of 33.67 ± 0.23 and 9.13 ± 0.25%, respectively. Meanwhile, PTKDOX/Cy’s hydrodynamic diameter changed from 107 nm in PBS, to 10 nm after treatment with H2O2, ClO− or ·OH (Figure 1c), indicating that it was degraded in response to ROS. Also, transmission electron microscopy (TEM) and scanning electron microscopy images recorded for PTKDOX/Cy are shown in Supporting Information Figures S33a and S33c, and the TEM image recorded for PTKDOX/Cy after treatment with 10 mM H2O2 and 10 μM papain is shown in Supporting Information Figure S33b. Furthermore, 1H NMR spectra revealed H2O2-triggered degradation of PTK-SS (Figure 1d) as well as the disassociation mechanism of PTKDOX ( Supporting Information Figure S35). Degradation of PTK-SS (15 mg mL−1) was detected using a commixture of DMSO-d6 and H2O2 (10 mM), while the thioketal of PTK-SS eventually turned into acetone (Figure 1d). The fluorescence of DOX in PTKDOX/Cy was inhibited by the Förster resonance energy transfer of DOX to CyNH2 and aggregation-caused quenching of DOX (Figure 1e). Next, we studied drug release of the PTKDOX/Cy and found almost no or moderate release of free DOX in the presence of PBS and 1 mM H2O2, respectively (Figure 1f). Conversely, large amounts of DOX were released in the presence of 10 mM H2O2, indicating that more drug could be released in cells with high H2O2 concentrations. The critical micelle concentration of PTKDOX PTKDOX can itself induce the formation of polymeric nanoparticles. To evaluate the critical micelle concentration (CMC), we measured the count rates of nanoparticles at different concentrations according to the previous literature.36 As shown in Supporting Information Figure S36a, the CMCs of PTKDOX nano-assembly is 0.0728 mg/mL. The dynamic light scattering data revealed that the size of obtained PTKDOX nanoparticles was about 73.2 nm ( Supporting Information Figure S36b). In vitro CyNH2 release and the cellular uptake mechanism for the PTKDOX/Cy Analysis of the release behavior of CyNH2 from PTKDOX/Cy in the presence of both ROS (10 mM H2O2) and enzyme (10 μM papain) ( Supporting Information Figure S37a) shows the cumulative release of CyNH2 was <5% in phosphate buffer at 37 °C after 48 h, which means negligible leakage before the prodrug reaches the tumor site. When H2O2 or papain was added, the release of CyNH2 was <10%. In contrast, the cumulative release amount of CyNH2 was 72.4% after 48 h in the presence of H2O2 and papain. These results indicate that CyNH2 only can be released when H2O2 and papain were simultaneously present. The endocytic pathway of PTKDOX/Cy by cancer cells was investigated in the presence of several endocytosis inhibitors including chlorpromazine (inhibitor of clathrin-mediated endocytosis), methyl-β-cyclodextrin (inhibitor of caveolae-mediated endocytosis), and amiloride (inhibitor of giant pinocytosis). As shown in Supporting Information Figure S37b, the cells treated with chlorpromazine at 4 °C showed lower nanoparticle internalization, suggesting that PTKDOX/Cy is susceptible to a clathrin-mediated endocytotic pathway in 4T1 cells in an energy-dependent manner. Intracellular fluorescence recovery of PTKDOX/Cy Next, we performed confocal image analysis on mouse breast cancer cell line 4T1 and the mouse embryonic fibroblast (MEF) cell line MEF to evaluate PTKDOX/Cy’s applicability in cancer therapy and imaging. Results revealed strong red fluorescence signals in both CyNH-Citval and PTKDOX/Cy-treated 4T1 cells, with ∼10- and 6-fold increases in the mean fluorescence intensity (MFI), respectively, relative to MEF cells (Figure 2a and Supporting Information Figure S38). In contrast, pretreatment of 4T1 cells with CA-074 methyl ester (CA-074-Me), a CTB inhibitor, resulted in diminished fluorescence. Overall, these results indicated selective activation of CyNH-Citval and PTKDOX/Cy