Acidic pH and High-H2O2 Dual Tumor Microenvironment-Responsive

Research Article Cite This: ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

www.acsami.org

Acidic pH and High‑H2O2 Dual Tumor Microenvironment-Responsive Nanocatalytic Graphene Oxide for Cancer Selective Therapy and Recognition Baoping Lin, Heting Chen, Danyang Liang, Wei Lin, Xiaoyang Qi, Hanping Liu,* and Xiaoyuan Deng* MOE Key Laboratory of Laser Life Science, College of Biophotonics, South China Normal University, Guangzhou 510631, Guangdong, China

ACS Appl. Mater. Interfaces Downloaded from pubs.acs.org by UNIV FRANKFURT on 03/16/19. For personal use only.

S Supporting Information *

ABSTRACT: It is well known that tumors have an acidic pH microenvironment and contain a high content of hydrogen peroxide (H2O2). These features of the tumor microenvironment may provide physiochemical conditions that are suitable for selective tumor therapy and recognition. Here, for the first time, we demonstrate that a type of graphene oxide nanoparticle (N-GO) can exhibit peroxidase-like activities (i.e., can increase the levels of reactive oxygen species (ROS)) under acidic conditions and catalyze the conversion of H2O2 to ROShydroxyl radicals (HO·) in the acidic microenvironment in Hela tumors. The concentrated and highly toxic HO· can then trigger necrosis of tumor cells. In the microenvironment of normal tissues, which has a neutral pH and low levels of H2O2, N-GOs exhibit catalase-like activity (scavenge ROS) that splits H2O2 into O2 and water (H2O), leaving normal cells unharmed. In the recognition of tumors, an inherent redox characteristic of dopamine is that it oxidizes to form dopamine− quinine under neutral (pH 7.4) conditions, quenching the fluorescence of N-GOs; however, this characteristic has no effect on the fluorescence of N-GOs in an acidic (pH 6.0) medium. This pH-controlled response provides an active targeting strategy for the diagnostic recognition of tumor cells. Our current work demonstrates that nanocatalytic N-GOs in an acidic and high-H2O2 tumor microenvironment can provide novel benefits that can reduce drug resistance, minimize side effects on normal tissues, improve antitumor efficacy, and offer good biocompatibility for tumor selective therapeutics and specific recognition. KEYWORDS: graphene oxide nanoparticles (N-GOs), dual tumor microenvironment response, hydroxyl radicals (HO·), nanocatalytic therapy, tumor recognition

1. INTRODUCTION

acidic pH) can be exploited as a strategy for selective tumor therapy and tumor cell recognition. In recent years, nanocatalysts for tumor therapy have attracted much interest. Nanocatalytic therapies offer many advantages, such as high enzymatic activity, multiple functionalities, and stability, which render them superior to natural enzymes such as horseradish peroxidase (HRP) and traditional biomimetic catalysis by simulated enzymes in the field of tumor therapy.9 Many metallic materials, such as Gd@ HRP nanodots,10 Fe3O4 NPs,11 MnO2 NPs,4 and rFeOxNs,12 have been used as nanocatalysts for tumor therapy. However, drug carrier mesoporous silica nanoparticles are difficult to degrade in vivo and aggregate spontaneously. The potential toxicity of the metal component for normal tissues also may hinder their further clinical usefulness. The instability of natural enzyme activities limits their application. However, a few metal-free materials (such as nitrogen-doped carbon

Multidrug resistance in tumors is a frustrating challenge in the treatment of cancer, including chemotherapy, immunotherapy, and drug-targeting therapy. Certain characteristics of the tumor microenvironment, such as hypoxia, acidic pH, and irregular tumor blood vessels, are important factors that contribute to multidrug resistance. Due to the unique microenvironment of tumors, certain factors, such as an acidic pH and a high concentration of hydrogen peroxide (H2O2), can be used as a trigger to overcome multidrug resistance.1 The abundance of hydrogen peroxide produced in tumor cells is due to the disproportionation of superoxide dismutase in mitochondria.2,3 Moreover, hypoxia and the rapid proliferation of tumor cells contribute to the generation of hydrogen peroxide.4 The acidic microenvironment of tumors (pH 5.6 to 6.8) is a hallmark of malignant tumor cells and is due to glycolysis in tumor cells, hypoxia, and insufficient blood perfusion.5,6 However, in normal cells, glycolysis is inhibited by the presence of oxygen.7 The pH in normal tissues is tightly controlled at approximately 7.4.5,8 Abundantly available tumor metabolites (H2O2 and © XXXX American Chemical Society

Received: December 26, 2018 Accepted: March 6, 2019

A

DOI: 10.1021/acsami.8b22487 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces Scheme 1. Schematic Illustration of Nanocatalytic N-GOsa

(a) Catalytic mechanisms for N-GOs in a tumor microenvironment and in a normal cell microenvironment. (b) Developed turn-on fluorescence sensor based on N-GOs with dopamine (DA) for diagnostic assays of cancer cells.

