Preliminary 19F-MRS Study of Tumor Cell Proliferation with 3′-deoxy-3′-fluorothymidine and Its Metabolite (FLT-MP)

The thymidine analogue 3′-deoxy-3′-[18F]fluorothymidine, or [18F]fluorothymidine ([18F]FLT), is used to measure tumor cell proliferation with positron emission tomography (PET) imaging technology in nuclear medicine. FLT is phosphorylated by thymidine kinase 1 (TK1) and then trapped inside cells; it is not incorporated into DNA. Imaging with 18F-radiolabeled FLT is a noninvasive technique to visualize cellular proliferation in tumors. However, it is difficult to distinguish between [18F]FLT and its metabolites by PET imaging, and quantification has not been attempted using current imaging methods. In this study, we successfully acquired in vivo 19F spectra of natural or nonradioactive 3′-deoxy-3′-fluorothymidine ([19F]FLT) and its monophosphate metabolite (FLT-MP) in a tumor xenograft mouse model using 9.4T magnetic resonance imaging (MRI). This preliminary result demonstrates that 19F magnetic resonance spectroscopy (MRS) with FLT is suitable for the in vivo assessment of tumor aggressiveness and for early prediction of treatment response.


Introduction
Tumor cell proliferation is a useful prognostic indicator of tumor aggressiveness, and proliferation may be evaluated to monitor and predict the response to antitumor therapy. Tumor cells and tissues with a high proliferation rate require a high rate of DNA synthesis [1][2][3][4][5]. Radiolabeled thymidine analogues are standard biomarkers for DNA synthesis and are generally used in nuclear medicine. One thymidine analogue, [ 11 C]-labeled thymidine, is well known as a radiotracer for positron emission tomography (PET) studies of tumor cell proliferation and DNA synthesis [6][7][8][9]. However, the short physical half-life (20 min) of [ 11 C]-thymidine and its rapid biodegradation are practical limitations to its use [4,10]. Consequently, the use of [ 18 F]-labeled 3 -deoxy-3fluorothymidine ([ 18 F]FLT) PET imaging to assess proliferation in various tumors has been reported in preclinical and clinical studies [11][12][13].  method, which is only used for in vitro drug screening at early stages. 19 F magnetic resonance imaging ( 19 F MRI) and spectroscopy (MRS) represent a promising in vivo quantitative imaging technique [16][17][18]. The nonradioactive isotope 19 F has a 100% natural abundance with 83% sensitivity of 1 H. The negligible background signal of endogenous 19 F in biological systems provides an extremely high signal-to-nose ratio and exceptional sensitivity, making 19 F MRI/MRS an ideal modality to monitor in vivo biochemical changes, in specific enzyme activity, cell tracking and migration, hypoxia, and quantitative neovascular responses [19,20].
In this study, we monitored TK1 activity by quantifying FLT and FLT-MP in vivo using 19 F MRI/MRS. Our aim was to develop and validate a suitable 19 F MRI/MRS imaging biomarker for cellular proliferation in tumors.

Results and Discussion
To detect the locations of FLT and FLT-MP, we investigated the 19 Figure 2(f) shows the 19 F MRS of the mixture; here, the former was FLT-MP and the latter was FLT. Because the area ratios of the spectra for the former and latter were approximately 60 and 100, respectively, the findings were consistent with the concentrations in the mixture of FLT-MP (60 mM) and FLT (100 mM).
We investigated whether the 19 F NMR or 19 F MRS signal of intracellular FLT-MP, produced as an FLT metabolite, could be detected in vitro. In the first group of cells that were not washed, both FLT and FLT-MP were clearly observed in the 19 F NMR spectra, although the FLT-MP peak was very weak ( Figure S3). However, the signal for FLT in the cells was very strong, and the concentration of FLT was 16.7 mM. In contrast, an FLT-MP peak in the first group of cells was not observed in the 19 F MRS; only an FLT peak was observed ( Figure S4). Figure 3 shows the 19 F NMR spectra of washed cells in the second group as a function of time. Both intracellular FLT and FLT-MP were clearly observed at −175.2 ppm and −174.5 ppm, respectively. Because the extra FLT was washed out, the FLT signal exhibited a moderate level relative to the cells in the first group. Although the extra FLT was washed out, the presence of FLT demonstrated that FLT and its metabolites were reversible in the cell [2]. The amounts of intracellular FLT-MP and FLT were therefore inconsistent over time. A relative ratio of FLT to FLT-MP, here, demonstrated the on-going phosphorylation of different spectra in various tumors unlike PET technology.
No peak was observed in the 19 F MRS signal for intracellular FLT-MP formed in the second group of cells because of its low concentration. These results demonstrated that a typical FLT concentration of 16.7 mM is required for in vivo detection by 19 F MRS. As previous reports, in vivo 19 F MR imaging is generally used for the high concentration of 89 mM due to the low sensitivity of that [21].
To chemically confirm the accuracy of quantitation and metabolite detection by 19 F MRS, an HPLC assay was performed. We, then, investigated the in vivo 19 F MR signals for FLT and FLT-MP. More precisely, we aimed to observe that the appearance of the FLT-MP signal is caused by metabolism of FLT in vivo. Figure 5 [12]. PET technology, high sensitivity, and the radiation of positron-emitting radioisotopes can easily permeate tissues, making PET a powerful molecular imaging modality to monitor the progression of cancer [22]. However, PET alone cannot readily distinguish between [ 18 F]FLT and [ 18 F]FLT-MP. Specifically, it is very difficult to simultaneously identify metabolites in vivo by kinetic analysis of FLT-PET images [8]. In that respect, our results show that 19 F MRS is a noninvasive and practical way to identify biomolecules in vivo, including fluorine atoms; it may, thus, be utilized to complement other imaging tools, such as PET.
MRI/MRS is also a promising molecular imaging method for cancer theragnostics [23,24]. For example, 13 C MRI/MRS study of hyperpolarized [ 13 C]pyruvate and its metabolite ([ 13 C]lactate) could be recently used to measure early responses to therapy, and the utilization of metabolite levels has been studied in clinical practice [25][26][27]. The hyperpolarized 13 C compounds, however, have restriction on the metabolism studies of DNA synthesis due to a time limit of hyperpolarization. The results of the present study, though 4 Contrast Media & Molecular Imaging

