Electron Paramagnetic Resonance Spectroscopy Investigation of


Apr 28, 2016 - Figure 4B shows the processed data (average and standard deviation) and fitted enhancement slope. We used the definition of enhancement...
1 downloads 11 Views 3MB Size


Article pubs.acs.org/JPCA

Electron Paramagnetic Resonance Spectroscopy Investigation of Radical Production by Gold Nanoparticles in Aqueous Solutions Under X‑ray Irradiation Joan Chang, Ryan D. Taylor, R. Andrew Davidson, Arjun Sharmah, and Ting Guo* Department of Chemistry, University of California, Davis, California 95616, United States ABSTRACT: Nanomaterials can enhance the effect of X-rays, but the mechanisms of enhancement can be complicated. Electron paramagnetic resonance (EPR) was used here to shed light on enhancement mechanisms by detecting the originally proposed physical enhancement of the effect of Xrays by as-made large gold nanoparticles. Specifically spin trap reagent 5-tertbutoxycarbonyl-5-methyl-1-pyrroline-N-oxide (BMPO) was used to trap radicals produced in aqueous solutions under X-ray irradiation. Even though only BMPO hydroxyl adducts were detected at the time of EPR measurement, both hydroxyl and superoxide radicals were found to contribute to the enhancement. The measured total enhancement was 0.7-fold per weight percent (wp) of Au in water using unfiltered X-rays. The theoretically predicted physical enhancement is 0.49 fold per wp of gold in water. This information, together with scavenging experimental results and the fact that the G-values are close for both radicals, suggest that hydroxyl and superoxide radicals contributing almost equally to the total measured enhancement. Further, the enhancement was found to be linearly dependent on the amount of large gold nanoparticles in water and no additional radical was produced beyond the amount predicted by type 1 physical enhancement, indicating that hydroxyl or superoxide radicals were not produced via catalytic pathways.



INTRODUCTION Nanomaterials made of heavy elements in aqueous solutions or biological samples can enhance the absorption of X-rays by these systems, and many applications have been developed based on the increased absorption and subsequently enhanced energy deposition in these systems. For example, DNA can be damaged by reacting with the increased amount of reactive oxygen species (ROS) produced in water under X-ray irradiation when gold nanoparticles (AuNPs) are added.1−4 Various cells have been shown to experience increased damage by the added nanomaterials under X-ray irradiation.5,6 In vivo studies have been carried out by multiple groups, and the methods and results are covered in several review articles.7−9 The basic science as well as applications of these enhancements led to the development of a new field called X-ray nanochemistry.10 Although many potential applications have been identified, future promise of the field depends on the understanding of detailed enhancement mechanisms so that the highest enhancement may be achieved with minimal X-ray dose. The originally designated enhancement is supposed to be caused by the increased absorption of X-rays by nanomaterials, which gives rise to increased emission of electrons from these nanomaterials that result in enhanced energy deposition in the medium such as water surrounding the nanomaterials. The enhanced energy deposition is called physical enhancement (PE) because the whole process is of physical origin,3,11,12 and © 2016 American Chemical Society

this description is broadly recognized and accepted in the literature.8,9 There are at least two types of PE, type 1 and 2, or T1PE and T2PE. T1PE is caused by energetic electrons and is an average effect: the enhancement occurs over the whole sample volume.13 In contrast, T2PE is enabled by low energy electrons emitted from nanomaterials and it only occurs near the surface of nanomaterials.14,15 For this reason, the volume fraction in which T2PE contributes is small for nanoparticles larger than 1 nm in diameter, so unless the probe reactions are localized only near the surface, T1PE dominates. However, it is not straightforward to measure T1PE because there are other processes besides PE that can interfere with PE measurement. Normally indirect methods such as dosimetric chemical reactions are used to determine the physical enhancement because it is nearly impossible to measure directly the enhanced absorption and emission of electrons from as-made nanomaterials.16,17 However, in many cases, the dosimetric reactions could not fiducially measure the increased absorption of X-rays by nanomaterials because these reactions are often influenced by other factors such as catalytic properties of nanomaterials and ROS produced in water under X-ray irradiation.10 Such catalytic processes are now called chemical enhancement (CE), which is caused by nanomaterials reacting Received: February 21, 2016 Revised: April 16, 2016 Published: April 28, 2016 2815

DOI: 10.1021/acs.jpca.6b01755 J. Phys. Chem. A 2016, 120, 2815−2823

Article

The Journal of Physical Chemistry A

Scheme 1. Pathways of BMPO Trapping •OH and •O2− and EPR Spectra of the Two Adducts BMPO−OH and BMPO-OOHa

a

Spectra of all four conformers are shown.

with ROS produced in water or other media under X-ray irradiation. This means that CE does not require more ROS be produced from the added nanomaterials and is therefore isolated from PE. Many multiple-step dosimetric reactions such as polymerization, hydroxylation, and DNA strand breaks may be catalyzed by nanomaterials and hence are susceptible to be influenced by CE. As demonstrated recently, one benefit of separating CE from PE is to be able to combine the two to achieve higher total enhancements than when they are separated.18 Nonetheless, it is essential to isolate or remove CE if PE is to be properly measured. The most straightforward way to detect the originally defined enhancement of T1PE without interference from CE is to use large nanoparticles with small surface areas. It has been shown that T1PE is 0.1−1.4-fold per weight percent (wp) of gold in water, depending on the X-ray energy.13 For 100 nm AuNPs, 1 wp is 2 nM and CE is negligible, so the enhancement is largely T1PE if the probes are uniformly distributed in solution. By contrast, for 3 nm AuNPs, 2 nM concentration corresponds to 37 ppm or 0.0037 wp, and T1PE is negligible; any measurable enhancement is most likely due to CE.16 When large amounts of nanomaterials are used, there is another effectantienhancement (AE)that may take place.16,19 This is because the surface of these nanomaterials may scavenge the ROS created by interactions between water and electrons emitted from nanomaterials. For example, hydroxyl radical (•OH) may be scavenged by AuNPs because gold has a lower reduction potential than •OH.16 CE and AE depend on nanomaterials and reactions, so it is possible to have these two processes coexist, or have one without the other because AE may be used to enable CE. An example of the latter is superoxide reacting with AuNPs to enable CE for hydroxylation reaction of 3-CCA.16

