Enhanced Radiosensitization of Gold Nanospikes via Hyperthermia in

Sep 30, 2016 - ... Cancer Radiotherapy through Synergistic External and Internal Regulations .... Advanced Healthcare Materials 2017 388, 1701211 ...
0 downloads 0 Views 3MB Size
Subscriber access provided by TRENT UNIV

Article

Enhanced Radiosensitization of Gold Nanospikes via Hyperthermia in Combined Cancer Radiation and Photothermal Therapy Ningning Ma, Yao-Wen Jiang, Xiaodong Zhang, Hao Wu, John Nicholas Myers, Peidang Liu, Haizhen Jin, Ning Gu, Nongyue He, Fu-Gen Wu, and Zhan Chen ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b10132 • Publication Date (Web): 30 Sep 2016 Downloaded from http://pubs.acs.org on October 5, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Enhanced Radiosensitization of Gold Nanospikes via Hyperthermia in Combined Cancer Radiation and Photothermal Therapy

Ningning Ma,† Yao-Wen Jiang,† Xiaodong Zhang,† Hao Wu,† John N. Myers,‡ Peidang Liu,§ Haizhen Jin,§ Ning Gu,† Nongyue He,† Fu-Gen Wu,*,† and Zhan Chen*,‡



State Key Laboratory of Bioelectronics, School of Biological Science and Medical

Engineering, Southeast University, Nanjing 210096, P. R. China

‡Department of Chemistry, University of Michigan, 930 North University Avenue, Ann Arbor, Michigan48109, United States §

Institute of Neurobiology, School of Medicine, Southeast University, Nanjing 210096, P. R.

China

*Address correspondence to [email protected], [email protected]

1

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT Metallic nanostructures as excellent candidates for nanosensitizers have shown enormous potentials in cancer radiotherapy and photothermal therapy. Clinically, a relatively low and safe radiation dose is highly desired to avoid damage to normal tissues. Therefore, the synergistic effect of the low-dosed X-ray radiation and other therapeutic approaches (or so-called “combined therapeutic strategy”) is needed. Herein, we have synthesized hollow and spike-like gold nanostructures by a facile galvanic replacement reaction. Such gold nanospikes (GNSs) with low cytotoxicity exhibited high photothermal conversion efficiency (η = 50.3%) and have excellent photostability under cyclic NIR laser irradiations. We have demonstrated that these GNSs can be successfully used for in vitro and in vivo X-ray radiation therapy and near-infrared photothermal therapy. For the in vitro study, colony formation assay clearly demonstrated that GNS-mediated photothermal therapy and X-ray radiotherapy reduced the cell survival fraction to 89% and 54%, respectively. In contrast, the cell survival fraction of the combined radio- and photothermal treatment decreased to 40%. The synergistic cancer treatment performance was attributable to the effect of hyperthermia, which efficiently enhanced the radiosensitizing effect of hypoxic cancer cells that were resistant to ionizing radiation. The sensitization enhancement ratio (SER) of GNS alone was calculated to be about 1.38, which increased to 1.63 when the GNS treatment was combined with the NIR irradiation, confirming that GNSs are effective radiation sensitizers to enhance X-ray radiation effect through hyperpyrexia. In vivo tumor growth study indicates that the tumor growth inhibition (TGI) in the synergistically treated group reached 92.2%, which was much higher than that of the group treated with the GNS-enhanced X-ray radiation (TGI = 29.8%) or the group treated with the GNS-mediated photothermal therapy (TGI = 70.5%). This research provides a new method to employ GNSs as multifunctional nanosensitizers for synergistic NIR photothermal and X-ray radiation therapy in vitro and in vivo. KEYWORDS: gold nanospikes, X-ray radiation therapy, radiosensitizing, photothermal therapy, synergistic effect 2

ACS Paragon Plus Environment

Page 2 of 54

Page 3 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

3

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1. INTRODUCTION Clinically, multiple methods have been adopted for tumor treatments, including surgery, chemotherapy, radiotherapy, immunotherapy, and some combined therapeutic strategies.1-5 Among various treatment methods, ionizing radiation is a common therapeutic strategy, which employs high-energy X- or γ-ray. It has been commonly and widely used during late-stage cancer treatment due to high tissue penetration.6 Ionizing radiation mainly induces water molecule ionization to produce reactive free oxygen radicals by photonic energy conversion. To improve the radiation tolerance of surrounding normal tissues, it is desired that a relatively low and safe radiation dose is used. The applications of multiple radiosensitizers are emerging in response to the needs of low-dosed radiation therapeutics. To date, various types of nanostructures, especially metal nanoparticles, have been used as tumor radiosensitizers due to their unique electronic and optical properties.7,8

Metallic nanostructures can be used as classical radiosensitizers because of their strong photoelectric absorption and secondary electron effects caused by high energy X- or γ-ray, which induce reactive oxygen species generation, DNA damage, and cell apoptosis and necrosis.9-13 Thus, metallic nanostructures make some radioresistant hypoxic cells easy to be killed by high energy rays. In detail, enhanced ionizing radiation therapy mediated by metallic nanostructures mainly includes the following underlying mechanisms: (1) Upon X-ray irradiation, more photoelectrons and Auger electrons are generated in the vicinity of the metal nanostructures, which ionizes water molecules to produce reactive free radicals. (2) Metallic nanostructures induce highly localized oxygen-enriched surrounding that prevent the 4

ACS Paragon Plus Environment

Page 4 of 54

Page 5 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

recombination between free electrons and reactive oxygen radicals in remaining cells.14,15

