Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

# Non-invasive imaging of radiocesium dynamics in a living animal using a positron-emitting 127Cs tracer

## Abstract

Visualizing the dynamics of cesium (Cs) is desirable to understand the impact of radiocesium when accidentally ingested or inhaled by humans. However, visualization of radiocesium in vivo is currently limited to plants. Herein, we describe a method for the production and purification of 127Cs and its use in visualizing Cs dynamics in a living animal. The positron-emitting nuclide 127Cs was produced using the 127I (α, 4n) 127Cs reaction, which was induced by irradiation of sodium iodide with a 4He2+ beam from a cyclotron. We excluded sodium ions by using a material that specifically adsorbs Cs as a purification column and successfully eluted 127Cs by flowing a solution of ammonium sulfate into the column. We injected the purified 127Cs tracer solution into living rats and the dynamics of Cs were visualized using positron emission tomography; the distributional images showed the same tendency as the results of previous studies using disruptive methods. Thus, this method is useful for the non-invasive investigation of radiocesium in a living animal.

## Introduction

The accident at Tokyo Electric Power Company’s Fukushima Daiichi Nuclear Power Station in March 2011 caused various radioactive materials to fall over a wide area1. In particular, 137Cs, which has a half-life of 30 years, had a serious impact on the affected area2,3. Subsequently, interest in the dynamics of radiocesium in animals, plants, and humans has increased. There are concerns about where in the body radiocesium ingested by humans accumulates, and how quickly it is excreted.

To more accurately calculate the internal exposure dose of radiocesium, it is necessary to analyze the dynamics of radiocesium in animals in more detail. The current calculation of internal exposure doses of radiocesium is based on data from animal studies, and human urine and feces. More than 90% of ingested soluble cesium (Cs) is absorbed in the gastrointestinal tract4, but the real-time behavior and kinetics immediately after ingestion remain unclear. For example, spherical Cs-bearing particles (Cs-balls) were emitted in the Fukushima nuclear accident5, and the dynamics of such Cs-balls in the human body are unknown.

Visualizing the dynamics of radiocesium in vivo is a powerful method for studying the dynamics of radiocesium. In the field of plant science, there have been reports of visualizing the dynamics of radiocesium in living plants. Kobayashi et al. have succeeded in visualizing the β rays of 137Cs using a real-time radioisotope imaging system6. Kurita et al. have succeeded in visualizing the dynamics of 137Cs by imaging Cherenkov light emitted from β rays with an ultra-sensitive CCD camera7. Although these methods are suitable for the visualization of radiocesium in plants with a small tissue thickness, they cannot visualize the dynamics of radiocesium in animals, which have a large tissue thickness. Kawachi et al. successfully visualized the 661 keV γ rays of 137Cs with a pinhole-type gamma camera8. However, this technique provides two-dimensional images, which are sufficient for plant research but are not suitable for animal research.

Positron imaging is widely used to non-destructively visualize the dynamics of positron-emitting radionuclides in vivo. The most common use of positron imaging is cancer screening by positron emission tomography (PET), which exploits the tendency of fluorodeoxyglucose labeled with the positron-emitting radionuclide 18F to accumulate in cancer cells. Additionally, PET has been used in medical research to analyze the kinetics of drugs labeled with 11C, 13N, and 15O. As for the positron-emitting nuclide of Cs, McKee et al. reported that 132Cs (half-life of 6.48 d, β+ intensity of 0.43%) was produced and administered to animals to verify the applicability of 132Cs as a tumor marker9. However, because 132Cs was produced by irradiating stable Cs, 132Cs could not be purified, resulting in low specific activity. Therefore, a 132Cs tracer is not suitable for non-invasive imaging of radiocesium in vivo where the tracer must exhibit high specific activity.

In this study, the positron-emitting nuclide 127Cs (half-life of 6.25 h, β+ intensity of 3.04%) without stable Cs was produced by irradiating iodine with a helium ion beam accelerated by a cyclotron. Then, we developed a unique purification method to adsorb and elute only 127Cs using a material that specifically collects Cs. Additionally, by administering the purified 127Cs tracer to living animals, we verified that 127Cs is useful for visualization of the short-term dynamics of radiocesium.

