Tumor Microenvironment Activable Self‐Assembled DNA Hybrids for pH and Redox Dual‐Responsive Chemotherapy/PDT Treatment of Hepatocellular Carcinoma

Smart self-assembled “Turn-ON” DNA hybrids are employed, which could respond to tumor microenvironment stimuli for cancer cell specific real-time fluorescence imaging, tumor-specific synergistic photodynamic therapy and chemotherapy in hepatocellular carcinoma. Photodynamic therapy (PDT), as a noninvasive therapeutic strategy, has been approved by the U.S. Food and Drug Administration for clinical usage to treat various tumoral diseases.1-3 Once exposed to appropriate light, photosensitizer (PS) as nontolerant therapeutic agent could transform endogenous oxygen to generate reactive oxygen species (ROS) then induce apoptosis of cancer cells to perform PDT.1 Chlorin e6 (Ce6), as an efficient photosensitizing (PS) agent, was widely applied for PDT because of rapid elimination from the body, high singlet oxygen generation efficiency and activation by NIR light.4-7 However, the nonspecific damage to vascular system and normal tissues once the injected PS exposed to sun light, can make serious side effects in patients including burn, swelling, and pain. In addition, the ROS generation of PS would rapidly consume tissue endogenous oxygen to produce severe hypoxia, while has no succeeding oxygen supply due to incomplete nature of the tumor vasculature, which could significantly hinder the PDT efficacy. Recently, tumor microenvironment triggered drug delivery system (DDS) has captured extensive attention due to their smart and finely controllable response, which could reduce the drug side effects and further increase their anticancer efficiency.8-10 Thus, design a stimuli-responsive DDS that combined PDT with controlled drug release behavior could be an effective strategy to avoid side-effects of PDT and increase the antitumor efficiencies. Nevertheless, the precise targeting of tumor cells with minimizing side effects in normal cells is a prerequisite to DDS for cancer treatment. Structured single-stranded DNA such as aptamer with low molecular weight is a promising molecular targeting element that has been extensively designed for targeted drug delivery applications, due to its advantages in terms of small size, synthetic accessibility, thermally stable, low immunogenicity, and flexible chemical modification.11-13 Additionally, aptamer was not only could act as targeting ligand but also could be applied as drugs delivery system for diagnosis and targeted therapy.14, 15 Herein, we designed a smart Ce6-DOX-DNA hybrid (named as Ce6-fDNADox probe) with tumor microenvironment responsive “Turn-ON” properties to perform cancer cell specific real-time fluorescence imaging, as well as increase the specificity of photodynamic therapy and chemotherapy. The hybrids were self-assembled by three types of “components” via complementary base pairing, and subsequently loaded with DOX (Doxorubicin) for real-time NIR fluorescence imaging, tumor-specific photodynamic therapy and chemotherapy via redox-responsive ROS generation, and pH-triggered DOX release in HCC (Hepatocellular Carcinoma). As illustrated in Scheme 1, the self-assembled Ce6-fDNADOX probe was prepared by hybridization using three types of functionalized single-strained DNA (f-ssDNA), termed as redox-responsive quencher ssDNA (RQD), Ce6 labeled ssDNA (CD), and cell-targeting aptamer “TLS11a” (TD), respectively, and each strand has a “sticky end” segment to assemble each other through a complementary base pairing with uniform size. The sequence information of these ssDNA was shown in Table 1. For selectively recognizing the hepatocellular carcinoma cells, the aptamer “TLS11a,” which could specifically recognize the membrane surface of HCC cells,16-19 was incorporated into “TD” chain. To achieve tunable switch “ON” for real-time fluorescence imaging and PDT treatment specifically in cancer cells with minimal side effects, we incorporated the BHQ2 quencher group to the 3′ end of RQD chain with a disulfide linkage; after the self-assembling of these single-strained DNAs, the fluorescence and ROS generation ability of Ce6 would be quenched by BHQ2 due to the energy transfer in normal cells; the BHQ2 quencher could be released due to the cleavage of disulfide linkage by the high level of GSH in cancer cells, then the fluorescence and the ROS generation ability of Ce6 could be recovered for NIR imaging and PDT treatment of cancer cells. Meanwhile, the loaded DOX in Ce6-fDNA probe through π–π stacking could be released due to the slightly acidic conditions of