fluorescence in 4T1 tumor cells, making it promising for tumor-specific intelligent images. Figure 2 | (a) Confocal microscopy images showing fluorescence of CyNH2 (red) in 4T1 and MEF cells after incubation with CyNH-Citval or PTKDOX/Cy for 6 h. Inhi. represents the CTB inhibitor CA-074-Me, which was preincubated with the cells for 2 h. (b) Confocal microscopy images showing intracellular ROS levels stained with DCFH-DA (green) in 4T1 and MEF cells after different treatments. VC represents ROS scavenger vitamin C. (c) Cytotoxicity of 4T1 and MEF cells incubated with CyNH2 or CyNH-Citval. (d) Cytotoxicity of 4T1 cells treated with PTKDOX, PTKDOX/Cy, and PTKDOX/Cy+VC. Statistical significance: *P < 0.05, **P < 0.01. Download figure Download PowerPoint Colocalization of CyNH2 with mitochondria and mitochondrial membrane potential study Furthermore, we found good colocalization between CyNH2 and MitoTracker Green-labeled mitochondria, as evidenced by a colocalization coefficient of 0.83 ( Supporting Information Figure S39). Next, we explored mitochondrial membrane potentials (MMPs) using in 4T1 cells the probe 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-imidacarbocyanine iodide (JC-1). Results showed that JC-1 could assemble into J-aggregates, with red fluorescence in high MMP mitochondrial matrix, but dispersed into the cytoplasm in a monomeric form with green fluorescence in low MMP mitochondrial matrix. Notably, cells treated with CyNH-Citval and CyNH2 exhibited marked green fluorescence ( Supporting Information Figure S40), indicative of a gradual decrease in MMP, although cells treated with CyNH-Citval+Inhi (a CTB inhibitor CA-074-Me) exhibited a weak green fluorescence. Intracellular ROS level measurements The decrease in MMP, caused by CyNH2, may be attributed to CyNH2-induced ROS production. We explored CyNH2- and PTKDOX/Cy-induced ROS production in 4T1 cells using the ROS sensor 2,7-dichlorofluorescin diacetate (DCFH-DA), which is oxidized by ROS and subsequently converted into dichlorofluorescin (DCF) with green fluorescence. Results revealed strong green fluorescence in cells treated with CyNH2 and PTKDOX/Cy, with an approximately eight- and sevenfold increase in the MFI, respectively, relative to PBS or DOX (Figure 2b and Supporting Information Figure S42a). Cells in the control group, which had been supplemented with ROS scavenger vitamin C (VC) and CTB inhibitor CA-074-Me, also exhibited a weak fluorescence signal. Taken together, these results confirmed that CyNH2 can elevate intracellular ROS production, which can be further utilized to activate DOX. Intracellular DOX release To confirm whether elevated ROS levels by released CyNH2 could amplify DOX activation, we employed confocal laser scanning microscope imaging to investigate intracellular DOX release from PTKDOX and PTKDOX/Cy in 4T1 cells. Results showed that 4T1 cells exhibited obvious red fluorescence (from DOX) after incubation with PTKDOX and PTKDOX/Cy for 6 h (Figure 2c). A further 12 h incubation resulted in a weak red fluorescence signal in the nucleus (stained with Hoechst 33342), indicating that only a small amount of DOX was released in cells treated with PTKDOX. Conversely, a strong red fluorescence signal was observed in the nucleus of cells treated with PTKDOX/Cy, indicative of enhanced activation of DOX with PTKDOX/Cy. Next, we employed HPLC to quantitatively evaluate concentrations of DOX in 4T1 cells, after incubation with PTKDOX and PTKDOX/Cy over time. Results showed that cells treated with PTKDOX/Cy had a 3.8-fold increase in the amount of released DOX relative to those treated with PTKDOX at 36 h ( Supporting Information Figure S42c). These results indicated that the released CyNH2 from PTKDOX/Cy with elevated ROS levels amplified the DOX activation in the tumor cells. In vitro cytotoxicity and