Mouse Positron Emission Tomography Study of the Biodistribution of

Mouse Positron Emission Tomography Study of the Biodistribution of Gold Nanoparticles with Different Surface Coatings Using Embedded Copper-64 Anders F. Frellsen,†,‡ Anders E. Hansen,‡,§ Rasmus I. Jølck,‡ Paul J. Kempen,‡ Gregory W. Severin,† Palle H. Rasmussen,† Andreas Kjær,§ Andreas T. I. Jensen,*,† and Thomas L. Andresen*,‡ †

The Hevesy Laboratory−Center for Nuclear Technologies, Technical University of Denmark, 4000 Roskilde, Denmark DTU Nanotech, Center for Nanomedicine and Theranostics, Technical University of Denmark, 2800 Lyngby, Denmark § Department of Clinical Physiology, Nuclear Medicine & PET and Cluster for Molecular Imaging, Rigshospitalet and University of Copenhagen, 2100 Copenhagen, Denmark ‡

S Supporting Information *

ABSTRACT: By taking advantage of the ability of 64Cu to bind nonspecifically to gold surfaces, we have developed a methodology to embed this radionuclide inside gold nanoparticles (AuNPs). 64Cu enables the in vivo imaging of AuNPs by positron emission tomography (PET). AuNPs have a multitude of uses within health technology and are useful tools for general nanoparticle research. 64Cu-AuNPs were prepared by incubating AuNP seeds with 64Cu2+, followed by the entrapment of the radionuclide by grafting on a second layer of gold. This resulted in radiolabeling efficiencies of 53 ± 6%. The radiolabel showed excellent stability when incubated with EDTA for 2 days (95% radioactivity retention) and showed no loss of 64Cu when incubated with 50% mouse serum for 2 days. The methodology was chelator-free, removing traditional concerns over chelator instability and altered AuNP properties due to surface modification. Radiolabeled 64Cu-AuNP cores were prepared in biomedically relevant sizes of 20−30 nm and used to investigate the in vivo stability of three different AuNP coatings by PET imaging in a murine xenograft tumor model. We found the longest plasma half-life (T1/2 about 9 h) and tumor accumulation (3.9%ID/ g) to result from a polyethylene glycol coating, while faster elimination from the bloodstream was observed with both a Tween 20-stabilized coating and a zwitterionic coating based on a mixture of sulfonic acids and quaternary amines. In the in vivo model, the 64Cu was observed to closely follow the AuNPs for each coating, again attributed to the excellent stability of the radiolabel. KEYWORDS: gold nanoparticles, copper-64, PET, intrinsic radiolabeling, coating materials dothelial system.12 Within biomedicine, AuNPs are furthermore useful in photodynamic therapy, where drugs coupled to the AuNPs are released due to laser-induced thermal heating or excitation.13,14 In addition, AuNPs represent a valuable imaging tool for in vivo studies of nanoparticle pharmacokinetics and biodistribution to obtain general information on nanoparticle behavior in living systems. Accurate knowledge of the in vivo location of AuNPs can be obtained by positron emission tomography (PET), which offers high sensitivity and real-time, noninvasive quantification of

N

anoparticles (NPs) have the ability to accumulate passively in a number of tumor types by the enhanced permeability and retention (EPR) effect.1,2 Gold nanoparticles (AuNPs) have a number of interesting properties for developing health technologies, as they have potential as imaging agents,3 are generally biologically inert and show low toxicity,4,5 have the ability to enhance the effect of radiotherapy6,7 and photothermal therapy,8 and show high in vivo stability.9,10 In addition, AuNP surfaces are easily modified, in particular by use of gold−sulfur bonds (Au−S), which are readily formed when AuNPs are mixed with compounds bearing thiol (SH) or disulfide (S−S) functionalities.11 Longcirculating AuNPs can be prepared by coating the surface with hydrophilic polymers, such as polyethylene glycol (PEG), which hinders aggregation and recognition by the reticuloen© 2016 American Chemical Society

Received: May 11, 2016 Accepted: October 18, 2016 Published: October 18, 2016 9887

DOI: 10.1021/acsnano.6b03144 ACS Nano 2016, 10, 9887−9898

Article

www.acsnano.org

Article

ACS Nano whole-body biodistribution.15 PET is based on the detection of photons that arise from the annihilation of an electron (e−) with a positron (e+) emitted by a PET radionuclide. The PET radionuclide 64Cu is especially useful when imaging NPs because of its straightforward chemistry and its considerable half-life of 12.7 h, which allows imaging in animals for up to 72 h.16 In addition, the low positron energy (Eav = 278 keV, Emax = 653 keV) of 64Cu offers good spatial resolution, almost identical to that of the most commonly used PET isotope 18F (Eav = 250 keV, T1/2 = 110 min).17 Further, 64Cu has possible uses beyond imaging, as it is considered a theranostic isotope due to its βemission,18 and as part of a matched theranostic pair with the therapeutic isotope 67Cu.19 Traditionally, 64Cu and other copper radioisotopes have been conjugated to NPs through bifunctional chelators, which complex the radionuclide and are tethered to the NP surface. 1,4,7,10-Tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA), 20 triethylenetetramine (TETA), 21 and 1,4,7triazacyclononanetriacetic acid (NOTA)22 have been widely employed as chelators for 64Cu, since they allow relatively fast and efficient radiolabeling at room temperature. However, the in vivo stability of such chelators has been questioned,23−25 and since free 64Cu accumulates significantly in tumors and liver, loss of 64Cu may seriously impact the interpretation of PET images.26 As molecular imaging of NPs in drug delivery applications primarily focuses on accurate and quantitative predictions of active substances in organs of interest, such inaccuracies are highly undesirable. Despite newer chelators, such as 2,2′-(1,4,8,11-tetraazabicyclo[6.6.2]hexadecane-4,11diyl)diacetic acid (CB-TE2A), appearing to offer somewhat increased in vivo stability, these usually require harsh labeling conditions that are incompatible with many NPs.27,28 Furthermore, as the conjugation between the AuNP surface and chelators is not always covalent or resistant to degradation, there is a risk of disassociation of the entire chelator−metal complex from the AuNPs. In addition, the very presence of macrocyclic chelators on NP surfaces has been implicated in altering their biodistribution.29,30 These factors have fueled a search for reliable, chelator-free alternative radiolabeling strategies, and the so-called intrinsic methods are quickly gaining momentum.31,32 Within the last couple of years, a few such methodologies for 64Cu-labeling of NPs have been reported: (1) incorporation of 64Cu into iron oxide NPs using a microwave technique;33 (2) formation by co-reduction of a nanoscale 64Cu-containing Cu−Au alloy in a 1:2.9 ratio;34 (3) reduction of 64Cu2+ onto the surface of AuNPs using hydrazine;35 (4) formation of AuNPs in the presence of 64 CuCl2 by reduction with sodium borohydride.36 In this article, we present methodology for chelator-free preparation of AuNPs embedded with 64Cu and demonstrate its excellent in vitro and in vivo stability. The AuNPs are prepared exclusively in aqueous media in biomedically relevant sizes. The method involves embedding of 64Cu in AuNPs, where the incorporated 64Cu is covered by a controllable layer of gold, thereby embedding it deep within the AuNPs. These characteristics eliminate the risk of disassociation of the radionuclide from the NP. As AuNPs are potentially useful as computerized tomography (CT) contrast agents, the radiolabeling methodology presented here also ensures that the CT contrast provided by the AuNPs is not diluted by the addition of significant amounts of lighter atoms, such as Cu. Finally, we have used the developed radiolabeling methodology to investigate the in vivo properties and biodistribution of 64Cu-

AuNPs with three different types of surface coating by PET imaging.

RESULTS AND DISCUSSION The radiolabeling methodology described in the present article was based on an observation that 64Cu2+ adhered nonspecifically to citrate-stabilized AuNPs when mixed. Initially, we imagined that this was due to electrostatic interactions and complexation between citrate molecules and 64Cu2+ on the gold surface. However, we observed that when a piece of solid gold was added to an aqueous solution of 64CuCl2 for 10 min, about 13% of the total 64Cu adhered to the gold and could not be washed off with additional water (84% of the associated 64Cu remained on the gold after 2 h in water). On this basis we speculated that the 64Cu2+ adhered or reacted with the gold surface itself. However, this nonspecific interaction was not stable, as the addition of ethylenediaminetetraacetic acid (EDTA) completely removed the radioactivity from the AuNP surfaces within 10 min. On this basis, we concluded that surface-bound 64Cu was only loosely bound to the AuNP surface. Therefore, we investigated possible methods by which to physically entrap the 64Cu2+ and prevent sequestering by exogenous Cu-binding agents, with the objective of preparing chelator-free 64Cu-AuNPs with no nonradioactive Cu added and with high in vivo radiolabel stability. Chelator-Free 64Cu Labeling of AuNPs. The Turkevich− Frens method was chosen for AuNP preparation since it allowed quick and simple synthesis of AuNPs in aqueous solution (Scheme 1).37,38 Using this methodology, firstScheme 1. Preparation and coating of AuNPs and 64CuAuNPs. Depicted is nonradioactive AuNP synthesis (1, 2, 3). The top horizontal lane depicts direct embedding (5, 6), the middle lane depicts seeded embedding (7, 8), and the bottom lane depicts passive surface labeling (4). (a) Sodium citrate, 65 °C to reflux. (b) Sodium citrate, dropwise addition of HAuCl4 at 65 °C, then reflux. (c) 1-Dodecanethiol and Tween 20. Radiolabeling of 4, 5, and 7 was achieved by adding [64Cu]CuCl2 directly to the reaction mixture.