The third problem encountered when large amounts of nanomaterials such as AuNPs are used is the need to separate the probe reaction products from the nanomaterials after irradiation if optical detection methods are used. This is because AuNPs may quench or attenuate fluorescence, and the commonly used fluorescent dosimetry requires purified samples with AuNPs removed from the sample.20,21 If no separation or purification is preferred, then these optical methods cannot be used. One way to overcome all these problems and determine exclusive T1PE, i.e., without T2PE/CE/AE, is to use electron paramagnetic resonance spectroscopy (EPR) to probe longlived spin adducts formed from spin traps reacting with radicals produced in water. EPR often detects ROS in one-step reactions that trap ROS in much more stable spin adducts.22,23 This simpler reaction pathway minimizes the chance of the measured enhancement being influenced by other processes such as CE. In addition, the concentration of trapping agents such as 5-tert-butoxycarbonyl-5-methyl-1-pyrroline-N-oxide (BMPO) is high enough to help eliminate the AE that often plagues the implementation and determination of enhancement in high concentrations of AuNPs.24 Last, EPR does not require the separation of probes or spin adducts from nanomaterials. As a result, EPR is one such method that may conveniently, quantitatively, and fiducially determine the amount of radicals and hence PE. Additional benefits of using EPR to detect enhancement include the following: (1) EPR can detect ROS in aqueous solution as well as in solid phase. (2) EPR spectra can reveal whether adding nanomaterials affects ROS generation or detection, a process that may be used to aid in elucidation of enhancement mechanisms. (3) EPR in principle can directly detect both •OH and superoxide radical (•O2−) at a known ratio. Many dosimetric reactions such as 3-carboxycourmarin 2816

DOI: 10.1021/acs.jpca.6b01755 J. Phys. Chem. A 2016, 120, 2815−2823

Article

The Journal of Physical Chemistry A

is not present at the time of EPR measurement for the reasons given in this work.

acid (3-CCA) to 7-hydroxy-CCA (7-OH−CCA) conversion can only probe one kind of radical such as •OH, and the reaction is often subjected to CE because it is a multiple-step reaction.25 Other probes may be used to probe either •OH or •O2−.17 EPR has been used to study ionizing radiation interaction with nanomaterials. There are two general ways of using EPR to probe enhanced production of radicals. The first is to rely on radicals produced from chemicals under X-ray irradiation without the use of spin traps.26−28 For instance, 2-methylalanine produced radicals CH3−•CH−COOH in the presence of AuNPs was studied with EPR recently to measure the enhancement in solid state matrices.29 A nonlinear response of enhancement as a function of the amount of gold was observed.30 The second way to perform EPR measurements is using spin traps. Commonly used EPR spin traps are nitrone or nitroxyl derivatives such as DMPO (5,5-dimethyl-1-pyrroline-Noxide),31 TEMPOL (4-hydroxy-2,2,6,6-tetramethylpiperidin-1oxyl),32 and BMPO (5-tert-butoxycarbonyl-5-methyl-1-pyrroline-N-oxide).33 These spin traps have been reported to detect both •OH and •O2− if these two radicals are created separately and chemically. However, it has been found that certain spin traps such as BMPO or CD-DIPPMPO only detects one of the two radicals when both radicals are present.34 Further, one spin adduct such as superoxide adduct DMPO-OOH can also decompose and convert to other spin adducts such as hydroxyl adduct DMPO−OH.35 These properties and shorter adduct lifetimes make it difficult to use certain spin traps to probe enhancement by nanomaterials exposed to X-rays.36,37 It is the intention of this work to explore the EPR method for more direct determination of the increased absorption and energy deposition by as-made nanoparticles. To date no spin traps were used to detect enhancement created by nanomaterials under X-rays, although enhancement by bulk materials under X-ray irradiation has been studied with spin traps.38 The results obtained here can also shed light on whether ROS are catalytically produced beyond what is predicted by physical enhancement. The outcome can therefore help differentiate types and mechanisms of enhancements. Particularly, this paper presents the results of the first EPR study of enhancement by irradiation of as-made polyethylene glycol (PEG)-coated AuNPs using spin trap BMPO in water with X-rays. BMPO has several advantages such as simple EPR spectra, long adduct lifetimes, stable in many solvents, and being able to spin trap both •OH and •O2−.24 No study on the trapping of hydrogen atoms using BMPO was reported. The results presented here demonstrate that EPR can be used to measure enhancement of the X-ray effect by nanomaterials in water. The outcomes are compared with the theoretically simulated enhancements. Because detection of enhancement using EPR with spin traps is based on the detection of secondary species such as •OH or •O2−, it is important to identify the mechanisms of reactions between the spin traps and these radicals. Possible spin trapping mechanisms using BMPO are given in Scheme 1. There are two types of radicals BMPO can trap, and each radical spin adduct has two conformers. These conformers are used as the basis sets to deconvolute the experimentally obtained EPR spectra. BMPO •O2− spin adduct takes the form of BMPO−OOH in pH 7 water. As it is shown here, BMPO superoxide adduct BMPO-OOH may be converted to BMPO hydroxyl adduct BMPO−OH in the X-ray irradiated water, and BMPO−OOH



