Surface modification-mediated biodistribution of 13C-fullerene C60 in vivo
- 1.5k Downloads
Functionalization is believed to have a considerable impact on the biodistribution of fullerene in vivo. However, a direct comparison of differently functionalized fullerenes is required to prove the hypothesis. The purpose of this study was to investigate the influences of surface modification on the biodistribution of fullerene following its exposure via several routs of administration.
13C skeleton-labeled fullerene C60 (13C-C60) was functionalized with carboxyl groups (13C-C60-COOH) or hydroxyl groups (13C-C60-OH). Male ICR mice (~25 g) were exposed to a single dose of 400 μg of 13C-C60-COOH or 13C-C60-OH in 200 μL of aqueous 0.9% NaCl solution by three different exposure pathways, including tail vein injection, gavage and intraperitoneal exposure. Tissue samples, including blood, heart, liver, spleen, stomach, kidneys, lungs, brain, large intestine, small intestine, muscle, bone and skin were subsequently collected, dissected, homogenized, lyophilized, and analyzed by isotope ratio mass spectrometry.
The liver, bone, muscle and skin were found to be the major target organs for C60-COOH and C60-OH after their intravenous injection, whereas unmodified C60 was mainly found in the liver, spleen and lung. The total uptakes in liver and spleen followed the order: C60 > > C60-COOH > C60-OH. The distribution rate over 24 h followed the order: C60 > C60-OH > C60-COOH. C60-COOH and C60-OH were both cleared from the body at 7 d post exposure. C60-COOH was absorbed in the gastrointestinal tract following gavage exposure and distributed into the heart, liver, spleen, stomach, lungs, intestine and bone tissues. The translocation of C60-OH was more widespread than that of C60-COOH after intraperitoneal injection.
The surface modification of fullerene C60 led to a decreased in its accumulation level and distribution rate, as well as altering its target organs. These results therefore demonstrate that the chemical functionalization of fullerene had a significant impact on its translocation and biodistribution properties. Further surface modifications could therefore be used to reduce the toxicity of C60 and improve its biocompatibility, which would be beneficial for biomedical applications.
KeywordsFullerene Hydroxylation Carboxylation Biodistribution Isotopic labeling
percentage of injected dose
percentage of injected dose per gram tissue
isotope ratio mass spectrometry
transmission electron microscopy
Vienna Pee Dee Belemnite
x-ray photoelectron spectroscopy
The physicochemical properties of nanomaterials, including their size, shape, aspect ratio, charge, composition, surface chemistry and exposure route, may have a significant impact on their molecular toxicity, cellular uptake, transport/fate, absorption, distribution, metabolism, excretion and toxicity in living organisms, thereby affecting their applications in biomedicine [1, 2, 3, 4, 5, 6]. The surface modification of nanomaterials plays an essential role in improving their biocompatibility, regulating their in vivo distribution and metabolism, and reducing their toxicity in the environment, which is especially important for poorly soluble or insoluble nanomaterials (e.g., carbon nanomaterials) [6, 7, 8]. Research directed toward exploring the surface chemistries of carbon nanomaterials can be used to develop a better understanding of their novel properties and functions for biomedical and numerous other applications [9, 10, 11]. For example, the water-soluble derivatives of C60 have been developed for cancer treatment, antioxidation against oxidative stress and free radicals, and bioimaging, etc. in biomedical applications [9, 10, 11]. However, the unique properties of nanomaterials may also pose unexpected hazards to human health and the environment [12, 13, 14, 15]. The development of simple, reliable and stable analytical methods for the quantitative in vivo detection of nanomaterials is therefore critical to evaluate their biosafety and provide information concerning nanofactor-relevant biological effects and toxicity [16, 17]. Quantitative data pertaining to the characteristics of nanomaterials may be beneficial for designing safe and effective nanomaterials/nanodrugs, establishing the predictive models of nanomaterials exposure, refining the toxicological theories, developing a thorough understanding of nanotoxicology mechanisms and regulating risk management strategies for nanomaterials [18, 19].