## Results and discussion

### Production of 127Cs

127Cs was produced by the 127I (α, 4n)127Cs reaction. A sodium iodide (NaI) target plate was irradiated with a 4He2+ beam from an AVF cyclotron at Takasaki Ion Accelerators for Advanced Radiation Application (TIARA). The incident energy on the target was set to 55 MeV with reference to the cross-section from the experimental data of Sathik et al.10 and data from the ACSELAM library11 (Supplementary Fig. 1). Under these irradiation conditions, 129Cs was also produced by the 127I (α, 2n)129Cs reaction. Supplemental Table 1 lists nuclides that were potentially generated.

Figure 1A shows the gamma spectrum of the mother solution, in which the irradiated NaI target was dissolved. Under the irradiation conditions in this study, the production yields of 127Cs and 129Cs at the end of irradiation were 80.1 ± 6.5 and 1.3 ± 0.04 MBq/µAh, respectively (n = 6). Additionally, 24Na, 124I, and 126I were also produced as by-products, which could be used effectively as radioactive tracers for sodium and iodide ions from the NaI target in the subsequent purification experiments. The decay data from the radionuclides produced in this study are summarized in Supplementaary Table 2. In the gamma-ray spectrum, the other peaks could not be identified, with the exception of 127Xe (daughter nuclide of 127Cs) among the nuclides listed in Supplementary Table 1.

### Purification of 127Cs tracer

The procedure for the separation of 127Cs from the irradiated NaI target is shown schematically in Fig. 2. Table 1 shows the radioactivity of the nuclides in each purification step when the irradiation condition was 0.5 µA × 2 h. First, we loaded the mother solution on to an anion-exchange column. The results showed that 124I and 126I were not detected in the effluent solution from the anion-exchange column, which indicated that iodide ions derived from the NaI target were removed.

Figure 3B shows the process of 127Cs elution from the Cs adsorbent column. First, to minimize re-adsorption of 127Cs to the Cs adsorbent, the column was turned upside down. Then, 10 mL of 1 mM (NH4)2SO4 was flowed from the left vial to the right vial via the column in the direction from the bottom to the top over 30 min. In the last 10 min, the column was turned upside down and the solution in the column was pushed out with air.

To verify the contamination of 18F into the final 127Cs tracer solution, we calculated the decay time of the peak counts of annihilation gamma ray (511 keV) with the data obtained by measuring the final solution every 2 h with the Ge detector (Fig. 4). By one-component fitting, we calculated that the half-life of the peak count was 6.279 ± 0.015 h, which is almost consistent with the half-life of 127Cs. Furthermore, using two-component fitting (127Cs and 18F) of this decay curve, the percentage of the 18F component in this peak count at the start time was calculated as 0.663 ± 1.273%. These results indicate that the contamination of 18F into the final 127Cs tracer solution was extremely low. Additionally, the estimate indicates that the radionuclidic purity of 127Cs in the 36 h after the end of irradiation was more than 99%.

The principle of Cs adsorption and elution from the Cs adsorbent can be described as follows: Cs is adsorbed by a substitution reaction with ammonia from the AMP in the column structure, and then the adsorbed Cs is replaced by ammonia from contact with the ammonium solution.

Figure 5 shows the elution rates of Cs from the adsorbent with various concentrations of different ammonium compounds in a batch experiment using non-radioactive Cs. We selected an eluate solution using two criteria to make the final tracer solution suitable for physiological study for both animal and tracer experiments in general biology, including plant experiments: the ion concentration should be as low as possible, and the pH should be close to neutral. Also, it was desired that the elution rate be at least 50% or more for efficient purification. When the ion concentration was limited to 1 mM, (NH4)2CO3 and (NH4)2SO4 were identified as candidates. The elution rate of 1 mM (NH4)2CO3 was higher than that of (NH4)2SO4, but the pH values were 9.2 and 5.7, respectively. Therefore, we used a solution of 1 mM (NH4)2SO4, for eluting 127Cs from the Cs absorbent column in the purification experiment.