EXPERIMENTAL AND THEORETICAL METHODS

Chemicals. Gold tetrachloride in HCl aqueous solution (30 wt % or 30 wp), sodium citrate, 3-carboxycoumarin acid (3-CCA), dimethyl sulfoxide (DMSO), and superoxide dismutase (SOD) were purchased from Sigma-Aldrich. The spin trap reagent 5-tert-butoxycarbonyl-5methyl-1-pyrroline-N-oxide (BMPO) was purchased from Dojindo (Japan). The capping ligand thiolated polyethylene glycol (mPEG-SH 5k MW) was purchased from Nanocs (Boston, MA). All water used was Milli-Q (MQ) water. All chemicals were used without further purification. Synthesis of AuNPs. AuNPs were prepared using a citrate reduction method. Briefly, 50 μL HAuCl4 was dissolved in 150 mL of MQ water. The solution was heated to reflux and 1.2 mL of 1-wp sodium citrate solution was rapidly injected into the solution under vigorous stirring. The solution was allowed to reflux for 15 min and then allowed to cool to room temperature. In order to prevent particle aggregation, up to 50 mg PEG-SH was added into the solution. PEG functionalized AuNPs (PEG-AuNPs) were incubated overnight after which the solution was purified by centrifugation four times, each at 6500 rpm for 15 min. Unless otherwise specified, all the AuNPs used in this work were PEG-AuNPs. After the final spin, a pellet was obtained and the relative wp of AuNPs in solution was determined using atomic absorption spectroscopy (AA). Size of the as-synthesized AuNPs was characterized by dynamic light scattering (DLS) using Zetasizer Nano S-90 equipped with 633 nm He−Ne laser and JEOL1230 transmission electron microscope operated at 100 keV. Sample Preparation. The solution was diluted to a final stock concentration of 1.33 wp of Au in water that was further diluted for use in the following experiments. A solution of 100 mM BMPO in water was mixed with different concentrations of AuNPs to prepare various AuNP samples between 0.1 and 1 wp Au in water. Control samples were prepared similarly, but with the AuNP solution replaced by MQ water. For samples containing DMSO for the purpose of scavenging OH radicals, 167 mM BMPO and 15 μL of 0.4 wp AuNPs was mixed with varying concentrations of DSMO. Control samples were prepared similarly, but with the DMSO solution replaced by MQ water. For samples containing SOD for the purpose of scavenging superoxide, no AuNPs were used. X-ray Irradiation Experiments. Twenty microliters of each sample was exposed to X-ray radiation for 4 min to obtain a total dose of 150 Gy using a microfocus X-ray source (PXS10-WB-10 mm, Thermo Scientific) operated at 100 kVp and 250 μA. Each time three samples with three different AuNP concentrations and a control sample without nanomaterials were irradiated simultaneously. 3-CCA dosimetric reaction was used to calibrate the effect of DMSO, and the method was identical to that used in a previous work.16 EPR Measurements. Ten microliters aliquots of all the irradiated samples were immediately placed in a capillary tube, capped with paraffin films, and frozen in a liquid nitrogen bath. Frozen samples were taken to the EPR instrument for analysis. EPR measurements were performed using a Bruker ECS-106 spectrometer at X band of 9.8 GHz. Prior to running any samples, a standardized weak pitch (0.00033% in KCL) sample was used to tune the instrument. Samples were then removed from the liquid nitrogen bath and allowed to thaw at room temperature for approximately 4 min before being placed in the cavity for measurement. The condition of the instrument was set as follows: center field: 3513 mT; sweep width: 140 mT; time constant: 40.96 ms; field sweep scan time: 41.94 s; two scans per spectrum; modulation frequency: 100 kHz; modulation amplitude: 0.974 mT; receiver gain: 2 × 104; microwave power: 5.02 mW. Data Processing. EasySpin was used to process the EPR data and simulate fitted spectra.39 Spectra acquired from the Bruker ECS-106 were loaded directly into the MatLab program. The fitting process identified individual radicals and conformers, which were responsible for the experimentally measured BMPO spin adduct spectra. Fitted 2817

DOI: 10.1021/acs.jpca.6b01755 J. Phys. Chem. A 2016, 120, 2815−2823

Article

The Journal of Physical Chemistry A spectra were then used to calculate the enhancement by comparing with control samples that contained no AuNPs. There are six methods that can be used to determine the enhancement. The first method is measuring the total peak height of the first derivative of EPR absorption spectra (as in all six methods shown here) and is the sum of the amplitude of all four peaks (Scheme 2A). The second method is measuring the peak height of the positive

Scheme 2. Six Methods to Determine EPR Signal Strengtha

Figure 1. BMPO EPR measurement results. (A) Typical EPR trace of BMPO spin trapping data in water. (B) Fitted spectra using the known conformers. Only BMPO−OH conformers are found in the detected EPR signal.

conditions employed in this work, only BMPO−OH existed at the time of EPR measurement. Table 1 shows the parameters used to fit the data shown in Figure 1B. The hyperfine coupling fitting parameters are similar Table 1. Fitting Parameters of EPR Patternsa BMPO−OH conformer I

a

(A) The sum of 4 peak heights; (B) the positive portion of the second peak; (C) the peak area under the positive side of the second peak; (D) the whole peak area of the second peak; (E) the whole peak area of the third peak; (F) the whole peak area.

hyperfine coupling constant [Lit.] (mT)24 hyperfine coupling constant [Exp.] (mT) line width [Exp., this work] (mT) fractional radical contribution (in %)

portion of the second peak only (Scheme 2B). The third (Scheme 2C) is measuring the area of the positive portion of the second peak, which corresponds to the peak of EPR absorption signal. The fourth is measuring the total area under the second peak (Scheme 2D). The fifth is measuring the total area under the third peak (Scheme 2E). The last method is measuring the combined area under all four peaks (Scheme 2F). Among these six methods, only two were used for processing the data and obtaining the enhancement. Enhancement Simulation. A Monte Carlo simulation was used to predict the enhancement by AuNPs using an established software package developed recently.3,14 The predicted enhancement given by the program was the ratio of energy deposited in water with AuNPs to that without AuNPs. Therefore, a high water background signal and a low increase in the signal after AuNP addition create a low enhancement. Subsequent ROS diffusion and reactions between ROS species and BMPO spin traps were not included in the predicted enhancement. The simulation also took into account X-ray spectra used in the experiment and attenuation of X-rays through the sample medium.