The development of efficient methods for the quantification of carbon nanomaterials in vivo is quite rightly regarded as the first and most important step in evaluating their biosafety . The biodistribution of carbon nanomaterials has been investigated extensively using a diverse variety of quantification techniques . Based on a comparison of the results generated to date by these different groups, it has been possible to devise several guidelines concerning the pharmacokinetic and biodistribution properties of carbon nanomaterials, which have been communicated to the wider scientific community [16, 20, 21, 22, 23, 24]. Among them, the surface functionalization of carbon nanomaterials is believed to have a significant influence on their biodistribution properties. For example, pristine carbon nanotubes (CNTs) and PEGylated CNTs showed completely different blood circulation and reticuloendothelial system (RES) accumulation levels after intravenous administration [25, 26, 27, 28, 29].
In light of the difficulties associated with the standardization of carbon nanomaterials, one could be forgiven for viewing proposed regularities in the biodistribution patterns of carbon nanomaterials obtained by comparing the results from different groups with some suspicion . Even identical materials prepared by different groups using the same protocol can generate very different biodistribution data and considerable differences may also be observed during the evaluation of otherwise identical carbon nanomaterials by different labs. For instance, Singh et al. and McDevitt et al. [30, 31, 32, 33] independently reported the preparation of amino-CNTs using the same protocol. While Singh et al. reported that the amino-CNTs were rapidly cleared from the body, McDevitt et al. reported much longer retention in RES organs. There is therefore an urgent need for the development of an analytical method for the direct quantitative comparison of the biodistribution profiles of differently functionalized carbon nanomaterials to address these issues.
Fullerene C60 has a well-defined structure that can be reliably quantified by stable isotope labeling [34, 35]. Stable isotope labelling avoids some of the drawbacks associated with radioactive labeling, such as the requirement for radioactive operating conditions and the generation of radioactive wastes. Furthermore, the introduction of radioactive or fluorophore tags can lead to pronounced changes in the surface chemistry of nanomaterials, thereby altering their transport and pharmacokinetic properties in vivo. In contrast, stable isotope 13C skeleton-labeled fullerene retains the intrinsic properties of normal fullerene and its preparation does not require a radioactive protocol. Moreover, stable isotopic tracers can be used to readily distinguish between the endogenous and exogenous sources of elemental building blocks. Stable isotopic tracers are also beneficial for differentiating between ultrafine carbon particles, carbon nanomaterials and carbon-rich biological samples with similar compositions and dimensions [36, 37, 38, 39, 40, 41]. The natural abundance of 13C is 1.1% (atom ratio), which results in high background 13C levels. However, the ratio of 13C to 12C is constant in all living organisms, which means that it does not interfere with the precision and accuracy of the test. Stable 13C has been incorporated into the carbon skeletons of numerous carbon nanomaterials, and the transportation and transformations profiles of the resulting isotope-labeled materials were readily detected by isotope ratio mass spectrometry (IRMS) [25, 29, 35, 42]. The enrichment or depletion of 13C in a sample can be expressed using the delta notation (δ 13C) to a high level of precision (within ±0.1‰) [43, 44]. The quantitative comparison of differently functionalized 13C-C60 materials could therefore provide a definitive understanding of the role played by functionalization in the biodistribution of carbon nanomaterials.
Characterization of C60-COOH and C60-OH
C60-COOH and C60-OH were both found to be highly dispersible in water and other aqueous systems. Furthermore, brown dispersions of C60-COOH and C60-OH were found to be stable for more than 6 months under ambient conditions without any visible precipitation. However, transmission electron microscopy (TEM) images revealed that very small aggregates were formed in aqueous solutions of C60-OH and C60-COOH, with diameters of 50 and 6–20 nm, respectively (Fig. 2). The size distribution patterns of both samples in solution are shown in Fig. 2b and d. The diameters of the C60-OH and C60-COOH particles in water were around 62 and 120 nm, respectively. The hydrodynamic radii of both samples were larger than the diameters observed in the TEM images.
