56Co-labelled radioactive Fe3O4 nanoparticles for in vitro uptake studies on Balb/3T3 and Caco-2 cell lines
- First Online:
- Cite this article as:
- Marmorato, P., Simonelli, F., Abbas, K. et al. J Nanopart Res (2011) 13: 6707. doi:10.1007/s11051-011-0577-7
- 160 Views
Magnetite nanoparticles (Fe3O4 NPs) are manufactured nanomaterials increasingly used in healthcare for different medical applications ranging from diagnosis to therapy. This study deals with the irradiation of Fe3O4 NPs with a proton beam in order to produce 56Co as radiolabel and also with the possible use of nuclear techniques for the quantification of Fe3O4 NPs in biological systems. Particular attention has been focused on the size distribution (in the range of 100 nm) and the surface charge of the NPs characterizing them before and after the irradiation process in order to verify if these essential properties would be preserved during irradiation. Moreover, X-ray diffraction studies have been performed on radioactive and non-radioactive NPs, to assess if major changes in NPs structure might occur due to thermal and/or radiation effects. The radiation emitted from the radiolabels has been used to quantify the cellular uptake of the NPs in in vitro studies. As for the biological applications two cell lines have been selected: immortalized mouse fibroblast cell line (Balb/3T3) and human epithelial colorectal adenocarcinoma cell line (Caco-2). The cell uptake has been quantified by radioactivity measurements of the 56Co radioisotope performed with high resolution γ-ray spectrometry equipment. This study has showed that, under well-established irradiation conditions, Fe3O4 NPs do not undergo significant structural modifications and thus the obtained results are in line with the uptake studies carried out with the same non-radioactive nanomaterials (NMs). Therefore, the radiolabelling method can be fruitfully applied to uptake studies because of the low-level exposure where higher sensitivity is required.
KeywordsFe3O4 NPs uptakeProton irradiationRadioactive nanoparticlesBalb/3T3 cell systemCaco-2 cell systemIn vitro uptakeNanomedicine
Nanotechnology has the potential for creating new materials with a great range of useful applications, from technology to industrial production, as well as medicine (e.g., drugs, diagnostics, and cancer therapy) (Liong et al. 2008; Lubbe et al. 1996). The use of NMs in healthcare offers the chance to further improve both diagnosis and therapy; in fact NMs are increasingly being applied to in vivo and in vitro diagnostics, medical imaging, targeted drug delivery, and regenerative medicine (McCarthy and Weissleder 2008; Gould 2006).
Magnetic NPs, in particular, are useful for drug targeting (Gupta and Hung 1989; Lubbe et al. 1996) and bioseparation (Sang et al. 2005) including cell sorting, because they experience a force in a magnetic field gradient directed toward the areas of high magnetic flux density. Moreover, Fe3O4 is useful for a wide variety of applications such as immuno-assays (e.g., for the detection of viruses), tissue repair, magnetic hyperthermia (heating of magnetic NPs by an external alternating radiofrequency system to cause necrosis of tumor cells) (Chan et al. 1993), and magnetofection (a new concept of drug delivery which makes use of cationic magnetic NPs to accelerate drug accumulation into cells). Finally, over the last years, there has risen great interest in the employment of magnetic NPs as contrast agents (Gupta and Gupta 2005a, b; Scherer et al. 2002), such as dextran magnetite in MRI (Magin et al. 1986; Gillis and Koenig 1987). As for biomedical applications the most important properties of magnetic particles are biocompatibility, injectability, non-toxicity and high-level accumulation in target tissues or target organs (Teja and Koh 2009).
On the other hand, nanotechnology, as an emerging technology, raises many issues like concerns about human toxicity and environmental impact (Service 2000; Colvin 2003). In this framework, the Joint Research Centre (JRC) of the European Commission is contributing to the development of testing strategies to assess the safety of NMs. To evaluate the implications deriving from the increasing use of NMs, the foremost needs concern the improvement of NMs characterization, the definition of methodologies to establish exposure to NMs and consequently the evaluation of potential hazards.
Concerning testing strategies, the in vitro approach can be a useful tool to screen and determine the possible toxicological effects related to the NP-cell interaction (Ponti et al. 2009).
