Evaluating Single-Cell DNA Damage Induced by Enhanced Radiation

Subscriber access provided by AUSTRALIAN NATIONAL UNIV

Article

Evaluating Single Cell DNA Damage Induced by Enhanced Radiation on Gold Nanofilm Patch Yong Qiao, Yuanshuai Zhou, Tongqian Xiao, Zhiwei Zhang, Liyuan Ma, Ming Su, and Guangli Suo ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b08460 • Publication Date (Web): 06 Oct 2017 Downloaded from http://pubs.acs.org on October 6, 2017

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 20

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

Evaluating Single Cell DNA Damage Induced by Enhanced Radiation on Gold Nanofilm Patch Yong Qiao1,2#, Yuanshuai Zhou1,4#, Tongqian Xiao1,4, Zhiwei Zhang1, Liyuan Ma2,3*, Ming Su2,3* and Guangli Suo1* 1

CAS Key Laboratory of Nano-Bio Interface, Suzhou Institute of Nano-Tech and Nano-Bionics, Chinese

Academy of Sciences, Jiangsu, China 215123. 2Department of Chemical Engineering, Northeastern University, Boston, Massachusetts, United States 02115. 3Wenzhou Institute of Biomaterials and Engineering, Wenzhou Medical University, Zhejiang, China 325001. 4University of Chinese Academy of Sciences, Beijing, China 100049. *Corresponding authors. E-mails: [email protected]; [email protected]; and [email protected]. #

These authors contributed equally to this work.

ABSTRACT Although radiotherapy is a general oncology treatment and often synergistically applied with surgery and chemotherapy, it can cause side effect during and after treatment. Gold nanoparticle was studied as a potential material to enhance radiation to induce damage in cancer cells. But few studies are conducted to examine the effects of gold nanofilm on cell impairment under X-ray treatment. This paper described a microfabrication based single cell array platform to evaluate DNA damage induced by enhanced X-ray radiation on gold nanofilm patches (GNFPes). Cancer cells were patterned on different diameter and thickness of GNFPes, where each cell was attached on one GNFP. The end-point DNA damage induced by X-ray was examined in situ at single cell level using halo assay. The preliminary data demonstrated that the enhancement of DNA damage was significantly related to the area and thickness of the GNFP. This platform may be hopefully used to establish the mathematic relation among DNA damage, X-ray dosage, thickness and area of GNFP, and further contribute to radiation dosage screening for personalized radiotherapy. KEYWORDS: radiation therapy, gold nanofilm patch, single cell array, halo assay, photoelectrons

INTRODUCTION Radiotherapy is one of the most effective treatment modality in cancer therapy. It is used alone or together with surgery, chemotherapy and immunotherapy to kill cancer cells or control them from growth using ACS Paragon Plus Environment


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 2 of 20

ionizing radiation that damages DNA of cancerous tissue.1-4 However, radiotherapy can cause damage to normal healthy tissue and cause side effects in patients during and after treatment. Advances in beam technology including intensity modulated radiotherapy and image-guided radiotherapy can decrease risk of side effects. But these methods are too expensive to be widely used, and can always be used on top of other techniques to achieve additional treatment benefits.5-8 Metal materials, such as gold, bismuth and platinum possess the characteristic to increase the absorption of X-ray with their high atomic (Z) number and generate more photoelectrons and Auger electrons.9-11 When incoming X-ray photons interact with nanoparticles or gold film, the photoelectrons and Auger electrons generated can cause radiolysis of water, produce free radicals, and end in increased DNA damage in cancer cells.9, 12 A body of studies reported that the radiation caused by X-ray could be enhanced by gold nanoparticles endocytosed in cells, and resulted in increased DNA damage.13-16 Dose enhancement of gold film surface was also studied. D. F. Regulla et al. preliminarily reported that the enhanced radiation at the surface of gold film with thickness of 150 µm significantly increased the rate of cell killing in vitro.12 Recently, Rakowski JT et al. derived the dose enhancement factor (DEF) models based on Monte Carlo modeling. The thickness of gold film (nanofilm) was regarded as the impact factors on DEF.17 However, previous works studied ensemble response of a large number of cells or just solid water to radiation and gold film, which exaggerates to the entanglement of experimental parameters in such a way that prevent identification of contribution of each one and cause artificial results. It is quite essential to establish a biological approach for systematic investigation to independently assess each parameter for enhanced radiation. This paper establishes a new micro-fabricated based single cell platform that allows disentangling different parameter involved in gold nanofilm patch (GNFP) enhanced radiation. We fabricated chips arrayed with uniformed round GNFPes possessing different thickness (10, 30, 50 and 100 nm) and diameter (7, 10, 15 and 25 µm). The area of each GNFP just allows one single cell to attach on. After exposing to controlled dose of X-ray radiation, DNA damage in each cell can be precisely quantified with halo assay. Apparently, unlike gold nanoparticle that can be endocytosed in cells and help X-ray to kill cancer cells, our platform may be potentially used for clinical diagnosis for radiotherapy owing to its distinct merits: (1) the material nature, ACS Paragon Plus Environment

Page 3 of 20

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


thickness and diameter of GNFP can be controlled independently and easily, this can ensure the homogeneity of gold material to each cell; (2) the effect of gold material itself on cells can be neglected because no endocytosis occurs; (3) a variety of cancer cells can be attracted onto GNFPes through different surface chemistry or ligand interaction, and exposed to X-ray radiation; (4) DNA damage in each cell is determined precisely in situ without transferring cells and at single cell level without ensemble mixing up; (5) different types of cells can be synchronously patterned onto the same chip and this possibly contributes to cancer heterogeneity research in radiotherapy; (6) the assay is performed at single cell level, thus needs only a small number of cells and can improve the utilization efficacy of cell, has single cell sensitivity, and does not have an overlapping issue; (7) the chip allows automated imaging analysis and robust statistical analysis and requires minimal user intervention or special tools; (8) this method can be made compatible with existing high throughput techniques (24 or 96 wells), which allows its use in drug screening.