9888


Article

ACS Nano

Figure 1. Characterization of AuNPs and 64Cu-AuNPs. Passive surface labeling is shown in red, direct embedding is shown in blue, and seeded embedding is shown in green. (A) Average radiolabeling yields of the three strategies given as the amount of 64Cu remaining associated with the AuNP fraction after incubation with EDTA for 30 min: passive surface labeling (4) (n = 3), direct embedding (6) (n = 6), and seeded embedding (8) (n = 5). (B) Mean sizes of AuNPs as determined by DLS. Uncoated 64Cu-AuNPs (5, 7) did not show significant change in size as compared to nonradiolabeled AuNPs prepared by an identical method (1, 2) (n = 5). Coated 64Cu-AuNPs (6, 8) showed an increase in size. (C) PDIs of AuNPs as determined by DLS. Embedding of 64Cu did not show a significant increase in PDI (n = 5). (D, E) UV−vis absorption spectra of AuNPs. The characteristic absorption peak for AuNPs is observed at 521 nm for 1, 523 nm for 5, and 532 nm for 6. This red-shift corresponds to an increase in the size of the AuNPs, correlating with DLS data. A similar pattern was observed for 2 (527 nm), 7 (526 nm), and 8 (531 nm). (F, G) Representative TEM images of first-generation AuNPs (1), with sizes of 14 ± 1.3 nm (G), and second-generation AuNPs (2) with sizes of 29 ± 2.4 nm (F). Size distributions are based on counts of n = 335 (1) and n = 443 (2) AuNPs. See Supporting Information (Figure S1) for size distribution histograms.

Following this rationale, direct embedding was achieved by adding 64Cu2+ to the HAuCl4 solution at the beginning of the AuNP synthesis to yield 5 (Scheme 1a). In addition, seeded embedding was achieved by adding 64Cu2+ to preformed AuNP seeds (1), onto which an additional layer of gold was grafted to yield 7. In seeded embedding the 64Cu-associated AuNP seeds would act as nucleation sites for further particle growth, controlling the particle size, while simultaneously facilitating the embedding of the 64Cu (Scheme 1b). In order to prevent aggregation of citrate-stabilized AuNPs during purification by filtration, 5 and 7 were coated with a mixture of 1dodecanethiol and Tween 20 to yield 6 and 8, respectively. The radiolabeling efficiencies of 4, 6, and 8 were measured after removing unbound 64Cu by incubation with EDTA for 30 min (Figure 1a). It was found that the optimal procedure for 5 and 7 was to incubate with EDTA to remove loosely bound 64 Cu, then coat the AuNPs to ensure colloidal stability to yield 6 and 8, whereafter 64Cu-EDTA was removed by filtration. Control experiments showed that 64Cu-EDTA did not associate with AuNPs coated with a 1-dodecanethiol/Tween 20 mixture, making EDTA scavenging a suitable way to remove unincorporated 64Cu. A radiolabeling efficiency of 26 ± 11% was obtained for the preparation of 4, and of the two embedding methods, 8 (53 ± 6%) showed a significantly higher radiolabeling efficiency than 6 (23 ± 7%). As the intention of 4 was to trap 64Cu under a layer of coating material, the EDTA was added after coating in this particular case. In addition, 64Cu had been observed to associate with Tween micelles, making it necessary to centrifuge 4 and remove the micelle-containing supernatant, before a radiolabeling efficiency could be determined. The 64Cu associated with 6 and 8 was assumed to be embedded within the AuNPs, since EDTA did not

generation AuNP seeds (1) and second-generation AuNPs grown from seeds (2) were prepared. Aqueous methods were preferred due to compatibility with 64Cu2+ and simplification of the final formulation for in vivo use. Initially, AuNPs (1, 2) and 64 Cu-AuNPs (5, 7) were prepared in refluxing water. The drawback of this approach was a polydisperse mixture of particle sizes.39 Other aqueous methods, such as reduction of Au3+ by the weak reductant sodium acrylate, provide very narrow size distributions, but take several days to carry out, which is incompatible with the 12.7 h half-life of 64Cu. However, we found that when forming the AuNPs at temperatures below boiling, as described by Nguyen et al.,40 the polydispersity index (PDI) was greatly improved. AuNP synthesis at 75 °C in water lowered PDIs from around 0.25 to around 0.10 for first-generation synthesis and from around 0.35 to about 0.15 for second-generation, without a significant influence on the radiolabeling efficiency or mean intensity weighted AuNP diameter. Three main radiolabeling strategies were investigated (Scheme 1): (1) Passive surface labeling (4); (2) direct embedding (5, 6); (3) seeded embedding (7, 8). Passive surface labeling by adhesion was conceived as a way in which 64Cu bound nonspecifically to AuNPs could be trapped by coating the AuNPs. This would potentially shield weakly bound 64Cu from transchelation. AuNPs (2) were synthesized and incubated with 64Cu2+ and subsequently coated with a 1dodecanethiol/Tween 20 mixture to yield 4. In addition to passive surface labeling, we investigated the embedding of 64Cu inside AuNPs. Wanting to utilize the affinity of 64Cu toward gold, we formed AuNPs in the presence of 64Cu2+. We imagined that the 64Cu2+ would adhere to the AuNP during formation and then be trapped inside the growing AuNP. 9889


Article

ACS Nano remove it from the citrate-stabilized 5 and 7 before coating. No difference in size (Figure 1b) and no increase in PDI (Figure 1c) were observed between 5 and 7 and their nonradioactive counterparts 1 and 2, respectively. This indicated that embedding of 64Cu2+ did not influence the size and stability of the AuNPs. Additionally, UV−vis analysis showed similar spectra for 1 and 5 as well as for 2 and 7, with the coated AuNPs 6 and 8 showing slightly higher absorption maxima (Figure 1d and e). All UV−vis spectra correlated well with sizes measured by DLS.41 The size and morphology of AuNPs prepared by the methods described (1, 2) were investigated by TEM (Figure 1f,g). TEM images are not capable of revealing surface functionalization of the AuNPs, only the size and shape of the solid gold core. These images confirmed the DLS measurements, as the diameters measured by TEM were similar (1: 14 ± 1.3 nm, 2: 29 ± 2.4 nm) to the DLS results. In addition, TEM showed that the methodology that was used resulted in AuNPs of quasi-spherical shape and narrow size distribution (for TEM-based size distribution histograms see Figure S1, Supporting Information). Stability of the 64Cu Radiolabels against EDTA and Serum. The stability of the radiolabels of 4 and 6 was investigated by incubation with EDTA for 2 days. The stability of 7 was investigated by coating it with all three coatings (8, 9, and 10, Chart 1) that were later investigated in vivo. 64Cu-

Figure 2. Stability of the radiolabels over 2 days against EDTA (10 mM, pH 7.5). The 64Cu associated with 64Cu-AuNPs prepared by direct embedding (6) and seeded embedding (8, 9, and 10) showed excellent stability against EDTA with 90 ± 3% (6), 95 ± 0% (8), 94 ± 4% (9), and 98 ± 2% (10) remaining associated with the 64CuAuNPs after 2 days. 64Cu-AuNPs prepared by passive surface labeling appeared less stable and showed substantial leak of 64Cu. Numbers are reported as mean ± SD (n = 3).

embedding, provided a remarkably stable radiolabel that was not influenced by the coating. Since seeded embedding (8) had the highest radiolabeling efficiency (53 ± 6%), this method was concluded to be superior. We suggest that this may be due to the 64Cu2+ having a longer time to associate with the AuNPs compared to the direct embedding method (6), which resulted in moderate labeling efficiency (23 ± 7%). All further studies were carried out using 64Cu-AuNP prepared by seeded embedding. As an additional measure of radiolabel stability, 8 was incubated with 50% mouse serum in phosphate-buffered saline (PBS) for 2 days at 37 °C. Mouse serum alone and 8 incubated with PBS were used as controls. The mixtures were separated by size-exclusion chromatography, and fractions were analyzed by liquid scintillation, while protein content and AuNPs were measured by UV−vis at 280 and 533 nm, respectively (Figure 3). Identical radioactivity profiles were observed for 64CuAuNPs with or without serum, confirming that no activity had

Chart 1. Molecular Structures of Gold Coatings Investigated in Vivo by PET/CT: 8, 1-Dodecanethiol/Tween 20; 9, MeOPEG5000SH; 10, Sulfonate/Quaternary Ammonium