To verify the desorption of AMP particles from the adsorbent, the molybdenum concentration of the solution eluted with 1 mM ammonium sulfate was measured and the desorption rate was calculated as 0.0086%. This indicates that the possibility of contamination of AMP particle into the final tracer solution is quite low when this adsorbent is used for purification.

### PET imaging

The purified 127Cs tracer solution was injected into the tail vein of a living rat, and the dynamics of Cs were visualized using PET, in duplicate. The doses of 127Cs for Experiments 1 and 2 were 2 and 11 MBq, and the imaging times were 3 and 4 h, respectively. To make the dose as low as reasonably achievable, we administered 2 MBq of 127Cs in Experiment 1. To obtain more clear PET images, we administered 11 MBq of 127Cs in Experiment 2, which we consider to be the more desirable dose. Figure 6A shows a 4-h integrated image of 127Cs in the whole body of the rat in Experiment 2. This result indicated that 127Cs passed or accumulated in the neck, heart, small intestine, and kidney. Figure 6B shows a cross-sectional view of the integrated image of 127Cs in the neck. From comparison with the CT image, the signal for 127Cs was found to be localized in the salivary glands in the neck. The localization pattern of the 127Cs signal was the same in Experiment 1 (Supplementary Fig. 4).

Figure 7 shows the time course of 127Cs PET counts in the whole body and in each organ. All counts were decay corrected at the start time of PET imaging and normalized to the dose of 127Cs and the volume of each region. Although there were differences in the normalized PET counts in the two experiments, the same trend was observed over each time course. The time course of the whole body indicated that the 127Cs tracer injected into the tail vein moved into the field of view of the PET in 30 min. The accumulation of 127Cs in the kidney reached a peak at 20–30 min and was rapidly excreted. In the salivary glands and heart, the accumulation of 127Cs reached a peak at 30–60 min and was slowly excreted. In contrast, although the molecular number of 127Cs administered in Experiments 1 and 2 differed five-fold, the kinetics of 127Cs exhibited the same trend and no saturation was observed even with the five-fold difference in molecular number, which reflects no region of specific affinity of 127Cs in rats. Moreover, our method can produce no-carrier-added 127Cs and no influence of injected radioactivity is expected.

Kaikkonen et al. have measured the radioactivity using a gamma counter in different tissues of goats 30 min after administrating 134Cs intravenously14. The results showed that the highest relative concentration of 134Cs was found in the kidney cortex, followed by the salivary gland, cardiac muscle, and small intestine. Sasaki et al. have reported that cattle fed with 134Cs and 137Cs for several weeks had more radiocesium accumulation in the kidney than in other organs15. Although the tested animal species and methods of radiocesium administration are different, these studies are consistent with the results of our study, which suggests that 127Cs can be used to correctly visualize Cs dynamics in the living animal body.

To the best of our knowledge, this is the first study to visualize the dynamics of radiocesium in living animals. In this study, we administered 127Cs to animals by intravenous injection. In the future, it should be possible to non-invasively calculate the transfer rate of radiocesium to each organ by administering 127Cs through ingestion or inhalation. The calculated results will contribute to the construction of more accurate models of Cs dynamics and improvements in the accuracy of effective dose coefficients. Furthermore, because of the short half-life of 127Cs (6.25 h), repeated experiments using the same animal are possible. For example, the effects of various types of internal decontamination agents could be evaluated accurately.

## Methods

### Production of 127Cs

127Cs was produced by the 127I (α, 4n)127Cs reaction. A target plate with a diameter of 10 mm and a thickness of 0.5 mm was made by pressing 125 mg of NaI reagent (Fujifilm Wako Pure Chemicals, Tokyo, Japan). The NaI target plate was irradiated with a 4He2+ beam for 2–3.25 h, at a beam current of 0.5 µA and an incident energy of 55 MeV, delivered from an AVF cyclotron at TIARA.