RESULTS AND DISCUSSION This paper describes a proof-of-concept study that GNFP with different thickness and area (diameter) can cause different enhancement effect of X-ray radiation, and therefore induce different levels of cell impairment. A biophysical model was developed to simulate the relationship between radiation enhancement of X-ray and gold film, and to allow extrapolation of enhanced impairment of cells attached on the film. In the model, we assume that one single cell, as a thin slab, is attached on the surface of one Gold Plate (GP), also namely GNFP, and exposed to a dose of X-ray. Considering the common size of A172 and HeLa cells in monolayer (diameter is about 10 to 30 µm), we simulated the relationship between diameter (from 7 to 30 µm) of gold film and emitted photoelectron. In terms of simulation, the number of photoelectron released will increase with the increase of surface area. However, only a certain area of GNFP (diameter ≤ 30 µm) was considered in our study to ensure one cell just attach one GNFP. Here, 15 µm of diameter of GP was defined for simulation of the relationship between thickness of GP and dose enhancement factor (DEF). The DEF was calculated through a computational pipeline based on Monte Carlo simulations. Figure S1 shows the variation of cellular DEF as a function of thickness of GP with defined 15 µm diameter and the DEF as a function of ACS Paragon Plus Environment


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 4 of 20

diameter of GP with defined 30 nm thickness. When cells are irradiated by a 50 kVp of X-ray source, DEF increases from 1-1.35 with the increase of thickness from 10-200 nm, and DEF increases from 1-1.07 with the increase diameter from 7-30 µm. In our study, the space between adjacent GNFPes on a chip is large enough (100 µm) to separate the single cells, so the effect of electrons from neighboring GNFPes is ignored in the model. In light of the simulation, we designed and fabricated the GNFP chip with different thickness and area by photolithography and metal deposition for single cell array. The design scheme is showed in figure 1. Briefly, a Su-8 photoresist covered micro-well pattern is formed by photolithography with proper photo mask. Then GNFPes are deposited on the substrate using thermo evaporator. After PEG-silane modification, cancer cells are selectively attached on GNFPes rather than silica surface. X-ray radiation can be enhanced by GNFP and causes increased DNA damage in cells. The damaged DNA is precisely measured through single cell halo assay that was used in previous studies such as drug screening, DNA damage induced by X-ray, cytotoxicity and genotoxicity caused by heavy-metal ions and nanomaterials.16, 18, 19 First of all, we evaluated the effect of GNFP on cell viability and cell cycle. The round GNFP with 30 nm thickness and 10 µm diameter was formed and arrayed on the silicon wafer surface (Figure 2A). Through modification with PEG on silicon surface, the silicon region was passivated with PEG. Afterwards, A172 cells were loaded onto the PEG-modified GNFP chip and were captured by the patch via cell characteristic of adherent growth. Figure 2B exhibits the formed single cell array on the modified GNFP chip, in which the patch area size is controlled to ensure just only one cell can be attached on one GNFP. After patterning the cells, 97% of patches were occupied by cells and 86% of patches were attached only by one cell per patch (Figure 2C). To evaluate the effect of GNFP on cells, the viability of patterned cells was first examined with live/dead assay (Live/Dead viability/cytotoxicity kit, Invitrogen, Carlsbad, CA). Here, dead cells were labeled as red and live cells were labeled as green. The representative fluorescence image (Figure 2D) shows that A172 cells attached on GNFPes are live well after 12 hours culture. To examine if the GNFP itself lead to A172 cell cycle variability or induce cell apoptosis, we detected PI stained DNA fragmentation with flow cytometer at 24 hours post-cell culture on GNFP chip or silicon wafer chip. The result demonstrates that, as compared with control ACS Paragon Plus Environment

Page 5 of 20

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


cells cultured on silicon wafer chip, cells on GNFP show no change of the populations of G0/G1 phase, S phase and G2/M phase. Small population of Sub-G1 indicates no apoptosis occurs in the cells on silicon wafer as well as GNFP (Figure 2E and F). In other words, GNFP shows no toxicity over cells cultured on it due to its excellent biocompatibility. Conversely, gold nanoparticles currently used in numerous medical applications cause cellular toxicity for their uptake in the cells and accumulation in vacuoles.20 Next, we examined the DNA damage caused by X-ray in cells on GNFP using single cell halo assay. A172 cells patterned on GNFP chip were exposed on 0 Gy (Figure 2G) and on 0.75 Gy of X-ray (Figure 2H), and then were embedded in agarose gel for halo assay. Figure 2G shows No DNA diffusion from nucleus. This indicates that no DNA damage occurs in cells without X-ray treatment (0 Gy). However, obvious halos are formed in the chip subjected to 0.75 Gy X-ray treatment (Figure 2H). Halo morphology is characterized by nucleus boundary becoming unclear and by diffused area growing bigger and brighter. The halo area is considered to consist of relaxed loops and damaged fragments. Since the DNA amount is indicated by fluorescence intensity of EB, the level of DNA damage can be determined according to the dimension of halo and nucleus. A typical halo from an X-ray treated cell is showed in figure 2I. Here, the damaged DNA diffuses out from nucleus and forms a smear circle surrounding the compact nucleus. Due to clear boundary, symmetric shapes of halos and nuclei, and non-overlapping nature of adjacent cells/halos, a relative nuclear diffusion factor (rNDF) can be derived accurately from the formula rNDF = (R2 - r2) / r2, where R and r are the diameters of large halo circle and small nucleus circle, respectively (Figure 2I). Fluorescence intensity is proportional to the yield of DNA in any given image, thus the level of DNA damage can be evaluated by measuring the radius of halos and nuclei in above equation. Although the halos form single cells appear to be not uniform due to cell heterogeneity (Figure 2J), the reliability of data could be ensured by acquiring and analyzing data from a large number of cells. Our research fundamentally bases on an assumption that halo assay can quantify the DNA damage at single cell level and then assess cell impairment. Here, we validated this assumption by comparison of halo assay (rNDF) with reactive oxygen species (ROS) and apoptosis. DNA damage response (DDR) plays an important role against detrimental effects of stress on cells, such as X-ray radiation. It coordinates many ACS Paragon Plus Environment