One promising technique to quantify NPs interactions with biological systems and cells is the use of radiotracers, because of the high sensitivity, accuracy and precision of radioactivity measurements. The method can be used to quantify the accumulation of NMs in in vivo and in vitro systems (Ponti et al. 2009; Abbas et al. 2010; Ghanem et al. 1993). In general, radioactive NPs can be created by making use of a radioactive precursor at the synthesis stage, or by direct activation of pre-existing NPs. The latter can be performed using neutrons (Abbas et al. 2009; Nagatani et al. 2000) or charged particle beams produced by cyclotrons.
The first aim of this study has been to produce radiolabelled Fe3O4 NPs, by irradiation of the NPs powder with a proton beam and to analyze the NPs before and after the irradiation process, verifying if they would maintain their main characteristics, and then to apply nuclear techniques to quantify their in vitro uptake (Xu et al. 2006; Cai et al. 2010; Zeng and Sun 2008; Simonelli et al. 2011).
We selected two cell lines, Balb/3T3 (immortalized mouse fibroblast cell line) as one of the most sensitive in vitro models for nanotoxicology (Ponti et al. 2010) and Caco-2 (human epithelial colorectal adenocarcinoma cell line) as potential target cells of Fe3O4 NPs in medical applications (Herrera et al. 2006). Quantitative information related to the cellular internalization was useful to select the minimum injection dose, in order to respect the as low as reasonably achievable (ALARA) concept and to follow the clearance process.
Materials and methods
Irradiation of Fe3O4
Characteristics and production routes of the Co radioisotopes produced by proton irradiation of Fe3O4 NPs
Decay mode and mean γ-ray peaks
1: 56Fe (p, 2n)
β+, 1.5 MeV
2: 57Fe (p, 3n)
Main γ-ray 931, 477, 1409 keV
1: 56Fe (p, n)
EC and β+, 1.5 MeV
2: 57Fe (p, 2n)
Main γ-ray 847, 1238, 2598, 1771 keV
3: 58Fe (p, 3n)
1: 57Fe (p, n)
2: 58Fe (p, 2n)
Main γ-ray 122, 136, 14 keV
1: 58Fe (p, n)
EC and β+, 0.5; 1.3 MeV
2: 58mCo decay
Main γ-ray 811 keV
Production of Fe3O4 radioactive NPs
Fe3O4 NPs in powder form have been supplied by Nanostructured and Amorphous Materials (Houston, USA). Their declared compound purity was 99.9%, containing traces of Co (40.2 μg/g) and Ni (21.5 μg/g) (http://www.nanoamor.com/i/u/300402/h/COA/COA_Fe3O4_2652FY.pdf).
An aluminum capsule, cleaned in ultrasonic bath, was loaded with 30 ± 3 mg of Fe3O4 NPs.
The cavity depth of the capsule was optimized on the basis of the 0.94 g/cm3 measured density of the Fe3O4 material. The energy of the proton beam was set to 23 MeV which resulted in a proton energy of 13.9 MeV when entering the NPs powder after having passed the aluminum window of the specimen holder, about 2 mm of cooling water and the aluminum entrance window of the capsule. After the passage of the protons through the Fe3O4 NPs target material, the proton energy was further reduced to about 12.7 MeV. These beam energies were calculated with the transport code SRIM (Ziegler et al. 2008) which enables to estimate the radiation damage caused by the incident proton beam and recoiling atoms. On the basis of the duration of the in vitro uptake experiments, the required minimum amount of radioactivity to achieve the desired detection sensitivity, the achievable radioactivity yield and the natural composition of iron NPs, the Fe3O4 sample was irradiated for 6 h with a beam current of 3 μA in order to obtain the theoretical radioactive 56Co yield of about 3,000 kBq. Radioactivity calculations were performed taking into account uncertainties about the proton initial energy, the nuclear reaction cross section, the mean beam current, the target density and the thickness of the various layers between the proton beam and the NPs powder. Finally the Fe3O4 NPs were analyzed inside the capsule and a radioactive yield of 3,400 kBq for 56Co (corrected at the end of the irradiation) was measured with HPGe (high purity germanium) detectors. After some days of cooling down, the capsule containing the Fe3O4 NPs powder was cleaned in an ultrasonic bath and was opened in a glove box to recover the NPs with some milliliters of deionised water. By comparing the radioactivity values of the whole capsule and the recovered sample, the amount of recovered Fe3O4 could be calculated exactly and then the stock solution concentration (recovered powder in water) could be determined.