BMPO−OH conformer II

[1.347; 1.531; 0.062]

[1.354; 1.23; 0.066]

[1.34; 1.531; 0.06]

[1.401; 1.278; 0.0655]

0.0275

0.0895

0.196

0.804

a

Literature values are shown, as well as the values obtained from this work.

to the literature values.24 The line widths for the two conformers are shown as well. Adding AuNPs may generate more ROS in water under Xray irradiation. The average size of AuNPs was 89 ± 23 nm based on TEM imaging and ranged from 41 to 60 nm (Zaverage) based on DLS detection. AuNPs of these sizes may produce chemical enhancement between 0.01 and 0.1 wp. A representative set of results are given in Figure 2, which includes the EPR spectra from BMPO adducts in pure water as well as in three solutions of different AuNP concentrations. The



RESULTS AND DISCUSSION Here it is shown that BMPO can trap X-ray radiation-generated radicals in water. Figure 1A shows the results of EPR detection of X-ray-irradiated BMPO aqueous solution. There are four EPR peaks, with the first and fourth having a shoulder. Figure 1B shows the fitted data using all four conformers, which are BMPO hydroxyl adduct (BMPO−OH) conformers I and II and BMPO superoxide adduct (BMPO−OOH) conformers I and II. The fitting of the experimental EPR results using these four conformers yielded a dominating conformer (80%) responsible for the experimental data: BMPO−OH conformer II. A second component of 20% is contributed by BMPO−OH conformer I. The ratio of these two conformers is similar to the literature value for BMPO−OH.24 Surprisingly, no BMPO− OOH was detected, which suggests that under the experimental

Figure 2. EPR results of enhancement for three gold nanoparticle concentrations. The EPR profiles (0.5 to 1.0 wp) remain the same as that of pure water (blank). The fitting parameters are given in Table 1. 2818

DOI: 10.1021/acs.jpca.6b01755 J. Phys. Chem. A 2016, 120, 2815−2823

Article

The Journal of Physical Chemistry A EPR traces are shifted horizontally by a small magnetic field strength for displaying purpose. The enhancements are apparent in Figure 2 as the signal amplitude increases as a function of the concentration of AuNPs. All four plots bear the same EPR pattern, suggesting that the conformers responsible for the EPR signals are identical in these measurements, with or without AuNPs. The enhancement is the ratio of the EPR signal strength of BMPO with AuNPs to that without AuNPs. Figure 3 shows the

Figure 3. Enhancement as a function of AuNP concentration from the processed signals obtained using the six methods shown in the Exerimental and Theoretical Methods section. Only three AuNP concentrations were used here.

Figure 4. Enhancement results. (A) Results of enhancement as a function of AuNP concentration using the area of the positive portion of the second peak. The lowest concentration of Au is 0.01 wp and the highest is 1.0 wp. (B) Average and standard deviation of the data shown in panel A. (C) Same data with the integrated areas of all four peaks. (D) Processed data (average and standard deviation) of panel C. The slopes in B and 4D are 0.63 and 0.70 wp−1, respectively.

results using the six methods described in the Experimental and Theoretical Methods section. The peak-to-peak processing method of all four peaks gives rise to the least consistent trend (solid circle).27 The peak-to-peak of the second peak (solid square), the integrated area under the positive portion of the second peak (solid diamond), the integrated area under the second or third peak (open circle and square), and the integrated area under all four peaks (open diamond) are shown in Figure 3 as well. The results of enhancement measurements using the integrated area of the second peak (either the positive portion or the whole peak) or all four peaks is the most consistent, and these two data processing methods were used to calculate the enhancement. The overall processed enhancement results using the integrated area of the positive half of the second peak area are shown in Figure 4A. A total of 16 sets of data, each with three AuNP concentrations, were obtained experimentally. The amount of gold in water ranged from 0.01 to 1.0 wp. Figure 4B shows the processed data (average and standard deviation) and fitted enhancement slope. We used the definition of enhancement per wp of gold in water (wp−1) to quantify enhancement. The enhancement is 0.63 wp−1 for Au in water in the form of AuNPs using unfiltered X-rays. This is close to the theoretically predicted T1PE, which is 1.0−1.4 wp−1 for high energy, filtered X-rays.13 Chemical enhancement, when existing, can be as high as a few fold between 0.01 and 0.1 wp. Here there is no indication of the existence of chemical enhancement because at low AuNP concentrations, the measured enhancements do not increase above the linear relationship predicted enhancement as a function of gold content. We also used the integrated total area of all four peaks to estimate the enhancement. Figure 4C shows the integrated area results as a function of concentration of AuNPs. Figure 4D shows similar processed data (average and standard deviation)

and the slope. The enhancement is 0.70 wp−1 of Au in water in the form of AuNPs, proving that the two methods of data processing yield similar results. Radical scavenging experiments were performed to determine the source of enhancement. The G-value (number of radicals per 100 eV of energy deposited) is 2.7 (steady state) for •OH, which is similar for •O2− for X-ray and γ-ray irradiation.40,41 Under X-ray irradiation of water, •O2− is created from solvated electrons reacting with molecular oxygen in water. However, pulsed radiolysis of water measurements shows that G-value for •OH is much higher than •O2− when measured at times earlier than microseconds.40 When BMPO is used, we assume that most •OH and •O2− are trapped because of the high concentrations of BMPO used here, so the relative amounts of these two radicals should be close to that defined by the steady-state G-values, i.e., equal amounts of •OH spin adduct BMPO−OH and superoxide spin adduct BMPO-OOH are expected. However, this is not what is shown in Figures 1 and 2 where only BMPO−OH is detected. Scavenging measurements were performed to investigate the cause. EPR results of the scavenging experiments are shown here, which include the outcome of using •OH scavenger dimethyl sulfoxide (DMSO) or •O2− scavenger superoxide dismutase (SOD) mixed with BMPO under X-ray irradiation. 0.3 wp AuNPs were used in the DMSO experiment, and the scavenging results were corrected for the enhancement caused by this amount of gold. Figure 5 shows the results, indicating that EPR signal began to decrease as DMSO was added (solid triangle). At the highest DMSO concentration, only 25% of the original EPR signal remained, suggesting a 75% BMPO−OH signal reduction. At 10 mM DMSO concentration, •OH was reduced by 50% according to our measurement using 3-CCA dosimetric reaction.16 These results imply that •OH is largely 2819