cellular apoptosis assay Next, we employed the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay to explore the cytotoxicity of CyNH-Citval and CyNH2 to more kinds of cells: human umbilical vein endothelial cell (HUVEC) line, human cervical cancer cell (Hela cells), mouse colon cancer cell line CT26, and mouse melanoma cell line B16. Tumor cells 4T1, Hela, CT26, and B16 cells had significantly lower viability than MEF and HUVEC cells following exposure (Figure 2d and Supporting Information Figure S41), indicating that CTB-activated prodrug CyNH-Citval exerts tumor-specific toxicity. Moreover, DOX and CyNH2 had a synergistic effect on 4T1 cells, as evidenced by a combination index of 0.75 ( Supporting Information Figure S42b). Furthermore, we investigated PTKDOX, PTKDOX/Cy, and PTKDOX/Cy+VC cytotoxicity on 4T1 cells, at different DOX concentrations, and found that PTKDOX exerted minimal cell-killing activity, whereas PTKDOX/Cy had higher toxicity than PTKDOX and the lowest half maximal inhibitory concentration (IC50) of 0.67 μg mL-1 in 4T1 cells (Figure 2e). In contrast, PTKDOX/Cy had low cell-killing activity in the presence of VC, indicating that CyNH2-induced increased activation of DOX was ameliorated. Based on this marked cytotoxicity of PTKDOX/Cy to 4T1 cells, we explored cellular apoptosis using the Annexin V-FITC/PI apoptosis detection assay. Results showed that PTKDOX/Cy treatment induced significantly higher cellular apoptosis than PTKDOX, whereas cell viability markedly increased following addition of either VC or the inhibitor ( Supporting Information Figure S43), which corroborated results from the cytotoxic test. In vivo bio-distribution study of PTKDOX/Cy Results from in vivo NIR fluorescence images of PTKDOX/Cy in 4T1 tumor-bearing nude mice revealed strong fluorescence in tumors of mice injected with PTKDOX/Cy compared to those treated with PTKDOX/Cy+Inhi, and the fluorescence was retained in the tumors up to 48 h (Figure 3a and Supporting Information Figure S44a). We collected the tumors and major organs 48 h after administration of PTKDOX/Cy and PTKDOX/Cy+Inhi to obtain ex vivo images. Results revealed stronger fluorescence intensity in tumors of mice treated with PTKDOX/Cy relative to the other organs, while a weak intensity was observed in the group treated with PTKDOX/Cy+Inhi ( Supporting Information Figure S44b). These results indicated that PTKDOX/Cy was selectively localized in tumor, and yielded precise in vivo images. To confirm the efficacy of PTKDOX/Cy in vivo, we analyzed fluorescence intensity of CyNH2 and ROS levels in tumor tissues after treatment with PTKDOX/Cy and PTKDOX/Cy+Inhi. Results showed strong fluorescence intensities in DCF and CyNH2 from PTKDOX/Cy-treated mice, but weak in PTKDOX/Cy+Inhi group ( Supporting Information Figure S45), confirming the release of CyNH2 and significant elevation of ROS levels in tumor tissues, which was consistent with results from the in vitro cell experiment. Taken together, these results provide further evidence that CyNH2 from PTKDOX/Cy is activated with concomitant NIR fluorescence for drug activation monitoring, and the activated CyNH2 causes mitochondria dysfunction in cancer cells and increases levels of intracellular ROS in tumor tissues. Figure 3 | (a) In vivo NIR fluorescence images of nude mice bearing 4T1 tumors, after intravenous injection with PTKDOX/Cy and PTKDOX/Cy+Inhi. Ex vivo fluorescence images. (b) Changes in tumor volumes in 4T1 tumor-bearing mice under different treatments. (c) Ex vivo tumor weights of mice under different treatments. (d) Images of tumors in mice after H&E and TUNEL staining at the end of the therapy in response to different treatments. Scale bar = 100 μm. Statistical significance: *P < 0.05, **P < 0.01, ***P < 0.001. Download figure Download PowerPoint In vivo antitumor evaluation