The infrared (IR) spectra of C60-COOH and C60-OH were also collected to confirm that the functionalization reactions had proceeded successfully (Fig. 3). The IR spectrum of C60-COOH contained a broad peak at 3375 cm−1, which was attributed to the stretching vibrations of the -COOH/-NH- groups. A tiny peak was also observed at 2931 cm−1, which was attributed to the stretching vibrations of the -CH2- groups. The peaks at 1644 and 1560 cm−1 were assigned to the stretching vibrations of the C = O and C = C moieties, respectively. Last, the characteristic peak of the -COOH groups was observed at 1410 cm−1. The IR spectrum of C60-OH contained a broad peak at 3406 cm−1 for the -OH groups, an intense peak at 1588 cm−1 for the C = C groups and a characteristic peak corresponding to the bending vibrations of the -OH groups at 1393 cm−1. The differences in functionalities between C60-COOH and C60-OH were also evaluated by X-ray photoelectron spectroscopy (XPS). As shown in Fig. 4, the C1s XPS spectrum of C60-COOH contained a strong C = O peak at 288.0 eV, whereas the spectrum of C60-OH contained a very weak C = O peak at 288.4 eV. The appearance of C-O-C (285.6 eV) is a common phenomenon for C60-OH. XPS demonstrated the chemical state of carbon atoms. The lack of C-O in C60-COOH indicated that there was no -OH in C60-COOH. Similarly, the very weak C = O in C60-OH indicated that there was very limited or no -COOH in C60-OH.
The 13C abundances of 13C-C60-COOH and 13C-C60-OH were measured by IRMS. Urea was used as a working standard to calibrate the IRMS facility, which gave a δ 13C value of about -40.5573 ± 0.01106. The value determined for urea by IRMS was found to be equal to that of the certified value for urea, indicating that our IRMS system could be used to make reliable measurements. The δ 13C values for 13C-C60-COOH and 13C-C60-OH were determined to be 7910.69 and 7914.92, respectively. It is noteworthy that the δ values for 13C-C60-COOH and 13C-C60-OH were much higher than that of the urea standard value. Based on the equation for determining the delta (δ) parameter [35, 43, 44, 45], the 13C abundances (the atom percentage of 13C atoms per all carbon atoms in a specific fullerene derivative) were calculated to be about 8.204% ± 0.013% for 13C-C60-COOH and 8.310% ± 0.017% for 13C-C60-OH, equaling to 5.3 13C-atoms per one 13C-C60-COOH and 5.0 13C-atoms per one 13C-C60-OH. The atom percentage of 13C reported for the unmodified C60 in Ref. 35 was 8.752%. Collectively, there were about 5 ~ 6 13C atoms per C60 cage, where the 13C atoms were incorporated in the skeleton of fullerene cage (Fig. 1). The amount of 13C found in the two fullerene nanomaterials was around 9.0%, which is higher than the weight percentage of 13C in natural carbon (1.2% ± 0.0013%). All of these characterization data indicated that the C60 derivative samples were suitable for the following in vivo quantification experiments.
Biodistributions of C60-COOH and C60-OH after i.v. injection
The biodistributions of C60-COOH and C60-OH were studied following their i.v. injection in mice. C60-COOH and C60-OH both showed low accumulation levels (expressed as the percentage of the injected dose per gram of tissue, %ID/g) in the bodies of the mice (Fig. 5). The distributions of C60-COOH and C60-OH were detected in the whole body with relatively high levels in the liver, bone, muscle and skin tissues. The values obtained for the accumulation of C60-COOH and C60-OH in the RES organs in the current study were much lower than those reported in our previous study for unmodified C60 (liver: 15.3%ID/g; spleen: 63.5%ID/g; lung: 18.2%ID/g) . Furthermore, the value for C60-COOH was slightly higher than that of C60-OH. During the first 5 min, it was clear that the distribution rate and accumulation levels of C60-OH were higher than those of C60-COOH. Considerable differences were observed in the accumulation levels of C60-COOH and C60-OH in the heart, liver, kidney, lung, small intestine and blood tissue samples. Furthermore, higher levels of accumulation were observed in the muscle and bone tissues for both C60-COOH and C60-OH compared with the other organs. The accumulation levels of C60-COOH and C60-OH increased over time with the accumulation levels of C60-COOH eventually catching up to those of C60-OH. At 24 h, we observed almost identical levels of trapping for C60-COOH and C60-OH, with very similar distribution patterns. C60-COOH resulted in slightly higher levels of accumulation in the liver compared with C60-OH following i.v. administration, whereas C60-OH resulted in higher accumulation levels in the bone and skin tissues. However, the only significant differences observed between the accumulation levels of C60-COOH and C60-OH were found in the bone. C60-COOH and C60-OH were almost completely eliminated from the test tissue samples at 7 d post exposure. Only low accumulation levels were detected in some of the tissues at this time point with no significant differences between the difference materials and tissues (p > 0.05 compared with the control group).