DOI: 10.1021/acs.jpca.6b01755 J. Phys. Chem. A 2016, 120, 2815−2823

Article

The Journal of Physical Chemistry A

these conformers or reaction pathways depends on the mechanisms of radical generation. We also estimated the magnitude of physical enhancement using a Monte Carlo method.14 As mentioned in the introduction, there are at least two types of physical enhancement: T1PE and T2PE. T1PE refers to an average enhancement occurring in the whole volume of a nanomaterial containing solution or sample.13 T2PE refers to nanoscale energy deposition that happens only near the surface of nanomaterials and is not an average enhancement over the whole volume.15 Since there is no attraction between AuNPs and BMPO or adducts, BMPO are uniformly distributed in the sample, and there is little T2PE. Therefore, only T1PE was theoretically simulated, which is the energy deposition enhancement that leads to the increased production of both •OH and •O2− based on G-values. Figure 6 shows the results of theoretical simulation of T1PE using an unfiltered X-ray spectrum from a 100 kVp tungsten X-ray source. Figure 6A shows the energy spectrum of the X-rays. Figure 6B shows the calculated energy deposition using the spectrum shown in Figure 6A. Figure 6C shows the total T1PE as a function of Au. The T1PE from AuNPs is 0.69 wp−1 for 3 mm thick samples. The low enhancements are mainly caused by the high absorption of 13−20 keV X-rays by water and the low enhancement by AuNPs in the same X-ray energy range. These results show that PE depends on X-ray energy spectrum, and 30−90 keV X-rays generate higher PE.18 It is important to note that the predicted enhancement is with respect to energy deposition, which is assumed to have a linear relationship with the enhancement of ROS. As a result, the •OH and •O2− production enhancement should both be at 0.69 wp−1 if there is no interference between the two radicals and the G-values for the two ROS are the same. This should be true for high concentrations of BMPO because all these radicals are readily trapped in adducts. If BMPO traps both radicals with equal efficiency and if all the BMPO−OOH adducts are converted to BMPO−OH adducts, then the total enhancement should be 0.69 wp−1, a value close to the experimentally measured total enhancement of 0.7 wp−1. The results obtained here show that BMPO can trap radicals generated from X-ray irradiating water in the presence of AuNPs and can be used to properly measure enhancement by AuNPs. The measured enhancement is attributed to •OH and •O2− by approximately the same magnitude. However, the spin adduct detected and used to obtain the enhancement is

Figure 5. Scavenging results. The solid triangles show the DMSO results, which indicate enhancement as a function of DMSO. Solid triangles in empty squares show the SOD measurement results.

and directly responsible for the EPR signal detected and BMPO−OH produced were largely destructed by DMSO. However, this physical picture is not entirely correct. SOD scavenging results are shown in Figure 5 as well (solid triangle in empty square), which surprisingly follow a similar trend as the DMSO concentration dependency. No AuNPs were used in SOD experiments to minimize interference or interaction between AuNPs and SOD. The SOD scavenging results suggest that some (∼50%) of the EPR signal was attributed to •O2− when a normal amount of SOD (∼5000 units/ml) were used. At this SOD amount, nearly all the •O2− is assumed to be scavenged.16,42 These results suggest that the two radicals both contributed to the total enhancement. A plausible explanation is that BMPO−OOH is readily converted to BMPO−OH when the aqueous solution is irradiated with X-rays, similar to those reported in the literature.43 Although further investigation is needed to completely explain these results, it is evident that both •O2− and •OH in X-ray irradiation of water contributes to the enhancement results shown in Figures 1, 2, and 4. The results shown here suggest that, although the detected EPR signal largely comes from BMPO−OH, both •OH and •O2− make significant contribution to the measured enhancements. Part of the detected BMPO−OH signal is attributed to •O2−, as BMPO−OOH may be converted to BMPO−OH during the process of X-ray irradiation. The exact weight of

Figure 6. Comparison of theory and experiment. (A) Spectrum of the unfiltered X-ray used for irradiation. (B) Calculated energy deposition profiles using the X-ray spectrum shown in panel A irradiating three different gold concentrations of 0.1, 0.5, and 1.0 wp in water. There is increased X-ray energy deposition with increasing gold concentration. (C) Absolute theoretical physical enhancements obtained with 0.1, 0.5, and 1.0 wp gold irradiated with the unfiltered X-ray spectrum. 2820

DOI: 10.1021/acs.jpca.6b01755 J. Phys. Chem. A 2016, 120, 2815−2823

Article

The Journal of Physical Chemistry A dominantly the •OH adduct, even though BMPO is shown to be able to trap either •OH or •O2− when they are produced separately using chemical methods.24 EPR measurements of enhancement using BMPO show linear dependency as a function of AuNP concentration, and the measured values match that of theoretically predicted enhancements when both •O2−and •OH are considered. In comparison, EPR measurements using alanine in the solid form show strong chemical enhancement at low AuNP concentrations.30 The linearity observed here suggests that EPR using BMPO can properly measure T1PE without interference from CE or AE between 0.01 and 1.0 wp. In this work, CE was not detected because the detected enhancement is found to be linearly dependent on the gold concentration. T2PE is negligible because BMPO is uniformly distributed in the sample. There is no indication that there is a measurable amount of extra •OH or •O2− than those predicted by physical enhancement or type 1 physical enhancement from large AuNPs at Au amounts up to 1 wp.17 There are several possible pathways through which superoxide spin adducts are either destroyed or converted to hydroxyl adducts when aqueous solution is irradiated by X-rays. For instance, we speculate that BMPO−OOH may react with solvated electrons through chemical reduction that converts BMPO−OOH into BMPO−OH when the spin traps are used in X-ray irradiated samples. A similar process has been observed with DMPO, which has a shorter adduct lifetime than BMPO.36,44 Other reagents such as light, transition metals, and cell components can convert BMPO−OOH to BMPO− OH.45 Another possible pathway is that different adducts may influence each other when they are together. For example, when only •O2− is produced chemically, BMPO−OOH is observed;24,46,47 but when both •OH and •O2− are present and measured with BMPO, only BMPO−OH is observed.34 If indeed 100% of BMPO−OOH is converted to BMPO−OH, then adding SOD would reduce the BMPO−OH signal by half, matching the results shown in Figures 4 and 5. Adding 10 mM DMSO should only reduce •OH by 50%, so it is possible DMSO also completely stops the conversion of BMPO−OOH to BMPO−OH, which then explains the additional 50% reduction shown in Figure 5. These explanations and the results shown in Figure 5 suggest that the contributions from •O2− and •OH to the final BMPO−OH adducts are approximately the same. If there no interconversion between •O2− adduct BMPO− OOH and •OH adduct BMPO−OH, then the lifetimes of the two radical adducts may play a role. In that case, BMPO−OOH may be more quickly destroyed due to the shorter lifetime of BMPO−OOH at room temperature in water. When measured separately, the lifetime of BMPO−OOH (16−22 min) is about half of BMPO−OH (30 min).22,48 Even though the samples were frozen after X-ray irradiation and before EPR measurement, if irradiation time is on the order of 5 min, the thaw time is 4 min and measurements takes 5 min then only one-third of the superoxide spin adducts are left and the •OH spin adducts decrease by only 30%. If both radicals are produced in equal amounts, then the BMPO−OOH signal is only half of that of BMPO−OH. However, because of the interconversion, the lifetime difference between the two adducts may not play a role. This work shows that it is possible to use EPR to measure enhancement, and the response is linear, therefore making it possible to use the method developed here to quantitatively determine the magnitude of enhancement. The conversion

from BMPO−OOH to BMPO−OH makes it interesting for future dynamic work to conclusively determine the mechanisms.



CONCLUSIONS EPR was used to characterize type 1 physical enhancement of the X-ray effect by gold nanoparticles using BMPO spin trap reagent, which can trap both •OH and •O2− produced in water as a result of X-ray irradiation of nanoparticles in water. It was found that •O2− and •OH each contributing ∼0.70 fold enhancement per weight percent of gold in water, giving rise to a total enhancement of 0.7 wp−1. The response of enhancement measured by BMPO spin adducts is linear with respect to AuNP concentration.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank Prof. R. David Britt and the CalEPR facility in the Department of Chemistry at the University of California Davis. We thank Dr. William Myers and Dr. Jeffrey Walton for their experimental assistance. This work was supported by the National Science Foundation (CHE-0955437 and CHE1307259).



REFERENCES

(1) Foley, E.; Carter, J.; Shan, F.; Guo, T. Enhanced Relaxation of Nanoparticle-Bound Supercoiled DNA in X-Ray Radiation. Chem. Commun. 2005, 3192−3194. (2) McMahon, S. J.; Hyland, W. B.; Brun, E.; Butterworth, K. T.; Coulter, J. A.; Douki, T.; Hirst, D. G.; Jain, S.; Kavanagh, A. P.; Krpetic, Z.; et al. Energy Dependence of Gold Nanoparticle Radiosensitization in Plasmid DNA. J. Phys. Chem. C 2011, 115, 20160−20167. (3) Carter, J. D.; Cheng, N. N.; Qu, Y. Q.; Suarez, G. D.; Guo, T. Nanoscale Energy Deposition by X-Ray Absorbing Nanostructures. J. Phys. Chem. B 2007, 111, 11622−11625. (4) Butterworth, K. T.; Wyer, J. A.; Brennan-Fournet, M.; Latimer, C. J.; Shah, M. B.; Currell, F. J.; Hirst, D. G. Variation of Strand Break Yield for Plasmid DNA Irradiated with High-Z Metal Nanoparticles. Radiat. Res. 2008, 170, 381−387. (5) Jain, S.; Coulter, J. A.; Hounsell, A. R.; Butterworth, K. T.; McMahon, S. J.; Hyland, W. B.; Muir, M. F.; Dickson, G. R.; Prise, K. M.; Currell, F. J.; et al. Cell-Specific Radiosensitization by Gold Nanoparticles at Megavoltage Radiation Energies. Int. J. Radiat. Oncol., Biol., Phys. 2011, 79, 531−539. (6) Butterworth, K. T.; Coulter, J. A.; Jain, S.; Forker, J.; McMahon, S. J.; Schettino, G.; Prise, K. M.; Currell, F. J.; Hirst, D. G. Evaluation of Cytotoxicity and Radiation Enhancement Using 1.9 Nm Gold Particles: Potential Application for Cancer Therapy. Nanotechnology 2010, 21, 295101. (7) Hainfeld, J. F.; Slatkin, D. N.; Smilowitz, H. M. The Use of Gold Nanoparticles to Enhance Eaidotherapy in Mice. Phys. Med. Biol. 2004, 49, N309−315. (8) Hainfeld, J. F.; Dilmanian, F. A.; Slatkin, D. N.; Smilowitz, H. M. Radiotherapy Enhancement with Gold Nanoparticles. J. Pharm. Pharmacol. 2008, 60, 977−985. (9) Retif, P.; Pinel, S.; Toussaint, M.; Frochot, C.; Chouikrat, R.; Bastogne, T.; Barberi-Heyob, M. Nanoparticles for Radiation Therapy Enhancement: The Key Parameters. Theranostics 2015, 5, 1030−1045.

2821

DOI: 10.1021/acs.jpca.6b01755 J. Phys. Chem. A 2016, 120, 2815−2823

Article

The Journal of Physical Chemistry A

Influence of Particle Size on the Detection Properties. Nanoscale 2012, 4, 2884−2893. (30) Guidelli, E. J.; Baffa, O. Influence of Photon Beam Energy on the Dose Enhancement Factor Caused by Gold and Silver Nanoparticles: An Experimental Approach. Med. Phys. 2014, 41, 032101. (31) Ionita, P.; Gilbert, B. C.; Chechik, V. Radical Mechanism of a Place-Exchange Reaction of an Nanoparticles. Angew. Chem., Int. Ed. 2005, 44, 3720−3722. (32) Zhang, Z. Y.; Berg, A.; Levanon, H.; Fessenden, R. W.; Meisel, D. On the Interactions of Free Radicals with Gold Nanoparticles. J. Am. Chem. Soc. 2003, 125, 7959−7963. (33) Zielonka, J.; Sarna, T.; Roberts, J. E.; Wishart, J. F.; Kalyanaraman, B. Pulse Radiolysis and Steady-State Analyses of the Reaction between Hydroethidine and Superoxide and Other Oxidants. Arch. Biochem. Biophys. 2006, 456, 39−47. (34) Abbas, K.; Hardy, M.; Poulhes, F.; Karoui, H.; Tordo, P.; Ouari, O.; Peyrot, F. Detection of Superoxide Production in Stimulated and Unstimulated Living Cells Using New Cyclic Nitrone Spin Traps. Free Radical Biol. Med. 2014, 71, 281−290. (35) Finkelstein, E.; Rosen, G. M.; Rauckman, E. J. Production of Hydroxyl Radical by Decomposition of Superoxide Spin-Trapped Adducts. Mol. Pharmacol. 1982, 21, 262−265. (36) Bacic, G.; Spasojevic, I.; Secerov, B.; Mojovic, M. Spin-Trapping of Oxygen Free Radicals in Chemical and Biological Systems: New Traps, Radicals and Possibilities. Spectrochim. Acta, Part A 2008, 69, 1354−1366. (37) Ueno, M.; Nakanishi, I.; Matsumoto, K. Method for Assessing X-Ray-Induced Hydroxyl Radical-Scavenging Activity of Biological Compounds/Materials. J. Clin. Biochem. Nutr. 2013, 52, 95−100. (38) Paudel, N.; Shvydka, D.; Parsai, E. I. Comparative Study of Experimental Enhancement in Free Radical Generation against Monte Carlo Modeled Enhancement in Radiation Dose Position Due to the Presence of High Z Materials During Irradiation of Aqueous Media. International Journal of Medical Physics, Clinical Engineering and Radiation Oncology 2015, 4, 300−307 300.. (39) Stoll, S.; Schweiger, A. Easyspin, a Comprehensive Software Package for Spectral Simulation and Analysis in Epr. J. Magn. Reson. 2006, 178, 42−55. (40) Farhataziz; Rodgers, M. A. J. Radiation Chemistry: Principles and Applications; VCH Publishers, Inc.: New York, 1987. (41) Pastina, B.; LaVerne, J. A. Effect of Molecular Hydrogen on Hydrogen Peroxide in Water Radiolysis. J. Phys. Chem. A 2001, 105, 9316−9322. (42) Mayer, B.; Klatt, P.; Werner, E. R.; Schmidt, K. Kinetics and Mechanism of Tetrahydrobiopterin-Induced Oxidation of NitricOxide. J. Biol. Chem. 1995, 270, 655−659. (43) Tsai, P.; Pou, S.; Straus, R.; Rosen, G. M. Evaluation of Various Spin Traps for the in Vivo in Situ Detection of Hydroxyl Radical. J. Chem. Soc., Perkin Trans. 2 1999, 1759−1763. (44) Takeshita, K.; Fujii, K.; Anzai, K.; Ozawa, T. In Vivo Monitoring of Hydroxyl Radical Generation Caused by X-Ray Irradiation of Rats Using the Spin Trapping/Epr Technique. Free Radical Biol. Med. 2004, 36, 1134−1143. (45) Shi, H. L.; Timmins, G.; Monske, M.; Burdick, A.; Kalyanaraman, B.; Liu, Y.; Clement, J. L.; Burchiel, S.; Liu, K. J. Evaluation of Spin Trapping Agents and Trapping Conditions for Detection of Cell-Generated Reactive Oxygen Species. Arch. Biochem. Biophys. 2005, 437, 59−68. (46) Wen, T.; He, W. W.; Chong, Y.; Liu, Y.; Yin, J. J.; Wu, X. C. Exploring Environment-Dependent Effects of Pd Nanostructures on Reactive Oxygen Species(ROS) Using Electron Spin Resonance(ESR) Technique: Implications for Biomedical Applications. Phys. Chem. Chem. Phys. 2015, 17, 24937−24943. (47) He, W. W.; Liu, Y. T.; Wamer, W. G.; Yin, J. J. Electron Spin Resonance Spectroscopy for the Study of Nanomaterial-Mediated Generation of Reactive Oxygen Species. J. Food Drug Anal. 2014, 22, 49−63. (48) Spasojevic, I.; Mojovic, M.; Stevic, Z.; Spasic, S. D.; Jones, D. R.; Morina, A.; Spasic, M. B. Bioavailability and Catalytic Properties of