AuNPs and 64Cu-EDTA were quantified by separation of the AuNPs from the EDTA fraction using filter centrifugation (molecular weight cut off (MWCO) 50 kDa, 7 nm) and by radioactivity−TLC analysis. Both methods gave comparable results. For 4 this revealed a constant leakage of 64Cu from the AuNP, leading to increasing amounts of 64Cu-EDTA (Figure 2). In contrast, long-term incubation of 6, 8, 9, and 10 with EDTA showed excellent stability of the 64Cu associated with these AuNPs. For 6, 95 ± 3% (n = 3) of the embedded radioactivity was retained in the 64Cu-AuNP fraction after 22.5 h and 90 ± 3% (n = 3) was retained after 44 h. For the coated versions of 7, 95 ± 0% (n = 3) was retained by 8 (41.5 h), 94 ± 4% (n = 3) was retained by 9 (44.0 h), and 98 ± 2% (n = 3) was retained by 10 (43.8 h) (Figure 2). This demonstrated that 64 Cu embedded inside AuNPs, by both direct and seeded

Figure 3. Serum stability study of 64Cu-AuNPs (8) incubated with PBS/mouse serum (1:1) for 2 days at 37 °C and then separated by size-exclusion chromatography. The left y-axis shows percent of total applied radioactivity in fractions of the 64Cu-AuNPs and 50% serum mixture (purple) and the 64Cu-AuNPs alone (red), while the right y-axis shows the absorbance at 280 nm for serum alone. The eluted radioactivity for the mixture and the 64Cu-AuNPs alone were virtually identical, indicating that no transfer of 64Cu to serum proteins had taken place. 9890


Article

ACS Nano Table 1. Characterization of 64Cu-AuNPs Used for in Vivo Investigationa 8 9 10 a

diameter (nm)

PDI

abs max (nm)

zeta pot (mV)

radioactivity (MBq/mL)

Au (mg/mL)

25 ± 5 73 ± 6 40 ± 9

0.18 ± 0.01 0.29 ± 0.01 0.20 ± 0.01

533 532 531

−7.0 ± 6.9 −22 ± 2.9 −31 ± 1.7

11.4 6.21 9.66

0.38 0.27 0.37

Size, absorption spectra, and zeta potentials were measured in PBS.

Figure 4. TEM characterization of coated 64Cu-AuNPs used in vivo (8, 9, 10). (A−C) Representative low-magnification TEM micrographs with high-resolution inset: A, 8; B, 9; C, 10. Average diameters of the 64Cu-AuNP cores were found to be 26.6 ± 3.3 nm (n = 327) (8), 28.3 ± 3.5 nm (n = 440) (9), and 26.9 ± 2.5 nm (n = 267) (10). (D−F) Representative negative-stained TEM images of coated 64Cu-AuNPs showing a corona of organic material surrounding the Au core: D, 8; E, 9; F, 10. No corona was seen for 8, whereas coronas were observed for both 9 and 10, measuring 5.9 ± 2.1 (n = 50) and 1.3 ± 0.3 nm (n = 50) in thickness, respectively. Scale bars equal 100 nm for A−C (10 nm for insets) and 50 nm for D−F.

Increasing the radiolabeling efficiency was attempted by adding the chelators EDTA and DOTA during seeded embedding. However, the Turkevich−Frens method of synthesizing AuNPs is relatively sensitive and the addition of EDTA or DOTA to the reaction mixture did not improve radiolabeling yield, but interfered with the formation of AuNPs, thus increasing the PDI to >0.30. Accordingly, our experiences were in contradiction with previous reports suggesting that addition of EDTA would improve the Turkevich synthesis of AuNPs.43 While we used as small amounts of gold as possible in order to achieve the highest possible specific activity, we investigated the scalability of the seeded embedding reaction. We obtained similar results at 100 mL/15 mg HAuCl4 to those at the 10 mL/1.5 mg HAuCl4 scale, indicating that the procedure is scalable. The highest specific activity we obtained was 438 MBq/mg of gold, and we have seen no indications of reaching an upper limit of how much activity it is possible to embed. The final gold concentration could be adjusted by concentrating the coated AuNPs on a centrifuge filter and redispersing in the desired volume, as described in the Materials and Methods. Preparation of Coated 64Cu-AuNPs for in Vivo Evaluation (8, 9, and 10). Utilizing the developed seeded embedding methodology, we prepared labeled 64Cu-AuNPs for in vivo investigations to study the pharmacokinetics and biodistribution of the AuNPs with three distinct surface

transferred to the serum proteins. Visually, the highest concentration of AuNPs was found in fractions 9 to 11 (these fractions also contained 87% of the total collected activity). Furthermore, the UV−vis absorbance at 533 nm, corresponding to the surface plasmon resonance peak of the AuNPs, matched the radioactivity in all fractions (see Figure S2, Supporting Information), which confirmed that the activity remained associated with the 64Cu-AuNPs. Further Investigations of Seeded Embedding. As a means to improve embedding and possibly elucidate the embedding mechanism, nonradioactive Cu2+ as a percentage of the total amount of gold was added during seeded embedding. This did not improve the radiolabeling; on the contrary, radiolabeling efficiencies decreased when increasing amounts of nonradioactive CuCl2 were added. This can be overcome by reduction to elemental Cu,35,42 but as no reductants for Cu2+ were added in our system, Cu2+ could be expected to retain its oxidation state. Even when nonradioactive Cu2+ was added in amounts that could not cover the entire AuNP surface, the radiolabeling efficiency still decreased. This indicates that 64 Cu2+ may associate only with specific areas of the AuNP surface and that these areas can be occupied by extra Cu2+. Alternatively, the increased presence of Cu2+ in the reaction mixture may disturb the embedding process (for an expanded discussion of this topic, see Supporting Information). 9891


Article

ACS Nano

Figure 5. 64Cu-AuNP biodistribution by PET/CT. 64Cu-AuNPs (8, 9, 10) were injected intravenously in FaDu tumor-bearing mice, and region-of-interest analysis was constructed on PET/CT images acquired 1, 4, and 24 h after injection. (A) Mean tissue accumulation given as %ID/g ± SD; n = 5 in all cases, except tumors where n = 10. For 10 at 4 and 24 h n = 4, and tumors, n = 8. (B) Representative PET/CT images of 8 (top), 9 (middle), and 10 (bottom). Organs are indicated as (H) heart and the large intracardiac blood volume, (L) liver, and (T) tumors. Yellow arrows illustrate the observed accumulation in lymph nodes for 8 at the 24 h scan (for magnification of lymph node accumulation, see Figure S4, Supporting Information). Color bar indicates activity level of 64Cu; for the 1 and 4 h scans the scale is 0−6.5% ID/g and 0−16%ID/g for the 24 h scans. (C) Left column: Representative axial PET/CT images at 24 h after intravenous injection of 8, 9, and 10. Tumors are indicated by white arrows (scale bar 0−16%ID/g). Right column: Autoradiography of 8, 9, and 10 in 8 μm tumor cryosections. The autoradiography illustrates the heterogeneous intratumoral distribution of the NPs. Scale bars represent semiquantiative accumulation level; the white line is 2 mm long. (D) Max tumor accumulation given as %ID/g ± SD (left three bars for each coating) and percent injected dose per tumor given as %ID/tumor ± SD (right three bars). n = 10 in all cases, except 10 (4 h, 24 h) where n = 8. The max tumor accumulation data mirror the heterogeneous distribution of the 64Cu-AuNPs in the tumors when compared to mean tumor accumulation.

PEG5000-SH was chosen since this polymer size has previously been described as optimal for obtaining intermediate circulation half-life AuNPs (10−20 h).45 Finally, 10 was prepared by coating 7 with a 1:1 mixture of sodium 10-mercaptodecanesulfonic acid and (10-mercaptodecyl)trimethylammonium bromide. Mixed-charge zwitterionic surface-modified AuNPs have shown excellent stability in physiological conditions in vitro and have been reported to be highly biocompatible.46 The phagocytic resistance of the zwitterionic AuNPs was shown to be superior compared to PEGylated AuNPs when evaluated in vitro.46,47 It has previously been reported that zwitterionic AuNPs have a circulation half-life of about 10−30 h.48,49 Accordingly, such AuNPs can be characterized as long circulating. Inclusion of these three distinct surface coatings with expected short-, intermediate-, and long-circulating properties enabled an efficient evaluation of the developed chelator-free seeded embedding methodology. For the in vivo study, 8, 9, and 10 (Chart 1) were incubated with EDTA and purified by filtration to remove unincorporated 64 Cu. Centrifugation was used to remove excess coating

coatings, the molecular structures of which are given in Chart 1. Please note that 8 in Chart 1 is identical to 8 in Scheme 1. A single batch of 7 was prepared, which was coated to give 8, 9, and 10 and compared in vivo by PET/CT imaging. This ensured that the observed biodistributions did not originate from secondary factors, such as differences in size, morphology, or embedding, due to batch-to-batch variations; that is, only the coatings are different on 8, 9, and 10. AuNP 8 was prepared by coating 7 with a 1:2 mixture of 1dodecanethiol and Tween 20 as previously described by Hainfeld.44 Surface coating with a combination of Tween 20 and 1-dodecanethiol has been reported to result in biocompatible AuNPs with excellent stability in media of high ionic strength as well as in thiolytic environments, due to the formation of a hydrophobic bilayer protecting the AuNP core.44 Such particles have been employed for in vivo use with a circulation half-life of about 2 h.44 Accordingly, the coating on 8 results in a relatively short in vivo half-life. AuNP 9 was prepared by coating 7 with MeO-PEG5000-SH. PEGylation of AuNPs and other NPs is a common technique to prevent opsonization and thereby enhance circulation half-life. MeO9892