Characterization of Fe3O4 NPs
The Fe3O4 NPs were characterized before and after irradiation. A Zetasizer Nano-ZS instrument (Malvern Instruments Ltd, UK) was used to determine the size distribution by dynamic light scattering (DLS) analysis and the ζ-potential. Samples were prepared by dispersion of 10 μg of Fe3O4 NPs in 10 mL of buffer solution for particle size distribution analyses (Phosphate Buffer Solution, pH 7.2, Gibco, Milan, Italy) and 10 mL of de-ionized water. In order to measure the ζ-potential, the NPs powder was recovered with de-ionized water and sonicated for at least 2 h in pulsed mode (Branson Digital Sonifier by Heinemann, Germany). The measurements were then performed at 25 °C in disposable plastic vials (Brand, Germany), with a sample concentration of 125 μg/mL.
The stability of the radiolabels, i.e., their adherence to the NPs, was studied by quantifying 56Co radiolabel release from Fe3O4 NPs in PBS and in the complete culture medium under standard cell culture conditions (37 °C, 5% CO2, 95% humidity, HERAEUS incubator, Germany) after 72-h incubation. The Fe3O4 concentration in both media was 500 μg/mL for an overall radioactivity of about 50 kBq. Free ions of 56Co, i.e., not bound to the NPs, were separated from the samples by ultracentrifugation (105.000×g, 90 min, Optima MAX 130.000 rpm Beckman, USA); then the supernatant was collected and the 56Co radioactivity was measured by γ-ray spectrometry. Moreover, an X-ray diffraction (XRD) study was performed on radioactive and non-radioactive Fe3O4 NPs to give evidence that no phase changes due to radiation or thermal damage occurred.
Cells line and exposure protocols
Caco-2 cells were purchased from Sigma-Aldrich (Salisbury, Great Britain). Balb/3T3 mouse fibroblasts stemming from the clone A31-1-1 were purchased from Hatano Research Institute, Japan. Both cell lines were certified mycoplasma free. The cultures for uptake experiments were prepared from deep-frozen stock vials and always kept in a sub-confluent state. Both Caco-2 and Balb/3T3 cells were maintained in complete culture medium (Dulbecco’s Modified Eagle Medium High Glucose, 10% fetal bovine serum, 5% penicillin/streptomycin, 4 mM l-Glutamine, 100 U/mL non essential aminoacid solution for Caco-2 cells; Modified Eagle Medium low glucose, 10% fetal bovine serum, 5% penicillin/streptomycin, Invitrogen, Milan, Italy for Balb/3T3 cells) in standard cell culture conditions.
As for the uptake experiments, on the first day 106 cells were seeded in 5 mL of complete culture medium in 75 cm2 flasks (Corning, Costar). After 24 h the cells were treated with suspensions of 5 mL culture medium containing selected concentrations of radiolabelled Fe3O4 NPs for exposure times of 72 h using exposure concentrations of 6, 30, and 60 μg/mL. Afterwards, the cells were washed twice with 5 mL of PBS (Gibco, Milan, Italy), detached with 1 mL of trypsin (Invitrogen, Milan, Italy) and harvested with 4 mL of complete culture medium. The cells were stained with a 1/10 dilution cell suspension in trypan blue (Sigma-Aldrich, Milan Italy) and counted using a Bürker chamber (Germany). In addition, the trypan blue exclusion assay was used as a fast and direct method to verify the Fe3O4 NPs potential toxicity of radioactive and non-radioactive NPs compared to the untreated controls.
Starting from the stock solution concentration, there were prepared samples of 6, 30, and 60 μg/mL whose concentrations were checked by γ-spectrometry measurements. Samples whose radioactivity content turned out to be higher than 20% over the expected concentration were reformulated. The fact that most of the fractions were obtained at the first attempt is evidence of the stock solution homogeneity in terms of radiolabelling.
All samples were measured at least with two different detectors calibrated with standard sources. The calibration sources were certified and purchased by ENEA (Italy), DAMRI, and CERCA (France). γ-Spectrometry analysis software packages of Canberra (Genie 2000 software) and EG&G Ortec (Gamma Vision Software) were used. In order to perform high accuracy measurements, various parameters (such as geometry of the samples, geometry of the calibration sources and distance from the Al-cup of the germanium detector) were changed according to radioactivity concentration and detector dead time so as to maintain measurement uncertainties below 12%.