### Column purification of 127Cs tracer

After the irradiation, the NaI target was dissolved in 1 mL of H2O. To remove iodide ions, the solution was loaded onto an anion-exchange column (Poly-Prep #731-6211; Bio-Rad, Hercules, CA, USA) and an additional 4 mL of H2O was loaded.

The produced radionuclides were characterized by a HPGe detector (crystal diameter of 58 mm and length of 67.3 mm) connected to a multichannel analyzer (SEIKO 7800 MCA). The radioactivity was determined by considering the γ-ray energy at 462.3 (127Cs), 371.9 (129Cs), 1369.6 (24Na), 602.7 (124I), and 666.3 keV (126I).

### Decay fitting analysis

The half-life of the radioactivity in the effluent solution (Supplementary Fig. 3) and the eluted solution (Fig. 4) from the Cs adsorbent column was estimated by fitting the data with the following one-component formula:

$$N = A \times 2^{{ - \frac{{\left( {t - t_{0} } \right)}}{\tau }}} ,$$

where $$t$$ and $$N$$ represent the measurement time and the count, respectively. The start time of the measurement and the count at the start time are represented by $$t_{0}$$ and $$A$$, respectively. The half-life is represented by $$\tau$$. Two parameters, $$A$$ and $$\tau$$, were estimated by fitting the data ($$t$$ and $$N$$) with the formula with fixing $$t_{0}$$ at the experimental value of the start time.

The effect of the 18F contamination in the eluted solution from the Cs adsorbent was also evaluated, by fitting the data with the following two-component formula:

$$N = A_{{{\text{Cs}}}} \times 2^{{ - \frac{{\left( {t - t_{0} } \right)}}{{\tau_{{{\text{Cs}}}} }}}} + A_{{\text{F}}} \times 2^{{ - \frac{{\left( {t - t_{0} } \right)}}{{\tau_{{\text{F}}} }}}} ,$$

where $$A_{{{\text{Cs}}}}$$ and $$A_{{\text{F}}}$$ and $$\tau_{{{\text{Cs}}}}$$ and $$\tau_{{\text{F}}}$$ represent the counts at the start time and the half-lives for Cs and F, respectively. Two parameters, $$A_{{{\text{Cs}}}}$$ and $$A_{{\text{F}}}$$, were estimated by fitting the data with the formula by fixing $$\tau_{{{\text{Cs}}}}$$ and $$\tau_{{\text{F}}}$$ at 6.25 h and 109.8 min, respectively, and then fixing the start time $$t_{0}$$ at 0.

### Cs elution rate analysis

The Cs-adsorbed adsorbents were prepared as follows. The fibrous Cs adsorbent was cut into 0.5-cm-diameter disk (8.3 mg). The adsorbent disk was placed in 10 mL of stable Cs solution with a concentration of 7.5 nM (1 ppb). After shaking for 4 h at 25 °C, the Cs adsorbent disk was removed from the Cs solution and the concentration of the remaining Cs was measured using inductively coupled plasma-mass spectrometry (7700 × ICP-MS, Agilent Technologies, Santa Clara, CA, USA). The calculated adsorption rate of Cs to the adsorbent was 84% ± 7% (n = 40).

The elution was carried out with aqueous solutions of (NH4)3C6H5O7, NH4HCO3, (NH4)2CO3, (NH4)2SO4, NH4HCO2, NH4NO3, CH3COONH4, and NH4Cl for the batch test. The concentrations of the ammonium compound solutions were adjusted to 0.5, 1.0, 2.0, 5.0, and 10 mM. The adsorbent disk was soaked in 5 mL of aqueous solutions of each ammonium compound, and was continuously shaken for 20 h at 25 °C. After elution, the concentration of Cs in the ammonium solutions was measured using ICP-MS. The elution rate of Cs was calculated by the following equation:

$${\text{Elution}}\,{\text{rate}}\, ({\text{\% }}) = \left( {C_{E} /C_{A} } \right) \times 100,$$

where CA and CE are the amount of adsorbed Cs in the adsorbent and eluted Cs in the ammonium solutions, respectively.