of PTKDOX/Cy Next, we evaluated anti-tumor efficacy of PTKDOX/Cy in vivo using mice bearing 4T1 tumors. The mice were kept until the tumors grew to ∼100 mm3, divided into five groups, and treated with PBS, free DOX, DOX+CyNH2, PTKDOX, and PTKDOX/Cy (5 mg kg−1 DOX), respectively, via tail vein injection. Thereafter, we measured and recorded the tumor volumes and body weights once every 2 days. Results showed that mice treated with DOX and PTKDOX exhibited restricted tumor growth to some extent, relative to those in the PBS group (Figure 3b). However, tumors in mice treated with PTKDOX/Cy and DOX+CyNH2 were markedly restricted. Notably, PTKDOX/Cy treatment exhibited the best therapeutic efficiency, as evidenced by a tumor inhibition rate of 53%, with the smallest incremental tumor volume (5.5 times) relative to the initial volume. Similar results were observed in tumor weights and images (Figure 3c and Supporting Information Figure S46a). Notably, body weights of mice treated with PTKDOX/Cy were almost constant during treatment, indicating great biosecurity of the prodrug PTKDOX/Cy to mice ( Supporting Information Figure S46b). Moreover, results from hematoxylin and eosin (H&E) staining of major organs revealed a negligible variation after PTKDOX/Cy treatment, relative to PBS, which also affirmed PTKDOX/Cy’s biosecurity ( Supporting Information Figure S47). Furthermore, histological analysis of tumors from mice in the PTKDOX/Cy group revealed that a great mass of tumor cells was dead and the nucleus destroyed (Figure 3d). Similar results were obtained after terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining, as evidenced by the observed green fluorescence. Conclusion We successfully designed a dual-prodrug delivery system, comprising a tumor-selective cascade amplified prodrug activation, for synergistic oxidation-chemotherapy. Enzyme-activated prodrug, CyNH2, not only endowed therapy with high tumor selectively but also elevated ROS production for efficient activation of prodrug DOX. Prodrug CyNH-Citval could be activated by CTB overexpressed in 4T1 tumor cells, relative to normal MEF cells, whereas activated CyNH2 caused a 7.2-fold increase in intracellular ROS production, which further enhanced DOX activation by 3.8-fold compared to PTKDOX. Moreover, PTKDOX/Cy resulted in the best in vivo therapeutic efficacy against 4T1 tumors, as evidenced by a 53% tumor inhibition rate, which was appreciably augmented in contrast to PTKDOX (10%). Taken together, these findings indicate that designing dual-prodrugs with enzyme-responsive cascade-amplification drug release is a promising strategy for improving selectivity and therapeutic efficacy of cancer therapy. Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Supporting Information Supporting Information is available and includes the details of the synthesis, characterizations, and TEM images as well as the evaluation of in vitro DOX release, the fluorescence change of CyNH-Citval in response to papain, confocal microscope image and flow cytometry studies, colocalization of CyNH2 with mitochondria, MMP study, intracellular ROS level measurements, intracellular DOX release, in vitro cytotoxicity and cellular apoptosis assay, in vivo bio-distribution study, in vivo antitumor evaluation of PTKDOX/Cy, and additional Figures S1–S47. Conflict of Interest The authors declare no competing financial interest. Acknowledgments This work was supported by the National Natural Science Foundation of China (nos. 52073101 and 51873072), the Science and Technology Program of Guangzhou (no. 202102010025), Guangdong Provincial Pearl River Talents Program (no. 2019QN01Y088), the Special Fund for the Construction of High-level Key Clinical Specialty (Medical Imaging) in Guangzhou, Guangzhou Key Laboratory of Molecular Imaging and Clinical Translational Medicine.