Labeling efficiency and labeling stability of 64Cu radioisotope on QT-NPs
To evaluate radiolabeling efficiency of 64Cu radioisotope depending on the amount of radioactivity, 64Cu-QT-NPs were first prepared with different doses of radioactivity of 74, 185, 370, and 740 MBq, corresponding to 8.19, 20.49, 40.97, and 81.94 pmol of 64Cu, respectively, those which were compared with 64Cu-ads-QT-NPs with 185 MBq (20.49 pmol of 64Cu). As expected, all the samples, except the physisorbed phase, showed very high labeling efficiencies beyond 99%, regardless of the amount of 64Cu radioisotope, as shown in Fig. 3; the labeling efficiency was obtained as 99.22 ± 0.16, 99.93 ± 0.08, 99.04 ± 0.99, and 99.25 ± 0.12%, respectively. On the other hand, the latter, where the 64Cu radioisotope was physically adsorbed on the surface of QT-NPs, was measured to be only 63.4 ± 0.99%; the low chemical stability of 64Cu-ads-QT-NPs could also be explained by the fact that the 64Cu radioisotope was weakly bound with physical adsorption on the external surface of QT-NPs in the form of unknown copper compound, which was found to be copper hydroxychloride, as shown in the Supplementary data Fig. S4. Different from the 64Cu-ads-QT-NPs sample, the 64Cu radioisotope atoms in the present 64Cu-QT-NPs were stabilized in the 2D QT lattice, resulting in high thermodynamic stability thanks to the gain of lattice energy as illustrated in Fig. 1a.
In the present study, the sample 64Cu-QT-NPs with 740 MBq was divided into 370 MBq of 64Cu-QT-NPs and 370 MBq of 64Cu-QT-NPs/BSA (Fig. 1) to explore 64Cu labeling efficiency and labeling stability. At first both the samples were centrifuged using filter tubes, and then radioactivities of the isolated precipitates and supernatants were measured with a radioisotope dose calibrator; the labeling efficiencies of 64Cu-QT-NPs and 64Cu-QT-NPs/BSA were determined to be 99.0 ± 0.99 and 99.7 ± 0.42%, respectively (Fig. 4a). It was, therefore, concluded that all the 64Cu2+ ions were stabilized in the 2D QT lattice without remaining in the synthesis solution. To cross-confirm the labeling efficiency for 64Cu-QT-NPs and 64Cu-QT-NPs/BSA, both the samples were studied on basis of radio-TLC, where the mobile phase was a 20 mM sodium citrate / 50 mM EDTA solution (pH = 5.5). As illustrated in Fig. 4b, the instant thin layer chromatograph (iTLC) paper, dropped with the radioactive sample, was placed into the mobile phase for 10 min. And then the unbound and unstable 64Cu radioisotopes were removed by the mobile phase. In the case of free 64Cu radioisotope, all the 64Cu ions were moved to the solvent front by the mobile phase (Fig. 4c). On the other hand, the radio-TLC peaks of 64Cu-QT-NPs and 64Cu-QT-NPs/BSA were obtained at the origin (starting line) as shown in Fig. 4d and 4e. However, from the radio-TLC results of 64Cu-ads-QT-NPs and 64Cu-ads-QT-NPs/BSA (Fig. S5), it was found that 49.1% of 64Cu radioisotopes adsorbed on the former and 90.8% of them on the latter were detached, due to the fact that the 64Cu radioisotopes were weakly bound on the external surface of QT-NPs as expected, and eventually removed by mobile phase easily. It is, therefore, concluded that the exellent chemical stability of 64Cu radioisotope in 64Cu-QT-NPs and 64Cu-QT-NPs/BSA can be ensured by the isomorphical substitution of 64Cu into the 2D QT lattice.
In addition, the 64Cu labeling stability for each sample was investigated not only in saline and PBS but also in biological media such as mouse serum and human one, and it was found that 64Cu-QT-NPs and 64Cu-QT-NPs/BSA exhibited very high stability beyond 99% in all media (Fig. 5). Such a high thermodynamic stability could only be explained by the gain of lattice energy, since the 64Cu radioisotopes were indeed incorporated into the octahedral sites of the QT lattice (Fig. 1).
In-vitro cytotoxicity and cellular uptake studies
Biocompatibility and toxicity of nanoparticles are one of important factors for their biomedical application. To evaluate the cell viability for QT-NPs containing 64Cu radioisotope, both the samples with 0.37 MBq, 64Cu-QT-NPs and 64Cu-QT-NPs/BSA, were treated on the MDA-MB-231 cell lines. After incubating them for 24 h, their cell viability was observed to be almost 100%, indicating their excellent biocompatibility and chemical stability (Fig. 6a), which is also well consistent with our previous studies; QT-NPs were extremely low in toxicity as confirmed by both in-vitro and in-vivo experiments, and internalized inside the cells very efficiently via clathrin-mediated endocytosis [42, 43]. In this study, the in-vitro cytotoxicity for the non-radioactive Cu-doped samples, Cu-QT-NPs and Cu-QT-NPs/BSA, was also evaluated in the concentration range of 1–100 µg/mL on the cell culture line of MDA-MB-231 for 24 and 48 h, and was found to be very low in toxicity (Fig. S6).