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 20

biological processes including DNA repair, regulation of cell cycle checkpoints, transcription of DDR related genes, and autophagy.21 When DNA damage is failed to be repaired, the cells execute programmed cell death, most often apoptosis. To evaluate if the DNA damage data derived from halo assay (rNDF) are eligible to indicate cell impairment, we designed paralleled experiments of X-ray treatment to compare the correlation between rNDF and ROS, and between rNDF and apoptosis. Here, A172 cells were seeded on (1) silicon wafer chip without radiation (Figure 3A, E and I); (2) GNFP chip without radiation (Figure 3B, F and J); (3) silicon wafer chip with 1.25 Gy of X-ray radiation (Figure 3C, G and K); (4) GNFP chip with 1.25 Gy of X-ray radiation (Figure 3D, H and L). After treatment of X-ray radiation, the cells were in parallel performed single cell halo assay (Figure 3A-D), ROS assay (Figure 3E-H) and apoptosis assay (Figure 3I-L). Without X-ray radiation treatment, A172 cells cultured no matter on silicon wafer chip or on GNFP chip show no DNA damage from single cell halo assay (Figure 3A and B) and almost no apoptosis from Annexin V/PI assay (Figure 3I and J). In addition, the ROS levels from A172 cells on silicon wafer chip or on GNFP chip are low and almost equal (Figure 3E and F). A certain level of DNA damage exhibited by smeared halo circle in halo assay is shown in A172 cells treated with 1.25 Gy of X-ray radiation show (Figure 3C and D). In terms of rNDF data, 1.25 Gy of X-ray radiation induces significantly more DNA damage in cells on GNFP chip than in cells on silicon wafer chip. In parallel, more ROS level and apoptosis proportion are detected in cells on GNFP chip. The analysis from Pearson correlation shows the high positive correlation between rNDF and ROS (R2 = 0.92) (Figure 3M), and between rNDF and apoptosis (R2 = 0.99) (Figure 3N). This distinctly indicates that enhanced radiation resulted from GNFP induces more DNA damage and therefore cause more cell impairment. In previous studies, the size and density of gold nanoparticle were regarded as key parameters to affect radiation enhancement of X-ray.9, 22 According to the model we developed, the area and thickness of GP are main factors to affect DEF. We hereby are attempted to validate the speculation that the area and thickness of GNFP play the similar role and cause increased DNA damage in cells exposed to X-ray. We fabricated a series of chips coated with different thickness and diameter of GNFPes and assessed how these parameters affect single cell DNA damage induced by X-ray radiation. Figure 4A-D shows optical images of single A172 cells attached onto GNFP chip with identical thickness (30 nm) and different diameter (7, 10, 15 and 25 µm). When ACS Paragon Plus Environment

Page 7 of 20

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


the diameters of GNFP are smaller than those of cells (figure 4A-C), the attached cells are imaged as dark shadows, while GNFPes are imaged as yellow disks in the bright field images. When the diameters of GNFP are larger than those of cells (figure 4D), dark ring that reflects contrast difference of cells and gold can indicate attached cells. After exposed to 1.25 Gy X-ray radiation, the single cell halo assay was performed. The curve graph of rNDF values in function of diameters of GNFP is shown in figure 4E, where DNA damage increases from 5 to 10 (rNDF values) with increasing diameters of GNFP from 0 to 25 µm. Here, 0 µm patch means that cells are just attached on silicon wafer chip. The increasing trend almost reaches plateau when the diameter of GNFP is 25 µm. In the absence of X-ray radiation, different diameters of GNFP causes no DNA damage in cells, the same as silicon wafer chip dose (0 µm diameter GNFP). To examine the effect of thickness of GNFP on DNA damage induced by X-ray, single A172 cells were attached on 15 µm diameter GNFP chip, where each cell occupied one GNFP. The thickness of GNFP was determined by atomic force microscope (AFM) to be 10, 30, 50 and 100 nm (Figure 4F). The figure S2 indicates nanoscale granular nature of film, suggesting the surface of GNFP could be treated as a group of nanoparticles assembled in two dimensions. These nanoparticles with < 10 nm height and < 5 nm width were simply characterized by AFM. After exposed to 1.25 Gy of X-ray radiation, DNA damage was examined with single cell halo assay. Figure 4G shows that rNDF values increase from 5 to 11.50 along with increasing thickness of GNFP from 0 to 100 nm. The GNFPes significantly enhance DNA damage, compared with silicon surface (thickness of 0 nm). The increasing trend reaches plateau when the thickness of GNFP exceeded 50 nm. Further, we examined DNA damage induced by different doses of X-ray (0.25, 0.75, 1.25, 2.5 Gy) in A172 and HeLa single cells attached on GNFP with fixed diameter (15 µm) and thickness (30 nm) (Figure 4H). The general trend is that cells attached onto GNFPes are experienced a large amount of DNA damage compared to those attached on silicon wafer chip; higher dose of X-ray radiation produces more DNA damage when other conditions are identical; and there is little variation in the radiosensitivity of two cell lines (A172 and HeLa cells) studied in this project. Notably, the increasing trend of rNDF in response to the increase of diameter or thickness of GNFP is not completely coincident with the model’s predicted result for DEF. The model we established cannot capture ACS Paragon Plus Environment


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 20

all the possible interactions in a complete manner. Numerous factors in experiment may contribute to this inconsistency. In the model, we assume that all the emitted photoelectrons from one GP are absorbed by one cell attached on the film. When GNFP is oversized and exceeds the coverage of one single cell, the photoelectrons emitted from the excess gold film cannot fully be absorbed by cell. Lots of complicated biological processes are involved in DDR induced by radiation. The relationship between cell DDR and photoelectron energy may not be simply linear. The characterization of gold nanoparticle film was reported in other studies.17, 23 But how the surface roughness affects the cells remains unclear. In nanoparticle enhanced radiation therapy, cells are decorated by a number of randomly distributed nanoparticles; while in the current platform, only the bottom side of cells contact with substrate. This asymmetrical placement of nanoparticles on surface of GNFP may affect DNA damage in different manners under X-ray treatment, as compared with nanoparticles in cells. Ongoing work about this platform is focused on this question. This platform may be useful to establish the mathematic relation for different kinds of human cells among DNA damage, X-ray dose, thickness and area of GNFP. When performing clinical diagnosis, single cancer cells and normal cells from one patient will be seeded on the chip patterned with different thickness and area of GNFPs (one single cell attached on one GNFP), after collecting DNA damage data from single cells subjected to known dosage of X-ray, we can deduce the optimal dosage of X-ray leading to desirable DNA damage of human tumors based on known parameters (dosage of X-ray used, desirable DNA damage, thickness and area of GNFPs). This personalized optimal dosage can maximize the curative effect of radiotherapy and simultaneously minimize the toxicity imposed on the patient body. Otherwise, heterogeneity is an obstacle to cancer radiotherapy. In the future, we may define optimal dosage of radiation for different type of cells in a bulk of cancer cells on the GNFP chip. By means of this platform, the accurate dosage of X-ray can be deduced to selectively eradicate the most obstinate cancer stem cells in the human body. We can deduce the optimal dosage of X-ray leading to desirable DNA damage of human tumors based on known parameters (dosage of Xray used, desirable DNA damage, thickness and area of GNFPs).

ACS Paragon Plus Environment

Page 9 of 20

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


CONCLUSIONS A micro-fabrication based single cell array platform was established to disentangle various experimental parameters in GNFP for DNA damage induced by enhanced radiation. Cancer cells are patterned on varying diameter and thickness of GNFPes, where each cell attaches on one GNFP. The end-point DNA damage upon X-ray exposure is examined in situ at single cell level using halo assay. This efficient method was confirmed to be qualified to evaluate the cell impairment by comparison with ROS and apoptosis assays. We found that, at some extend, the area and thickness of GNFP were two key factors to affect the enhancement of DNA damage induced by x-ray.

EXPERIMENTAL SECTION Chip fabrication An array of GNFP chip is generated on a silicon substrate using photolithography and patch deposition as follows. A layer of positive photoresist (thickness of 1.1 µm, AZ4620, Fisher Scientific, Pittsburgh, PA) is spun on a silicon substrate (P-type boron-doped resistivity of 8–25 Ωcm-2, University Wafer, South Boston, MA). An array of circles with different diameter (7, 10, 15 and 25 µm) and constant distance of 100 µm between the closest two neighbor patches is generated on the substrate by exposing to ultraviolet light (365 nm, UVP, Upland, CA) through mask with different feature sizes and washed by AZ400K developer (Fisher Scientific, Pittsburgh, PA). After deposited 2 nm of chromium, gold film (GNFP) of different thickness (10, 30, 50 and 100 nm) is deposited on the substrate using thermal evaporator, which is followed by lifting off photoresist in sonicated water bath. The thickness of GNFPes is determined by atomic force microscope (AFM) (Dimension 3100, Veeco, NY, USA). After rinsing and drying, the uncovered silicon region is modified with polyethylene glycol (PEG) to avoid non-specific binding by incubating in 3 mM of PEG-silane (molecular weight 472-604, Gelest, Tullytown, PA) in toluene, containing 1% triethylamine (TEA, Sigma-Aldrich, St. Louis, MO) (v/v) as catalyst for 2 hours at 60°C in nitrogen.24, 25 After removing unbound PEG-silane by sonication in toluene or ethanol, the substrate is washed with water and dried in a gentle flow of nitrogen. PEGcovered silicon surface can effectively prevent cell attachment, while GNFPes can attract cells. As a control ACS Paragon Plus Environment


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 10 of 20

group, the cells were arrayed on the silicon wafer chip which is fabricated by surface chemistry modification and laser lithography followed by our previous study.16 As a control chip, the bare silicon wafer is modified by PEG-silane as above. After PEGylation, a conformal contact is done on the silicon by PEI modified PDMS stamp25. This control chip is named silicon wafer chip. Cell culture Two human cancer cell lines, HeLa (human cervical cancer) and A172 (human brain glioblastoma), were obtained from American Type Culture Collection (ATCC Manassas, VA), and cultured in standard conditions (5% CO2 in air at 37°C) in RPMI-1640 medium (Thermo Scientific, Logan, UT) supplemented with 10% (v/v) fetal bovine serum (Thermo Scientific, Logan, UT) and 1% (v/v) penicillin/streptomycin (Thermo Scientific, Logan, UT). After cell monolayer reached 70-80% confluency, cells were trypsinized with 0.25% trypsin/0.53 mM EDTA solution (Thermo Scientific, Logan, UT) at 37°C for 3 minutes, followed by adding fresh medium at room temperature to neutralize trypsin. After centrifugation and resuspension in fresh medium, cell viability was determined by staining with Trypan blue (VWR, West Chester, PA) and cell number was counted with hemocytometer (Horsham, PA). Cell cycle and Sub-G1 assay To examine the effect of gold surface on cell cycle, the A172 cells patterned on the GNFP chip and silicon wafer chip were cultured for 24 hours and detached, collected by centrifugation at 1200 rpm for 4 minutes at room temperature, and resuspended in 1 ml ice cold PBS buffer. Cell suspension was added dropwise to 1 ml of 70 % ethanol and stored at 4 °C to for 2 hours. Cells were collected from ethanol by centrifugation at 1200 rpm for 10 minutes at 4 °C. The collected cells were stained with 500 µl propidium iodide (PI) (20 µg/ml, Thermo Scientific, Logan, UT) containing 0.1 % Triton X-100 for 15 minutes at 37 °C and assessed with a BD Accuri C6 flow cytometer (BD Biosciences). Cells collected from silicon wafer chip were used as a control. The percentage of cells in Sub-G1 (apoptotic cells), G0/G1, S phase, and G2/M phase was calculated using CellQuest software. X-ray treatment and single cell halo assay


Page 11 of 20

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


Single cell halo assay was previously used to quantify DNA damage level.16, 18 Cells are seeded onto silicon wafer chip (control group) or GNFP chip at a density of 106 cells/ml and incubated for 30 minutes. After incubation, unattached cells are rinsed away by phosphate buffer saline (PBS). The viability of patterned cells is tested with live/dead assay (Invitrogen, Carlsbad, CA). The substrate with cells is placed onto a large piece of gelbond (Lonza, Rockland, ME), followed by dropping a 1.5 ml 1% low melting point (LGT)-agarose (Invitrogen, Carlsbad, CA) on top of the substrate. The gelbond is used to cover the silicon substrate together with agarose. The substrate is kept at room temperature for 5 minutes to allow gel solidification. A Mini-X Xray tube from Amptek (Bedford, MA) with a silver anode operating at 50 kV and 100 µA is used to produce Xray. The X-ray tube is fitted with a brass collimator (2 mm diameter pinhole) to focus X-rays onto the target. For single cell halo assay, cells treated or not treated with X-ray are immersed in 0.3 M NaOH for 30 minutes at room temperature, and stained in 10 µg/ml ethidum bromide (EB, Alfa Aesar, Ward Hill, MA) solution for 15 minutes. The stained slide is incubated in deionized water for 5 minutes to remove unbound dye, and imaged using an Olympus BX51M fluorescent microscope (Olympus) coupled with an Olympus ColorView CCD camera (Hunt Optical & Imaging, Pittsburgh, PA). The images are analyzed by ImageJ (NIH), and used to calculate relative nuclear diffusion factor (rNDF). ROS assay The A172 cells were cultured on the silicon wafer chip and GNFP chip at 24 hour after cell seeding, the culture medium was renewed and cells are exposed to X-ray. After exposed, medium was removed. Cells were then washed once with warm PBS gently, applied with a sufficient amount of the 25 µM carboxy-H2DCFDA (Invitrogen, Carlsbad, CA) working solution, and incubated for 30 minutes with dye at 37°C. The silicon wafer chip or GNFP chip with patterned cells was gently washed for three times by warmed PBS. The A172 cells were trypsinized, washed and resuspended in PBS, pH 7.3 at a concentration of approximately 1×106 cells/ml. The stained suspension was shielded from light and stored on ice until analysis. Flow cytometry measurements were performed on samples using a BD Accuri C6 System flow cytometry system (BD, San Jose, CA) using the 488-nm laser measuring forward and orthogonal light scatter and green fluorescence. Fluorescence data were obtained from 10,000 viable cells per sample. ACS Paragon Plus Environment


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 12 of 20

Annexin V/PI assay Apoptosis was determined by staining A172 cells with Annexin V-FITC apoptosis detection kit (Invitrogen, Carlsbad, CA). In brief, the cells on silicon wafer chip or GNFP chip were incubated overnight after treatment with X-ray. The cells were trypsinized and washed twice by cold PBS then resuspended in 1× binding buffer (10 mM HEPES/NaOH pH 7.4, 140 mM NaCl, 2.5 mM CaCl2) at a concentration of 1×106 cells/ml. Five µl of annexin V-FITC and 10 µl propidium iodide solution were then added to those cells, and incubated cells at room temperature for exactly 10 minutes and protect from light. At least 10,000 cells were counted and analyzed using flow cytometer. Statistical analysis The results represented the mean of at least three independent experiments. For the single cell halo assay, each data point was averaged from at least 100 individual cells. The mean and error values were calculated using OriginPro 8.5. All data were presented as the mean with standard deviation. The statistical significance of results was determined by means of an analysis of variance using the SPSS software (SPSS 19.0, IBM, Armonk, NY). Comparisons between control group and treatment group were based on t-test. A result was considered statistically significant difference when P≤0.05. The Pearson correlation test was performed using GraphPad Prism 4 package (GraphPad Software) to determine the correlation between the data from halo assay and data from ROS or apoptosis assay. Here, Pearson’s correlation coefficient, R2, rang from 0-1, where 1 is a total positive and 0 is no correlation. Supporting Information. Simulations of dose enhancement factor with different parameters of GNFP; the roughness surface structure of GNFP

ACKNOWLEDGEMENTS This work has been generously supported by Ministry of Science and Technology (MOST) (Grant No. 2017YFA0104301), Ministry of Science and Technology (MOST) (Grant No. 2014CB965003), National Natural Science Foundation of China (Grant No. 31471307) and One Hundred Person Project from Chinese Academy of Sciences to Guangli Suo, a New Investigator Research Award from Bankhead-Coley Cancer ACS Paragon Plus Environment

Page 13 of 20

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


Research Program to Liyuan Ma, and a Director’s New Innovator Award from National Institute of Health to Ming Su.

FIGURE CAPTIONS Figure 1. Scheme of GNFP chip fabrication and single cell halo assay. Figure 2. The feature of GNFP chip and single cell halo assay. A) Optical image of arrayed GNFPes on the silicon wafer. B) Optical image of single A172 cell array on the GNFP chip, in which A172 cells only can be attached on the gold surface but are repelled by modified silicon surface. C) The histogram of radios of no cell, one cell and two or more cells attached on one gold thinfilm patch on chip. D) Fluorescence image of viability of cells tested with live/dead assay, in which dead cells are labeled as red, and living cells are labeled as green. Flow cytometer assay of cell cycles and Sub-G1 assay for A172 cells on silicon wafer chip (E) and on the GNFP chip (F). The proportions of Sub-G1, G0/G1, S and G2/M phases are labeled in histograms. G) Fluorescent image of A172 cells on GNFP exposed on 0 Gy of X-ray irradiation and H) on 0.75 Gy of X-ray radiation. I) The image of a typical halo. R and r are the diameters of large halo circle and small nucleus circle, respectively. Here, 10 µm diameter and 30 nm thickness of gold patch was used for these experiments. J) The histogram of heterogeneity analysis of cell with different cell cycle and DNA content based on single cell halo assay on one chip. Figure 3. Comparison of single cell halo assay with ROS and apoptosis assays. Single cell halo assay of A172 cells attached on silicon wafer (A and C) or on GNFPes (B and D) are subjected to 1.25 Gy of X-ray treatments (C and D) or no X-ray treatment (A and B). The rNDF values exhibited in the A-D images are average data and standard deviation (SD) data calculated from 100 halos. (E-H) The histogram of flow cytometry of paralleled ROS assay. The proportion of ROS positive cells are shown in the V1-R region in the histogram. (I-L) The dot flow cytometry of paralleled apoptosis assay by Annexin V-FITC/PI staining. The proportion of apoptosis cells are shown in the Q1-LR region in the dot plot images. (M) Pearson correlation between halo assay and ROS assay, and (N) between halo assay and apoptosis assay. R2 means coefficient of determination. The error bars



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 14 of 20

mean SDs from three independent repeats. Here, 10 µm diameter and 30 nm thickness of gold patch is used for these experiments. Figure 4. The effect of thickness and diameter of gold patch on DNA damage induced by X-ray. A-D) Arrayed A172 cells on the different diameter (0, 7, 10, 15 and 25 µm) of GNFPes. E) DNA damage of arrayed A172 cells on the different diameter of GNFPes exposed to 1.25 Gy X-ray radiation. The DNA damage is evaluated using single cell halo assay. The arrayed cells not exposed to X-ray radiation are used as control. Here, as the control group, “0” diameter means A 172 cells on the silicon wafer chip. F) The different thickness of single GNFP is characterized by AFM. G) DNA damage of arrayed A172 cells on the different thickness (0, 10, 30, 50 and 100 nm) of GNFPes exposed to 1.25 Gy X-ray radiation. The arrayed cells not exposed to X-ray radiation are used as negative control. Here, as the control group, “0” means cells on the silicon wafer chip. H) Evaluation of DNA damage of different cancer cell lines (A 172 and HeLa) patterned on the GNFP chip (15 µm diameter and 30 nm thickness of gold patches) and silicon wafer chip, which are exposed to various doses of Xray radiation (0, 0.25, 0.75, 1.25 and 2.5 Gy). The error bars mean SDs from three independent repeats.