By multiplying the %ID/g data by the weights of the organs or tissues, it was possible to obtain the corresponding %ID (percentage of injected dose) data (Fig. 6). The total uptakes in the liver and spleen were of the order C60 > > C60-COOH > C60-OH. The distribution rate at 24 h followed the order C60 > C60-OH > C60-COOH. The liver and intestine were found to be the major target organs for C60-COOH and C60-OH at 24 h after i.v. injection, whereas unmodified C60 was mainly found in the liver (23.75%ID), spleen (17.85%ID) and lung (4.56%ID) . It is noteworthy that the uptakes of C60-COOH and C60-OH in bone and muscle were found to be very large at 24 h, considering that the total weights of the bone and muscle tissues were high. However, we did not obtain the %ID values for the bone and muscle tissues because it was not possible to accurately determine the weights of the bone and muscle tissues and it would be unreasonable to infer that the accumulation levels were the same in all of the bone and muscle tissues.
Uptakes of C60-COOH, C60-OH and unmodified C60 in RES organs after i.v. injection 24 h
Unmodified C60 (%ID) a
8.35 ± 2.37
7.27 ± 1.26
23.75 ± 3.81*,§
0.33 ± 0.01
0.27 ± 0.10
17.85 ± 4.35*,§
0.44 ± 0.23
0.64 ± 0.23
4.56 ± 1.96*,§
Biodistributions of C60-COOH and C60-OH after gavage and i.p. injection
The absorption and translocation properties of C60-COOH and C60-OH varied considerably depending on their route of administration (Fig. 7). For example, the results revealed that C60-COOH showed high levels of tissue accumulation following gavage administration, with high concentrations of C60-COOH observed in the stomach and intestine. Several other organs, including the heart, liver, spleen, lung and bone, were also targeted by this material. In contrast, the accumulation levels for C60-OH were all rather low at less than 1.0%ID/g. However, these differences in the accumulation levels of C60-COOH and C60-OH were found to be insignificant because of the large variabilities observed in the experiments, as well as the low accumulation levels.
Following i.p. exposure, C60-COOH was only detected in the bone, spleen and kidneys at very low concentrations, which indicated low levels of absorption in these organs. In contrast, C60-OH translocated into a much wider range of tissues after i.p. injection, including the stomach, kidneys, lung, intestine, blood, muscle and bone, where it showed high accumulation levels. Notably, C60-OH was detected in tissues both in- and outside of the abdominal cavity. In a similar manner to gavage exposure, the differences observed between C60-COOH and C60-OH following i.p. exposure were determined to be insignificant.
According to the experimental results described above, the surface modification of fullerene nanomaterials led to changes in their accumulation levels in specific organs. At 24 h post exposure via i.v. administration, unmodified C60 was found to be mainly distributed in the liver, spleen and lung, whereas C60-COOH and C60-OH were mainly trapped in liver, bone, muscle and skin following the same time period. These findings therefore implied that the chemical functionalization of C60 with carboxyl or hydroxyl groups resulted in the formation of materials that could be eliminated from the liver or deposited in bone with greater ease than the parent fullerene.