cancer cells and its lysosomes to perform chemotherapy, which then could combined with the “on” state of PDT for synergistic treatment of HCC. As a proof of concept, we chose 24-bases sticky ends to obtain a stable enough self-assembled structure and load the DOX in this paper. To demonstrate the precise self-assemble of these three functionalized ssDNA, agarose gel electrophoresis was used to examine the formation of each building unit. As shown in Figure S1 (Supporting Information) and Figure 1A, we could clearly see that the three types of functionalized ssDNA were indeed completely self-assembled into new high-molecular weight products (Ce6-fDNA) in Lane 7 with the molar ratio of RQD, CD, and TD as 1:1:1, according to the gradually increasing of the product's molecular weight. Meanwhile, the Ce6 labeling of CD chain did not affect the complementary base pairing and the self-assemble efficiency of Ce6-fDNA probe (as shown in Figure 1A). To confirm the successful assembling of Ce6-fDNA probe, UV–vis spectra of CD chain, RQD chain, and Ce6-fDNA probe was analyzed. As shown in Figure 1B, comparing with the CD chain (red line) and RQD chain (green line), the Ce6-fDNA probe (black line) exhibited two absorption peaks at 420 nm corresponding to the characteristic absorption peak of Ce6 at 405 nm with slightly red-shift that was consistent with previous reported works,12 and at 585 nm corresponding to the characteristic absorption peak of BHQ2. In addition, the fluorescence intensity of Ce6 in Ce6-fDNA probe was significantly quenched roughly 9.4-folds (89.4%) by the nearby BHQ2, which was calculated from the emission peak at 675 nm, compared with the same amount of CD chain (Ce6 concentration) in TM buffer, due to the efficient energy transfer between Ce6 and BHQ2 (Figure 1C and Figure S2, Supporting Information). To further confirm whether reducing agent could specifically cleave the disulfide linker at the 5′ end of Ce6-fDNA probe, which could recover the florescence of Ce6, 5 × 10−6 m of Ce6-fDNA probe was incubated with or without 10 × 10−3 m dithiothreitol (DTT) in TM buffers at room temperature, respectively. As shown in Figure 1D and Figure S3 (Supporting Information), the fluorescence intensity of Ce6-fDNA probe was gradually increased along with time after adding DTT to cleave the disulfide linker, and the fluorescence intensity was increased more than tenfolds after 2 h comparing with the beginning (Figure 1D and Figure S4, Supporting Information). In contrast, no obvious fluorescence increase was detected in Ce6-fDNA probe without DTT incubation after 2 h. Furthermore, the singlet oxygen generation ability of Ce6-fDNA probe was investigated through using the ROS indicator (anthracenediylbis (methylene) dimalonic acid, ABDA).4 Due to the efficient quenching of Ce6 by nearby BHQ2 inside the Ce6-fDNA probe, there is no significant decrease of the ABDA absorbance upon laser irradiation (670 nm, 0.2 W cm−2, as shown in Figure S5, Supporting Information), which clearly indicates the suppression of singlet oxygen production by energy transfer between Ce6 and BHQ2. However, the ABDA absorbance showed a sharp decline along with time when Ce6-fDNA probe incubated with DTT and then irradiated with 670 nm laser (0.2 W cm−2, Figure 2A,B), due to the cleavage of disulfide linker and release the quencher BHQ2, which could recover the singlet oxygen production of Ce6. Afterward, the chemotherapy drug “DOX” was loaded into the Ce6-fDNA probe in TM buffer through π–π interaction within “GC” and “CG” double helix to obtain the final products “Ce6-fDNADOX probe.” To confirm DOX payload, both UV–vis spectra and fluorescence spectra were analyzed. As shown in Figure 2C,D, the Ce6-fDNADOX probe (black line) exhibited a new absorption peak at 505 nm, in accordance with the characteristics absorption peak of DOX; meanwhile, the fluorescence of DOX (black line) was significantly quenched in Ce6-fDNADOX probe, compared with the free DOX (red line), with a quenching efficiency of 89.41% (Figure S6, Supporting Information) due to the aggregation of DOX inside the probe. Furthermore, the loaded DOX could be fast released from the Ce6-fDNADOX probe by responsive to the external low pH. As shown in Figure S7 (Supporting Information), DOX fluorescence of Ce6-fDNADOX probe was significantly increased at pH 5.0, compared with Ce6-fDNADOX probe at pH 7.0, which was attributed to the increased hydrophilicity and solubility of DOX at this pH.20 Active targeting and selectively internalized into cancer cells are of great importance for minimizing toxicity to normal cells. The HCC cell specific targeting ability of our Ce6-fDNA probe and Ce6-fDNADOX probe were further examined. As shown in Figure 3A, the Ce6-fDNA probe could be selectively internalized into HepG2 cells, but not Hela cells (only with minimal background nonspecific uptake) observed by confocal laser scanning microscopy analysis (CLSM), due to the specific HCC cell targeting ability of aptamer “TLS11a” in the TD chain. Meanwhile, the Ce6 fluorescence signals of Ce6-fDNA probe could gradually increase along with time inside the cytoplasm of HepG2 Cells, because of the cleavage of disulfide linker by intracellular GSH. Furthermore, the controlled release of DOX from Ce6-fDNADOX probe was further analyzed by CLSM. The HepG2 cells were first incubated with 5 × 10−6 m Ce6-fDNADOX probe for 2 h, then the free probes were removed and the cells were further incubated with fresh culture medium for additional 2 h. As shown in Figure 3B, the fluorescence of DOX (ex 488 nm, em 595 nm) and Ce6 (ex 633 nm, em 650 nm) were clearly detected after the 2 + 2 h incubation in HepG2 cells, respectively; although the DOX and Ce6 fluorescence also could be detected in Hela cells under the same conditions, these fluorescence signals are much lower than that in HepG2 cells, which is due to the specific internalization of the Ce6-fDNADOX probe in HepG2 cells but only nonspecific background uptake in Hela cells. In HepG2 cells, the Ce6 fluorescence (red) is mainly diffused in cytoplasm due to the fluorescence recover, while partial of the DOX fluorescence (green) is appeared in the nucleus of HepG2 cells due to the DOX release from Ce6-fDNADOX probe and subsequently enter into the cell nucleus. To further confirm the controlled release behavior of Ce6-fDNADOX probe, HepG2 cells were incubated with Ce6-fDNADOX probe or Ce6-fNDNADOX (the probe without the specific HCC cell targeting element) for 2 h, then washed and incubated with fresh culture medium for additional 2, 4, 6, and 24 h, respectively. The amount of released DOX was evaluated by recording the fluorescence intensity of the whole cells in 96 wells (n = 5). As shown in Figure S8 (Supporting Information), the fluorescence intensity of DOX released from Ce6-fDNADOX probe (ex 488, em 595 nm) was presented a concomitant increasing manner through expanding the incubation time. Furthermore, the fluorescence intensity of DOX from Ce6-fDNADOX was significantly higher than that of Ce6-fNDNADOX, indicating that the Ce6-fDNADOX probe could be more efficiently uptaken by HepG2 cells due to the specific recognizing, and could serve as an effective drug delivery system for selective targeting HepG2 cells with controlled drug release properties. Since the Ce6-fDNADOX probe has both the redox responsive PDT effects and pH responsive chemotherapy effects, the synergistic antitumor efficacy of Ce6-fDNADOX was further investigated by using a Cell Counting Kit-8 (CCK8) assay. As shown in Figure 3C, under the 670 nm laser irradiation with the power intensity of 0.2 W cm−2, Ce6-fDNA (Ce6, 2.05 × 10−6 m) inhibited 37.9% of the proliferation; Ce6-fDNADOX (Ce6, 2.05 × 10−6 m; Dox, 39.5 × 10−6 m) without laser irradiation inhibited 33.9% of the proliferation; however, Ce6-fDNADOX with 670 nm laser irradiation significantly inhibited 85.1% of cell proliferation. In contrast, the equivalent Ce6-fNDNADOX (Ce6, 2.05 × 10−6 m; Dox, 39.5 × 10−6 m) under the same condition only inhibited 26.1% of cell growth, which due to none sufficient probe uptake. These results clearly demonstrated remarkable increased synergistic antitumor effects of our Ce6-fDNADOX through PDT and chemotherapy in HepG2 cells. Furthermore, the synergistic antitumor efficacy of Ce6-fDNADOX probe was determined by flow cytometry using Annexin-V-fluoroisothiocyanate (FITC)/propidium iodide(PI) staining. The stained cells were divided into four subgroups, the viable group, the early apoptotic group, the late apoptotic/necrotic group, and the dead cells/debris group, which were localized in the lower left, lower right, upper right, and upper left quadrants, respectively. As shown in Figure 3D, the majority of HepG2 cells were localized in the lower left quadrant with more than 91.72% of the viable cells in the control, and the percentage of apoptotic and necrotic cells was only 8.28% for the control group. Meanwhile, compared with the control, most of the cells without Ce6-fDNADOX treatment still remained alive (90.01% of the viable cells) under the laser irradiation at 670 nm (0.2 W cm−2). However, the percentage of viable cells was significantly decreased (26.98% of the viable cells), and the percentage of apoptotic and necrotic cells was obviously increased to 73.02% in the Ce6-fDNADOX treated groups with laser irradiation, which was much higher than PDT treatment alone (25.84%, only incubated with Ce6-fDNA with laser irradiation) or chemotherapy alone (23.74%, only incubated with Ce6-fDNADOX without laser irradiation). In contrast, the percentage of apoptotic and necrotic cells was only 18.18% for the Ce6-fNDNADOX group under laser irradiation, which may be due to the insufficient amount of Ce6-fNDNADOX by nonspecific uptake. These results suggested the selectivity and combined cell killing efficiency or the combined PDT and chemotherapy effects of our Ce6-fDNADOX. To further investigate the Ce6-fDNADOX targeting ability in vivo, a systematic comparative investigation was performed to validate the selective accumulation of Ce6-fDNADOX in HepG2-tumor bearing mice model. As shown in Figure 4A, a weak red fluorescence signal of Ce6 from Ce6-fDNADOX could be seen in the tumor site (white ring) in the first 1 h after intravenous injecting of Ce6-fDNADOX into mice, followed by a significant increase of the Ce6 fluorescence signal at the tumor site post 2 h of injection. However, the Ce6 fluorescence signal was obviously decreased after 4 h of injection. In contrast, no Ce6 fluorescence signal from Ce6-fNDNADOX could be seen at the tumor site during the entire procedure after the injection of Ce6-fNDNADOX into HepG2-tumor bearing mice model (Figure 4A). Moreover, the biodistribution of Ce6-fDNADOX or Ce6-fNDNADOX in HepG2-tumor bearing mice model was also examined after 2 h of injection (Figure 4B). It could be clearly seen that the tumor from Ce6-fDNADOX injected mice presented significantly higher fluorescence signal of Ce6 (red) and DOX (green) than that of Ce6-fNDNADOX injected mice, and the Ce6-fDNADOX injected mice presented more accumulation in tumor than heart, lung, or spleen. These results clearly suggested that our Ce6-fDNADOX had specific tumor targeting ability for hepatocellular carcinoma in vivo. Next, we evaluated the in vivo synergistic antitumor efficacy of Ce6-fDNADOX in HepG2-tumor bearing mice model. After dorsal tumor nodules grew to ≈100 mm3, mice were randomly divided into four groups for studying the therapeutic efficacy; changes in tumor volumes of these four groups that received various treatments as indicate (as mentioned in the Experimental Section) were monitored for 12 d. As shown in Figure 4C,D, mice experienced a rapid tumor growth in PBS buffer (phosphate buffered saline) treated mice with 670 nm laser irradiation for 10 min (0.2 W cm−2), indicating that only the use of laser irradiation has no therapeutic effects. In contrast, Ce6-fDNA, Ce6-fNDNADOX, and Ce6-fDNADOX administration/irradiation groups showed obvious delay of tumor growth. Moreover, the mice received Ce6-fDNADOX treatment with 670 nm laser irradiation exhibited significant slower tumor growth, compared with the mice that received Ce6-fDNA treatment with 670 nm laser irradiation, which due to the synergistic therapeutic effects of PDT and chemotherapy. Noteworthy, the Ce6-fDNADOX treatment had much better therapeutic effects than the Ce6-fNDNADOX treatment since the Ce6-fDNADOX probe could more efficiently accumulate in the tumor site due to the targeting ability of TLS11a aptamer. Taken together, these results clearly demonstrated that our Ce6-fDNADOX probe could selectively target to HepG2 tumor cells in vivo to perform the synergistic PDT and chemotherapy with excellent therapeutic efficacy. High drug toxicity usually leads to weight loss, we thus further measured the body weight of HepG2-tumor mice model for all groups during the treatment to evaluate the toxicity of our probe. As shown in Figure 4D, no obvious weight loss of all groups was observed, indicating that the toxicity or side effects of our Ce6-fDNADOX probe was not serious at all. Moreover, we also evaluated the tumor tissues by hematoxylin and eosin (H&E) after 48 h of indicated treatment (Figure 4E). Compared with PBS treated group, the tumors that received Ce6-fDNA treatment with 670 nm laser irradiation or Ce6-fNDNADOX treatment with 670 nm laser irradiation showed a certain degree of tissue and cellular damage, due to the apoptosis of cancer cells induced by the PDT or nonspecific accumulation of Ce6-fNDNADOX for limited PDT and chemotherapy. However, the Ce6-fDNADOX treated tumors with 670 nm laser irradiation showed obvious cell and as indicated by decrease intensity of cancer cells. These results the significantly synergistic therapeutic efficiency of of our Ce6-fDNADOX probe. In we reported an self-assembled DNA hybrid system probe) for real-time NIR fluorescence imaging, tumor-specific photodynamic and chemotherapy in and in vivo. In this the smart Ce6-fDNADOX probe was to the side-effects by “ON” and of the PDT as and the treatment by with synergistic chemotherapy. were by and by using Chlorin e6 (Ce6), and were from and apoptosis were from with a of was from a and used for all all other were and used as Cell HepG2 cells and cells were from cells were in medium with and of Ce6-fDNA an of and Ce6 were in in the for Ce6 was then to 5′ end ssDNA in at pH 7 by in the The Ce6 was removed by of and for four of the DNA and Ce6 was by the absorbance at and The has a with Ce6 from the of = + = 3′ end disulfide and BHQ2 ssDNA (RQD), TLS11a aptamer (TD), and none targeting aptamer were by of the three of ssDNA RQD chain, chain (CD), and cell-targeting or were to three with a buffer TM buffer × 10−3 m 10 × 10−3 m pH buffer was prepared with with of 5 × 10−3 m of of and 1 of The molar ratio of RQD to CD to TD is each was to for 5 min and to room 4 h to the self-assemble of Ce6-fDNA probe or probe × 10−6 of Ce6-fDNADOX of DOX were into the 10 × 10−6 m Ce6-fDNA probe or probe. the was with for 12 h, and the free DOX was removed by of Ce6-fDNADOX probe or probe for four The amount of loaded DOX was determined by the absorbance at nm in TM buffer The has a with DOX from the of = + = DNA was with and analyzed using agarose gel at for min in buffer × 10−3 m 2 × 10−3 m and × 10−3 m acid, pH The were stained with by and by from of Ce6-fDNA 5 × 10−6 m of Ce6-fDNA probe was incubated with 10 × 10−3 m DTT for The fluorescence spectra were at room in a a The was 405 nm, and the emission were in the from 650 to nm with both and emission of 10 nm under a of ROS of Ce6-fDNA under 670 nm ROS generation of Ce6-fDNA probe was measured through using ABDA as an 15 the Ce6-fDNA probe in 10 × 10−3 m dithiothreitol was incubated for 4 h, then the Ce6-fDNA probe was into the × 10−6 m ABDA under the 670 nm laser irradiation with the power intensity of 0.2 W for and respectively; the absorbance of ABDA was measured by an UV–vis of HCC In The selective uptake of Ce6-fDNA and Ce6-fDNADox probe by HepG2 cells and cells were investigated using confocal respectively. HepG2 cells × or cells × were and for 24 h at in the the medium was with fresh culture medium Ce6-fDNA probe, and further incubated for and respectively. the HepG2 cells or Hela cells were washed three with PBS at room the cells were by a confocal with 405 nm laser for Ce6. Ce6-fDNADOX was into the cells for 2 h, and then the medium was with fresh culture medium for an additional 2 h. the cells were by a confocal with 488 nm laser for DOX nm) and 405 nm laser for Ce6 of Ce6-fDNADOX In Cell Counting (CCK8) was used to the cell toxicity of Ce6-fDNA (Ce6, 2.05 × 10−6 Ce6-fDNADOX (Ce6, 2.05 × 10−6 m; 39.5 × 10−6 and Ce6-fNDNADOX (Ce6, 2.05 × 10−6 m; 39.5 × 10−6 m) HepG2 cells and cells. The cells were in a at a of 1 × cells well and incubated in a for 24 h. the medium was with fresh culture medium Ce6-fDNA, Ce6-fDNADOX or Meanwhile, the cells incubated with cell culture medium only were prepared as The medium was after 4 h then were washed with PBS at room to Afterward, fresh culture medium was and the cells were exposed to 670 nm laser irradiation (0.2 W for 10 After laser irradiation, the cells were incubated with fresh culture medium at for 48 h. Cell was as Cell = × The and are the absorbance of the treated cells (as and the control cells respectively. The was the absorbance of at were performed in of Cell by Ce6-fDNADox The is used to drug induced cell HepG2 cells was first into a at a of 1 × cells well at in a for 24 h. The cells then were washed three with PBS to dead cells, followed by incubation with Ce6-fDNADox the probe as that were in a culture medium at for 4 h. Next, the cells were washed by PBS to the Ce6-fDNADox probe, and then exposed to 670 nm laser irradiation (0.2 W for 10 Afterward, the cells were incubated with fresh culture medium at for 48 h. the cells were and in buffer, and the and were the Afterward, were incubated in for 15 min at room and then analyzed using flow analysis of was performed using of or the was as the were shown as through at three was by the of and the and of the and of the of and the of and the of and and the of and As a to our and this information by the are and may be for but are not or from information than be to the The is not for the or of information by the than be to the corresponding for the