Gold nanoparticles have been widely used since 1950 and issues related to their biological safety have been examined in vivo and in vitro.16-19 Good biocompatibility and low cytotoxicity of gold nanostructures make them possible for clinical radiotherapy. Besides, it was found that gold as a high-Z material (Z = 79) will result in higher efficiency for X-ray mediated cell damages.20-24 The optimal nanostructures for cancer treatment should be large enough (more than 5 nm) to avoid rapid clearance by macrophages but small enough (less than 100 nm) to go through the capillary vessels.25 In theory, the radiosensitizing effect is size-dependent and the larger-sized nanostructures should lead to better effect. However, it can be very complex in real situations.6 For example, it was found that the 50 nm gold nanoparticles had stronger radiosensitizing ratio than the 14 and 74 nm ones since the 50 nm gold nanoparticles showed greater cellular uptake efficiency than others.26 It was also found that the very small 15 nm polyvinyl pyrrolidone-coated Ag nanoparticles have been used as effective radiosensitizers in radiotherapy for gliomas and the underlying mechanisms involved reactive oxygen species generation and autophagic stress.27 In addition, in vivo and in vitro investigations indicated that the radiation enhancement effects of the 12 and 27 nm PEG-coated gold nanoparticles were larger than the 4.8 and 46.6 nm ones.6 The poly(ethylene glycol) (PEG)-conjugated nanoparticles have prolonged circulation periods in blood flow so that they were widely employed for passive targeting delivery.28-31

One of the difficulties of radiation therapy is the low efficiency to kill hypoxic cancer cells.32-34 However, the emerging hyperthermia therapeutic approach is able to kill cells 5

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 54

which are resistant to ionizing radiation. An appropriate hyperpyrexia level could create a continuous oxygen-enriched situation and subsequently improve oxygenation status in the tumor, which makes hypoxic cells more sensitive to ionizing radiation.35-39

Photothermal therapy takes full advantages of the near-infrared (NIR, e.g., λ = 700–900 nm) light including high tissue penetration, low cost and less side effects. In recent years, various types of photo-absorbing agents have been investigated such as metallic nanostructures,40-42 carbon-based

materials,43-45

copper

chalcogenide

semiconductors,46

and

organic

compounds,47 which accumulated in tumor sites and could produce hyperpyrexia and selectively ablate tumor by absorbing external NIR light. Therefore, more and more new nanomaterials were designed to expand the absorption band to the near-infrared region by altering the nanomaterial shape for potential photothermal therapy. For example, by changing the shape of spherical gold nanoparticles to spiky gold nanostructures and gold superstructures, the absorption wavelength could be changed from visible to NIR region. In addition, the spiky tips provided higher local electromagnetic field.48,49 However, sometimes NIR photothermal therapy fails to ablate tumors sufficiently.50-54 Consequently, synergistic therapy is preferable to improve tumor therapeutic efficiency and avoid the respective shortcomings of each therapeutic approach. It is necessary to develop new nanostructures as both efficient photothermal absorbing agents and radiosensitizers for tumor treatment. Nanostructures have been developed and used as photo-absorbing agents or radiosensitizers, but rarely for both functions.36,55-59

In this work, we successfully synthesized gold nanostructures with a high density of 6

ACS Paragon Plus Environment

Page 7 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

surface nanospikes and hollow interiors. Such unique gold nanostructures presented potential applications in tumor therapy since their sharp tips and high surface area broaden the surface plasmon resonance and extend such resonance to the near-infrared region. For the first time, the prepared gold nanospikes (GNSs) were utilized as both radiosensitizers and photothermal absorbing agents for combined photothermal and X-ray radiation therapy (Scheme 1). Compared to the use of photothermal therapy or radiotherapy alone, the as-prepared GNSs remarkably decreased cancer cell survival rate and efficiently inhibited cancer cell growth via synergistic X-ray ionizing radiation therapy and NIR photothermal therapy.

Scheme 1. Schematic showing the PEGylated GNSs for combined photothermal and X-ray radiation therapy.

7

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 54

2. MATERIALS AND METHODS

2.1. Materials. In this study all chemicals of analytical grade were used. Silver nitrate (AgNO3), trisodium citrate dihydrate, gold (III) chloride hydrate (HAuCl4·3H2O), L-ascorbic acid (LAA), dimethyl

sulfoxide

(DMSO),

fluorescein

isothiocyanate

(FITC),

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), 4% glutaraldehyde and Triton X-100 were purchased from Sigma-Aldrich (St. Louis, MO) and used as received unless noted otherwise. Thiol-terminated monomethoxy poly(ethylene glycol) (HS-PEG, M.W. ~2000 Da) was ordered from JenKem Technology Co., Ltd. (Beijing, China). All aqueous solutions were prepared using ultrapure water with a resistivity of 18.2 MΩ·cm obtained from a Milli-Q ultra purified water system (Millipore Corp., Billerica, USA). Human oral epidermoid carcinoma (KB) cell lines and mouse uterine cervical cancer cell line (U14) were obtained from the Type Culture Collection of the Chinese Academy of Science, Shanghai, China. Female athymic nude mice (BALB/c-nu) were purchased from the comparative medicine center of Yangzhou University (Jiangsu, China).

2.2. Synthesis of PEGylated GNSs. GNSs were prepared via the galvanic replacement reaction between Ag nanoparticles (AgNPs) and HAuCl4 according to the method reported by Wang et al.60 with slight modifications. Typically, 180 µL of 1% AgNO3 aqueous solution was mixed with 25 mL of deionized water and heated. Then the mixture was kept boiling for 10 min. 5 mL 1% trisodium citrate aqueous solution was added rapidly into the above boiling AgNO3 solution under vigorous stirring, keeping boiling for additional 30 min. The reaction mixture finally formed citrate-stabilized AgNPs in the solution. The cooled citrate-stabilized AgNPs solution was centrifuged and then re-suspended in 25 mL of ultrapure water and stored at 4 °C for further experiments. 140 µL of freshly prepared 1% 8

ACS Paragon Plus Environment

Page 9 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

HAuCl4·3H2O aqueous solution was diluted with 1 mL ultrapure water. Subsequently, 5 mL of above prepared citrate-stabilized AgNPs solution was immediately added into the diluted HAuCl4 solution with vigorous stirring followed by the addition of 1 mL 10 mM L-ascorbic acid aqueous solution. The color of the solution changed to dark blue from light yellow within a short period of time. Eventually, GNSs were formed.