Low accumulation levels of C60-OH after i.v. administration have also been reported by several other groups. For example, Li et al. reported the biodistribution of 67Ga-C60(OH)x after i.v. exposure . In this particular case, the liver (12.9%ID/g), spleen (17.1%ID/g), skull (23.7%ID/g) and thighbone (17.6%ID/g) showed relatively high accumulation levels of the modified fullerene at 24 h, whereas all of the other organs showed accumulation levels of less than 4%ID/g. Using another radioactive isotope 99mTc, Li et al. found that 99mTc-C60(OH)x was distributed in the liver (9.4%ID/g), spleen (7.2%ID/g), bone (8.4%ID/g), lung (1.7%ID/g), kidneys (5.2%ID/g), intestine (2.1%ID/g) and heart (0.7%ID/g) of mice with low accumulation levels at 24 h . Similarly, Song et al. reported that 99mTc(CO)3-C60(OH)20 distributed in the whole body except for the brain, with accumulation levels of less than 1.0%ID/g at 24 h . Maksin et al. reported the biodistribution of radiolabeled fullerenol after i.v. exposure . The results of this study revealed that fullerenol was distributed into the lung (2.1%ID), liver (20.1%ID), kidney (5.3%ID), muscle (0.16%ID/g), bone (0.59%ID/g), thyroid (0.2%ID), spleen (0.66%ID) and intestine (16%ID) at 2 h. Li et al. reported the biodistribution of C60(OH)x(O)y after i.v. exposure, and showed that fullerenol was distributed into the heart, lung, liver, spleen, kidney, muscle, bone, intestine, fur and brain . Furthermore, relatively high accumulation levels were found in the liver (4.3%ID/g), spleen (2.3%ID/g), kidney (4.1%ID/g) and bone (8.0%ID/g) at 24 h post exposure. Ji et al. labeled C60(OH)x with 125I and followed the translocation of 125I-C60(OH)x in tumor-bearing mice . The results showed that 125I-C60(OH)x was distributed through the whole body of the mice, except for their brain, and that relatively high accumulation levels were found in the liver (2.0%ID/g), stomach (1.2%ID/g), kidney (1.6%ID/g) and bone (1.7%ID/g) at 24 h. These results from the literature therefore confirm the reliability of our distribution results for C60-OH. Furthermore, based on the experimental reports described above, it is clear that the surface hydroxylation of C60 leads to improved water-solubility and bioavailability. These changes also enhance the ability of these materials to cross biological barriers in vivo, and therefore reduce the accumulation and retention levels of these compounds in the tissues/organs.
Furthermore, hydroxylation led to slightly lower levels of RES capture than those encountered following the carboxylation of the fullerene core. The low RES capture of C60-OH observed in this study was consistent with the results of several other studies. For example, 67Ga-, 125I- and 99mTc-labeled fullerenols all showed low levels of RES capture [46, 47, 48, 49, 50, 51]. Pristine C60 nanoparticles are hydrophobic xenobiotic substances, which are easily trapped in the RES organs as part of an organism’s natural defense mechanism. While surface chemical modifications can lead to enhanced hydrophilicity and biocompatibility properties in fullerene nanomaterials, as well as enhancing their ability to cross biological barriers in mice, they can also lead to reduced uptake into the RES organs.
Our biodistribution results for C60-COOH were different from those reported by Yamago’s group . In Yamago’s study, C60 was functionalized via a cycloaddition reaction with dipolar trimethylenemethane to give a water-soluble fullerene derivative bearing terminal carboxyl groups. These fullerene derivatives were predominantly trapped in the liver (91.7%ID at 16 h), with the accumulation levels in the heart, spleen, lung, kidneys, brain and testicles being much lower. The difference between our results and those of Yamago’s groups could be attributed to differences in the functionalization density and the length of chains attached to the carboxylic acid groups. In our study, each fullerene cage had two carboxyl groups, which were directly linked on the pyrrolidine ring. In Yamago’s study, each fullerene cage had only one carboxyl group, which was linked to the five-membered ring with a small linker. The solubility of C60-COOH was therefore much higher than that of the fullerene derivatives described in Yamago’s study. The lower hydrophilicity of Yamago’s compounds would lead to higher levels of protein binding. Yamago’s fullerene derivatives would also be recognized by opsonin and more easily trapped in the RES organs.
The surface functionalization of fullerene also affected its bioavailability and distribution following the gavage and i.p. administration routes, although the differences observed in these cases were not found to be significant because of their low accumulation levels. The high absorption of C60-COOH following gavage exposure could be attributed to the hydrophobicity of the fullerene cage and the small size of the C60-COOH aggregates. There have been several reports in the literature that show that small nanoparticles with good hydrophobicity show high bioavailability after oral exposure [53, 54, 55, 56, 57, 58, 59]. However, following i.p. exposure, the tissue translocation of C60-OH was much more widespread and pronounced than that of C60-COOH. The low translocation of C60-COOH observed in this case could be attributed to the negative charges on the carboxylate groups and its hydrophobicity. The free translocation of hydroxylated carbon nanomaterials has also been observed for CNTs. For example, Wang’s group reported that hydroxylated CNTs were detected in the whole body after i.p. injection [60, 61]. The results of the current study suggest that the translocation and distribution properties of functionalized fullerenes are regulated by their surface-modified groups. Moreover, accumulated fullerene derivatives could be cleared from the body during long-term observations, since the i.v. injection data clearly indicated that the fullerene derivatives had been cleared completely at 7 d post exposure.
Mechanistically, how the surface regulates bioavailability and biodistribution might be related to the opsonization of fullerene. It is well known that nanoparticles entering systemic circulation can be tagged with opsonin. The subsequent recognition of opsonin can lead to these nanoparticles being captured by phagocytic cells in the RES organs. We previously demonstrated that surface functionalization can affect the interactions formed between fullerenes and proteins [62, 63]. For example, increasing the degree of functionalization (hydroxylation) led to a decrease in the extent of several fullerene–protein interactions, most likely because of the reduced opsonization of C60-OH. The lower accumulation levels of C60-COOH and C60-OH could therefore be attributed to reduced levels of opsonization following their surface modification.
It is noteworthy that C60-COOH and C60-OH formed different-sized aggregates because of the differences in their functionalization. The differences in the sizes of these aggregates could also affect their biodistribution behaviors. However, the influence of their size would be much less pronounced in this case, because the particulate sizes of C60-COOH and C60-OH were smaller than the pulmonary capillary diameter. When particulate size is smaller than the pulmonary capillary diameter, the particles can escape pulmonary filtration and be captured by the RES via opsonization [20, 21]. Opsonization is mainly dependent on the surface of a given nanomaterial, whereas particle size plays a much smaller role.
It is also noteworthy that the accumulation levels of C60-COOH and C60-OH found in bone and muscle (hind leg) were both relatively high following short-term observation (up to 24 h), suggesting that the total amounts of accumulation in bone and muscle were large. To estimate the total uptakes, we assumed that the accumulation levels in whole bone, muscle and skin were homogenous and that the tissue indices (tissue weight/body weight) were 10, 25 and 10% for bone, muscle and skin, respectively. The bone uptake was 25 ± 3%ID for C60-COOH and 36 ± 1%ID for C60-OH. The muscle uptake was 48 ± 21%ID for C60-COOH and 40 ± 3%ID for C60-OH. The skin uptake was 14 ± 3%ID for C60-COOH and 21 ± 5%ID for C60-OH. The percentage of C60-COOH increased from 42%ID at 5 min to 112%ID at 24 h, whereas the percentage of C60-OH increased from 45%ID at 5 min to 123%ID at 24 h. This implied that both C60-COOH and C60-OH were rapidly cleared from the blood and distributed in the interstitial fluid, from which point they gradually migrated into the organs and tissues during the first 24 h. The total percentages at 24 h were all larger than 100%ID, suggesting that while our original estimation was inaccurate, the trends (not the accurate values) were meaningful.
The results of the current study clearly show that the functionalization shifted the uptake of fullerene from the RES organs to the soft tissue and skeleton. Similar trends to this have been reported elsewhere in the literature, where the hydroxylation of fullerene led to high levels of accumulation in the bone and muscle tissues [46, 47, 48, 49, 50, 51]. These differences could be attributed to the functionalized fullerenes being better equipped to escape capture by the RES and more capable of making their way into the soft tissues and bone. The retention of these fullerene derivatives in the body remained transient, with no statistically significant levels of accumulation being detected after 7 d. From a toxicological perspective, this shift in the accumulation of fullerene derivatives from the RES toward the soft tissues and bone would definitely have an impact on their toxicity. Reduced levels of RES capture would relieve the toxicity of fullerenes toward the liver, lung and spleen, although the translocation of these materials to other organs and tissues could lead to toxicity elsewhere. The good news in this case is that these fullerene derivatives were rapidly cleared from the body. The long-term toxicities of fullerene derivatives could be lower than those of the corresponding pristine materials. Future studies should therefore focus on the toxicological evaluation of pristine and functionalized fullerene systems to develop a better understanding of the toxicological impact of functionalization.