Furthermore, the in-vitro cellular uptake behavior for the biocompatible 64Cu-labeled QT-NPs (64Cu-QT-NPs and 64Cu-QT-NPs/BSA) was also studied on the cell culture line of MDA-MB-231 for 24 h. As represented in Fig. 6b, the cellular uptake efficiencies (%dose/cells) of 64Cu-QT-NPs and 64Cu-QT-NPs/BSA were obtained as 23.0 ± 1.36 and 7.9 ± 1.24, respectively, within 30 min, but improved remarkably as 43.3 ± 5.02 and 31.2 ± 2.44, respectively, after 24 h, those which were higher than that of other reported radioisotope labeled samples. In the case of 64Cu-DOTA conjugated dendrimers, a very low uptake efficiency of 1.11% was observed after 2 h of incubation time on human nasopharyngeal cancer (KB) cell line [57]. For 64Cu-NOTA-PA1 formed by conjugating 64Cu-NOTA with pasireotide derivative (PA1), its intercellular uptake efficiency was only 3.67 ± 0.36 and 2.97 ± 0.09% at 2 h on human breast cancer (MCF-7) and human lung cancer (A549) cell lines, respectively [58]. In the case of 64Cu-Sur-NGR2,64Cu-labeled dimeric NGR (asparagine-glycine-arginine) peptide based on sarcophagine (Sur) cage, its uptake value was observed to be 0.72 ± 0.01 and 1.72 ± 0.24% upon treating on the human fibrosarcoma (HT-1080) cell line at 1 h and 2 h, respectively, after incubation [59]. We, therefore, came up with a conclusion that our designed samples, 64Cu-QT-NPs and 64Cu-QT-NPs/BSA, can be a promising radioisotope delivery and imaging technology platform not only because of their high labeling efficiency and excellent chemical stability, but also because of their extremely high biocompatibility and low toxicity.
In-vivo PET imaging and biodistribution studies
To demonstrate the targeting property of 64Cu-QT-NPs and 64Cu-QT-NPs/BSA in in-vivo, the PET imaging and biodistribution studies were carried out on the MDA-MB-231 tumor-bearing xenograft mouse model (Fig. 7 and Fig. 8); they were visualized using a small-animal PET scanner (Inveon™), where the reconstructed images could be made by detecting gamma rays (energy, 511 keV) generated from the decay of 64Cu radioisotopes (\({}_{29}{}^{64}\text{C}\text{u}\to {}_{28}{}^{64}\text{N}\text{i} + {}_{1}{}^{0}{\beta } \left(\text{p}\text{o}\text{s}\text{i}\text{t}\text{r}\text{o}\text{n}\right)+ {{\nu }}_{\text{e}} \left(\text{n}\text{e}\text{u}\text{t}\text{r}\text{i}\text{n}\text{o}\right)\)) [60, 61] as shown in Fig. 7a and 7b. The coronal and transverse PET images were made at various progressive time using a PET scanner, and then the time activity curves for tumors, muscles, liver, and blood were obtained by performing the quantitative region-of-interest (ROI) analysis as shown in Fig. 7c and 7d and Table S2. A prompt and persistent uptake of nanoplates into tumor tissues was achieved in both cases due to the passive targeting effect of QT-NPs (50 to 200 nm) as well described as the enhanced permeability and retention (EPR) effect. Therefore the tumor region (white dot circles and yellow arrow) in PET images becomes brighter for both, but the image of 64Cu-QT-NPs/BSA appears to be much brighter than that of 64Cu-QT-NPs with respect to the time. As presented in Fig. 8a and Table S2, the 64Cu-QT-NPs contents uptaken by tumor tissues and cells were increased gradually from 0.71 ± 0.10, 1.40 ± 0.29, and 2.13 ± 0.25, to 2.43 ± 0.60 %ID/g with respect to the given time (2, 6,24, and 48 h). After BSA coating, however, 64Cu-QT-NPs/BSA seemed to be more efficiently uptaken by tumor tissues, since their contents were even more increased from 0.96 ± 0.36 to 1.80 ± 0.20, and 4.53 ± 0.51, to 4.93 ± 0.81 %ID/g with respect to the same period of tie of 2, 6, 24, and 48 h, indicating that the BSA coated sample exhibited 2-fold higher than the uncoated after 24 h from the injection. Such a difference could be explained by the fact that the blood circulation time of 64Cu-QT-NPs was surely improved by the BSA coating. In practice, %ID/g of 64Cu-QT-NPs in blood pool was observed to be 0.55 ± 0.03, 0.56 ± 0.03, 0.55 ± 0.10, and 0.45 ± 0.02 for 2, 6, 24 and 48 h, respectively, while that of 64Cu-QT-NPs/BSA was measured to be 0.70 ± 0.10, 0.93 ± 0.18, 1.25 ± 0.10, and 1.23 ± 0.19, respectively, for the same period of time. This is an evidence that the uptake