Article

ACS Nano materials and to concentrate the 64Cu-AuNPs. For all three coatings, the pellets containing the 64Cu-AuNPs were easily redispersed in aqueous media. Characterization of the final in vivo formulations is shown in Table 1, with TEM images shown in Figure 4. The TEM images revealed homogeneous populations of 64Cu-AuNPs in all three cases with mean core diameters ranging from 26.6 to 28.3 (Figure 4A−C), which was comparable to the DLS-derived mean intensity-weighted diameter for the uncoated second-generation 64Cu-AuNP (7) of 24.2 ± 7.7 nm. In addition, visualization by negative staining showed the coating coronas, with the zwitterionic coating (10) having a thickness of 1.3 ± 0.3 nm (n = 50) and the PEG coating (9) a thickness of 5.9 ± 2.1 nm (n = 50). The PEG coating being slightly thicker than the zwitterionic coating is expected from the size of the coating molecules. The large diameter of 9 observed by DLS (73 ± 6 nm) is considered to be caused by a larger hydrodynamic diameter due to the PEG hydration. The 1-dodecanethiol/Tween 20 coating (8) was not visible, which is considered to be due to the sample drying process during TEM imaging. All coated 64Cu-AuNPs exhibited increased PDI and hydrodynamic diameter as compared to 7, when measured by DLS. This was most pronounced for 9, as a result of the comparatively larger PEG5000 polymer, while 10 was larger than 8, possibly due to an increased hydrated shell retained by the more charged nanoparticle surface (for UV−vis spectra see Figure S3, Supporting Information). Modest PDI values and the TEM analysis for all three preparations did not show signs of aggregation of the nanoparticles. After coating, all 64 Cu-AuNPs were in the range 10−100 nm, where longcirculating, untargeted NPs can be expected to show passive tumor accumulation and limited initial clearance by the liver, spleen, and kidneys.23,29,50 In Vivo Investigation of 8, 9, and 10 (Chart 1). 64CuAuNPs were administered intravenously in FaDu tumorbearing mice and evaluated by PET/CT (Figure 5). Accumulation values are reported as percent injected dose per gram tissue (%ID/g). The injected amounts of Au and 64Cu were 76 μg and 2.28 MBq for 8, 54 μg and 1.24 MBq for 9, and 74 μg and 1.93 MBq for 10, respectively. This corresponds to each animal receiving 2.79 × 1011 NPs for 8, 1.98 × 1011 NPs for 9, and 2.71 × 1011 NPs for 10. The amount of Au was slightly lower for 9, which could slightly increase the first-pass clearance due to opsonization.51 However, we did not see any evidence of this in our results, as will be seen below. PETderived accumulation in tumors, blood, liver, and spleen at 1, 4, and 24 h postinjection (Figure 5a) mirrored the ex vivo biodistribution data obtained by well counting of organs excised at 24 h (Figure 6). Notable differences were observed between the three coatings. The 1-dodecanethiol/Tween 20-coated 64Cu-AuNPs (8) showed relatively rapid elimination from the bloodstream, being almost completely cleared at 4 h (2.3 ± 1.1%ID/g), when compared to long-circulating NPs, such as liposomes, polymeric micelles, and other AuNPs, where plasma half-lives of 10−30 h in murine models are observed.23,29,35,52 Hepatic uptake was substantial and rapid (29.9 ± 1.2%ID/g at 4 h), while tumor and spleen accumulation were both modest after 24 h (1.29 ± 0.1%ID/g and 2.2 ± 0.2%ID/g, respectively), likely as a result of the limited circulation time. Prolonged circulation is a prerequisite for passive accumulation of NPs in tumors.2 The observed rapid clearance is in accordance with the reported plasma half-life on the order of a few hours.44 In addition, Hainfeld reported low liver contrast immediately following

Figure 6. Ex vivo analysis of 64Cu-AuNP biodistribution. Tissue accumulation (%ID/g) based on radioactivity was determined by well counting of excised organs. In addition, tumor, liver, spleen, and muscle were analyzed for Au content by ICP-MS. We saw good correlation between 64Cu and Au accumulation in each organ, underlining the in vivo stability of the radiolabel. The Au and corresponding 64Cu accumulation data were subjected to a twotailed, unpaired t test, with a hypothesized mean difference of 0. Significant difference is shown as (−) p > 0.25, (*) p < 0.25, (**) p < 0.05, (***) p < 0.01. Number given as mean ± SD, n = 5, except tumors, where n = 10.

injection,44 which is also in accordance with our observation that hepatic accumulation of 8 is a gradual process with an increase from 1 h to 4 h. It may be speculated that the relatively high critical micelle concentration of 99 mg/L and hydrophilic−lipophilic balance number of 16.7 of the stabilizing Tween 2053 could cause it to detach from the 64Cu-AuNPs under in vivo sink conditions. This would expose the lipophilic surface to interaction with bloodstream components, such as opsonins, causing relatively rapid liver uptake and clearance from blood. As the only 64Cu-AuNPs investigated here, 8 showed accumulation in lymph nodes (for magnification of lymph node accumulation, see Figure S4, Supporting Information). The PEG-coated 64Cu-AuNPs (9) showed improved circulation compared to 8 with an approximate plasma halflife (exponential function estimate based on three blood data points) of 9 h. This corresponds well with reported values for long-circulating NPs in murine models52 and literature data for this type of coating.45 Liver uptake of 9 was limited and relatively constant throughout the imaging period (9.0−9.8% ID/g), while uptake in the spleen was low at first but reached a high level of 29.3 ± 2.2%ID/g at 24 h. The tumor uptake of 9 rose throughout the imaging period and reached a level of 3.89 ± 0.1%ID/g at 24 h. This also corresponds well with reported values for long-circulating NPs such as liposomes.23,52 Accordingly, the PEG-coating (9) appears to be stable in vivo and to confer long-circulating properties to the 64Cu-AuNPs. The sulfonate/quaternary ammonium-coated 64Cu-AuNPs (10) exhibited the fastest elimination from the bloodstream of the three coatings that were investigated. At the 1 h time point, only 5.1 ± 1.2%ID/g remained in the bloodstream, which was markedly lower than the other two coatings. At later time points, very little radioactivity was present in the blood. 10 appeared to rapidly locate to the liver, with accumulation reaching 41.6 ± 3.3%ID/g within 1 h and thereafter remaining relatively constant throughout the imaging period. Virtually no accumulation in the spleen or tumor was observed, suggesting 9893


Article

ACS Nano

CONCLUSION By adding [64Cu]CuCl2 to AuNPs during their formation, we have developed a chelator-free method for stable embedding of the PET isotope 64Cu inside AuNP cores. Seeded embedding by incubation of 64Cu with gold seeds followed by growth of a further gold layer gave superior radiolabeling yields of 53% compared to 23% for direct radiolabeling of AuNP seeds. The labeling was stable against EDTA and serum challenge as well as in vivo. This method allows for 64Cu-AuNPs to be imaged by PET without interference from unbound 64Cu. In addition, we used the developed 64Cu-AuNPs to investigate the in vivo properties of three distinct coatings for AuNPs in FaDu tumorbearing mice. In our hands, a PEG5000 coating attached by thiols performed better in terms of long circulation and tumor accumulation than a coating stabilized by Tween 20 or a zwitterionic coating based on a 1:1 mixture of sulfonates and quaternary ammonium. Nanoparticle coatings are attracting increasing attention as the role of protein absorption to nanoparticle surfaces forming a protein corona is being recognized as highly important and may drive a range of biological effects.56 Future work should aim to couple studies on nanoparticle biodistribution as a function of surface coatings with quantitative analysis of protein binding and carona formation. It has been found that zwitterionic surfaces can lower protein binding and may furthermore effect tumor targeting;57 however, in the present study we observed PEG coatings were superior. Another important aspect is potential antibody responses, IgM and IgG, which will be dependent on the nanoparticle surface coatings and may prevent multiple injections of the same particle type.

that hepatic sequestration was too rapid for uptake in these organs to take place. These results are in contradiction with the literature, where a circulation half-life of 30 h has been reported for a system identical to the one we have tested48 and a half-life about 6−8 h being observed for a similar system using carboxylic acids instead of sulfonates.49 The reason for this discrepancy is currently unknown, but further research may clarify the matter. In addition to PET/CT and well counting, we measured the gold content in selected organs by digestion in aqua regia followed by ICP-MS measurement. We hypothesized that if the 64 Cu was stably embedded within the AuNPs, we could expect a correlation between %ID/g of 64Cu and %ID/g of Au. This correlation was found to exist for all three coatings. The stable association of the 64Cu with the AuNPs is further underlined by the relatively different biodistributions of 8, 9, and 10, which were reflected by both 64Cu and Au (Figure 6). Free 64Cu would be expected to primarily accumulate in the liver (20− 30%ID/g after 1 h, falling to 10−15%ID/g at 22 h) and tumor (around 5%ID/g after 22 h in FaDu tumors).26 This is fairly similar to long-circulating NPs, which is often a point of critique when using 64Cu to image these, although free 64Cu is removed faster from the bloodstream than long-circulating NPs. However, here we demonstrate that the 64Cu closely follows the Au in three unique biodistributions of AuNPs, which underlines that the 64Cu is indeed stably embedded within the AuNPs. Final Remarks. In the present article, we have described methods to stably embed 64Cu within AuNPs. Using this system, we showed that 64Cu-AuNPs coated with PEG thiols (9) exhibited superior long-circulating properties and better passive tumor accumulation than 64Cu-AuNPs coated with 1dodecanethiol/Tween 20 (8) or a zwitterionic sulfonate/ quartenary ammonium mixture (10). Since our prepared 64CuAuNPs do not have chelators attached to the surface nor are subjected to reductive conditions postformation, they can be freely modified with other compounds, such as targeting ligands, and their biodistribution can be expected to mirror that of nonradiolabeled counterparts. 64Cu-AuNPs have potential uses as a theranostic tool to predict or monitor the biodistribution of AuNPs administered for other purposes, such as photothermal ablation. In thermal ablation, however, a uniform distribution of AuNPs within the tumor tissue may be desirable in order to obtain a homogeneous delivery of heat. Since our results indicate relatively large intratumoral variation in accumulation (Figure 5c) with high local maxima (Figure 5d), the effects of using systemically administered AuNPs for this purpose require detailed investigations. As a biologically inert system, 64Cu-AuNPs will remain intact in vivo, and when the 64Cu is stably embedded, the system will accurately report the general in vivo trafficking of NPs, both whole-body and interorgan, making it a very useful tool in NP research. Another possible use for AuNPs is as CT agents. However, the amount of gold needed to obtain sufficient CT contrast by systemic administration is most likely prohibitive in human use, because large amounts would accumulate in the liver and spleen and remain there for extended periods of time. 54 A way to circumvent this problem is local administration of the AuNPs, e.g., fixed in a gel, in order to obtain a sufficient gold concentration.55

MATERIALS AND METHODS Further experimental details are available in the Supporting Information, including materials, characterization of AuNPs, and synthesis of coating materials. Synthesis of First-Generation AuNP Seeds (1). The method was adapted from literature procedures.37,40 A solution of HAuCl4· 4H2O (14.8 mg, 35.9 μmol) in Milli-Q water (100 mL) was prepared. The solution was heated to 75 °C under vigorous stirring, and aqueous trisodium citrate (8.00 mL, 288 μmol, pH 7.0) was added. The mixture was stirred for 60 min at 75 °C. The resulting dark red dispersion was heated to reflux for 30 min followed by cooling to room temperature, furnishing citrate-stabilized first-generation AuNPs (0.066 mg/mL) (1). Synthesis of Second-Generation AuNPs (2), Followed by C12H25-SH/Tween 20 Coating (3). The method was adapted from literature procedures.37,40 AuNP seeds (1) (20 mL, 0.14 mg/mL, pH 6.0) were mixed with Milli-Q water (80 mL). To the dispersion was added aqueous trisodium citrate (2.0 mL, 72 nmol, pH 7.0). The reaction mixture was heated to 75 °C under vigorous stirring. Then aqueous HAuCl4·4H2O (15.0 mL, 18.5 mg, 45.0 μmol) was added dropwise to the warm dispersion. The reaction mixture was then heated to reflux for 30 min and subsequently cooled to room temperature, furnishing a dispersion of citrate-stabilized secondgeneration AuNPs (0.087 mg/mL) (2). Coating was achieved by quick addition of a mixture of 1% (v/v) 1-dodecanethiol in ethanol (1.00 mL, 41.7 nmol) and 1% (v/v) Tween 20 in ethanol (2.00 mL, 17.8 nmol) under vigorous stirring. Stirring was continued for 10 min, furnishing the coated second-generation AuNPs (3). Preparation of 64Cu-AuNPs by Passive Surface Labeling with C12H25-SH/Tween 20 Coating (4). To a vial containing dried [64Cu]CuCl2 were added aqueous NH4OAc buffer (100 μL, 0.1 M, pH 5.5), Milli-Q water, (700 μL) and a dispersion of citrate-stabilized second-generation AuNPs (2) (200 μL, 0.087 mg/mL). The reaction mixture was magnetically stirred for 10 min at room temperature, and a 1% (v/v) 1-dodecanethiol solution in ethanol (24 μL, 1.00 nmol) 9894


Article

ACS Nano and a 1% (v/v) Tween 20 solution in ethanol (48 μL, 0.428 nmol) were mixed and added quickly under vigorous stirring. Stirring was continued for 10 min, after which the reaction mixture was mixed with aqueous tripotassium EDTA (50 μL, 100 mM, 5 μmol, pH 7.5) and incubated for 30 min, before measuring the labeling yield and stability. Preparation of 64Cu-AuNPs by Direct Embedding (5), Followed by C12H25-SH/Tween 20 Coating (6). A solution of HAuCl4·4H2O (1.5 mg, 3.6 μmol) and [64Cu]CuCl2 in Milli-Q water (10 mL) was prepared. The solution was heated to 75 °C under vigorous stirring. To the warm solution was added aqueous trisodium citrate (800 μL, 28.8 μmol, pH 7.0). The yellow solution changed color through faint pink to a red dispersion. After stirring for 40 min at 75 °C the dispersion was heated to reflux for 20 min, followed by cooling to room temperature. The resulting 64Cu-AuNPs (5) were mixed with aqueous tripotassium EDTA (100 μL, 10.0 μmol, pH 7.5) and incubated for 30 min. Then a mixture of 1% (v/v) 1dodecanethiol in ethanol (100 μL, 4.17 nmol) and 1% (v/v) Tween 20 in ethanol (200 μL, 1.78 nmol) were added under vigorous stirring, furnishing C12-SH/Tween 20-coated first-generation 64Cu-AuNPs (6). Stirring was continued for 10 min before measuring the labeling yield and stability. Preparation of 64Cu-AuNPs by Seeded Embedding (7), Followed by C12H25-SH/Tween 20 Coating (8). AuNP seeds (1) (20.0 mL, 0.14 mg/mL, pH 6.0) were mixed with Milli-Q water (80 mL) and aqueous trisodium citrate (2.0 mL, 72 μmol, pH 7.0) followed by addition of [64Cu]CuCl2 to the resulting mixture. The mixture was incubated for 20−30 min during heating to 75 °C under vigorous stirring. This was followed by dropwise addition of HAuCl4· 4H2O (18.5 mg, 44.9 μmol) in Milli-Q water (15 mL). During the addition the color changed from red to pink. The dispersion was heated to reflux for 20 min, causing a color change to purple and then dark red, followed by cooling to room temperature. The resulting 64 Cu-AuNPs (7) were then incubated with aqueous tripotassium EDTA (100 μL, 100 mM, 10.0 μmol, pH 7.5) for 30 min. Then 1% (v/v) 1-dodecanethiol in ethanol (1.00 mL, 41.7 nmol) and 1% (v/v) Tween 20 in ethanol (2.00 mL, 17.8 nmol) were mixed and added quickly under vigorous stirring, furnishing C12-SH/Tween 20-coated second-generation 64Cu-AuNPs (8). Stirring was continued for 10 min before measuring the labeling yield and stability. Test of Radiolabeling Stability against EDTA. After preparation, 64Cu-AuNPs were mixed with aqueous tripotassium EDTA (100 μL, 100 mM, 10.0 μmol, pH 7.5) and incubated for 30 min to sequester any unincorporated 64Cu. Separation was obtained by centrifugation (15 min, 4500g) in Amicon Ultra centrifuge filters (MWCO 50 kDa, 7 nm). The radioactivities (A) in the filtrate (small molecules including 64Cu-EDTA) and the filtrand (64Cu-AuNPs) were compared to provide the radiolabeling efficiency (rad eff = Afiltrand/ (Afiltrate + Afiltrand)). The radiolabeled AuNPs were then redispersed in aqueous tripotassium EDTA (10 mL, 10 mM, pH 7.5). For approximately 48 h the stability was monitored by spinning down the AuNPs through centrifuge filters and measuring the activities in filtrand and filtrate, to show any disassociated 64Cu as 64Cu-EDTA. After each centrifugation the AuNPs were redispersed in aqueous tripotassium EDTA (10 mL, 10 mM, pH 7.5). In addition to centrifuge filtration, the labeling stability was determined by analyzing the mixture on radio-TLC using 5% (w/v) NH4OAc in H2O−MeOH (1:1). Radioactivity at the origin was reported as associated with the 64 Cu-AuNPs, while 64Cu-EDTA represented unbound 64Cu. Test of Radiolabeling Stability in Serum. The method was adapted from Seo et al.29 Freshly prepared 64Cu-AuNPs (8) (1.8 mg) dispersed in PBS (0.5 mL, pH 7.4) were mixed with mouse serum (0.5 mL) and incubated for 2 days at 37 °C. Then an aliquot of the mixture (100 μL) was applied to a size-exclusion chromatography column, packed with Sephacryl-300HR, and eluted with PBS-buffer; 40 fractions (∼1.5 mL/each) were collected. The UV absorption at 280 nm of each fraction was measured. From each fraction was taken a sample (100 μL), which was mixed with liquid scintillation cocktail (10 mL) and counted on a liquid scintillator. Two control experiments were carried out in the same way as described above: (1) mouse serum

(0.5 mL) mixed with PBS (0.5 mL); (2) 64Cu-AuNPs (8) (1.8 mg) dissolved in PBS (1.0 mL, pH 7.4). Preparation of Coated 64Cu-AuNPs for in Vivo Evaluation (8, 9, and 10). Citrate-stabilized 64Cu-AuNPs (7) were prepared as described above and then split into three portions prior to coating. Three different coatings were used: (8) a 2:1 mixture of 1dodecanethiol and Tween 20, (9) MeO-PEG5000-SH, (10) a 1:1 mixture of sodium 10-mercaptodecane-sulfonic acid and (10mercaptodecyl)trimethylammonium bromide. The materials for coating 10 were synthesized as described in the Supporting Information (for 1H NMR spectra see Figures S4 and S5). Formulation 8 was obtained as previously described by premixing 1% (v/v) ethanol solutions of 1-dodecanethiol (205 μL, 8.55 nmol) and Tween 20 (410 μL, 3.65 nmol), then mixing with the 64Cu-AuNPs (8.2 mL, c(Au) 0.075 mg/mL). Formulation 9 was obtained by stirring the 64CuAuNPs (8.2 mL, c(Au) 0.075 mg/mL) with aqueous MeO-PEG5000SH (85 μL, 85 nmol), which corresponds to eight PEG molecules per nm2 of gold surface. According to Wuelfing et al. the “footprint” of one PEG-S ligand is 0.35 nm2, while the “footprint” of one 1-dodecanethiol ligand is 0.21 nm2.58 This gives a moderate excess of 1.7 times the surface saturation for 8 and 2.8 times the surface saturation for 9. Formulation 10 was prepared by mixing the 64Cu-AuNPs (8.2 mL, c(Au) 0.075 mg/mL) with a 1:1 aqueous solution (82 μL) of the sulfonate (50 mM) and quaternary ammonium lipids (50 mM), which corresponds to 495 thiols per nm2 or 100 times the surface saturation of thiols. The large excess of thiols used for 10 was due to the observed instability when using less material. After coating, the 64Cu-AuNP centrifuge filters (MWCO 50 kDa, 7 nm) were used to remove 64CuEDTA. Excess coating material was removed by centrifugation (2 × 35 min at 17000g), each time removing the supernatant and redispersing the AuNP pellet in Milli-Q water. Finally the 64Cu-AuNP dispersions were concentrated 6.5 times and redispersed in physiological phosphate buffer solution (NaCl (8.0 g), KCl (0.20 g), Na2HPO4· 12H2O (2.9 g), KH2PO4 (0.2 g) in 1000 mL of Milli-Q water (pH 7.4) to give a final concentration of gold of approximately 0.5 mg/mL. Animal Model. Female Naval Medical Research Institute (NMRI) nude mice, 5 weeks old (Taconic Europe, Lille Skensved, Denmark), were acclimatized for 1 week and had free access to water and chow. Human pharyngeal squamous cell carcinoma (FaDu) cells, purchased from American Type Culture Collection, were cultured in Minimum Essential Medium (MEM) with Earle’s salts and sodium bicarbonate supplemented with 2 mM L-glutamine (both from Sigma-Aldrich), 1 mM sodium pyruvate, 1% MEM nonessential amino acids solution (100×), 10% fetal calf serum, 100 units/mL penicillin, and 100 μg/mL streptomycin (all from Invitrogen) at 37 °C in 5% CO2. Mice were s.c. inoculated with 5 × 106 FaDu cells suspended in 100 μL (1:1 cell culture medium and BD Matrigel (VWR, Søborg, Denmark)) on both flanks. Tumors were allowed to grow for 18 days, reaching sizes of 393 ± 221 mm3. Mice were randomized into three groups of five mice to receive 8, 9, or 10. All animal experiments were approved by the Danish Animal Welfare Council, Ministry of Justice. PET/CT Imaging Procedures. PET/CT imaging was performed 1, 4, and 24 h after injection of 8, 9, or 10. Mice were anesthetized by inhalation anesthesia (∼3.5% sevoflurane in 40% O2 and medial grade air), and 64Cu-AuNPs were injected into a tail vein. Mice had 2.76 mg/ kg Au (8), 1.98 mg/kg Au (9), and 2.72 mg/kg Au (10) injected, equaling a mean (±SD) [64Cu] activity of 16.7 ± 0.46 MBq (8), 9.1 ± 0.60 MBq (9), and 14.0 ± 0.41 MBq (10). 64Cu-AuNPs were allowed to distribute for 1 h before commencing a 5 min PET scan (termed 1 h scan) with the mouse under anesthesia and positioned on a scanning bed. After completion the scanning bed was moved into the CT fieldof-view for CT scanning. After the scanning procedure mice were allowed to cage rest until performing an identical PET/CT scan at 4 h (termed 4 h scan, 5 min PET acquisition) and again at 24 h (termed 24 h scan, 15 min PET acquisition) after injection of 64Cu-AuNPs. All imaging procedures were performed using inhalation anesthesia as described above. PET/CT imaging was performed on a dedicated Inveon small animal PET/CT system with CT-based PET image attenuation (Siemens Medical Systems, PA, USA). PET scans were reconstructed using a maximum a posteriori reconstruction algorithm 9895


Article

ACS Nano (0.815 × 0.815 × 0.796 mm, resolution; 1.2 mm). Emission data were corrected for dead time and decay, and attenuation correction was performed based on the corresponding CT scan. Image analysis was performed using commercially available Inveon software (Siemens Medical Systems, PA, USA). 3D regions of interest were manually constructed and decay-corrected data (% injected dose per gram of tissue (%ID/g)) reported. Ex Vivo Autoradiography and [64Cu] Gamma Counting. After completion of the 24 h PET/CT scans, mice were terminally perfused under inhalation anesthesia (∼5% sevoflurane in 40% O2 and medial grade air) with >40 mL of isotonic saline solution. Tumors were removed by dissection, and one-half of two tumors from each 64CuAuNP group was snap frozen in dry ice cooled isopentane and mounted in freezing media (Tissue Tek Sakura Finetek, USA). Tumors were subsequently cryosectioned in multiple 8 μm sections and thaw mounted on Microscopy Plus slides (Superfrost Ultra Plus Thermo Scientific, Germany). Semiquantitative autoradiography images of microregional 64Cu-AuNP distribution were acquired by exposing tumor sections to a phosphor imaging screen for 90 min and developed using a phosphor imaging system (Cyclone Plus phosphor imager, PerkinElmer, Waltham, MA, USA) to form semiquantitative photostimulated luminescence images. Tumor sections were subsequently stored in PBS until further analysis. The 64Cu levels in tumors and tissues of interest were determined by gamma counting the excised tissues. The following tissues were analyzed: blood, tumor, liver, spleen, kidney, lung, muscle, heart, brain, intestine. Tissues were placed in preweighed well-counting tubes and well counted for 120 s using a [64Cu] calibrated automatic gamma counter (Wizard2TM3, PerkinElmer). Tumors and tissues were stored at −80 °C until further analysis. Ex Vivo ICP-MS Measurements of Gold in Tissue Samples. Tissue samples from tumor, liver, spleen, and muscle from each mouse were analyzed. Tissue samples of approximately 100 mg were cut and weighed in 15 mL plastic tubes. To each tube was added HNO3 (500 μL, 16 M), H2O2 (300 μL, 10 M), and HCl (50 μL, 11 M). The samples were incubated at 65 °C for 12 h to dissolve the tissue. To the digested samples were added H2O (10 mL) per tube. The water contained 0.5 ppb iridium as an internal reference. Dilutions of the digested tissue samples were prepared. For tumor and muscle a 10× dilution was prepared. For liver and spleen a 100× dilution was prepared. Diluting was done with a 2% (w/v) HCl solution containing 0.5 ppb iridium as an internal standard. A standard curve for gold was prepared, and each sample was measure by ICP-MS. The gold concentration in each kind of tissue was calculated as an average %ID/ g for each of the three groups of mice injected with 8, 9, or 10, respectively.

measurements and Dennis Ringkjøbing Elema (DTU Nutech) for supplying 64CuCl2.

REFERENCES (1) Davis, M. E.; Chen, Z. G.; Shin, D. M. Nanoparticle Therapeutics: An Emerging Treatment Modality for Cancer. Nat. Rev. Drug Discovery 2008, 7, 771−782. (2) Matsumura, Y.; Maeda, H. A New Concept for Macromolecular Therapeutics in Cancer Chemotherapy: Mechanism of Tumoritropic Accumulation of Proteins and the Antitumor Agent Smancs. Cancer Res. 1986, 46, 6387−6392. (3) Kim, D.; Park, S.; Lee, J. H.; Jeong, Y. Y.; Jon, S. Antibiofouling Polymer-Coated Gold Nanoparticles as a Contrast Agent for In Vivo X-Ray Computed Tomography Imaging. J. Am. Chem. Soc. 2007, 129, 7661−7665. (4) Connor, E. E.; Mwamuka, J.; Gole, A.; Murphy, C. J.; Wyatt, M. D. Gold Nanoparticles Are Taken up by Human Cells but Do Not Cause Acute Cytotoxicity. Small 2005, 1, 325−327. (5) Goodman, C. M.; McCusker, C. D.; Yilmaz, T.; Rotello, V. M. Toxicity of Gold Nanoparticles Functionalized with Cationic and Anionic Side Chains. Bioconjugate Chem. 2004, 15, 897−900. (6) Lee, J.; Chatterjee, D. K.; Lee, M. H.; Krishnan, S. Gold Nanoparticles in Breast Cancer Treatment: Promise and Potential Pitfalls. Cancer Lett. 2014, 347, 46−53. (7) Hainfeld, J. F.; Slatkin, D. N.; Smilowitz, H. M. The Use of Gold Nanoparticles to Enhance Radiotherapy in Mice. Phys. Med. Biol. 2004, 49, N309−N315. (8) Kennedy, L. C.; Bickford, L. R.; Lewinski, N. a; Coughlin, A. J.; Hu, Y.; Day, E. S.; West, J. L.; Drezek, R. a A New Era for Cancer Treatment: Gold-Nanoparticle-Mediated Thermal Therapies. Small 2011, 7, 169−183. (9) Alkilany, A. M.; Lohse, S. E.; Murphy, C. J. The Gold Standard: Gold Nanoparticle Libraries to Understand the Nano-Bio Interface. Acc. Chem. Res. 2013, 46, 650−661. (10) Dykman, L.; Khlebtsov, N. Gold Nanoparticles in Biomedical Applications: Recent Advances and Perspectives. Chem. Soc. Rev. 2012, 41, 2256−2282. (11) Yeh, Y.-C.; Creran, B.; Rotello, V. M. Gold Nanoparticles: Preparation, Properties, and Applications in Bionanotechnology. Nanoscale 2012, 4, 1871−1880. (12) Klibanov, a L.; Maruyama, K.; Torchilin, V. P.; Huang, L. Amphipathic Polyethyleneglycols Effectively Prolong the Circulation Time of Liposomes. FEBS Lett. 1990, 268, 235−237. (13) Sershen, S.; Westcott, S. Temperature-sensitive PolymerNanoshell Composites for Photothermally Modulated Drug Delivery. J. Biomed. Mater. Res. 2000, 51, 293−298. (14) Tuchina, E. S.; Tuchin, V. V.; Khlebtsov, B. N.; Khlebtsov, N. G. Phototoxic Effect of Conjugates of Plasmon-Resonance Nanoparticles with Indocyanine Green Dye on Staphylococcus Aureus Induced by IR Laser Radiation. Quantum Electron. 2011, 41, 354−359. (15) Willmann, J. K.; van Bruggen, N.; Dinkelborg, L. M.; Gambhir, S. S. Molecular Imaging in Drug Development. Nat. Rev. Drug Discovery 2008, 7, 591−607. (16) Cai, W.; Wu, Y.; Chen, K.; Cao, Q.; Tice, D. a; Chen, X. In Vitro and In Vivo Characterization of 64Cu-Labeled Abegrin, a Humanized Monoclonal Antibody against Integrin Alpha v Beta 3. Cancer Res. 2006, 66, 9673−9681. (17) Williams, H. A.; Robinson, S.; Julyan, P.; Zweit, J.; Hastings, D. A Comparison of PET Imaging Characteristics of Various Copper Radioisotopes. Eur. J. Nucl. Med. Mol. Imaging 2005, 32, 1473−1480. (18) Yoshii, Y.; Furukawa, T.; Kiyono, Y.; Watanabe, R.; Mori, T.; Yoshii, H.; Asai, T.; Okazawa, H.; Welch, M. J.; Fujibayashi, Y. Internal Radiotherapy with Copper-64-Diacetyl-Bis (N4-Methylthiosemicarbazone) Reduces CD133+ Highly Tumorigenic Cells and Metastatic Ability of Mouse Colon Carcinoma. Nucl. Med. Biol. 2011, 38, 151− 157. (19) Muller, C.; Bunka, M.; Haller, S.; Koster, U.; Groehn, V.; Bernhardt, P.; van der Meulen, N.; Turler, A.; Schibli, R. Promising Prospects for 44Sc-/47Sc-Based Theragnostics: Application of 47Sc

ASSOCIATED CONTENT S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.6b03144. Figures S1, S2, S3, S4, and S5 along with more experimental details (PDF)

AUTHOR INFORMATION Corresponding Authors

*E-mail (A. T. I. Jensen): [email protected]. *E-mail (T. L. Andresen): [email protected]. Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS This research was funded by the Danish Council for Strategic Research, the European Research Council Excellence (ERC) Programme, and the Lundbeck Foundation. We would like to thank Søren Bredmose Simonsen (DTU Energy) for TEM 9896


Article

ACS Nano for Radionuclide Tumor Therapy in Mice. J. Nucl. Med. 2014, 55, 1658−1664. (20) Shokeen, M.; Anderson, C. J. Molecular Imaging of Cancer with Copper-64 Radiopharmaceuticals and Positron Emission Tomography (PET). Acc. Chem. Res. 2009, 42, 832−841. (21) Jones-Wilson, T. M.; Deal, K. a; Anderson, C. J.; McCarthy, D. W.; Kovacs, Z.; Motekaitis, R. J.; Sherry, a D.; Martell, a E.; Welch, M. J. The In Vivo Behavior of Copper-64-Labeled Azamacrocyclic Complexes. Nucl. Med. Biol. 1998, 25, 523−530. (22) Zhang, Y.; Hong, H.; Engle, J. W.; Bean, J.; Yang, Y.; Leigh, B. R.; Barnhart, T. E.; Cai, W. Positron Emission Tomography Imaging of CD105 Expression with a 64Cu-Labeled Monoclonal Antibody: NOTA is Superior to DOTA. PLoS One 2011, 6, e28005. (23) Jensen, A. I.; Binderup, T.; Kumar EK, P.; Kjær, A.; Rasmussen, P. H.; Andresen, T. L. Positron Emission Tomography Based Analysis of Long-Circulating Cross-Linked Triblock Polymeric Micelles in a U87MG Mouse Xenograft Model and Comparison of DOTA and CBTE2A as Chelators of Copper-64. Biomacromolecules 2014, 15, 1625− 1633. (24) Bartholomä, M. D. Recent Developments in the Design of Bifunctional Chelators for Metal-Based Radiopharmaceuticals Used in Positron Emission Tomography. Inorg. Chim. Acta 2012, 389, 36−51. (25) Anderson, C. J.; Ferdani, R. Copper-64 Radiopharmaceuticals for PET Imaging of Cancer: Advances in Preclinical and Clinical Research. Cancer Biother.Radiopharm. 2009, 24, 379−393. (26) Jørgensen, J. T.; Persson, M.; Madsen, J.; Kjær, A. High Tumor Uptake of (64)Cu: Implications for Molecular Imaging of Tumor Characteristics with Copper-Based PET Tracers. Nucl. Med. Biol. 2013, 40, 345−350. (27) Boswell, C. A.; Sun, X.; Niu, W.; Weisman, G. R.; Wong, E. H.; Rheingold, A. L.; Anderson, C. J. Comparative In Vivo Stability of Copper-64-Labeled Cross-Bridged and Conventional Tetraazamacrocyclic Complexes. J. Med. Chem. 2004, 47, 1465−1474. (28) Sprague, J. E.; Kitaura, H.; Zou, W.; Ye, Y.; Achilefu, S.; Weilbaecher, K. N.; Teitelbaum, S. L.; Anderson, C. J. Noninvasive Imaging of Osteoclasts in Parathyroid Hormone-Induced Osteolysis Using a 64Cu-Labeled RGD Peptide. J. Nucl. Med. 2007, 48, 311−318. (29) Seo, J. W.; Mahakian, L. M.; Kheirolomoom, A.; Zhang, H.; Meares, C. F.; Ferdani, R.; Anderson, C. J.; Ferrara, K. W. Liposomal Cu-64 Labeling Method Using Bifunctional Chelators: Poly(ethylene Glycol) Spacer and Chelator Effects. Bioconjugate Chem. 2010, 21, 1206−1215. (30) Fonge, H.; Huang, H.; Scollard, D.; Reilly, R. M.; Allen, C. Influence of Formulation Variables on the Biodistribution of Multifunctional Block Copolymer Micelles. J. Controlled Release 2012, 157, 366−374. (31) Goel, S.; Chen, F.; Ehlerding, E. B.; Cai, W. Intrinsically Radiolabeled Nanoparticles: An Emerging Paradigm. Small 2014, 10, 3825−3830. (32) Sun, X.; Cai, W.; Chen, X. Positron Emission Tomography Imaging Using Radiolabeled Inorganic Nanomaterials. Acc. Chem. Res. 2015, 48, 286−294. (33) Wong, R. M.; Gilbert, D. A.; Liu, K.; Louie, A. Y. Rapid SizeControlled Synthesis of Oxide Nanoparticles. ACS Nano 2012, 6, 3461−3467. (34) Zhao, Y.; Sultan, D.; Detering, L.; Cho, S.; Sun, G.; Pierce, R.; Wooley, K. L.; Liu, Y. Copper-64-Alloyed Gold Nanoparticles for Cancer Imaging: Improved Radiolabel Stability and Diagnostic Accuracy. Angew. Chem., Int. Ed. 2014, 53, 156−159. (35) Sun, X.; Huang, X.; Yan, X.; Wang, Y.; Guo, J.; Jacobson, O.; Liu, D.; Szajek, L. P.; Zhu, W.; Niu, G.; Kiesewetter, D. O.; Shouheng, S.; Chen, X. Chelator-Free (64)Cu-Integrated Gold Nanomaterials for Positron Emission Tomography Imaging Guided Photothermal Cancer Therapy. ACS Nano 2014, 8, 8438−8446. (36) Zhao, Y.; Sultan, D.; Detering, L.; Luehmann, H. Facile Synthesis, Pharmacokinetic and Systemic Clearance Evaluation, and Positron Emission Tomography Cancer Imaging of 64Cu−Au Alloy Nanoclusters. Nanoscale 2014, 6, 13501−13509.

(37) Turkevich, J.; Stevenson, P. C.; Hillier, J. A Study of the Nucleation and Growth Processes in the Synthesis of Colloidal Gold. Discuss. Faraday Soc. 1951, 11, 55. (38) Frens, G. Controlled Nucleation for the Regulation of the Particle Size in Monodisperse Gold Suspensions. Nature, Phys. Sci. 1973, 241, 20−22. (39) Zhao, P.; Li, N.; Astruc, D. State of the Art in Gold Nanoparticle Synthesis. Coord. Chem. Rev. 2013, 257, 638−665. (40) Nguyen, D. T.; Kim, D.-J.; So, M. G.; Kim, K.-S. Experimental Measurements of Gold Nanoparticle Nucleation and Growth by Citrate Reduction of HAuCl4. Adv. Powder Technol. 2010, 21, 111− 118. (41) Bastús, N. G.; Comenge, J.; Puntes, V. Kinetically Controlled Seeded Growth Synthesis of Citrate-Stabilized Gold Nanoparticles of up to 200 Nm: Size Focusing versus Ostwald Ripening. Langmuir 2011, 27, 11098−11105. (42) Alvarez-Paneque, A. F.; Rodríguez-González, B.; PastorizaSantos, I.; Liz-Marzán, L. M. Shape-Templated Growth of Au@Cu Nanoparticles. J. Phys. Chem. C 2013, 117, 2474−2479. (43) Schulz, F.; Homolka, T.; Bastu, N. G.; Puntes, V.; Weller, H.; Vossmeyer, T. Little Adjustments Significantly Improve the Turkevich Synthesis of Gold Nanoparticles. Langmuir 2014, 30, 10779−10784. (44) Hainfeld, J. F. Functional Associative Coatings for Nanoparticles. US 20080089836 A1, 2008. (45) Zhang, G.; Yang, Z.; Lu, W.; Zhang, R.; Huang, Q.; Tian, M.; Li, L.; Liang, D.; Li, C. Influence of Anchoring Ligands and Particle Size on the Colloidal Stability and In Vivo Biodistribution of Polyethylene Glycol-Coated Gold Nanoparticles in Tumor-Xenografted Mice. Biomaterials 2009, 30, 1928−1936. (46) Liu, X.; Huang, H.; Jin, Q.; Ji, J. Mixed Charged Zwitterionic Self-Assembled Monolayers as a Facile Way to Stabilize Large Gold Nanoparticles. Langmuir 2011, 27, 5242−5251. (47) Liu, X.; Jin, Q.; Ji, Y.; Ji, J. Minimizing Nonspecific Phagocytic Uptake of Biocompatible Gold Nanoparticles with Mixed Charged Zwitterionic Surface Modification. J. Mater. Chem. 2012, 22, 1916. (48) Liu, X.; Li, H.; Chen, Y.; Jin, Q.; Ren, K.; Ji, J. Mixed-Charge Nanoparticles for Long Circulation, Low Reticuloendothelial System Clearance, and High Tumor Accumulation. Adv. Healthcare Mater. 2014, 3, 1439−1447. (49) Liu, X.; Chen, Y.; Li, H.; Huang, N.; Jin, Q.; Ren, K.; Ji, J. Enhanced Retention and Cellular Uptake of Nanoparticles in Tumors by Controlling Their Aggregation Behavior. ACS Nano 2013, 7, 6244− 6257. (50) Maeda, H.; Wu, J.; Sawa, T.; Matsumura, Y.; Hori, K. Tumor Vascular Permeability and The EPR Effect in Macromolecular Therapeutics: A Review. J. Controlled Release 2000, 65, 271−284. (51) Laverman, P.; Brouwers, A. H.; Dams, E. T.; Oyen, W. J.; Storm, G.; van Rooijen, N.; Corstens, F. H.; Boerman, O. C. Preclinical and Clinical Evidence for Disappearance of Long-Circulating Characteristics of Polyethylene Glycol Liposomes at Low Lipid Dose. J. Pharmacol. Exp. Ther. 2000, 293, 996−1001. (52) Petersen, A. L.; Binderup, T.; Jølck, R. I.; Rasmussen, P.; Henriksen, J. R.; Pfeifer, A. K.; Kjær, A.; Andresen, T. L. Positron Emission Tomography Evaluation of Somatostatin Receptor Targeted 64 Cu-TATE-Liposomes in a Human Neuroendocrine Carcinoma Mouse Model. J. Controlled Release 2012, 160, 254−263. (53) Kim, C.; Hsieh, Y. Lo. Wetting and Absorbency of Nonionic Surfactant Solutions on Cotton Fabrics. Colloids Surf., A 2001, 187− 188, 385−397. (54) Sadauskas, E.; Danscher, G.; Stoltenberg, M.; Vogel, U.; Larsen, A.; Wallin, H. Protracted Elimination of Gold Nanoparticles from Mouse Liver. Nanomedicine 2009, 5, 162−169. (55) Jølck, R. I.; Binderup, T.; Hansen, A. E.; Scherman, J. B.; Munch af Rosenschold, P.; Kjær, A.; Andresen, T. L. Injectable Colloidal Gold in A Sucrose Acetate Isobutyrate Gelating Matrix with Potential Use in Radiation Therapy. Adv. Healthcare Mater. 2014, 3, 1680−1687. (56) Monopoli, M. P.; Åberg, C.; Salvati, A.; Dawson, K. A. Biomolecular Coronas Provide The Biological Identity of Nanosized Materials. Nat. Nanotechnol. 2012, 7, 779−786. 9897


Article

ACS Nano (57) Mizuhara, T.; Saha, K.; Moyano, D. F.; Kim, C. S.; Yan, B.; Kim, Y.; Rotello, V. M. Acylsulfonamide-Functionalized Zwitterionic Gold Nanoparticles for Enhanced Cellular Uptake at Tumor pH. Angew. Chem., Int. Ed. 2015, 54, 6567−6570. (58) Wuelfing, W. P.; Gross, S. M.; Miles, D. T.; Murray, R. W. Nanometer Gold Clusters Protected by Surface-Bound Monolayers of Thiolated Poly(ethylene Glycol) Polymer Electrolyte. J. Am. Chem. Soc. 1998, 120, 12696−12697.

9898


Mouse Positron Emission Tomography Study of the Biodistribution of

Recommend Documents