a

nanomaterials (N-CNMs)13) have been developed and used as nanocatalysts for treating tumors. N-CNMs require H-ferritin nanoparticles to target tumor cells. These nanocatalysts are limited by their targeting efficiency and low therapeutic selectivity. Therefore, the development of improved nanocatalytic materials for tumor cell therapy that are not toxic and have excellent selectivity is required. Two-dimensional hexagonal lattice graphene materials have attracted much interest in nanobiomedicine for their outstanding biocompatibility and high specific surface areas.14 Graphene oxide (GO), as one important graphene derivative, has shown potential for use in biosensors and bioimaging15 and gene and drug delivery,16 as well as for photothermal therapy.17 GO as a substrate can accelerate cell adhesion and proliferation for wound healing.18 GO materials have been effective against bacterial depositions due to superoxide-anionindependent oxidation.19 Other researchers have further demonstrated that the antibacterial activity of GO increases as the GO area decreases.20 Moreover, ultrasmall GO

nanoparticles exhibited lower cytotoxicity toward cells than many large-sized GO.21 In addition, nanoscale GO may facilitate its biodegradation in cellular microenvironments.22 Here, we indicate for the first time that graphene oxide nanoparticles (N-GOs) mimic peroxidase-like catalytic activity; that is, they produce reactive oxygen species (ROS)-hydroxyl radicals (HO·) in an acidic tumor microenvironment. The presence of high levels of H2O2 in the tumor microenvironment increases the therapeutic effect of N-GOs on tumor cells. Thus, N-GOs may markedly and efficiently suppress tumor growth in a dual tumor microenvironment response (Scheme 1a). However, N-GOs perform a catalase-like activity (scavenge ROS) in the neutral microenvironment of normal cells and convert H2O2 to oxygen (O2) and water (H2O), leaving normal cells unharmed. The efficient identification of tumor cells within a large population of cells is of immense importance for the diagnosis and prognosis of cancers. The acidic tumor microenvironment is a vital consideration for the identification and detection of tumor cells. The inherent redox B


Research Article

ACS Applied Materials & Interfaces

Figure 1. Characterization of N-GOs. (a) Image of suspensions of GO, N-GOs, and rGO. (b,c) TEM images of N-GOs. (d) UV−vis absorption spectra of GO and N-GOs dispersed in water. (e) Raman spectra of GO and N-GOs. (f) FTIR spectra of GO and N-GOs.

Figure 2. 3D spatial Raman mapping for GO and N-GOs. (a) GO at ∼1358 cm−1. (b) GO at ∼1608 cm−1. (c) N-GOs at ∼1337 cm−1. (d) N-GOs at ∼1613 cm−1. The field of view is 28 μm × 31 μm with a pixel-to-pixel distance of 4 μm.

C


Research Article


Figure 3. In vitro characterization of the catalytic performance of N-GOs. (a) Catalytic oxidation of 2.0 mM 3,3′,5,5′-tetramethylbenzidine (TMB); absorption spectrum in 1.97 M H2O2 with 0.15 mg·mL−1 N-GOs. (b) pH sensitive peroxidase-like catalytic activity of 0.11 mg·mL−1 NGOs. (c) Change in the N-GO catalytic absorbance ratio before and after adding H2O2 at different pH values. A is the absorbance value in the presence of H2O2, and A0 is the absorbance value in the absence of H2O2. (d) Spin-trapped ESR spectrum of HO· radicals in the H2O2/DMPO system with N-GOs. (e) Time-dependent absorbance spectra for the detection of 1.97 M H2O2 in 0.15 mg·mL−1 N-GOs. (f) Comparison of the time dependent absorption at 650 nm with 0.226 and 0.300 mg·mL−1 N-GOs. (g) N-GO absorbance spectrum after the addition of different concentrations of H2O2 (0, 0.49, 0.78, 1.08, 1.66, and 1.97 M N-GOs). (h) Linear calibration plot for the N-GO peroxidase-like catalytic reaction for H2O2 detection.

reveals strong G bands at 1608 cm−1 for GO and at 1613 cm−1 for N-GOs that are due to vibration of the CC bond in the lattice and weak D bands at 1358 cm−1 for GO and at 1337 cm−1 for N-GOs that are attributed to pronounced margin defects in graphene oxide.23 The ID/IG ratio for GO (1.11) is slightly lower than that for N-GOs (2.23), confirming that the N-GOs possess more structural defects than GO.23,26 Atomic concentrations of the GO and the N-GOs were determined by FTIR (Figure 1f). The infrared absorption spectrum of N-GOs is more intense than that of GO, which may be due to the breaking of symmetry in the N-GO structure or to an increase of the molecular dipole moment, which increases the intensity of the absorption peak. An absorption peak for the C−H bond at 816 cm−1 for N-GOs indicates that C−C bonds of GO were broken during the synthesis of the N-GOs. 2.2. Raman Mapping of N-GOs and GO. Figure 2a,b shows the Raman mapping of GO from 28 μm × 31 μm regions. The peak strengths for GO approximately 1358 and 1608 cm−1 were very similar. As shown in Figure 2, the symmetry of the GO structure is better, indicating that the graphene oxide nanostructure array has better structural symmetry and uniformity.27 However, each point in the NGO region shows significant differences in peak strengths between ∼1337 and ∼1613 cm−1, indicating that the symmetry differences for the N-GO structures may be explained by oxidation of the surrounding peaks, resulting in more defects. 2.3. In Vitro Peroxidase-like Catalytic Activity of NGOs. As shown in Figure 3a, the N-GO + H2O2 + TMB solution displayed a blue color, and a characteristic absorption

characteristic of dopamine (DA) is oxidation to dopamine− quinine, which can quench the fluorescence of N-GOs in the neutral microenvironment of normal cells, but this characteristic has almost no effect on the fluorescence of N-GOs in an acidic tumor microenvironment (Scheme 1b). Fluorescent NGOs that contain dopamine exhibit an outstanding capacity for specifically recognizing tumor cells in a tumor microenvironment. Unlike many targeted drug treatments, N-GOs as nanocatalysts use a dual tumor microenvironment response to improve the efficacy of the treatment and minimize side effects. Furthermore, N-GOs are metal-free materials with good biocompatibility and low toxicity and have no requirement for a carrier.