In summary, we have quantitatively investigated the impact of different surface modifications and routes of administration on the biodistribution of fullerene in vivo using a 13C-labeled carbon cage tracer. We specifically compared the biodistribution behaviors of C60-COOH and C60-OH in mice following their exposure via three different pathways, where C60-COOH and C60-OH showed distinct distribution patterns. The surface modification of fullerene C60 led to a decrease in its accumulation and rate of distribution, as well as altering its target organs. These differences clearly indicated that the chemical functionalization of fullerene has a significant influence on its translocation and biodistribution. In particular, the differences between C60-COOH and C60-OH were more distinguishable following gavage and i.p. administration. Undoubtedly, specific surface modifications could lead to improvements in the biocompatibility of C60, by reducing its toxicity, which could be of some benefit in terms of the biomedical application of these materials. We therefore believe that our results could benefit specific biomedical applications, as well as having an impact on biosafety studies concerning other carbon nanomaterials.
Amorphous carbon powder of 99% 13C was purchased from Cambridge Isotopes Laboratories, Inc. (Andover, MA, USA). 13C-C60 was prepared using the arc discharge method according to our previously reported procedure [34, 35]. Unlabeled C60 was obtained from Nanjing XFNANO Materials Tech Co., Ltd (Nanjing, China). All of the other chemicals used in this study were purchased as the analytical degree from various chemical suppliers.
Preparation of 13C-C60-COOH and 13C-C60-OH
The carboxylation of 13C-C60 was conducted according to a modified version of the method reported by Gan et al . Briefly, Me2HNCH2COOEt•HCl (279 mg, 2.0 mmol) and sodium hydroxide (228 mg, 5.7 mmol) were added to 15 mL of methanol, and the resulting mixture was stirred until it formed a solution. This solution was then added to a solution of 13C-C60 in toluene (60 mL), and the resulting mixture was irradiated with a nondestructive xenon lamp (300 W) under continuous stirring for 1 h, during which time the solution changed color from purple to brown. The solution was then concentrated under reduced pressure to approximately 20 mL and treated with sodium hydride (1 g), before being heated at 70 °C under continuous stirring for 6 h. The mixture was then cooled to room temperature and added to 10 mL of methanol before being evaporated to dryness under reduced pressure to give a residue. The residue was then added to water (50 mL) to dissolve the 13C-C60-COOH, and the resulting suspension was centrifuged (3000 × g, 5 min). The supernatant was then collected and purified by column chromatography over Sephadex G-25 gel to separate the desired 13C-C60-COOH material from any unwanted small molecules. The first fraction was collected and lyophilized to give 13C-C60-COOH.
The hydroxylation of 13C-C60 was conducted according to a procedure from the literature [65, 66, 67]. Briefly, a toluene solution of 13C-C60 was vigorously stirred with powdered NaOH and 50% tetrabutylammonium hydroxide (TBAH) at room temperature in an open vessel for 24 h. The biphasic reaction mixture was then separated and the aqueous phase was filtered to remove trace quantities of water-insoluble residues. The filtrate was then added with methanol, which resulted in the formation of a brown precipitate. The methanol was subsequently removed in vacuo to give a residue, which was dissolved in water and purified by column chromatography over Sephadex G-25 gel (1 × 30 cm) eluting with neutralized water. Fractions containing the desired product were freeze-dried to give 13C-C60-OH as a yellow powder.
The 13C-C60-COOH and 13C-C60-OH samples generated above were characterized by IRMS (ThermoFisher Delta V advantage; Waltham, MA, USA), TEM (Tecnai G2 F20 U-TWIN; Hillsboro, Oregon, USA), Nanosizer analysis (Nanobrook-Omni, Bruker; Holtsville, New york, USA), IR spectrometry (Mir-IR, Nicolet iN10 MX, ThermoFisher; Waltham, MA, USA) and XPS (Thermo Scientific ESCALab250Xi; Waltham, MA, USA).