### PET imaging

PET imaging experiments using two living rats were performed (Experiments 1 and 2). The eluted solution containing 127Cs was evaporated to dryness in a glass vial and transported from TIARA to Tohoku University. Approximately 300 µL of physiological saline solution (154 mM NaCl) was added to the glass vial, and the 127Cs was recovered. The final concentration of (NH4)2SO4 was 33 mM. A tracer solution containing 2 or 11 MBq of 127Cs was injected to a 6-week-old male Wistar rats (132 or 136 g) via tail vein under inhalation anesthesia with 2% isoflurane in air. The distribution of 127Cs in the living rat was imaged using a small-animal PET instrument (Clairvivo PET16; Shimadzu, Japan) for 3 or 4 h. The tracer was injected 15.3 or 14.2 h after the end of irradiation. A list-mode DRAMA algorithm17 was used for image reconstruction with an iteration of 1 and sub-iterations of 100. The PET image had a matrix size of 128 × 128 × 213 and each voxel size was 0.7 mm3. No attenuation correction was performed, but we did conduct scatter correction using emission scatter convolution subtraction18. 18F solution was used for scanner calibration. The energy window for the scanner was set to 250–750 keV. A CT scan of the rat was also performed using a small-animal CT instrument (Clairvivo CT; Shimadzu) before the injection of the 127Cs tracer solution. Quantitative analysis of the PET images was performed with Amide software19.

### Ethics approval

All experimental procedures conformed to "Regulations for Animal Experiments and Related Activities at Tohoku University", and were reviewed by the Institutional Laboratory Animal Care and Use Committee of Tohoku University, and finally approved by the President of University. Ethical approval in this study has been obtained from the Institutional Laboratory Animal Care and Use Committee of Tohoku University.

## References

1. 1.

Endo, S. et al. Measurement of soil contamination by radionuclides due to the Fukushima Dai-ichi Nuclear Power Plant accident and associated estimated cumulative external dose estimation. J. Environ. Radioact. 111, 18–27. https://doi.org/10.1016/j.jenvrad.2011.11.006 (2012).

2. 2.

Yasunari, T. J. et al. Cesium-137 deposition and contamination of Japanese soils due to the Fukushima nuclear accident. Proc. Natl. Acad. Sci. 108, 19530–19534. https://doi.org/10.1073/pnas.1112058108 (2011).

3. 3.

Hirose, K. 2011 Fukushima Dai-ichi nuclear power plant accident: Summary of regional radioactive deposition monitoring results. J. Environ. Radioact. 111, 13–17. https://doi.org/10.1016/j.jenvrad.2011.09.003 (2012).

4. 4.

4ICRP. ICRP Publication 56: Age-dependent Doses to Members of the Public from Intake of Radionuclides. Vol. 56 (Elsevier Health Sciences, Amsterdam, 1990).

5. 5.

Adachi, K., Kajino, M., Zaizen, Y. & Igarashi, Y. Emission of spherical cesium-bearing particles from an early stage of the Fukushima nuclear accident. Sci. Rep. 3, 2554. https://doi.org/10.1038/srep02554 (2013).

6. 6.

6Kobayashi, N. I. In Agricultural Implications of the Fukushima Nuclear Accident (eds Tomoko, M. N.& Keitaro, T.) 37–48 (Springer Japan, Tokyo, 2013).

7. 7.

Kurita, K. et al. Development of a Cherenkov light imaging system for studying the dynamics of radiocesium in plants. J. Nucl. Sci. Technol. 54, 662–667. https://doi.org/10.1080/00223131.2017.1299051 (2017).

8. 8.

Kawachi, N. et al. Imaging of radiocesium uptake dynamics in a plant body by using a newly developed high-resolution gamma camera. J. Environ. Radioact. 151(Part 2), 461–467. https://doi.org/10.1016/j.jenvrad.2015.04.009 (2016).

9. 9.

McKee, J. S. et al. Dynamic studies of positron-emitting putative tumor marker 132Cs in mice show differential tumor and regional uptake. Cancer Detect. Prev. 8, 87–93 (1985).

10. 10.