of 64Cu-QT-NPs/BSA in tumor tissues was more efficiently made after BSA coating due to the enhanced blood circulation time. In addition, the tumor-to-liver ratio (%) of 64Cu-QT-NPs/BSA (21.86 ± 5.05 %) was also examined to check its targetingfunction, and was found to be 1.7-fold higher than that of 64Cu-QT-NPs (12.67 ± 6.38 %) after 48 h from the injection (Fig. 8b).It is surely due to an enhanced colloidal stability of Cu-QT-NPs/BSA (186 ± 38 nm of size distribution in saline, PDI: 0.356) compared to Cu-QT-NPs (585 ± 102 nm of size distribution in saline, PDI: 0.669) as shown in Table S3. From the above results, the particle size of 64Cu-QT-NPs/BSA could also be expected to be ~ 200 nm in the blood enabling the enhanced tumor penetration by the EPR effect. And it is, therefore, not that surprising that the BSA coated, 64Cu-QT-NPs/BSA, showed 2-fold higher tumor uptake rate than 64Cu-QT-NPs.
Even though the present radio-labeled samples, 64Cu-QT-NPs and 64Cu-QT-NPs/BSA, could effectively deliver the radioisotopes to tumor tissues, thanks to the passive targeting effect, there were some toxicity issues due to a high uptake rate in the liver. After intravenous administration of inorganic nanoparticles within the size of 10 ~ 200 nm, they were known to be rapidly sequestered from the blood and severely accumulated in reticuloendothelial system (RES) organs including liver, spleen, and lymph nodes [62–64]. Thus the liver uptake and clearance of injected nanoparticles have been usually considered as major concerns for bio-applications. Therefore, the U.S. Food and Drug Administration (FDA) regulates that all injected agents have to be cleared completely in a reasonable period of time [62–64]. According to the present in-vivo biodistribution study for 64Cu-QT-NPs and 64Cu-QT-NPs/BSA as shown in Fig. 9, the uptake rates into liver for the former (49.1 ± 8.7%ID/g) and the latter (49.5 ± 8.0%ID/g) were determined to be rather high at 2 h after injection, but decreased repectively to 58.0 and 56.4% after 48 h in liver, resulting in 20.6 ± 3.7 and 21.6 ± 3.1%ID/g, respectively. And the uptake rates into spleen for the former (134.1 ± 51.1%ID/g) and the latter (88.9 ± 41.4%ID/g) were also reduced down to 58.9 and 68.2% after 48 h, giving rise to 55.0 ± 31.8 and 28.3 ± 3.6%ID/g, respectively. As well documented [62], inorganic nanoparticles could be mainly trapped in the reticuloendothelial system (RES) organs (liver and spleen) in the beginning, and then significantly decreased over time due to the excretion mechanism by hepatic (bile to feces) pathway. As expected, the present samples, 64Cu-QT-NPs and 64Cu-QT-NPs/BSA, were uptaken more into liver and spleen than the other organs (Fig. 7 and Fig. 9). In addition, intact 64Cu ions (free 64Cu radioisotopes from 64CuCl2) were also uptaken up into liver tissues but usually cleared within 24 h by the renal (kidney) clearance (Fig. S6).
Nevertheless, the liver uptake of 64Cu-QT-NPs and 64Cu-QT-NPs/BSA over 24 h may be posed its toxicity, which is always one of the major concern for the use of nanoparticles in nanomedicine [63]. However, according to our previous in-vivo liver toxicity results for QT-NPs as summarized in Table S4, the serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST), as the recognition index of liver toxicity, liver disease, or liver damage, were found to be within the normal ranges (7-227 and 37–329 U/L for ALT and AST, respectively) after 72 h from the intraperitoneal administration of QT-NPs [44], indicating the biocompatibility and non-toxicity of the present samples, 64Cu-QT-NPs and 64Cu-QT-NPs/BSA. Furthermore, the concentrations of Mg and Al ions, the major elements of QT, found in liver tissue for the QT treated animal group were determined to be very similar with those for the QT non-treated one at one week after the final injection (Table S3) [44]. As described above, the uptake rate in liver for 64Cu-QT-NPs and 64Cu-QT-NPs/BSA was decreased ~ 60% from 2 to 48 h after injection, expecting its clearance further from the body over time. We, therefore, propose that the present biocompatible nanoplates, 64Cu-QT-NPs and 64Cu-QT-NPs/BSA, could be the promising new nanoagent for PET imaging.