(10) Davidson, R. A.; Guo, T. An Example of X-Ray Nanochemistry: Sers Investigation of Polymerization Enhanced by Nanostructures under X-Ray Irradiation. J. Phys. Chem. Lett. 2012, 3, 3271−3275. (11) Cho, S. H. Estimation of Tumor Dose Enhancement Due to Gold Nanoparticles During Typical Radiation Treatments: A Preliminary Monte Carlo Study. Med. Phys. 2005, 32, 2162−2162. (12) Pradhan, A. K.; Nahar, S. N.; Montenegro, M.; Yu, Y.; Zhang, H. L.; Sur, C.; Mrozik, M.; Pitzer, R. M. Resonant X-Ray Enhancement of the Auger Effect in High-Z Atoms, Molecules, and Nanoparticles: Potential Biomedical Applications. J. Phys. Chem. A 2009, 113, 12356− 12363. (13) Davidson, R. A.; Guo, T. Average Physical Enhancement by Nanomaterials under X-Ray Irradiation. J. Phys. Chem. C 2014, 118, 30221−30228. (14) Lee, C.; Cheng, N. N.; Davidson, R. A.; Guo, T. Geometry Enhancement of Nanoscale Energy Deposition by X-Rays. J. Phys. Chem. C 2012, 116, 11292−11297. (15) Sharmah, A.; Yao, Z.; Lu, L.; Guo, T. X-Ray-Induced Energy Transfer between Nanomaterials under X-Ray Irradiation. J. Phys. Chem. C 2016, 120, 3054−3060. (16) Cheng, N. N.; Starkewolf, Z.; Davidson, A. R.; Sharmah, A.; Lee, C.; Lien, J.; Guo, T. Chemical Enhancement by Nanomaterials under X-Ray Irradiation. J. Am. Chem. Soc. 2012, 134, 1950−1953 1950.. (17) Misawa, M.; Takahashi, J. Generation of Reactive Oxygen Species Induced by Gold Nanoparticles under X-Ray and Uv Irradiations. Nanomedicine 2011, 7, 604−614. (18) Davidson, R. A.; Guo, T. Multiplication Algorithm for Combined Physical and Chemical Enhancement of X-Ray Effect by Nanomaterials. J. Phys. Chem. C 2015, 119, 19513−19519. (19) Ionita, P.; Spafiu, F.; Ghica, C. Dual Behavior of Gold Nanoparticles, as Generators and Scavengers for Free Radicals. J. Mater. Sci. 2008, 43, 6571−6574. (20) Reineck, P.; Gomez, D.; Ng, S. H.; Karg, M.; Bell, T.; Mulvaney, P.; Bach, U. Distance and Wavelength Dependent Quenching of Molecular Fluorescence by [email protected] Core-Shell Nanoparticles. ACS Nano 2013, 7, 6636−6648. (21) Jennings, T. L.; Singh, M. P.; Strouse, G. F. Fluorescent Lifetime Quenching near D = 1.5 Nm Gold Nanoparticles: Probing Nset Validity. J. Am. Chem. Soc. 2006, 128, 5462−5467. (22) Mitchell, D. G.; Rosen, G. M.; Tseitlin, M.; Symmes, B.; Eaton, S. S.; Eaton, G. R. Use of Rapid-Scan Epr to Improve Detection Sensitivity for Spin-Trapped Radicals. Biophys. J. 2013, 105, 338−342. (23) Tsai, P.; Marra, J. M.; Pou, S.; Bowman, M. K.; Rosen, G. M. Is There Stereoselectivity in Spin Trapping Superoxide by 5-TertButoxycarbonyl-5-Methyl-1-Pyrroline N-Oxide? J. Org. Chem. 2005, 70, 7093−7097. (24) Zhao, H. T.; Joseph, J.; Zhang, H.; Karoui, H.; Kalyanaraman, B. Synthesis and Biochemical Applications of a Solid Cyclic Nitrone Spin Trap: A Relatively Superior Trap for Detecting Superoxide Anions and Glutathiyl Radicals. Free Radical Biol. Med. 2001, 31, 599−606. (25) Louit, G.; Foley, S.; Cabillic, J.; Coffigny, H.; Taran, F.; Valleix, A.; Renault, J. P.; Pin, S. The Reaction of Coumarin with the Oh Radical Revisited: Hydroxylation Product Analysis Determined by Fluorescence and Chromatography. Radiat. Phys. Chem. 2005, 72, 119−124. (26) Lund, E.; Gustafsson, H.; Danilczuk, M.; Sastry, M. D.; Lund, A.; Vestad, T. A.; Malinen, E.; Hole, E. O.; Sagstuen, E. Formates and Dithionates: Sensitive Epr-Dosimeter Materials for Radiation Therapy. Appl. Radiat. Isot. 2005, 62, 317−324. (27) Guidelli, E. J.; Ramos, A. P.; Zaniquelli, M. E. D.; Nicolucci, P.; Baffa, O. Synthesis and Characterization of Gold/Alanine Nanocomposites with Potential Properties for Medical Application as Radiation Sensors. ACS Appl. Mater. Interfaces 2012, 4, 5844−5851. (28) Waldeland, E.; Hole, E. O.; Sagstuen, E.; Malinen, E. The Energy Dependence of Lithium Formate and Alanine Epr Dosimeters for Medium Energy X Rays. Med. Phys. 2010, 37, 3569−3575. (29) Guidelli, E. J.; Ramos, A. P.; Zaniquelli, M. E. D.; Nicolucci, P.; Baffa, O. Synthesis and Characterization of Silver/Alanine Nanocomposites for Radiation Detection in Medical Applications: The 2822

DOI: 10.1021/acs.jpca.6b01755 J. Phys. Chem. A 2016, 120, 2815−2823

Article

The Journal of Physical Chemistry A Copper and Iron for Fenton Chemistry in Human Cerebrospinal Fluid. Redox Rep. 2010, 15, 29−35.

2823

DOI: 10.1021/acs.jpca.6b01755 J. Phys. Chem. A 2016, 120, 2815−2823