REFERENCES (1) Guandalino, M.; Dupre, A.; Francois, M.; Leroy, B.; Antomarchi, O.; Buc, E.; Dubois, A.; Guy, L.; Pezet, D.; Gagniere, J. Previous Radiation for Prostate Neoplasm Alters Surgical and Oncologic Outcomes after Rectal Cancer Surgery. J. Surg. Oncol. 2015, 112, 802-808. (2) Teng, F.; Kong, L.; Meng, X.; Yang, J.; Yu, J. Radiotherapy Combined with Immune Checkpoint Blockade Immunotherapy: Achievements and Challenges. Cancer Lett. 2015, 365, 23-29. (3) Haikerwal, S. J.; Hagekyriakou, J.; MacManus, M.; Martin, O. A.; Haynes, N. M. Building Immunity to Cancer with Radiation Therapy. Cancer Lett. 2015, 368, 198-208. (4) Ree, A. H.; Redalen, K. R. Personalized Radiotherapy: Concepts, Biomarkers and Trial Design. Br. J. Radiol. 2015, 88, 1-13. (5) Graf, R.; Boehmer, D.; Budach, V.; Wust, P. Interfraction Rotation of the Prostate As Evaluated by Kilovoltage XRay Fiducial Marker Imaging in Intensity-Modulated Radiotherapy of Localized Prostate Cancer. Med. Dosim. 2012, 37, 396-400. (6) Yao, Q.; Zheng, R.; Xie, G.; Liao, G.; Du, S.; Ren, C.; Li, R.; Lin, X.; Hu, D.; Yuan, Y. Late-Responding Normal Tissue Cells Benefit from High-Precision Radiotherapy with Prolonged Fraction Delivery Times via Enhanced Autophagy. Sci. Rep. 2015, 5, 1-7. (7) Kadoury, S.; Cheriet, F.; Labelle, H. Personalized X-Ray 3-D Reconstruction of the Scoliotic Spine From Hybrid Statistical and Image-Based Models. IEEE T. Med. Imaging 2009, 28, 1422-1435. (8) Zhang, Y.; Knopf, A.; Tanner, C.; Boye, D.; Lomax, A. J. Deformable Motion Reconstruction for Scanned Proton Beam Therapy Using On-Line X-Ray Imaging. Phys. Med. Biol. 2013, 58, 8621-8645. (9) Hossain, M.; Su, M. Nanoparticle Location and Material Dependent Dose Enhancement in X-Ray Radiation Therapy. J. Phys. Chem. C 2012, 116, 23047-23052. (10) Rahman, W. N.; Bishara, N.; Ackerly, T.; He, C. F.; Jackson, P.; Wong, C.; Davidson, R.; Geso, M. Enhancement of Radiation Effects by Gold Nanoparticles for Superficial Radiation Therapy. Nanomed.- Nanotechnol. 2009, 5, 136-142. ACS Paragon Plus Environment

Page 15 of 20

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


(11) Su, M. Multiplexed Detection of Molecular Biomarkers with Phase-Change Nanoparticles. Nanomedicine (London, U. K.) 2013, 8, 253-263. (12) Regulla, D. F.; Hiebert, L. B.; Seidenbusch, M. Physical and Biological Interface Dose Effects in Tissue due to XRay-Induced Release of Secondary Radiation from Metallic Gold Surfaces. Radiat. Res. 1998, 150, 92-100. (13) Liu, C. J.; Wang, C. H.; Chien, C. C.; Yang, T. Y.; Chen, S. T.; Leng, W. H.; Lee, C. F.; Lee, K. H.; Hwu, Y.; Lee, Y. C.; Cheng, C. L.; Yang, C. S.; Chen, Y. J.; Je, J. H.; Margaritondo, G. Enhanced X-Ray Rrradiation-Induced Cancer Cell Damage by Gold Nanoparticles Treated by A New Synthesis Method of Polyethylene Glycol Modification. Nanotechnology 2008, 19, 1-5. (14) Wang, C.; Sun, A.; Qiao, Y.; Zhang, P.; Ma, L.; Su, M. Cationic Surface Modification of Gold Nanoparticles for Enhanced Cellular Uptake and X-Ray Radiation Therapy. J. Mater. Chem. B 2015, 3, 7372-7376. (15) Wood, K. D.; Weingeist, M. D.; Bhatia, N. S.; Engelward, P. B. Single Cell Trapping and DNA Damage Analysis Using Microwell Arrays. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 10008-10013. (16) Qiao, Y.; Wang, C.; Su, M.; Ma, L. Single Cell DNA Damage/Repair Assay Using HaloChip. Anal. Chem. 2012, 84, 1112-1116. (17) Rakowski, J. T.; Laha, S. S.; Snyder, M. G.; Buczek, M. G.; Tucker, M. A.; Liu, F.; Mao, G.; Hillman, Y.; Lawes, G. Measurement of Gold Nanofilm Dose Enhancement Using Unlaminated Radiochromic Film. Med. Phys. 2015, 42, 59375944. (18) Ma, L.; Xun, X.; Qiao, Y.; An, J.; Su, M. Predicting Efficacies of Anticancer Drugs Using Single Cell HaloChip Assay. Analyst 2016, 141, 2454-2462. (19) Qiao, Y.; An, J.; Ma, L. Single Cell Array Based Assay for In Vitro Genotoxicity Study of Nanomaterials. Anal. Chem. 2013, 85, 4107-4112. (20) Mironava, T.; Hadjiargyrou, M.; Simon, M.; Jurukovski, V.; Rafailovich, M. H. Gold Nanoparticles Cellular Toxicity and Recovery: Effect of Size, Concentration and Exposure Time. Nanotoxicology 2010, 4, 120-137. (21) Czarny, P.; Pawlowska, E.; Bialkowska-Warzecha, J.; Kaarniranta, K.; Blasiak, J. Autophagy in DNA Damage Response. Int. J. Mol. Sci. 2015, 16, 2641-2662. (22) Verma, A.; Stellacci, F. Effect of Surface Properties on Nanoparticle-Cell Interactions. Small 2010, 6, 12-21. (23) Lansåker, P. C.; Hallén, A.; Niklasson, G. A.; Granqvist, C. G. Characterization of Gold Nanoparticle Films: Rutherford Backscattering Spectroscopy, Scanning Electron Microscopy with Image Analysis, and Atomic Force Microscopy. AIP Adv. 2014, 4, 1-6. (24) Lan, S.; Veiseh, M.; Zhang, M. Surface Modification of Silicon and Gold-Patterned Silicon Surfaces for Improved Biocompatibility and Cell Patterning Selectivity. Biosens. Bioelectron. 2005, 20, 1697-1708. (25) Wang, Z.; Zhang, P.; Kirkland, B.; Liu, Y.; Guan, J. Microcontact Printing of Polyelectrolytes on PEG Using an Unmodified PDMS Stamp for Micropatterning Nanoparticles, DNA, Proteins and Cells. Soft Matter 2012, 8, 7630-7637.