The two cell lines were treated with the established NPs concentrations for 72 h. After 3 days, as described in the exposure protocol, from each selected starting concentration, four samples were generated: the “cellular pellet”, that is the cellular fraction, the “End of Treatment” (EoT), that is the recovered culture medium at the end of the 72-h exposure, the first and the second wash of the pellet fraction to remove the leftover culture medium. The four generated samples were measured to determine cellular uptake as precisely as possible. The experiments were replicated at least four times for each of the three concentrations and results have been expressed as pg/cell, while the radioactive content of each fraction has been presented as percentage referring to the starting concentration, the so-called Start of Treatment (SoT).
Results and discussion
Leaching tests were performed to quantify the ionic fraction of 56Co released in NP suspensions both in PBS and in culture medium (for 72 h at 37 °C). 500 μg of Fe3O4 NPs was diluted in 1 mL of solution. This quantity was selected only for practical reasons as an easily detectable radioactivity of 5 Bq should correspond to a 56Co ions leaching of 0.01% from the NPs.
After leaching and subsequent ultracentrifugation, in case of PBS solution, no 56Co was found in the liquid fraction; while in the culture medium supernatant, 175 Bq could be detected, corresponding to a leaching of 0.35% of the 56Co ions from the NPs. This can be considered as negligible.
Fe3O4 uptake results on the Balb/3T3 and the Caco-2 cell lines expressed as pg/cell for the three exposure concentrations
Expected concentration (μg/mL)
Balb/3T3 cell line
Caco-2 cell line
Experimental concentration (μg/mL)
Experimental concentration (μg/mL)
5.9 ± 1.2
2.8 ± 0.4
6.1 ± 0.8
1.2 ± 0.5
32.6 ± 5.5
13.3 ± 4.4
32.8 ± 3.6
3.1 ± 1.3
64.1 ± 7.4
30.0 ± 3.3
60.9 ± 4.9
5.5 ± 1.8
The uptake analyses for the 6, 30, and 60 μg/mL NPs suspensions show that the uptake rate is related to the concentration tested in both the Balb/3T3 and the Caco-2 cell lines (Fig. 6). Data highlight a Fe3O4 NPs accumulation in mouse fibroblast (Balb/3T3) higher than in the human epithelial colorectal cell line (Caco-2) in all the concentrations tested.
A previous study (Berry et al. 2004) has showed that the mechanism of uptake for uncoated and dextran coated Fe3O4 NPs is most likely via fluid phase endocytosis, with a similar rate of uptake. Transmission electron microscopy study (Gupta and Gupta 2005a, b) shows that Fe3O4 NPs are internalized within the fibroblast 12 h after the start of the treatment. However, more studies will be required to understand the biological mechanism of Fe3O4 NPs interaction and internalization within cells (Osaka et al. 2009).
This study demonstrates that Fe3O4 NPs in dry powder form can be effectively radiolabelled by irradiation with protons because of the favorable combination of high natural abundance of 56Fe and a high yield for the nuclear reaction 56Fe(p, n) 56Co. Thermal effects that might cause aggregation and agglomeration of NPs or morphological changes can sufficiently be suppressed by irradiation of only thin layers of encapsulated NPs powders, which enhances cooling efficiency and limits the beam energy deposition in the target volume. An early determination of the required radioactivity levels and the optimization of irradiation parameters are helpful in this respect. Direct collision radiation damage can be considered insignificant under the present irradiation conditions.
The achieved radioactive yield of irradiated NPs was high enough to perform various uptake experiments with one batch of radioactive NPs. DLS and ζ-potential measurements, as well as XRD analysis, before and after irradiation, strongly support the conclusion that the irradiation process did not alter the physico-chemical properties of the NPs. Concerning both buffer solution and culture medium, we have shown that radioactivity was retained in the NPs structure, and consequently no radiolabel release occurred after irradiation.
A special advantage of the applied nuclear detection technique consists of the possibility of monitoring individual elements without any additional sample preparation due to the γ-ray emission specificity of the radioisotopes involved.
There have been presented results of two complete uptake studies with 56Co-labelled Fe3O4 NPs on two different cell lines. It thence appears that the levels of Fe3O4 NPs uptake in mouse fibroblast (Balb/3T3) and human epithelial colorectal adenocarcinoma (Caco-2) cell lines are closely dependent on the concentration of the NPs the cells are exposed to. Data of the two cell lines tested show a semi-linear dose-uptake response. Owing to the high resolution γ-ray spectrometry measurements, it was possible to precisely establish the concentration levels of the exposure and to determine the concentrations of NPs in the cells in response to this exposure with very limited effort in sample preparation, thus minimizing possible sources of measurement errors and other artifacts.