Conclusion
In this study, FLT and its metabolite were measured for the first time in an in vivo mouse model using 19

Chemicals.
All reagents were purchased from commercial sources, and the following agents were FLT and FLT-MP (Research Center FutureChem Co., Ltd., Seoul, Korea) and trifluoroacetic acid (TFA) (Sigma-Aldrich, St. Louis, MO).

Cell
Culture. The MCF-7 human breast cancer cell line expressing the HSV-tk gene was maintained in RPMI-1640 medium supplemented with 10% FBS, 1% antibiotics, and 100 g/mL of G418 (Invitrogen). Cultures were maintained in a 37 ∘ C incubator with 5% CO 2 , and the medium was changed every 3 days. For 19 F MRS, MCF-7 cells were plated, and 5 × 10 7 cells were suspended in 500 L of serum-free RPMI medium containing FLT (16.7 mM) before being incubated at 37 ∘ C for different time periods (5 min, 30 min, 60 min, and 120 min). The cells were divided into two groups. The first group of cells was not washed and was used for 19 F NMR and 19 F MRS. The second group of cells was washed three times with PBS, scraped from the plate, centrifuged at 1,000 rpm for 3 min, and then collected for use in 19 F NMR, 19 F MRS, and HPLC analyses.
For HPLC, the pellets were resuspended in PBS to a final volume of 1 mL and were then lysed by three cycles of freezing and thawing; the lysates were centrifuged at 14,000 rpm for 5 min at 4 ∘ C. The supernatant was used for HPLC analysis.
FLT and FLT-MP were extracted from the samples after growth for 5 min, 30 min, 60 min, or 120 min by three cycles 8  of freezing and thawing. After centrifugation (14,000 rpm for 5 min at 4 ∘ C), the samples comprising a 90 : 10 mixture of supernatant: D 2 O, were placed in 5 mm nuclear magnetic resonance (NMR) tubes for data acquisition.

NMR.
The 19 F NMR measurements were conducted at 28 ∘ C on a Bruker 400-MHz NMR spectrometer, equipped with a 5-mm BBFO probe. The experimental parameters were as follows: pulse angle, 90 ∘ (18.32 sec); repetition rate, 1 sec; 172 K data set; 2,000 scans. All 19 F data were processed using TopSpin and analyzed with MestReNova. 1 H MR images were acquired with a fast spin echo sequence using the following settings: repetition time (TR) 2500 ms, echo time (TE) 25 ms, matrix 256 × 256, field of view (FOV) 5 × 5 cm 2 , slice thickness 2 mm, gap 0 mm, averages 2, and scan time 2 min 45 sec. 19 F MR images of phantom were acquired with a gradient echo sequence using the following settings: TR 100 ms, TE 4.0 ms, matrix 64 × 64, FOV 5 × 5 cm 2 , slice thickness 2 mm, gap 0 mm, averages 1200, and scan time 2 h 8 min. 19 F MRS of phantoms and in vivo data were acquired with Point-REsolved Spectroscopy (PRESS) using the following settings: TR 3000 ms, TE 15 ms, voxel size 5 × 5 × 5 mm 3 , averages 512, and scan time 25 min.

Conflicts of Interest
The authors declare that they have no conflicts of interest.