Animal experiments were performed humanely in accordance with the Animal Care and Use Program Guidelines of the Sichuan Province, China. Male ICR mice (25 g) were obtained from Sichuan University Animal Center, Chengdu, China. The mice were raised in plastic cages (three mice/cage) on a 12-h light/dark cycle with ad libitum access to food and water. Following acclimation, the mice were randomly divided into 15 groups (three mice for each group) for the biodistribution studies.
Biodistributions of 13C-C60-COOH and 13C-C60-OH after i.v. injection
The mice were injected via their tail vein with 13C-C60-COOH or 13C-C60-OH at a single dose of 400 μg in 200 μL of 0.9% aqueous NaCl solution. Blood samples were collected periodically from the mice, which were subsequently scarified at designated time intervals. Several tissue samples, including heart, liver, spleen, stomach, kidney, lung, brain, large intestine, small intestine, bone (hind leg), muscle (hind leg) and skin (abdomen), were collected from the mice and dissected, homogenized and lyophilized before being analyzed by IRMS using our previously reported procedure . The control animals were injected with 200 μL of 0.9% aqueous NaCl solution and sampled in the same way. The results of these analyses have been presented as %ID/g values for all of the tissue samples, as well as %ID values for the heart, liver, spleen, stomach, kidneys, lungs, brain, large intestine and small intestine samples.
Translocation profiles of 13C-C60-COOH and 13C-C60-OH after gavage and i.p. administration
For gavage exposure, the mice were exposed to 13C-C60-COOH or 13C-C60-OH at a single dose of 400 μg in 200 μL of 0.9% aqueous NaCl solution via gavage. For i.p. exposure, the mice were intraperitoneally injected with 13C-C60-COOH or 13C-C60-OH at a single dose of 400 μg in 200 μL of 0.9% aqueous NaCl solution. The control groups were exposed to 200 μL of 0.9% aqueous NaCl solution according to the same protocols as those described above. All of these animals were sampled in the same way as those subjected to the i.v. injection protocol.
All data have been expressed as the mean values of three individual observations together with the associated standard deviation (mean ± SD). Significant differences between the biodistribution data of C60-OH and C60-COOH were calculated using the Student’s t-test. Differences were considered significant for p < 0.05. Comparisons were made in a time-dependent manner between the different exposure pathways. The %ID data of 13C-C60-OH and 13C-C60-COOH were compared with those of the unmodified C60 to reveal the impact of the surface functionalization on the biodistribution of fullerene.
This work was supported by the National Natural Science Foundation of China (Nos. 11475194, 21307101 and 61378040), Beijing Natural Science Foundation (No. 2152038), Top-notch Young Talents Program of China, and the Project of Postgraduate Degree Construction, Southwest University for Nationalities (No. 2015XWD-S0703).
- 9.Meng H, Xing GM, Blanco E, Song Y, Zhao LN, Sun BY, et al. Gadolinium metallofullerenol nanoparticles inhibit cancer metastasis through matrix metalloproteinase inhibition: imprisoning instead of poisoning cancer cells. Nanomedicine: NBM. 2012;8(2):136–46.Google Scholar
- 35.Chang X-L, Ruan LF, Yang S-T, Sun BY, Guo CB, Zhou LJ, et al. Quantification of carbon nanomaterials in vivo: Direct stable isotope labeling on the skeleton of fullerene C60. Environ Sci: Nano. 2014;1(1):64–70.Google Scholar
- 49.Maksin T, Djokic D, Jankovic D, Djordjevic A, Neskovic O. Comparison of some physico-chemical parameters and biological behaviour of fullerenol labelled with technetium-99m. J Optoelectro Adv Mater. 2007;9(8):2571–7.Google Scholar
- 55.Lv L-Z, Tong C-Q, Yu J, Han M, Gao J-Q. Mechanism of enhanced oral absorption of hydrophilic drug incorporated in hydrophobic nanoparticles. Int J Nanomed. 2013;8(1):2709–17.Google Scholar
- 60.Wang HF, Wang J, Deng XY, Sun HF, Shi ZJ, Gu ZN, et al. Biodistribution of carbon single-wall carbon nanotubes in mice. J Org Chem. 2004;4(8):1019–24.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.