To prevent the change of surface morphology and make as-synthesized GNSs highly water-soluble, HS-PEG was conjugated on the surface of GNS through thiol–gold coupling. Typically, 200 µg of HS-PEG was added to the GNS solution and the mixture solution was continuously stirred for 2 h at room temperature. Excessive HS-PEG was removed by centrifugation at 10,000 rpm and washed three times with deionized water to obtain PEGylated GNSs. Then the prepared GNS solution was stored at 4 °C for quantitative analysis or further experiments.

2.3. Characterization of PEGylated GNSs. The morphology of synthesized GNSs was investigated using a JEOL-2100 transmission electron microscope (TEM) operating at an accelerating voltage of 200 kV and a field emission scanning electron microscope (FESEM, Zeiss Ultra Plus, Germany) at an accelerating voltage of 20 kV. Meanwhile, Zeiss Ultra Plus emission scanning electron microscope was equipped with energy-dispersive spectrometer (EDS) for element component analysis. Samples for TEM study were prepared by drop-casting GNS solution onto a carbon-coated copper grid. The mean size of GNSs was calculated from representative images and the size histogram was generated simultaneously by Nano Measurer software (Version 1.2). The hydrodynamic diameter and zeta potential of GNSs were measured by dynamic light scattering (DLS) using a Malvern Zetasizer (Nano ZS, Malvern, UK). UV–vis spectra were collected using a UV–vis spectrophotometer 9

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(UV-2600, Shimadzu, Japan) with a wavelength range from 300 to 900 nm. The GNS surface coating was investigated by a Fourier transform infrared spectrometer (FTIR, Nicolet iS50, Thermo scientific, USA). FTIR spectra of samples in KBr were determined at room temperature in the spectral region 400–4000 cm–1. An average of 30 scans per sample with a nominal resolution of 4 cm–1 was used for FTIR spectra collection. Experiments of X-ray photoelectron spectroscopy (XPS) were conducted using an X-ray photoelectron spectrometer (PHI Quantera II, Ulvac-PHI, Japan). The fluorescence spectrum of GNS before and after FITC labeling was determined by spectrofluorophotometer (RF-5301PC, Shimadzu, Japan).

2.4. Quantitative Assay of GNS Concentration by Atomic Absorption Spectroscopy (AAS). The final concentration of the prepared GNS solution was determined with atomic absorption spectroscopy (AAS, AAnalyst 400, USA). Briefly, GNS solution was digested by freshly prepared chloroazotic acid, transferred to a 10-mL volumetric flask and then diluted with ultrapure water to 10 mL. The standard stock solution of gold element (1000 µg/mL) was purchased from the National Research Center for Certified Reference Material in China. The stock solution was diluted to a series of solutions with different concentrations (e.g., 1, 2, 3, 4, and 5 µg/mL) and the absorbance values of those solutions were measured by AAS to create the standard curve. According to the standard curve, the concentration of the prepared GNS solution was determined by AAS under the same condition.

2.5. Cytotoxicity Assay. Cell culture reagents were purchased from Gibco, USA. KB cells were cultured in high-glucose complete Dulbecco's Modified Eagle Medium (DMEM) supplemented with

10

ACS Paragon Plus Environment

Page 10 of 54

Page 11 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

10% fetal bovine serum (FBS), streptomycin (0.1 mg/mL) and penicillin (100 U/mL) at 37 °C in a humidified incubator with 5% CO2.

The cytotoxicity of the GNS solution against KB cells was evaluated by MTT assay. In brief, KB cells were seeded in a 96-well plate at a concentration of 5 × 103 cells per 100 µL of culture medium and continuously cultured for 24 h until ~80% confluency in 5% CO2 incubator at 37 °C. After exposure to GNS solutions with various concentrations for 24 h, the cell culture medium in each well was replaced with 10 µL of 5 mg/mL MTT agents in phosphate buffered saline (PBS) buffer, followed by incubation for 4 h at 37 °C. The formed formazan crystals were dissolved in 150 µL of DMSO. Then the absorbance at 570 nm in each well was measured using a Multiskan FC microplate photometer (Thermo). Each group was measured in triplicate.

2.6. Live-Dead Staining. To further determine cell viability of GNS-treated cancer cells, KB cells were analyzed using live/dead viability/cytotoxicity kit (Molecular Probe, Life technologies) by confocal laser scanning microscopy (CLSM, Leica, TCS SP8, Germany). After exposure to media or GNS solution for 24 h, cell culture medium was removed and replaced with PBS buffer consisting of 5 µL of Calcein AM (Ex/Em: 495 nm/520 nm) and 10 µL of PI (Ex/Em: 530 nm/620 nm) for 20 min at room temperature followed by fluorescence microscopic study.

2.7. Cellular Uptake and BioTEM Observation. To assess the intracellular uptake of GNSs, fluorescein isothiocyanate (FITC), a green fluorescence labeled probe, was employed to estimate the internalization amount of GNSs. FITC-labeled GNSs (GNS-FITC) were synthesized via the covalent binding of HS-PEG-FITC onto the GNS surface. Briefly, HS-PEG-FITC was prepared by the addition

11

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

reaction between FITC and HS-PEG-NH2. The HS-PEG and HS-PEG-FITC were mixed in molar ratio of about 99:1 and immediately added into the GNS solution under vigorous stirring for 2 h at room temperature. Excessive PEG derivatives were removed by centrifugation and washed three times with water to obtain GNS-FITC. Meanwhile, KB cells were seeded onto 8-chambered coverglass (Lab-Tek Chambered Coverglass System) at a density of 3 × 104 cells/mL (7,000 cells per well) in a 37 °C incubator overnight. Then cell culture medium was replaced with freshly prepared GNS-FITC in DMEM (50 µg/mL) and incubated for additional 24 h. The intracellular GNS content was determined using confocal fluorescence microscopy and flow cytometry (NovoCyte, ACEA).