2. RESULTS 2.1. Characterization of N-GOs. The N-GOs were synthesized using a chemical oxidation method. The N-GO images, shown in Figure 1a, demonstrate the solubility and stability in water after 1 month. However, larger nanometersized graphene oxide and graphene exhibit aggregation in an aqueous solution. This behavior is an indication that the additional oxygen functionalities render the N-GOs more polar.23 Figure 1b,c shows TEM images of N-GOs with a size range from 100 to 220 nm. N-GOs and GO display absorption peaks at 233 and 235 nm (Figure 1d), representing absorption peaks for the π bond and a functional group on the benzene ring, respectively.24 The position of the peak shifts to shorter wavelengths with decreasing nanosheet thickness or with breaking of CC double bonds.25 Raman spectra (Figure 1e) D


Research Article


Figure 4. In vitro cytotoxicity profiles and intracellular catalytic mechanisms. (a) In vitro Hela tumor cell cytotoxicity profiles for N-GOs under acidic and neutral conditions. (b) Raman spectroscopy of Hela tumor cells at 0.58 mg·mL−1 N-GOs under acidic and neutral conditions. (c) Live/ dead assay of Hela tumor cells incubated with N-GOs at 0, 0.11, 0.30, and 0.58 mg·mL−1 under acidic and neutral conditions for 24 h. The cells were stained using calcein AM/PI. (d) Hela tumor cells were contained with 0.057 and 0.11 mg·mL−1 N-GOs in acidic and neutral media for 24 h and contained with the ROS fluorescent probe DCFH-DA.

spectra at 650 nm of chromogenic TMB after oxidation by the N-GOs is shown in Figure 3e. The absorption time course curves change at 650 nm (Figure 3f), demonstrating that the absorption peak increases as the reaction time increases within a certain range. The 650 nm absorption peak reaches an optimum at 350 s and then decreases afterward. We also detected a relationship between the absorbance and the hydrogen peroxide concentration (Figure 3g). The absorbance of chromogenic TMB at 650 nm increased gradually with the TMB concentration (from 0 to 1.969 M) indicating that the peroxidase-like reaction catalyzed by the N-GOs is dependent on the hydrogen peroxide concentration. As shown in Figure 3h, the absorbance change at 650 nm displays a linear correlation with the H2O2 concentration from 0 to 1.969 M. The fitted curve equation was A = −0.00673 + 0.14156 M CH2O2 with a correlation coefficient of 0.97061. The oxidation of TMB to the chromogenic TMB by N-GOs is affected by the H2O2 concentration, which exhibited a good linear relationship. Corresponding Michaelis−Menten kinetics of N-GOs was analyzed (Figure S1). The N-GOs catalyzed the disproportionation of H2O2 to generate highly toxic hydroxyl radicals (HO·) at the tumor site under acidic conditions, while generating nontoxic H2O and O2 under neutral pH conditions. Thus, N-GOs perform a peroxidase-like activity under acidic conditions that can produce therapeutic effects on most tumor cells. 2.4. In Vitro Therapeutic Efficacy of N-GOs. Hydrogen peroxide is generated as a by-product of cellular metabolism and causes minor damage to normal/tumor cells under normal

peak for chromogenic TMB was observed at 650 nm, while there was no significant color change and negligible absorption at 650 nm in the TMB + H2O2 solution, the TMB + N-GO solution, or the GO + H2O2 + TMB solution. These results indicate that the N-GOs dramatically enhance the peroxidaselike catalytic activity compared with GO. The peroxidase-like catalytic activity of the N-GOs was dependent upon the pH as shown in Figure 3b. However, the N-GOs did not display peroxidase-like catalytic activity under neutral conditions. This result indicates that the N-GOs display peroxidase-like activity under acidic conditions. Figure 3c further shows the change in absorbance ratios as a function of pH. At pH 6.0, the absorbance value increased 17-fold compared with pH 7.4. The catalytic mechanism of peroxidase-like reactions includes electron transfer and the production of HO·.12 The electron spin resonance (ESR) technique was employed to investigate free radical production that was induced by N-GOs in the presence of H2O2 in acidic conditions. To trap hydroxyl radicals that are short-lived, 5,5-dimethyl-1-pyrroline N-oxide (DMPO) can be used to react with free radicals to form radical adducts that can be detected by ESR.12 An intense characteristic HO·/DMPO peak (1:2:2:1) was observed under acidic conditions (pH 6.0) (Figure 3d), indicating that hydroxyl radicals were generated by the disproportionation of H2O2, which oxidized TMB to chromogenic TMB under acidic conditions. There was a Fenton-like reaction that the peroxidase-like activity of the N-GOs catalyzed. The N-GOs greatly improved the peroxidase-like activity under acidic conditions. The time-dependent increase in the absorption E


Research Article


Figure 5. (a) Fluorescence emission (λex = 400 nm) spectra of N-GOs (0.20 mg·mL−1) and GO (0.20 mg·mL−1) dispersed in water. (b) Fluorescence of an N-GO solution upon addition of a dopamine (DA) solution (0.20 mg·mL−1) (λex = 400 nm) in pH 6.0 or 7.4 buffered solutions. (c) An inherent redox characteristic of dopamine (DA) involves oxidation to form dopamine−quinine under acidic pH conditions. CLSM images of the probe 0.20 mg·mL−1 N-GOs and 0.20 mg·mL−1 dopamine (DA) incubated with living Hela tumor cells for 24 h in (d) an acidic (pH 6.0) culture medium and (e) in a neutral (pH 7.4) culture medium.

conditions. N-GOs can catalytically degrade H2O2 to produce free radical reactive oxygen species (ROS). The ROS can cause oxidative stress, resulting in oxidative damage to biological systems such as lipids, proteins, and DNA.24,28 To evaluate the cytotoxicity of N-GOs, Hela tumor cells were coincubated with N-GOs at varying concentrations (0, 0.11, 0.30, and 0.58 mg· mL−1) in an acidic culture medium at pH 6.0 and in a neutral culture medium at pH 7.4 for 24 h. Toxicity assessment was conducted using a Cell Counting Kit-8 (CCK-8) assay. The results indicate that cell viability is strongly dependent on the concentration of the N-GOs and the pH (Figure 4a). The cell viability assays demonstrated that the N-GOs were not cytotoxic to cells in a neutral (pH 7.4) culture medium. At normal substrate concentrations, the N-GOs could enhance cell growth in a culture medium at pH 7.4. However, when cells were treated simultaneously with the N-GOs in acidic (pH 6.0) culture media, high concentrations of N-GOs caused a significant loss of cellular activity. The results indicate that the N-GOs induce significantly higher cytotoxicity under acidic conditions. Raman spectroscopy can be used to determine a chemical fingerprint of cells under different conditions to evaluate

cellular activity, because they can provide chemical information on cellular membranes.29,30 Raman signals for Hela cells corresponding to typical Raman peaks at ∼1601, ∼1448, ∼1328, and ∼1195 cm−1 were assigned to CC stretching, CH2 bending, in-phase CH2 twisting, and CC−H in-plane bending, respectively.31,32 Hela cells were incubated with NGOs for 24 h and then subjected to Raman spectroscopy to obtain information on cellular molecules (Figure 4b). Compared with the neutral medium (pH 7.4), the Raman shifts for cells in an acidic medium (pH 6.0) were weaker at ∼1601, ∼1448, ∼1328, and ∼1195 cm−1, indicating that incubation of Hela cells with N-GOs for 24 h in an acidic (pH 6.0) culture medium can destroy the molecular structure of Hela cells and induce necrosis of tumor cells. The distribution of viable and dead Hela tumor cells can be investigated visually using calcein AM (green fluorescence) and propidium iodide (PI, red fluorescence) (Figure 4c). Hela cells were treated with varying concentrations of N-GOs (0, 0.11, 0.30, and 0.58 mg·mL−1) in a culture medium at pH 7.4 or in a culture medium at pH 6.0 for 24 h. Then, living and dead cells were stained with calcein AM or PI to produce green or red fluorescence, respectively. Images obtained by confocal F


Research Article


analyses were performed on BALB/c mice 3 days after injection. Changes in WBC, RBC, HGB, and RDW counts may reflect any abnormality in the immune system. The normal saline control group and the N-GO experimental group were compared. No significant difference was observed for the WBC, RBC, HGB, and RDW counts between the normal and the N-GO groups (P < 0.05), indicating that N-GOs were not toxic to the immune system of mice as a result of the treatment (Table 1). To evaluate therapeutic performance in vivo, N-

laser scanning microscopy (CLSM) revealed that no significant or only a minor damage was evident in the control group with 0.11 or 0.30 mg·mL−1 N-GOs in a pH 7.4 culture medium and in a pH 6.0 culture medium, respectively. Large numbers of dead cells were observed when N-GOs were incubated at certain concentrations (0.58 mg·mL−1) in a pH 6.0 culture medium. By comparison, in a pH 7.4 culture medium, a vast majority of Hela cancer cells were still alive with 0.58 mg·mL−1 N-GOs. This result can be explained by N-GOs catalyzing the breakdown of H2O2 in tumor cells under physiological acidic conditions to generate HO·, which is highly toxic to cancer cells. 2′,7′-Dichlorofluorescein diacetate (DCFH-DA), a green fluorescence-emitting indicator that can be used to measure intracellular ROS levels,24 was used to stain Hela tumor cells after the cells were treated with 0.057 or 0.11 mg·mL−1 N-GOs or an untreated control group in pH 6.0 or pH 7.4 culture media to determine the levels of ROS (Figure 4d). Green fluorescence was barely visible in the control group or under neutral (pH 7.4) conditions, revealing the insignificant extent of ROS generation in the control group at neutral pH (7.4). Conversely, the intensity of green fluorescence can be clearly observed under acidic conditions (pH 6.0), implying high intracellular levels of ROS. Corresponding fluorescence intensity distribution was collected and analyzed (Figure S2). Therefore, the peroxidase-like activity of the N-GOs can be attributed to a burst of ROS in cells under acidic (pH 6.0) conditions while leaving normal cells unharmed. 2.5. Fluorescence of N-GOs and Recognition of Tumor Cells. We recorded the UV−vis and Raman spectra of N-GOs/DA (Figures S3 and S4). Upon excitation at 400 nm, the N-GOs show a strong fluorescence spectrum with a maximum at 510 nm (Figure 5a). However, GO does not display a fluorescence absorption peak. Normalized PL emission spectra of the N-GOs and GO at different excitation wavelengths are shown in Figure S5. Figure 5b shows typical emission spectra for the N-GOs. Upon addition of dopamine (DA) to the N-GOs in pH 6.0 and pH 7.4 buffered solutions, the fluorescence of the N-GOs is quenched significantly at pH 7.4 compared with that observed at pH 6.0, indicating that NGOs/DA can act as an efficient fluorescence sensor in the acidic microenvironment of Hela tumor cells. N-GO fluorescence quenching was enhanced under neutral (pH 7.4) conditions, which is due to the generation of dopamine− quinine under neutral (pH 7.4) conditions (Figure 5c).33 To study the potential of N-GOs with DA for recognition of the acidic microenvironment of Hela tumor cells, N-GOs and DA were incubated with Hela cells in acidic (pH 6.0) and neutral (pH 7.4) culture media. The confocal laser scanning microscopy (CLSM) images obtained are shown in Figure 5d,e. Laser confocal images of Hela cells (λex = 480 nm) are shown in Figure S6. As can be seen, Hela tumor cells show strong fluorescence in an acidic (pH 6.0) culture medium. In sharp contrast, very weak fluorescence was observed with Hela cells in a neutral (pH 7.4) culture medium, indicating that the N-GOs/DA can differentiate the acidic microenvironment of tumor cells from the neutral microenvironment of normal cells (pH 7.4). These data provide strategies for the recognition of tumor cells based on extracellular pH and show great potential as biological labels for tumor diagnosis. 2.6. In Vivo Cancer Therapeutic Efficacy. To assess the toxicity of the N-GOs in vivo, N-GOs were intravenously injected into healthy female BALB/c mice. Routine blood

Table 1. Determination of the Blood Components of BALB/c Micea group index WBC (109/L) RBC (1011/L) HGB (g/L) RDW (%)

N-GOs 8.15 10.75 177.67 18.3

± ± ± ±

1.2 0.08 20.33 1.09

control 6.16 10.46 180.33 18.97

± ± ± ±

0.56 0.15 30.33 0.56

a

WBC: leukocyte; RBC: red blood cell; HGB: hemoglobin; RDW: coefficient of variation in RBC distribution width. Statistical analyses show that there are no significant differences between N-GOs and control group (P < 0.05, n = 4).

GOs were tested on Hela cervical tumor cell xenografts in BALB/c nude mice. Saline (control group) was administered intratumorally to evaluate therapeutic performance. To qualitatively investigate the antitumor efficacy of the N-GOs, the growth rates of tumors were continuously monitored after injection of N-GOs (Figure 6a,b). The tumors in the salinetreated control group exhibited the most significant growth. However, as shown in Figure 6a, the tumors in mice treated with N-GOs exhibited the greatest extent of growth inhibition, suggesting that the N-GOs efficiently suppressed tumor growth. The inhibitory mechanism for the N-GOs could be attributed to the generation of highly toxic HO· free radicals by N-GO-catalyzed degradation of hydrogen peroxide in an acidic tumor microenvironment. To further prove the therapeutic effect of the N-GOs, microscopic images of hematoxylin and eosin (H&E)-stained tumor slices were obtained for histological investigation (Figure 6c). As expected, the membrane and nuclei of tumor cells from the saline-treated control group exhibited a normal morphology. However, tumors that received the combination therapy with the N-GOs experienced severe destruction of their membrane morphology and nuclear structures, further indicating that nanocatalytic therapy with N-GOs is very effective in inhibiting tumor growth. To evaluate the safety of the N-GOs in nude mice, histological examinations of H&E-stained organ slices were conducted in N-GO-injected and saline control groups after 8 days. Figure 6d indicates that there are no noticeable signs of toxic side effects, demonstrating that the N-GOs exerted low toxicity in these animals.

3. CONCLUSIONS In this paper, a type of graphene oxide nanoparticle (N-GO) that exhibits peroxidase-like activity (i.e., can increase the levels of reactive oxygen species (ROS)) on acidic pH was synthesized. In the tumor microenvironment with acidic pH and high H2O2, N-GOs catalyze H2O2 into highly toxic hydroxyl radicals (HO·), which trigger necrosis of tumor cells. On the other hand, in a normal cell microenvironment with neutral pH, N-GOs exhibit catalase-like activity (scavenge G


Research Article


Figure 6. In vivo therapeutic efficacy of N-GOs against Hela tumor xenografts. (a) Tumor growth curves for N-GO-treated and saline control groups. (b) Photographs of each group of dissected tumors after 8 days of therapy. (c) Histological H&E analyses of the tumor sections from saline and N-GO-treated tumor-bearing mice. All scale bars are 200 μm. (d) Histological H&E staining of major organs in nude mice after 8 days. The scale bar is 200 μm (P < 0.001, n = 4).

ROS) and cause no damage to normal cells. N-GOs, therefore, are highly tumor microenvironment selective. With dopamine (DA) conjugation, N-GOs become a novel probe for tumor cell recognition. The inherent redox characteristic of dopamine (DA) determines it will be oxidized to form dopamine− quinine under neutral (pH 7.4) normal tissues, thus quenching the fluorescence of N-GOs; however, under an acidic (pH 6.0) tumor microenvironment, no dopamine−quinine is formed and the fluorescence in N-GOs is not affected. This pHcontrolled response provides an active targeting strategy for tumor cell recognition. N-GOs were synthesized by chemical oxidation and were shown to have good solubility and stability in water compared to graphene and graphene oxide. Moreover, the N-GOs disproportionate H2O2 into highly toxic hydroxyl radicals (HO·) and exhibit peroxidase-like activity under acidic conditions. This catalytic reaction demonstrates the high catalytic performance of the N-GOs. The high concentration of N-GOs catalyzes the disproportionation of H2O2 into highly toxic hydroxyl radicals in the acidic microenvironment of tumor cells and triggers tumor cell necrosis. The N-GOs/DA probe possesses very weak fluorescence under neutral conditions but can increase fluorescence under acidic conditions. Thus, we can design a fluorescence probe based on N-GOs with dopamine for the detection of tumor cells. Our findings demonstrate that N-GOs can provide effective tumor therapy without causing adverse side effects on normal tissues/

organs. This platform can act as a biosensor and can enhance the effectiveness of cancer treatment.

4. METHODS 4.1. Synthesis of N-GOs. Five milligrams of graphene oxide was mixed with 10 mL of H2O and then ultrasonicated at room temperature for 30 min. Then, 10 mL of 30% H2O2 and 150 μL of 25% NH3·H2O were added to the mixture, and it was stirred for 30 h at 80 °C. The N-GOs were evaporated and condensed at 80 °C to remove excess H2O2 and NH3·H2O. The solution was filtered with a PVDF (0.22 μm) syringe filter (Millipore), and the N-GOs were rinsed three times with deionized water. 4.2. Characterization of N-GOs. TEM images were obtained with a JEM-2010HR transmission electron microscope (JEOL, Japan). The fluorescence emission spectra were recorded by a PerkinElmer LS55 luminescence spectrophotometer. Raman spectra were obtained with a Raman spectrometer (Derbyshire, England). Ultraviolet−visible−near-infrared (UV−vis−NIR) absorption spectra were obtained using a UV−vis−NIR spectrometer (UV-3200S, MAPADA, China). Infrared spectroscopy of N-GOs was conducted with a Fourier transform infrared spectrometer (Nicolet 6700, America). Electron spin resonance (ESR) spectra were obtained with an ESR spectrometer (E-Scan, USA). 4.3. Michaelis−Menten Kinetics of N-GOs. Four hundred microliters of 5 mM TMB was added to monitor the chromogenic reaction (λ = 650 nm) of 0.15 mg mL−1 N-GOs or 0.15 mg·mL−1 GO upon the addition of 1.969 M H2O2 and 20 mmol·L−1 KH2PO4/ NaOH buffer solution at 35 °C for 10 min. 4.4. Raman Spectra of GO, N-GOs, and N-GOs/DA. The Raman spectrum of GO was recorded by 514 nm laser. The Raman spectra of N-GOs and N-GOs/DA were recorded by 785 nm laser. H



■

The laser power of the sample was 10 mW, the exposure time was 8 s, and the accumulation number was 4. 4.5. Fluorescence Measurements. Further, a 0.20 mg·mL−1 NGO suspension containing 0.20 mg·mL−1 dopamine and 20 mmol·L−1 KH2PO4/NaOH buffer solution was used for all experiments at 35 °C for 2 h, and fluorescence spectra were recorded. The emission intensity was measured at 510 nm with an excitation wavelength of 400 nm. 4.6. In Vitro Cytotoxicity of N-GOs. Hela cells were added to the wells of a 96 well plate at a density of 3000 cells per well and incubated for 12 h. The medium was discarded, and the cells were rinsed twice with phosphate-buffered saline (PBS). The concentrations of N-GOs were 0, 0.11, 0.30, and 0.58 mg·mL−1, and 10% FBS containing DMEM high-glucose medium was added. Then, a buffer solution (KH2PO4/NaOH, pH 6.0) was added, and the pH was adjusted to 6.0. After 24 h of incubation at 37 °C, the cell culture medium was discarded, and the cells were washed three times with PBS. Cell counts were obtained with a Kit-8 (CCK-8 assay) containing DMEM high-glucose medium at 100 μL per well and read in a microplate reader after 2 h. 4.7. Raman Spectra of Hela Cells. The Raman spectra of Hela cells were obtained by 785 nm excitation. The laser power of the sample was 10 mW, the exposure time was 20 s, and the accumulation number was 8. 4.8. Observations of Viable and Dead Cells by CLSM. Hela cells were cultured in φ 15 CLSM-exclusive culture disks for 12 h, and then N-GOs and GO were added. The cells were cultured for 24 h, stained with 3′,6′-di(O-acetyl)-4′,5′-bis(N,N-bis(carboxymethyl)aminomethyl)fluorescein tetraacetoxymethyl ester (calcein AM)/ propidium iodide (PI), and then incubated for 15 min. Confocal fluorescence spectra were recorded by 480 nm and 545 nm lasers. 4.9. Observation of ROS by CLSM. 1 × 105 Hela cells were cultured in a φ 15 CLSM-exclusive culture disk with DMEM medium for 12 h; the medium was then discarded, the cells were washed twice with PBS, and the DMEM (pH 7.4 or 6.0) containing N-GOs was replaced. After 24 h, the fluorescence probe 2′,7′-dichlorofluorescin diacetate (DCFH-DA) was added, and 15 min later, the fluorescence substance DCF (λex = 480 nm, λem = 525 nm) could be observed under a confocal microscope. 4.10. Confocal Fluorescence Imaging. Hela cells were placed on the φ 15 CLSM-exclusive culture disk with DMEM medium and incubated for 12 h. The medium was replaced with a fresh medium containing 0.20 mg·mL−1 N-GOs and 0.20 mg·mL−1 dopamine (DA) and then incubated for 24 h. Finally, the cells were washed three times with PBS and imaged by confocal microscopy (excitation at 545 nm for the red channel (560−650 nm)). 4.11. Toxicology Experiment in Vivo. Eight healthy female BALB/c mice were divided into two groups (10 mg·kg−1, and saline control group). N-GOs were injected intravenously into the female BALB/c mice. The mice were intravenously injected with N-GOs. After 3 days, blood from the eyes of nude mice was taken for routine blood testing. 4.12. Tumor Mouse Model. All female BALB/c nude mice were purchased from the Experimental Animal Center of the Foshan Medical Animal Experimental Center in Guangdong Province and were approved by the Animal Protection and Utilization Committee of the South China Normal University. To establish a 6 week BALB/c mouse tumor model, 4 × 106 Hela cells were suspended in 200 μL of serum-free DMEM (containing 1% pen/strep, 100 U/mL penicillin, and 100 μg/mL streptomycin) and injected subcutaneously into the necks of nude mice. Two weeks later, when the tumor volume reached 200 mm3, saline (200 μL) (n = 4) and N-GOs (200 μL, 10 mg/mL) (n = 4) were injected intratumorally into the nude mice. Tumor sizes were measured with digital Vernier calipers every 2 days. The following formula was used: tumor volume estimation (tumor length) × (tumor width)2/2. The normalized tumor volume was calculated as V/V0, where V0 was the initial tumor volume before treatment. The main organs of each group were taken and histologically analyzed by hematoxylin and eosin (H&E) staining.

Research Article

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.8b22487. Lineweaver−Burk plot of N-GOs; intensity profiles of confocal images in Figure 4d representing ROS generation in Hela cells; UV−vis absorption spectra of N-GOs, N-GOs/DA, and DA; Raman spectroscopy of N-GOs/DA under acidic (pH 6.0) and neutral (pH 7.4) buffer solutions; normalized PL emission spectra of the N-GOs and GO at different excitation wavelengths; CLSM images of the probe N-GOs/dopamine (DA) in living Hela cells (λex = 480 nm) (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (X.D.). *E-mail: [email protected] (H.L.). ORCID

Baoping Lin: 0000-0001-8883-9238 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This study was supported by the National Natural Science Foundation of China (Nos. 81671729 and 81171379) and the Science and Technology Project of Guangzhou (No. 201805010002).

■

REFERENCES

(1) Chen, S.; Yang, K.; Tuguntaev, R. G.; Mozhi, A.; Zhang, J.; Wang, P. C.; Liang, X.-J. Targeting Tumor Microenvironment with PEG-based Amphiphilic Nanoparticles to Overcome Chemoresistance. Nanomedicine 2016, 12, 269−286. (2) López-Lázaro, M. Dual Role of Hydrogen Peroxide in Cancer: Possible Relevance to Cancer Chemoprevention and Therapy. Cancer Lett. 2007, 252, 1−8. (3) Buettner, G. R.; Ng, C. F.; Wang, M.; Rodgers, V. G. J.; Schafer, F. Q. A New Paradigm: Manganese Superoxide Dismutase Influences the Production of H2O2 in Cells and Thereby Their Biological State. Free Radical Biol. Med. 2006, 41, 1338−1350. (4) Prasad, P.; Gordijo, C. R.; Abbasi, A. Z.; Maeda, A.; Ip, A.; Rauth, A. M.; DaCosta, R. S.; Wu, X. Y. Multifunctional AlbuminMnO2; Nanoparticles Modulate Solid Tumor Microenvironment by Attenuating Hypoxia, Acidosis, Vascular Endothelial Growth Factor and Enhance Radiation Response. ACS Nano 2014, 8, 3202−3212. (5) Justus, C. R.; Dong, L.; Yang, L. V. Acidic Tumor Microenvironment and pH-Sensing G protein-Coupled receptors. Front. Physiol. 2013, 4, 354. (6) Shan, L. 64Cu-1,4,7,10-Tetraazacyclododecane-1,4,7-Tris-Acetic Acid-10-Maleimidoethylacetamide-ACEQNPIYWARYADWLFTTPLLLLDLALLVDADEGTG. In Molecular Imaging and Contrast Agent Database; National Center for Biotechnology Information: Bethesda, MD, 2004. https://www.ncbi.nlm.nih.gov/ books/NBK23129/. (7) Gatenby, R. A.; Gillies, R. J. Why do Cancers have High Aerobic Glycolysis? Nat. Rev. Cancer 2004, 4, 891−899. (8) Dai, Y.; Ma, P.; Cheng, Z.; Kang, X.; Zang, X.; Hou, Z.; Li, C.; Yang, D.; Zhai, X.; Lin, J. Up-Conversion Cell Imaging and pHInduced Thermally Controlled Drug Release from NaYF4:Yb3+/ Er3+@Hydrogel Core−Shell Hybrid Microspheres. ACS Nano 2012, 6, 3327−3338. (9) Gao, L.; Yan, X. Nanozymes: an Emerging Field Bridging Nanotechnology and Biology. Sci. China: Life Sci. 2016, 59, 400−402. I


Research Article

ACS Applied Materials & Interfaces (10) Yang, W.; Shi, X.; Shi, Y.; Yao, D.; Chen, S.; Zhou, X.; Zhang, B. Beyond the Roles in Biomimetic Chemistry: An Insight into the Intrinsic Catalytic Activity of an Enzyme for Tumor-Selective Phototheranostics. ACS Nano 2018, 12, 12169−12180. (11) Huo, M.; Wang, L.; Chen, Y.; Shi, J. Tumor-Selective Catalytic Nanomedicine by Nanocatalyst Delivery. Nat. Commun. 2017, 8, 357. (12) Wang, L.; Huo, M.; Chen, Y.; Shi, J. Iron-Engineered Mesoporous Silica Nanocatalyst with Biodegradable and Catalytic Framework for Tumor-Specific Therapy. Biomaterials 2018, 163, 1− 13. (13) Fan, K.; Xi, J.; Fan, L.; Wang, P.; Zhu, C.; Tang, Y.; Xu, X.; Liang, M.; Jiang, B.; Yan, X.; Gao, L. In Vivo Guiding NitrogenDoped Carbon Nanozyme for Tumor Catalytic Therapy. Nat. Commun. 2018, 9, 1440. (14) Rong, P.; Yang, K.; Srivastan, A.; Kiesewetter, D. O.; Yue, X.; Wang, F.; Nie, L.; Bhirde, A.; Wang, Z.; Liu, Z.; Niu, G.; Wang, W.; Chen, X. Photosensitizer Loaded Nano-Graphene for Multimodality Imaging Guided Tumor Photodynamic Therapy. Theranostics 2014, 4, 229−239. (15) Yoon, H. J.; Kim, T. H.; Zhang, Z.; Azizi, E.; Pham, T. M.; Paoletti, C.; Lin, J.; Ramnath, N.; Wicha, M. S.; Hayes, D. F.; Simeone, D. M.; Nagrath, S. Sensitive Capture of Circulating Tumour Cells by Functionalized Graphene Oxide Nanosheets. Nat. Nanotechnol. 2013, 8, 735−741. (16) Weaver, C. L.; LaRosa, J. M.; Luo, X.; Cui, X. T. Electrically Controlled Drug Delivery from Graphene Oxide Nanocomposite Films. ACS Nano 2014, 8, 1834−1843. (17) Jung, H. S.; Kong, W. H.; Sung, D. K.; Lee, M.-Y.; Beack, S. E.; Keum, D. H.; Kim, K. S.; Yun, S. H.; Hahn, S. K. Nanographene Oxide-Hyaluronic Acid Conjugate for Photothermal Ablation Therapy of Skin Cancer. ACS Nano 2014, 8, 260−268. (18) Lee, W. C.; Lim, C. H. Y. X.; Shi, H.; Tang, L. A. L.; Wang, Y.; Lim, C. T.; Loh, K. P. Origin of Enhanced Stem Cell Growth and Differentiation on Graphene and Graphene Oxide. ACS Nano 2011, 5, 7334−7341. (19) Liu, S.; Zeng, T. H.; Hofmann, M.; Burcombe, E.; Wei, J.; Jiang, R.; Kong, J.; Chen, Y. Antibacterial Activity of Graphite, Graphite Oxide, Graphene Oxide, and Reduced Graphene Oxide: Membrane and Oxidative Stress. ACS Nano 2011, 5, 6971−6980. (20) Perreault, F.; de Faria, A. F.; Nejati, S.; Elimelech, M. Antimicrobial Properties of Graphene Oxide Nanosheets: Why Size Matters. ACS Nano 2015, 9, 7226−7236. (21) Zhang, H.; Peng, C.; Yang, J.; Lv, M.; Liu, R.; He, D.; Fan, C.; Huang, Q. Uniform Ultrasmall Graphene Oxide Nanosheets with Low Cytotoxicity and High Cellular Uptake. ACS Appl. Mater. Interfaces 2013, 5, 1761−1767. (22) Bianco, A. Graphene: Safe or Toxic? The Two Faces of the Medal. Angew. Chem., Int. Ed. 2013, 52, 4986−4997. (23) Botas, C.; Á lvarez, P.; Blanco, C.; Santamaría, R.; Granda, M.; Ares, P.; Rodríguez-Reinoso, F.; Menéndez, R. The Effect of the Parent Graphite on the Structure of Graphene Oxide. Carbon 2012, 50, 275−282. (24) Chong, Y.; Ge, C.; Fang, G.; Tian, X.; Ma, X.; Wen, T.; Wamer, W. G.; Chen, C.; Chai, Z.; Yin, J.-J. Crossover Between Anti- and ProOxidant Activities of Graphene Quantum Dots in the Absence or Presence of Light. ACS Nano 2016, 10, 8690−8699. (25) Backes, C.; Paton, K. R.; Hanlon, D.; Yuan, S.; Katsnelson, M. I.; Houston, J.; Smith, R. J.; McCloskey, D.; Donegan, J. F.; Coleman, J. N. Spectroscopic Metrics Allow in Situ Measurement of Mean Size and Thickness of Liquid-Exfoliated Few-Layer Graphene Nanosheets. Nanoscale 2016, 8, 4311−4323. (26) Cheng, M.; Yang, R.; Zhang, L.; Shi, Z.; Yang, W.; Wang, D.; Xie, G.; Shi, D.; Zhang, G. Restoration of Graphene from Graphene Oxide by Defect Repair. Carbon 2012, 50, 2581−2587. (27) Zhu, C.; Meng, G.; Huang, Q.; Wang, X.; Qian, Y.; Hu, X.; Tang, H.; Wu, N. ZnO-Nanotaper Array Sacrificial Templated Synthesis of Noble-Metal Building-Block Assembled Nanotube Arrays as 3D SERS-Substrates. Nano Res. 2015, 8, 957−966.

(28) Halliwell, B. Free Radicals, Antioxidants, and Human Disease: Curiosity, Cause, or Consequence? Lancet 1994, 344, 721. (29) Vendrell, M.; Maiti, K. K.; Dhaliwal, K.; Chang, Y.-T. SurfaceEnhanced Raman Scattering in Cancer Detection and Imaging. Trends Biotechnol. 2013, 31, 249−257. (30) Keisham, B.; Cole, A.; Nguyen, P.; Mehta, A.; Berry, V. Cancer Cell Hyperactivity and Membrane Dipolarity Monitoring Via Raman Mapping of Interfaced Graphene: Toward Non-Invasive Cancer Diagnostics. ACS Appl. Mater. Interfaces 2016, 8, 32717−32722. (31) Huang, Y.-S.; Karashima, T.; Yamamoto, M.; Hamaguchi, H.-o. Molecular-Level Investigation of the Structure, Transformation, and Bioactivity of Single Living Fission Yeast Cells by Time- and SpaceResolved Raman Spectroscopy. Biochemistry 2005, 44, 10009−10019. (32) Kaliaperumal, V.; Hamaguchi, H.-o. Casting New Physicochemical Light on the Fundamental Biological Processes in Single Living Cells by Using Raman Microspectroscopy. Chem. Rec. 2012, 12, 567−580. (33) Zhao, J.; Zhao, L.; Lan, C.; Zhao, S. Graphene Quantum Dots as Effective Probes for Label-Free Fluorescence Detection of Dopamine. Sens. Actuators, B 2016, 223, 246−251.

J


Acidic pH and High-H2O2 Dual Tumor Microenvironment-Responsive

Recommend Documents