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

Photoresist coated

Halo

Exposed and developed

Page 16 of 20

Au patches array Au

Single cell

Photoresist Silicon wafer Gold surface PEGylation Coated agarose Cell

X-ray

Halo assay

Single cell array

PEG modification


Qiao, et al. Figure 1

Page 17 of 20

Single cell array (GNFP:10µm of diameter, 30nm of thickness) GNFP chip

Single cell on chip

C

100

B Percentage (%)

A

Live/Dead assay

100 µm

100 µm

86%

D

50 µm

75 50 25 0

11%

3%

200 µm

1 0 2 The cell number on one patch

Cell cycle and Sub-G1 assay G0/G1 72.2%

F S 5.4%

S 5.0%

G2/M 12.0%

0

G2/M 12.3%

Apoptotic 8.7%

10,000

Count

20,000

20,000

Apoptotic 8.5%

10,000

Count

G0/G1 71.9%

E

GNFP chip

25,000

25,000

Silicon wafer chip

0 0

500,000

1000,000

1,500,000

2,000,000

0

500,000

DNA content-A

1000,000

1,500,000

2,000,000

DNA content-A

Single cell halo assay (GNFP chip: 10µm of diameter, 30nm of thickness) X-ray dose: 0Gy

0.75Gy

G

0.75Gy

ID

H 50 µm

0.75Gy

J8

50 µm

6

R

rNDF

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


r

4 2

100 µm

100 µm

10 μm

0

1

2

3

4

5

6

7

8

9

Defined cell from left to right




105

FL2-H

104 102

107.2

V1-R 1.6%

J

105

Q1-UR 0.3%

104

103

105

FL1-H

Q1-UL 0.0%

106

103

104

105

FL1-H

Q1-UL 0.0%

106

107.2

Q1-UR 2.1%

FL2-H

104

100

103

Q1-LR 86.3%

0

102

100 μm 101

102

103

104

FL1-A

105

106

107.2

102

103

104

105

106

Apoptotic cells (%)

10

Q1-LR 44.7%

105

L

0 1

2

3

4

5

6

7

5

6

7

N

FL2-H5

V1-R 94.6%

20

100

102

107.2

40

rNDF

104

V1-L 5.4%

106

60

107.2

Q1-UR 6.6%

102

FL1-A

105

107.2

104

106

103

R2=0.92615

80

0

103

100 300

102

100

Q1-LR 0.1% 102

K

200

H Count

107.2

FL2-H 107.2

0

GNFP chip (10 µm, 30nm)

Q1-UL 0.1%

106

104

106

V1-R 28.9%

V1-L 77.1%

101

D

105

FL1-H

102 105

107.2

104

FL1-A

106

103

100 μm rNDF=6.18±0.49

104

103

103

Count

100 300

102

200

G Count

rNDF=3.24±0.27

M

Q1-LR 0.0%

102

V1-L 98.4%

106

107.2

FL1-A

105

106

104

0

Silicon wafer

103

100 μm

C

Q1-UR 0.3%

103

Count

300

F

102

200

B

rNDF=1±0.034

Q1-UL 0.0%

ROS positive cells (%)

I

100 μm

101

1.25 Gy of X-ray

V1-R 2.0%

107.2

V1-L 98.0%

106

300

E 200

rNDF=1±0.077

100

Silicon wafer

A

Annexin V/PI

ROS assay

101

GNFP chip (10 µm, 30nm)

No X-ray

Single cell halo

0

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

Page 18 of 20

R2=0.9914

80 60 40 20 0 0

1

2

3

4

rNDF

107.2

FL1-H



Page 19 of 20

GNFP thickness: 30nm 10 μm 15 μm

GNFP diameter : 7 μm

B

C

D

50 µm

50 µm

50 µm

50 µm

12

8 6

25

30nm of thickness 15

10

rNDF

25 μm

A

7

10

0

0 Gy 1.25 Gy

4 2 0

5

10

15

20

25

F 120

Thickness by AFM(nm)

E

10μm of diameter 80

40

0

G 12

H

30 0 Gy 1.25 Gy

8 6 0

10μm of diameter

A172 on Au HeLa on Au

8 6

2 20

40

60

80

100

0

Thickness of gold patch (nm) ACS Paragon Plus Environment

2.5

0.75

4

2

A172 on Si HeLa on Si 1.25

0.25

4

0

100

12 10

10

50

Designed thinckness (nm)

rNDF

10

100

50

30

10

Diameter of gold patch(m)

rNDF

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


10μm of diameter 30nm of thickness

0 0.0

0.5

1.0

1.5

2.0

2.5

X-ray dose (Gy) Qiao, et al. Figure 4


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

Table of Contents (TOC)


Page 20 of 20

Evaluating Single-Cell DNA Damage Induced by Enhanced Radiation

Recommend Documents