For intracellular biological electron microscopic study, KB cells were cultured in the 6-well plates overnight prior to exposure to GNSs in DMEM (50 µg/mL). After 24 h incubation, cells were fixed in 4% glutaraldehyde for 30 min and washed three times with PBS buffer. Subsequently, the fixed cells were stained with 1% osmium tetroxide and uranyl acetate at room temperature for 1 h followed by ethanol series dehydration before embedding samples in epoxy resin. Ultrathin sections (~70 nm) were cut by an ultramicrotome, mounted on copper grids and stained doubly with uranyl acetate/lead citrate. The prepared samples were then imaged using a BioTEM (H-600(4), Hitachi, Japan).

2.8. Photothermal Evaluation Tests and NIR Photothermal Therapy. To assess the photothermal response, the GNS stock solution was diluted to a series of solutions with different concentrations (6, 12, 25, and 50 µg/mL). The solutions were placed respectively in 200 µL centrifugal tubes and illuminated with an 808 nm high power laser diode at a power density of 2 W/cm2for 10 min (beam spot 0.8 × 0.8 cm2). An Ai50 near–infrared thermal imaging camera was simultaneously used to record the temperature changes and thermal images of the solution over time. 12

ACS Paragon Plus Environment

Page 12 of 54

Page 13 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Meanwhile, to further explore the photothermal response of GNS-treated cells, KB cells were exposed to GNSs in DMEM (50 µg/mL) for 24 h and then washed twice with PBS buffer. Finally, GNS-treated cells were harvested, placed in 200 µL centrifugal tubes and irradiated with an 808 nm laser at a power density of 2 W/cm2 for 10 min. The photothermal heating curves were evaluated in a 37 °C incubator.

To further evaluate the NIR photostability of GNSs, the following experiments were performed. In a typical study, a 150 µL 50 µg/mL GNS aqueous solution in 200 µL centrifuge tubes was irradiated with an 808 laser at a power density of 2 W/cm2 for 10 min and then the laser was turned off. The UV–vis spectra of the solution before and after an 808 nm laser exposure were taken to assess the stability of the GNSs. Furthermore, laser on/off cycle assays were carried out to evaluate the NIR photostability of GNSs. Typically, a GNS aqueous solution (25 µg/mL) was exposed to an 808 nm laser under previous conditions for 180 s. Then the laser was shut off for 180 s and turned on again. The laser was on and off with an interval of 180 s for 6 cycles. The temperature of the GNS solution was recorded continuously with a NIR thermal imaging camera.

For the NIR photothermal therapy groups, cells were exposed to GNSs (50 µg/mL) for 24 h (or not exposed to GNSs as control) and then washed twice with PBS buffer. After exposure, cells were irradiated with an 808 nm laser at a power density of 2 W/cm2 for 5 min. After the treatment, the cells were further cultured overnight at 37°C without light interference.

2.9. Radiation Treatment. For X-ray radiation sample groups, KB cells were washed twice with 1 × PBS after exposure to GNS solution for 24 h. Next, cells were irradiated with X-ray (6 MV, 200

13

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cGy/min) from an electron linear accelerator (Primus-M, Siemens, Germany) for different periods of time.

2.10. Colony Formation Assay. KB cells were cultured in 6-well plates at a density of 5 × 104 per well for 24 h. When the cells had grown to ~80% confluency in plates, the cells were divided into eight groups: (a) Control, (b) GNS only, (c) NIR only, (d) GNS + NIR, (e) X-ray only, (f) GNS + X-ray, (g) NIR + X-ray, and (h) GNS + NIR + X-ray. The group a had no dispose consistently as a comparison. The groups b, d and h were treated with GNSs in DMEM (50 µg/mL) for 24 h. After 24 h treatment, GNSs in DMEM was replaced with fresh culture medium and washed twice with PBS buffer. The groups c and d were irradiated with an 808 nm laser (2 W/cm2, 5 min). Concurrently, the groups e and f received the administration of X rays by a linear accelerator at various doses of 2, 4, 6, and 8 Gy, respectively. Particularly, the groups g and h were irradiated with an 808 laser (2 W/cm2, 5 min) firstly and then immediately with X-ray treatment (2, 4, 6, and 8 Gy). Finally, KB cells were trypsinized, counted and seeded 1500, 1500, 3000, 3000, and 3000 cells at corresponding doses of 0, 2, 4, 6, and 8 Gy in 35 mm dishes with 2 mL of fresh culture medium. There were three replicates for each sample group. Subsequently, cells were incubated for additional 10 days followed by Giemsa staining (KeyGen Biotech, China). The colonies formed with more than 50 cells were counted to evaluate the effects of respective treatments.

2.11. Calculation of the Sensitization Enhancement Ratio. Cell survival fraction of each sample group was calculated by the ratio of the seeded cells following treatments to form colonies vs. the untreated cells as described above. Cell survival fraction and sensitization enhancement ratio (SER) were determined by a classical multi-target single-hit model (S = 1– (1– eD/D0)n), where S is the 14

ACS Paragon Plus Environment

Page 14 of 54

Page 15 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

survival fraction, D is the radiation dose, D0 is the mean lethal dose, and n is the extrapolation number.6 In the above model formula, D0and n were calculated by a “Statistical Program for Social Sciences” software (SPSS, version 19.0). The final survival fraction curve of each group was generated via a nonlinear fitting using Origin 8.0 and the SER was expressed with the following formulae: SER = Dq (control group)/Dq (treated group) and Dq = ln(n) × D0, where Dq is the quasi-threshold dose.

2.12. Real-Time Cell Analysis. After respective treatments for the previously discussed eight sample groups, the KB cells were trypsinized and counted by flow cytometry. Subsequently, 15,000 cells re-suspended in 500 µL of culture medium were seeded onto an electronic 8-well plate of RTCA iCELLigence analyzer (ACEA Biosciences, Inc.) for cell proliferation and growth assays. The RTCA instrument was placed in a standard CO2 cell culture incubator where it transmitted data wirelessly to an iPad housed outside the incubator. The software in the iPad friendly allowed for real-time interfacing with electronic 8-well plate and included real-time data display and analysis functions.

2.13. Evaluation of Reactive Oxygen Species levels. To assess the generation of intracellular reactive oxygen species (ROS) in the previously introduced eight sample groups, the reactive oxygen species assay kit (KeyGenBiotech, China) was employed. Briefly, KB cells with respective treatments were harvested and loaded with fluorescent probes, 10 µM dichlorodihydrofluorescein diacetate (DCFH-DA) KeyGen Biotech, China), in DMEM without phenol red for 15–30 min in dark (at 37 °C in 5% CO2). Meanwhile, cells were pretreated with the ROS positive control reagent (ROSup) for 30 min. All cells were trypsinized and immediately analyzed by flow cytometry. The fluorescence signals

15

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

were detected in FL-1 channel using a FITC filter (Ex/Em: 492 nm/517 nm). For each sample, the mean fluorescence intensity of 10,000 cells was recorded to verify intracellular generation of ROS.

2.14. Immunofluorescence Assay for Phosphorylated Ser 139 on Histone H2AX (γ-H2AX). After different treatments for 24 h, cells were washed three times with 1 × PBS, fixed with 4% glutaraldehyde for 10 min and permeated with 1% Triton X-100. Next, cells were blocked in 1% bovine serum albumin for 60 min to avoid nonspecific protein adsorption and incubated with Anti-Human Phospho-H2AX (S139) Alexa Fluor®488 mouse mAb (eBioscience, USA) overnight at 4 °C. Hoechst 33342 was used to stain the nuclei.

2.15. Cell Apoptosis and Necrosis Assays.The Annexin V–FITC/PI (KeyGenBiotech, China) apoptosis detection assay was performed to compare the apoptosis-inducing capabilities of GNSs before and after irradiation. Briefly, after 24 h treatment, KB cells in eight sample groups were trypsinized and re-suspended in 500 µL 1 × binding buffer at the concentration of 1 × 106 cells/mL. Afterwards, 5 µL of Annexin V–FITC and 5 µL of PI were consecutively added into the above 1 × binding buffer and the cells were incubated for another 20 min at room temperature in dark. Finally, the quantitative analysis of cell apoptosis and necrosis were performed immediately by flow cytometry.

2.16. In VivoStudies. All animal experimental procedures were performed under protocols approved by the Animal Care and Use Committee of Southeast University. 1 × 106 U14 cells (mouse uterine cervical cancer cells) suspended in 100 µL of PBS were subcutaneously implanted into the right hind legs of BALB/c female mice to generate the U14 xenograft tumor model. The mice were

16

ACS Paragon Plus Environment

Page 16 of 54

Page 17 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

used for further experiments when the tumor had reached approximately 90 mm3 in volume. For the in vivo tumor growth study, the mice were divided into the following seven groups on day 0 and each group included three mice: (1) Control, (2) GNS only, (3) NIR only, (4) GNS + NIR, (5) X-ray only, (6) GNS + X-ray, and (7) GNS + NIR + X-ray. Briefly, the mice in the control group 1 received intratumoral injection with sterilized PBS solution (100 µL/mouse). The mice in groups 2, 3, 6, and 7 were intratumorally injected with 100 µL of 400 µg/mL GNSs in PBS solution. After 3 h post-injection, the mice in groups 3, 4, and 7 were irradiated with an 808 nm laser at a power density of 0.75 W/cm2 for 5 min in the tumor site, and simultaneously an Ai50 infrared thermal imaging camera was used to record the temperature change of the tumor region. Subsequently, the mice in group 7 received X-ray radiation (6 MV, 6 Gy) after an 808 nm laser irradiation. The mice in group 5 were exposed to 6 MV X-ray radiation (6 Gy). After the above treatments, the tumor volume and mouse weight were measured daily until day 9. The tumor volume was measured using a vernier caliper and calculated as the volume = (tumor length) × (tumor width)2/2. The relative tumor volume was calculated as V/V0, where V0 was the tumor volume when the treatment was initiated, and V is the volume of the tumor measured daily after the treatment. Finally, after the entire experiment was completed, all the U14 tumor-bearing BALB/c nude mice were sacrificed under protocols approved by the Animal Care and Use Committee of Southeast University.

2.17. Statistics. The paired Student's t-test was used for statistical analysis of clonogenic assay using the SPSS software. All obtained data were repeated three or more times and presented in the format of mean ± standard deviation (Mean ± SD). The results were subjected to one-way ANOVA using Tukey’s test to analyze differences between the treated and control groups in which a P-value 17

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

less than 0.05 was considered as a significant difference. Images are representative of three or more experiments.

3. RESULTS AND DISCUSSION 3.1. Preparation and Characterization of GNSs. The underlying formation mechanism of spike-like gold nanostructures has been clearly presented by Wang et al.60 Basically, the galvanic replacement reaction between AgNPs and HAuCl4 is very fast, leading to the formation and later dissolution of silver chloride (AgCl), which helps for the formation of branched tips and hollow interiors of spike-like gold nanostructures. More details can be found in the original reference.60 Figure1a shows a simple schematic displaying the change from the initial Ag seeds (Ag NPs) to the resultingGNSs.

18

ACS Paragon Plus Environment

Page 18 of 54

Page 19 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

19

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. Characterization of the prepared GNSs. (a) A schematic showing the evolution process of the spike-like gold nanostructures prepared from Ag seeds.For the whole process, the reaction solution color changed from light yellow to dark blue. Insert images are pictures of the solutions of initial Ag seeds and resulting GNSs, respectively. (b) TEM image and size distribution histogram of the prepared GNSs. The hollow and spike-shaped structures of GNSs were characterized by TEM. Scale bar: 20 nm. The insert size distribution histogram in b was obtained by size analysis of over 200 particles. The mean diameter was 54 ± 9 nm. (c) The surface morphology of GNSs was characterized by SEM. Scale bar: 20 nm. (d) UV–vis absorption spectra of Ag seeds and GNSs. 1: Absorption spectrum of Ag seeds (peak at 420 nm); 2: Absorption spectrum collected from the HAuCl4 aqueous solution after the addition of Ag seeds (peak at 570 nm); 3: Absorption spectrum from the solution in “2” above after the addition of L-ascorbic acid (peak at 670 nm). (e) FTIR spectra of GNSs before and after PEGylation. (f) XPS spectrum of the PEGylated GNSs. (g) XPS spectrum of Au4f7/2 and Au4f5/2 for the PEGylated GNSs.

To date, numerous methods have been explored for controllable synthesis of gold nanostructures with different morphologies, including chemical, physical and combined methods. The precise control of the size distribution of gold nanostructures can still be a challenge. In this study, uniform and colloidal GNSs were synthesized by a galvanic replacement reaction.HS-PEG, a commonly used stabilizer and protective agent, was employed here to maintain dispersed (not aggregated) GNSs in a PBS buffer or culture medium and also prevent changes of GNS surface morphology.TEMand SEM images showed that the prepared GNSs had high density of branched tips on their surfaces and hollow interiors (Figure 1b and c). Results from energy dispersive spectroscopy (EDS) confirmed that no Cl was detected in resulting GNS (Figure S1, the Al signal comes from the aluminized 20

ACS Paragon Plus Environment

Page 20 of 54

Page 21 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

paper substrate). However, the existence of a very small quantity of silver was detected, resulting from the incomplete dissolution of Ag on the surface. The size distribution histogram (shown in the insert of Figure 1b) obtained from the TEM images revealed that the mean diameter of GNSs was 54 ± 9 nm, which was smaller than those reported previously.60 The interior size was in the range of ~20 nm (Figure 1b) and smaller compared to the Ag seeds (Figure S2 in supporting information, dav = 30 ± 8 nm). Table S1 depicts that the PEGylated GNSs have negative surface charges measured by DLS. GNSs in PBS buffer could be stable for up to one month due to the electrostatic repulsion between the charged GNSs and the steric hindrance of the surface PEG segments on GNSs. The hydrodynamic diameter of GNSs in PBS buffer was measured by DLS experiment (Table S1), which was larger than the mean size obtained from the TEM images because of the extension of the PEG molecular chain and the thickness of the hydration shell. Herein, we want to mention that the GNSs below used for tumor treatment refers to the PEGylated GNSs unless noted otherwise.

UV–vis absorption spectra were collected from the sample solution at different stages during the GNS preparation process (Figure 1d). We believe that these UV–vis spectra were attributed to the surface plasmon resonance of each nanostructure. The maximum surface absorption peaks of the reactants, the reaction intermediates, and the final products are clearly visible, centered at 420, 570, and 670 nm corresponding to the initial Ag NPs, reaction intermediates, and the resulting GNSs, respectively. After the addition of L-ascorbic acid into the HAuCl4 aqueous solution with AgNPs, the resulting product exhibited enhanced absorption in the desired 700–900 nm region, suitable for effective photothermal therapy.The 21

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

shift of the maximum absorption bands may originate from the formation of nanoshells and branched structures on the GNS surface. It is well known that surface plasmon absorption is closely related to the physical dimensions of the metallic nanostructures and the surrounding medium.61 Herein, a broader absorption band of GNSs likely indicated a broader size distribution, which was confirmed by the measured hydrodynamic diameter and the higher polydispersity index (PDI) from DLS experiments.

FTIR spectroscopy is a powerful tool to identify functional groups of organic molecules. Figure 1e shows the FTIR spectra from GNS samples before and after PEGylation. Before PEGylation, the FTIR spectrum of GNSs exhibited the characteristic peaks of L-ascorbic acid molecules which may be nonspecifically adsorbed onto the prepared GNS surface. While the PEGylated GNSs presented strong absorption peaks near 1100 cm–1, corresponding to the – C–O–C– stretching vibrations in PEG molecules.This confirms the successful covalent grafting of HS-PEG onto GNS surfaces.

The XPS spectrum of the PEGylated GNSs is shown in Figure 1f. The binding energies of various elements including C, O, S, Au were detected from synthesized PEGylated GNSs. The presence of the S2p peak at 162.3 eV demonstrated the Au–S bond formation between GNS with HS-PEG. This conclusion was also supported by the XPS spectrum of Au0 collected from the PEGylated GNSs shown in Figure 1g. We believed that the binding energies of Au4f7/2 (84.2 eV) and Au4f5/2 (87.9 eV) could be assigned to Au0. Compared to the standard binding energy of Au0 (Au4f7/2 : 84.0 eV; Au4f5/2 : 87.7 eV),62 the binding energy of Au0 shifts towards the higher energy side, which was likely caused by the strong 22

ACS Paragon Plus Environment

Page 22 of 54

Page 23 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

interaction between GNS and HS-PEG molecules. The XPS data verified the formation of the PEG-modified GNSs.

3.2. Cellular Uptake Efficiency and Biocompatibility of GNSs. The amount of nanostructures that are internalized by cancer cells is critical for inducing irreversible damages to cancer cells. Extensive investigations have shown that the cellular adhesion and uptake of nanomaterials is strongly dependent on the sizes and surface modification of those nanomaterials.63-68 Cellular internalization of gold nanomaterials between 10 nm and 50 nm in diameter had an increasing trend, whereas the 50 nm-sized gold nanoparticles were internalized more quickly than smaller and larger ones.6

In this study, two methods were employed to study intracellular uptake efficiency of the prepared GNSs. In the method 1, KB cells were treated with freshly prepared GNSs (50

µg/mL) for 24 h and then observed by biological TEM (BioTEM). The BioTEM results show that large aggregations of GNSs in the cytoplasm were concentrated around the cell nucleus. GNS clusters were mostly accumulated in endosomes or lysosomes (Figure 2a) and few clusters were located in the cytoplasm. Generally, lysosomes are the eventual endpoints of the degraded materials, so it is not surprising that only very few GNS clusters can leak from the lysosomes. To some degree, gold nanomaterials, because of their unique metallic features,may cause impairment of lysosome degradation capacity.69 In the method 2, the confocal laser scanning microscopy (CLSM) was used for visual assessment of cellular uptake efficiency of GNSs. Nuclei were stained blue with Hoechst 33342 and green fluorescence signals indicated intracellular FITC-labeled GNSs (GNS-FITC). Before this 23

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

experiment, as depicted in Table S1 in the supporting information, the zeta potential of GNS-FITC was measured, which has a negative charge at – 16 ± 8 mV. Such a zeta potential was lower than that of the GNS before FITC labeling. We then compared the fluorescence intensity from GNS-FITC and GNS (Figure S3, supporting information). We observed that the fluorescence of FITC anchored on the surfaces of GNS was very strong. The zeta potential and fluorescence spectroscopic results both demonstrate the successful conjugation of FITC on the GNSs. Next, to exclude the signal interference of cell spontaneous fluorescence, cells without any treatment were imaged as the control group. As shown in the fluorescence images (Figure 2b), cells exhibit significantly high green fluorescence signals in GNS-FITC treated sample group. Large aggregates of GNS shown as bright green spots in the cytoplasm were detected.FITC, as a fluorescence probe, can roughly reveal the location and amount of intracellular GNSs. Internalized GNS-FITC in a cell mostly accumulated around the cell nucleus because of its large dimensions, which is consistent with previous BioTEM images. To quantify the cellular uptake of KB cells, we also carried out the flow cytometric analysis. The mean fluorescence intensity of the GNS-FITC treated cells is 2.5fold larger than that of the untreated control sample group after 24 h incubation (Figure 2c).

24

ACS Paragon Plus Environment

Page 24 of 54

Page 25 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

25

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. Assessment of cellular uptake efficiency of GNSs. (a) Biological TEM images of cellular localization of GNSs in KB cells after treatment with PBS (control) or GNSs (50 µg/mL) for 24 h. N, Nuclei; C, Cytoplasm. Scale bars: 0.5 µm. Large aggregates of GNSs were clustered around the cell nucleus after the GNS treatment and were not observed in the untreated control sample group. Intracellular GNSs were mostly enclosed within endosomes (black arrows) and lysosomes (white arrows) as black and white arrows indicated. (b) Typical fluorescence images of KB cells incubated with PBS or FITC-labeled GNSs (50 µg/mL) for 24 h. Cells incubated with FITC-labeled GNSs showed significant intracellular fluorescence signals. Large aggregates of GNSs as bright green spots in the cytoplasm can also be seen. Nuclei were stained with Hoechst 33342 (blue) and green fluorescence signal from FITC-labeled GNSs. Scale bars: 25 µm. (c) Flow cytometric analysis of the mean fluorescence intensity (n = 10,000 cells) for KB cells incubated with PBS or GNSs (50 µg/mL) for 24 h. (d) Viabilities of KB cells were assessed by MTT assay (with absorbance at 570 nm) after treatment with GNS solutions at various concentrations ranging from 3 to 200 µg/mL.

It is well known that the cytotoxicity of nanostructures mostly depends on their size and surface modification.6 These cytotoxicity properties are determined by many factors such as the electrostatic interaction between the cationic nanomaterials and negatively charged cell membranes, biochemical reaction inside the plasma membrane, etc. Currently, PEGylation is commonly used as an effective method to improve the monodispersity and biocompatibility of naked gold nanomaterials. Although the biocompatibility of PEGylated gold nanoparticles is well-known, the biosafety of PEGylated spike-shaped gold nanostructures has been less examined. For biomedical applications, it is essential to evaluate the cytotoxicity of GNSs. Herein, the thiol-terminated poly (ethylene glycol) was conjugated on the surface of naked GNS through thiol–gold coupling.70 Then, the cytotoxic effect induced by GNSs on KB cells 26

ACS Paragon Plus Environment

Page 26 of 54

Page 27 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

was firstly evaluated by MTT assay. As shown in Figure 2d, GNS solutions with different concentrations ranging from 3 to 200 µg/mL led to decreased cell viability in a dose-dependent manner after 24 h incubation, indicating that the cytotoxic effect of GNSs increased as a function of GNS concentration. Even at a high concentration up to 200 µg/mL, the cytotoxicity of GNSs towards the KB cells is not significant. Secondly, to demonstrate visually the ratio between the live and the dead cells corresponding to PBS or GNS treatment, the Calcein AM/PI double-staining experiment was performed to evaluate cancer cell ablation. Immediately after the treatment, cells in each sample group were examined under a fluorescence microscope using a live-dead staining procedure. Membrane-impermeable PI dye entered dead cells and displayed enhanced red fluorescence when bound to DNA double helices, whereas green fluorescence spots by Calcein AM staining indicated the live cells (Figure S4 in supporting information).Our results showed that the prepared 54 nm GNSs exhibited excellent biocompatibility and low cytotoxicity at the concentrations required (e.g., ~50 µg/mL) for hyperthermia and radiation therapy enhancement which will be discussed below.

3.3. Photothermal Effect Evaluation and Long-Term Photostability of GNSs. As discussed above, the surface plasmon resonance of GNSs was broadened and shifted to the near-infrared region. Therefore, the synthesized GNSs should present strong photothermal effect upon an 808 nm laser irradiation. As a comparison, the spherical gold nanoparticles (GNPs) with 50 nm in size were examined. We collected the absorption spectra and studied the photothermal effect of GNSs and GNPs with similar diameters. The UV–vis absorption 27

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

spectrum of GNPs shows that the surface plasmon resonance is peaked at 528 nm with negligible absorption in the near infrared region, whereas the absorption band of GNSs was detected in the range of 500–900 nm (Figure S5a). As depicted in Figure S5b, the temperature of GNP solution increased slightly upon laser irradiation, reaching 34 °C after 10 min. Under the same condition, the temperature of the GNS solution of the same concentration rose to a much higher temperature of 71 °C.

Figure 3. Photothermal effect evaluation and photostability of GNSs. (a) Temperature responses and infrared thermal images for GNS solutions with various concentrations under an 808 nm laser irradiation at a power density of 2 W/cm2 for 10 min. PBS was used as the negative control group. (b) Temperature plot of KB cells treated with GNS solution (50 µg/mL) irradiated with an 808 nm laser at a power density of 2 28

ACS Paragon Plus Environment

Page 28 of 54

Page 29 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

W/cm2 for 10 min. Cells in DMEM were used as the negative control group for comparison. (c) UV-vis absorption spectra of GNS solution in PBS buffer before and after 10 min continuous laser irradiation (808 nm, 2 W/cm2). The inset showed the photograph of GNSs after exposure to an 808 nm laser for 10 min. The insert scale bar: 20 nm. (d) The temperature changes of a GNS solution (25 µg/mL) over six laser on/off cycles upon NIR laser irradiation.

Next, we examined the photothermal effects of GNS solutions with various concentrations, which were exposed to an 808 nm laser irradiated at a power density of 2 W/cm2 for 10 min at room temperature. As shown in Figure 3a and Figure S6 in supporting information, the solution temperature increase and infrared thermal images were quantified by real-time thermal imaging using a thermal camera. The temperatures of GNS solutions with various concentrations increased under the NIR laser radiation and rapidly rose within the first five minutes. The temperature of each solution then reached a plateau. The negative control group with no GNSs only exhibited negligible temperature increase under the laser irradiation. The rate of temperature rise and the final temperature were directly related to the GNS concentration; a slower increase was observed for a solution with a lower GNS concentration. The temperature of the GNS solution (50 µg/mL) can reach to 88 °C after five minutes laser irradiation. Furthermore, the photothermal conversion efficiency (η) of the GNSs (25 µg/mL) was calculated to be 50.3% (see Supporting Information for more details), which was lower than that of gold nanostars and nanorods reported previously but higher than that of Cu2-xSe nanocrystals and WS2 quantum dots.36,71,72 In addition, the photothermal effect of internalized GNSs in KB cells was also evaluated. Cells were incubated with GNSs (50 µg/mL) overnight and then harvested in 200 µL 29

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

centrifuge tubes (2 ×105 cells/mL, 150 µL) before exposure to an 808 nm laser at a power density of 2 W/cm2 for 10 min in a 37 °C incubator. Figure 3b shows that the temperature of the GNS-treated cells rose more slowly compared to that of GNS solution alone within the continuous 808 nm laser irradiation period. Finally, the average temperature of GNS-treated cells rose by 13 °C. These results clearly demonstrated that the synthesized GNSs could induce photothermal effects after exposure to an 808 nm laser.

Photostability evaluation of GNSs is necessary since photodegradation may greatly limit the photothermal treatment efficiency. We evaluated the photostability of GNS solutions, by means of continuously exposing a solution to the NIR laser irradiation (808 nm, 2 W/cm2). The UV–vis absorption spectrum was measured after an 808 nm laser irradiation, meanwhile the morphological change of GNSs was also observed by TEM. Figure 3c shows that the UV–vis absorption spectrum did not vary after the 10 min laser irradiation.The TEM image exhibited that GNS solution still maintained the same morphology and size after exposure to an 808 nm laser, indicating their photostability under the NIR laser irradiation. Moreover, an additional experiment was carried out to further evaluate the stability of GNS solution. Six cycles of the NIR laser on/off experiments were performed, where a 25 µg/mL GNS solution was irradiated with a NIR laser for 3 min, followed by naturally cooling to room temperature without the NIR laser radiation for another 3 min, and the process was repeated (Figure 3d). No significant decrease for the temperature elevation was observed in six cycles. The excellent photostability demonstrated that the GNS is an excellent candidate for photothermal therapy. 30

ACS Paragon Plus Environment

Page 30 of 54

Page 31 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

3.4. Synergistic Photothermal Effect and Radiation Sensitizing Effect of GNSs in Vitro. The low cytotoxicity and high NIR photothermal effect of GNSs make them appropriate photothermal agents. In a different study which will be reported in the future, we found that GNSs showed excellent radiosensitizing effect. Herein, we will emphasize the use of GNSs in synergistic photothermal and X-ray radiation therapy with KB cells. Colony formation assays were conducted to determine the survival fractions of KB cells after the different treatments.

In this experiment, KB cells were divided into eight sample groups (Control, GNS only, NIR only, GNS+NIR, X-ray only, GNS + X-ray, NIR + X-ray, and GNS + NIR + X-ray). It is worth mentioning that the synergistic photothermal/radiotherapy sample group (GNS + NIR + X-ray) was irradiated with an 808 laser firstly and then was immediately exposed to X-ray radiation. Finally, colonies were counted to evaluate the self-renewal efficiency of KB cells after 10 days post-irradiation. The dose-dependent radiation sensitizing effects of GNSs with or without NIR photothermolysis are shown in Figure 4.

31

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. GNS-enhanced photothermal therapy and radiotherapy decreased survival rate of KB cells. (a) Colony formation images of KB cells treated with GNSs (50 µg/mL) combined with 808 laser irradiation (2 W/cm2, 5 min) and/or X-ray radiation (4 Gy). (b) Survival fractions of KB cells after respective 32

ACS Paragon Plus Environment

Page 32 of 54

Page 33 of 54

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

treatment. Data represents a Mean ± SD of at least three identically and independently prepared samples. (c) Sensitization enhancement ratios of NIR, GNS and GNS+NIR were calculated by the multi-target single-hit model. (d) GNSs in combined NIR photothermal and X-ray radiation exposure decreased KB cell survival rate detected by colony formation assay. Data represented were Mean ± SD of three identical experiments made in three replicates. Error bars were based on standard deviation of three parallel samples. P values were calculated by the student’s test: **p < 0.01, ***p