Sathik, N. P. M., Ansari, M., Singh, B. P. & Prasad, R. Measurement and analysis of excitation functions for alpha induced reactions on iodine and cesium. Pramana 47, 401–410. https://doi.org/10.1007/BF02847827 (1996).

11. 11.

Table of Isotope Production Cross Sections (ACSELAM Library). https://wwwndc.jaea.go.jp/ftpnd/sae/acl.html. Accessed Aug 2020.

12. 12.

Shibata, T. et al. Evaluation of a cesium adsorbent grafted with ammonium 12-molybdophosphate. Radiat. Phys. Chem. 119, 247–252. https://doi.org/10.1016/j.radphyschem.2015.10.026 (2016).

13. 13.

Krasnov, N. N., Dmitriev, P. P., Dmitrieva, Z. P., Konstantinov, I. O. & Molin, G. A. Yield of F18 in the bombardment of sodium, magnesium, and aluminum by He3 ions and sodium by α-particles. Soviet. At. Energy 28, 333–334. https://doi.org/10.1007/bf01403935 (1970).

14. 14.

Kaikkonen, M., De Gritz, B. & Eriksson, L. Short-term distribution of 134Cs in relation to 51Cr-EDTA after intravenous dose in goats. Acta Physiol. Scand. 183, 321–332. https://doi.org/10.1111/j.1365-201X.2004.01391.x (2005).

15. 15.

Sasaki, K. et al. Radiocesium distribution in the tissues of Japanese black beef heifers fed fallout-contaminated roughage due to the Fukushima Daiichi nuclear power station accident. Biosci. Biotechnol. Biochem. 76, 1596–1599. https://doi.org/10.1271/bbb.120424 (2012).

16. 16.

Sato, K. et al. Performance evaluation of the small-animal PET scanner ClairvivoPET using NEMA NU 4-2008 Standards. Phys. Med. Biol. 61, 696–711. https://doi.org/10.1088/0031-9155/61/2/696 (2015).

17. 17.

Kinouchi, S. et al. GPU implementation of list-mode DRAMA for real-time OpenPET image reconstruction. IEEE Nucl. Sci. Symp. Med. Imaging Conf. https://doi.org/10.1109/NSSMIC.2010.5874188 (2010).

18. 18.

Nai, Y.-H., Ose, T., Shidahara, M. & Watabe, H. 137Cs transmission imaging and segmented attenuation corrections in a small animal PET scanner. Radiol. Phys. Technol. 10, 321–330. https://doi.org/10.1007/s12194-017-0407-4 (2017).

19. 19.

Loening, A. M. & Gambhir, S. S. AMIDE: A free software tool for multimodality medical image analysis. Mol. Imaging 2, 15353500200303132. https://doi.org/10.1162/15353500200303133 (2003).

## Acknowledgements

We thank Mr. Koji Imai (Beam Operation Co., Ltd.) for assistance in irradiation for 127Cs production. Victoria Muir, PhD, from Edanz Group (https://en-author-services.edanzgroup.com/) edited a draft of this manuscript.

## Funding

This work was supported by JSPS KAKENHI (Grant numbers 16K06962 and 19H04296).

## Author information

Authors

### Contributions

N.Su., Y.-G.Y., S.F., and N.K. conceived and designed the research. N.Su., Y.-G.Y. and K.K. conducted the 127Cs purification experiments. T.S., H.H. and N.Se. produced the fibrous Cs adsorbent and conducted the Cs elution rate analysis. Y.F. and H.W. conducted the PET imaging experiments. N.Su., K.K., H.W. and N.K. analyzed the PET imaging data. M.Y. conducted the decay fitting analysis. N.Su. wrote the manuscript, which was reviewed by the other authors.

### Corresponding author

Correspondence to Nobuo Suzui.

## Ethics declarations

### Competing interests

The authors declare no competing interests.

### Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

## Rights and permissions

Reprints and Permissions

Suzui, N., Shibata, T., Yin, YG. et al. Non-invasive imaging of radiocesium dynamics in a living animal using a positron-emitting 127Cs tracer. Sci Rep 10, 16155 (2020). https://doi.org/10.1038/s41598-020-73351-2

• Accepted:

• Published: