Development of Technetium-99m-Labeled BODIPY-Based Probes Targeting Lipid Droplets Toward the Diagnosis of Hyperlipidemia-Related Diseases

Hyperlipidemia causes systemic lipid disorder, which leads to hepatic steatosis and atherosclerosis. Thus, it is necessary to detect these syndromes early and precisely to improve prognosis. In the affected regions, abnormal formation and growth of lipid droplets is observed; therefore, lipid droplets may be a suitable target for the diagnosis of hyperlipidemia-related syndromes. In this study, we designed and synthesized [99mTc]Tc-BOD and [99mTc]Tc-MBOD composed of one technetium-99m and two BODIPY scaffolds with hydroxamamide (Ham) or N-methylated hydroxamamide (MHam) in radiochemical yields of 54 and 35%, respectively, with a radiochemical purity of over 95%. [99mTc]Tc-BOD showed significantly higher accumulation levels in foam cells than in non-foam cells (foam cells: 213.8 ± 64.8, non-foam cell: 126.2 ± 26.9 %dose/mg protein, p < 0.05) 2 h after incubation. In contrast, [99mTc]Tc-MBOD showed similar accumulation levels in foam cells and non-foam cells (foam cells: 92.2 ± 23.3, non-foam cell: 83.8 ± 19.8 %dose/mg protein). In normal mice, [99mTc]Tc-BOD exhibited gradual blood clearance (0.5 h: 4.98 ± 0.35, 6 h: 1.94 ± 0.12 %ID/g) and relatively high accumulation in the liver 6 h after administration (15.22 ± 1.72 %ID/g). Therefore, [99mTc]Tc-BOD may have potential as an imaging probe for detecting lipid droplets in disease lesions of hyperlipidemia.


Introduction
Hyperlipidemia or dyslipidemia is a chronic lipid disorder in which the lipid content in blood, such as cholesterol, increases abnormally [1]. This disease causes systemic lipid disorder, with manifestations such as fatty infiltration of the liver and lipid deposition in vessels, which leads to hepatic steatosis and atherosclerosis [2,3]. Such hyperlipidemia-related syndromes are risk factors for severe diseases, such as liver dysfunction, liver cancer, and cerebral and myocardial infarction [4,5]. Therefore, it is necessary to detect these syndromes early and precisely to improve prognosis.
Lipid droplets are an intracellular organelle where neutral lipids of cells are stored [6,7]. In hepatic steatosis and atherosclerosis lesions, the abnormal formation and growth of lipid droplets is observed, and this is suspected to play a key role in the progression of these syndromes [8,9]. Thus, abnormally enlarged lipid droplets may be a suitable target for the diagnosis of hyperlipidemia-related syndromes. To detect lipid droplets, several hydrophobic dyes, including fluorescent probes such as Nile red and boron dipyrromethene (BODIPY), have been reported for histological or microscopic visualization [10,11]. Among these dyes, BODIPY493/503 demonstrated specific staining of lipid droplets and has been examined as an in vitro staining agent for the detection of lipid droplets in living cells [12]. For non-invasive in vivo imaging and clinical diagnosis, imaging modalities, such as positron emission tomography (PET), single photon emission computed tomography (SPECT), and magnetic resonance imaging (MRI), are widely used [13]. Although several MRI probes based on BODIPY scaffolds have been developed for the diagnosis of vulnerable plaques [14,15], no PET or SPECT imaging probe targeting lipid droplets has been reported to exist. Therefore, we aimed to develop novel nuclear medical imaging probes based on BODIPY scaffolds for the visualization of lipid droplet regions in hyperlipidemia-related syndrome. Technetium-99m ( 99m Tc) is one of the major radioisotopes for clinical SPECT imaging because the gamma rays emitted from 99m Tc with a photon energy of 141-keV are readily detectable by the SPECT scanner, and 99m Tc is easily produced using a molybdenum-99/technetium-99m generator. To label the BODIPY scaffold with 99m Tc, we selected two types of chelators: hydroxamamide (Ham) and N-methylated hydroxamamide (MHam). Ham is a thiol-free chelating agent that produces the 99m Tc-complex with a high radiochemical yield under moderate reaction conditions [16,17]. In addition, Ham forms the 99m Tc-complex with a metal-to-ligand ratio of 1:2, which makes it possible to easily produce 99m Tc-labeled compounds with a bivalent targeting scaffold [18,19]. MHam is also a thiol-free chelating agent with similar properties to Ham, and the 99m Tc-complex with MHam exhibited higher stability in murine plasma than that with Ham [20,21].
Molecules 2019, 24, x FOR PEER REVIEW 2 of 10 related syndromes. To detect lipid droplets, several hydrophobic dyes, including fluorescent probes such as Nile red and boron dipyrromethene (BODIPY), have been reported for histological or microscopic visualization [10,11]. Among these dyes, BODIPY493/503 demonstrated specific staining of lipid droplets and has been examined as an in vitro staining agent for the detection of lipid droplets in living cells [12]. For non-invasive in vivo imaging and clinical diagnosis, imaging modalities, such as positron emission tomography (PET), single photon emission computed tomography (SPECT), and magnetic resonance imaging (MRI), are widely used [13]. Although several MRI probes based on BODIPY scaffolds have been developed for the diagnosis of vulnerable plaques [14,15], no PET or SPECT imaging probe targeting lipid droplets has been reported to exist. Therefore, we aimed to develop novel nuclear medical imaging probes based on BODIPY scaffolds for the visualization of lipid droplet regions in hyperlipidemia-related syndrome. Technetium-99m ( 99m Tc) is one of the major radioisotopes for clinical SPECT imaging because the gamma rays emitted from 99m Tc with a photon energy of 141-keV are readily detectable by the SPECT scanner, and 99m Tc is easily produced using a molybdenum-99/technetium-99m generator. To label the BODIPY scaffold with 99m Tc, we selected two types of chelators: hydroxamamide (Ham) and Nmethylated hydroxamamide (MHam). Ham is a thiol-free chelating agent that produces the 99m Tccomplex with a high radiochemical yield under moderate reaction conditions [16,17]. In addition, Ham forms the 99m Tc-complex with a metal-to-ligand ratio of 1:2, which makes it possible to easily produce 99m Tc-labeled compounds with a bivalent targeting scaffold [18,19]. MHam is also a thiolfree chelating agent with similar properties to Ham, and the 99m Tc-complex with MHam exhibited higher stability in murine plasma than that with Ham [20,21].

In Vitro Cellular Uptake Study
The cellular uptake of [ 99m Tc]Tc-BOD and [ 99m Tc]Tc-MBOD is expressed as the %dose/mg protein (Figure 2A), and foam cell formation was confirmed using Oil red O staining ( Figure 2B,C). The cellular uptake of [ 99m Tc]Tc-BOD into foam cells was significantly higher than that into non-foam cells (foam cell: 213.8 ± 64.8, non-foam cell: 126.2 ± 26.9 %dose/mg protein, p < 0.05). On the other hand, the difference between the uptake of [ 99m Tc]Tc-MBOD into foam cells and non-foam cells was not significant (foam cell: 92.2 ± 23.3, non-foam cell: 83.8 ± 19.8 %dose/mg protein) ( Figure 2A).

In Vitro Cellular Uptake Study
The cellular uptake of [ 99m Tc]Tc-BOD and [ 99m Tc]Tc-MBOD is expressed as the %dose/mg protein (Figure 2A), and foam cell formation was confirmed using Oil red O staining ( Figure 2B,C). The cellular uptake of [ 99m Tc]Tc-BOD into foam cells was significantly higher than that into non-foam cells (foam cell: 213.8 ± 64.8, non-foam cell: 126.2 ± 26.9 %dose/mg protein, p < 0.05). On the other hand, the difference between the uptake of [ 99m Tc]Tc-MBOD into foam cells and non-foam cells was not significant (foam cell: 92.2 ± 23.3, non-foam cell: 83.8 ± 19.8 %dose/mg protein) ( Figure 2A).

In Vitro Cellular Uptake Study
The cellular uptake of [ 99m Tc]Tc-BOD and [ 99m Tc]Tc-MBOD is expressed as the %dose/mg protein (Figure 2A), and foam cell formation was confirmed using Oil red O staining ( Figure 2B,C). The cellular uptake of [ 99m Tc]Tc-BOD into foam cells was significantly higher than that into non-foam cells (foam cell: 213.8 ± 64.8, non-foam cell: 126.2 ± 26.9 %dose/mg protein, p < 0.05). On the other hand, the difference between the uptake of [ 99m Tc]Tc-MBOD into foam cells and non-foam cells was not significant (foam cell: 92.2 ± 23.3, non-foam cell: 83.8 ± 19.8 %dose/mg protein) ( Figure 2A).

Stability in Murine Plasma
The in vitro stability of [ 99m Tc]Tc-BOD was evaluated by incubation in murine plasma at 37 • C for 6 h. [ 99m Tc]Tc-BOD had a radiochemical purity of over 80% in murine plasma 6 h after incubation ( Figure 3).

Stability in Murine Plasma
The in vitro stability of [ 99m Tc]Tc-BOD was evaluated by incubation in murine plasma at 37 °C for 6 h. [ 99m Tc]Tc-BOD had a radiochemical purity of over 80% in murine plasma 6 h after incubation ( Figure 3).

Stability in Murine Plasma
The in vitro stability of [ 99m Tc]Tc-BOD was evaluated by incubation in murine plasma at 37 °C for 6 h. [ 99m Tc]Tc-BOD had a radiochemical purity of over 80% in murine plasma 6 h after incubation ( Figure 3).

Discussion
In this study, we developed novel 99m Tc labeled probes, [ 99m Tc]Tc-BOD and [ 99m Tc]Tc-MBOD, for the purpose of imaging of lipid droplets. Ham-BODIPY (2) and MHam-BODIPY (5), the precursors of [ 99m Tc]Tc-BOD and [ 99m Tc]Tc-MBOD, respectively, were synthesized with total yields of 9.7 and 1.0%, respectively (Scheme 1). [ 99m Tc]Tc-BOD and [ 99m Tc]Tc-MBOD were obtained under moderate conditions (at room temperature for 10 min) with radiochemical yields of 54 and 35%, respectively. In general, the chelation reactions of technetium-99m and hydroxamamide/N-methylated hydroxamamide have been evaluated only by confirming the disappearance of the peak derived from [ 99m Tc]TcO 4 − in HPLC analysis. This is because it remains challenging to synthesize the analogues of the rhenium hydroxamamide/N-methylated hydroxamamide complexes [18,19]. In this study, when the reaction mixture of Ham-BODIPY (2) was analyzed using RP-HPLC, two 99m Tc-labeled compounds were found at retention times of 17.7 and 20.6 min ([ 99m Tc]Tc-BOD), which were different from the retention time of [ 99m Tc]NaTcO 4 (Supplementary Figure S1E). This result was consistent with previous reports on 99m Tc-Ham complexes, suggesting that two isomers of [ 99m Tc]Tc-BOD were produced by the 99m Tc complexation reaction using Ham ( Figure 1) [18]. There is no report stating that BODIPY itself is a ligand or substrate for specific receptors or enzymes, and it is suspected to be delivered to lipid droplets via its lipophilicity. Thus, we considered the two isomers of [ 99m Tc]Tc-BOD not to have different properties for targeting lipid droplets, and we used a mixture of both isomers for the in vitro and in vivo evaluations of [ 99m Tc]Tc-BOD. On the other hand, analysis of the 99m Tc complexation reaction with MHam-BODIPY (5) by RP-HPLC revealed a single radioactive peak at a retention time of 20.0 min ([ 99m Tc]Tc-MBOD). This suggested that N-methylation of Ham is a useful strategy to obtain a single 99m Tc-MHam complex by preventing isomerism, as described in a previous report [20].
We next performed an in vitro study to elucidate whether [ 99m Tc]Tc-BOD and [ 99m Tc]Tc-MBOD are able to detect lipid droplets. As lipid-rich model cells, we selected foam macrophages, which are used for studies on atherosclerosis [22,23]. Previously reported optical imaging probes have hydrophobic characteristics to detect lipid droplets [10]. [ 99m Tc]Tc-BOD accumulation in foam cells was significantly higher than that in non-foam cells, suggesting that [ 99m Tc]Tc-BOD may be taken up specifically by cells with lipid droplets. However, [ 99m Tc]Tc-MBOD accumulated similarly in both foam cells and non-foam cells. The reason as to why [ 99m Tc]Tc-MBOD did not accumulate in foam cells specifically is still unclear, but the hydrophobicity of the BODIPY scaffold may not be as necessary for specific accumulation in foam cells.
We next evaluated the distribution of [ 99m Tc]Tc-BOD in normal (ddY) mice. In murine plasma, although the rate of the two [ 99m Tc]Tc-BOD isoforms was changed, [ 99m Tc]Tc-BOD exhibited high stability for 6 h, which is suitable for in vivo imaging ( Figure 3). In normal mice, [ 99m Tc]Tc-BOD levels gradually decreased in the blood, which is suitable for SPECT imaging (Table 1). However, [ 99m Tc]Tc-BOD also had relatively high uptake and retention in the stomach and liver. To visualize hepatic steatosis, [ 99m Tc]Tc-BOD (with its high uptake and retention in the liver) is not suitable. One strategy to solve this problem is to connect two Ham groups with a propylene linker, which was reported to reduce liver retention in a previous report [24]. Regarding the diagnosis of atherosclerosis, some previously reported probes also had high accumulation in the liver [25]. Therefore, [ 99m Tc]Tc-BOD holds potential for imaging lipid droplets in atherosclerotic lesions.
Several kinds of probes labeled with fluorine-18 and iodine-123/125 have been reported as radiolabeled probes with a BODIPY scaffold [26][27][28]. However, fluorine-18 is a cyclotron-produced isotope; this limits the production of fluorine-18-labeled probes because installation of a cyclotron system is expensive and requires a large facility. On the other hand, 99m Tc is easily produced using a molybdenum-99/technetium-99m generator without a large facility, and probes labeled with 99m Tc can be produced via a one-step reaction with 99m Tc. Some BODIPY-based compounds labeled with 99m Tc have been reported [11]. Compared with those compounds, the radiolabeling condition of [ 99m Tc]Tc-BOD is mild (only requiring incubation for 10 min at room temperature), which makes it more convenient for the acquisition of 99m Tc-labeled BODIPY. Thus, [ 99m Tc]Tc-BOD may be useful as a radiolabeled probe with a BODIPY scaffold.
A multimodal imaging technique combining nuclear medical imaging and fluorescence imaging has been developed. It is expected to integrate each advantage, such as a high spatial resolution in fluorescence imaging and a high signal permeability in nuclear medical imaging [29]. The BODIPY scaffold we used as a ligand targeting lipid droplets has suitable properties for fluorescence imaging (Supplementary Table S1 and Figure S2). Therefore, we tried to perform a microscopic fluorescence imaging study. In this study, we used Ham-BODIPY instead of [ 99m Tc]Tc-BOD because it is quite difficult to acquire enough 99m Tc-radiolabeled compound (i.e., approximately a femtomole level) for fluorescence imaging (Supplementary Figure S3). As a result, a high florescent signal was observed in the foam cells treated with Ham-BODIPY (Supplementary Figure S3B), whereas the fluorescent signal was observed not only in the lipid droplets but also in the other cell components. This was also observed in the non-foam cells treated with Ham-BODIPY. Although the chemical properties of [ 99m Tc]Tc-BOD and Ham-BODIPY presumed from the HPLC analysis were different (Figure 1 and Supplementary Figure S1), this result may correlate with the non-specific accumulation of [ 99m Tc]Tc-BOD in the non-foam cells (Figure 2A). Therefore, we are now trying to optimize 99m Tc-labeled BODIPY probes for achieving more specific imaging of lipid droplets.

Radiolabeling
To a solution of 2 (Ham-BODIPY) or 5 (MHam-BODIPY) (1 mg) in an acetic acid/ethanol mixture (1/4, 100 µL), 100 µL of [ 99m Tc]NaTcO 4 solution (37 MBq) and 30 µL of tin(II) tartrate hydrate solution (2 mg of tin(II) tartrate hydrate (7.5 µmol) dissolved in water (2.5 mL)) were added. The reaction mixture was incubated at room temperature for 10 min and purified by RP-HPLC on a Cosmosil 5C 18 -AR-II column (Nacalai Tesque, diameter: 4.6 mm, length: 150 mm, particle size: 5 µm, pore size: 120 Å) using the mobile phase (10 mM phosphate buffer (pH of 7.4)/acetonitrile = 60/40 (0 min) to 30/70 (30 min)) at a flow rate of 1.0 mL/min. To acquire [ 99m Tc]Tc-BOD, the two main peaks (retention time: around 17 to 18 min and around 20 to 21 min) were collected, combined, and then concentrated to remove acetonitrile. The mixture of those two fractions was used as [ 99m Tc]Tc-BOD in the subsequent examination. As for [ 99m Tc]Tc-MBOD, the main peak (retention time: around 20 to 21 min) was collected, then concentrated. The radiochemical yields of [ 99m Tc]Tc-BOD and [ 99m Tc]Tc-MBOD were calculated by dividing the radioactivity of the collected fraction (the mixture of the two peaks in the radiolabeling study of [ 99m Tc]Tc-BOD) by the radioactivity of the cludes measured before HPLC purification. The radiochemical purities were analyzed using RP-HPLC with the same method as the purification procedures.

Preparation of Foam Cells
Animal experiments were conducted in accordance with our institutional guidelines and approved by the Kyoto University Animal Care Committee. ddY mice were purchased from Japan SLC, Inc. (Shizuoka, Japan) and were housed with food and water supplied ad libitum under a 12-h light/12-h dark cycle. To serve as lipid droplet-rich model cells, we prepared mouse foam cells following a previously reported method [22]. In brief, 10% thioglycolate medium (10 mL) was injected intraperitoneally to ddY mice (females, 5 weeks old). After being housed for 4 days, the mice were sacrificed and the peritoneal cavity was washed using an injection of phosphate-buffered saline (PBS) (pH of 7.4) (Nacalai Tesque, Kyoto, Japan) to isolate macrophages. The macrophages were plated on 6-well plates (8 × 10 5 cells/well) and incubated in Dulbecco's modified Eagle's medium (DMEM) (4500 mg/L of Glucose; Nacalai Tesque) supplemented with 10% fetal bovine serum and 100 U/mL of penicillin and streptomycin at 37 • C in an atmosphere containing 5% CO 2 for 24 h. After incubation, the medium was removed and acetyl low density lipoprotein (LDL) (50 µg/mL) was added to the new medium (2 mL). Macrophages were incubated to form foam cells for 48 h. The formation of foam cells was confirmed by Oil red O staining. After removing the medium and washing twice with 1 mL of PBS (pH of 7.4), 2 mL of staining solution (Oil red O 1.8 mg/mL in 2-propanol/water = 3/2) was added to each well, and incubated at room temperature for 20 min. The staining solution was removed, and cells were washed twice with 1 mL of PBS (pH of 7.4). Bright field images were obtained using FSX-100 (Olympus, Tokyo, Japan).

In Vitro Cellular Uptake Study
[ 99m Tc]Tc-BOD (the mixtures of two isomers) or [ 99m Tc]Tc-MBOD (37 kBq, 2 mL in DMEM supplemented with 0.1% DMSO) was added to foam cells or non-foam cells cultured in 6-well plates (n = 6), and the plates were incubated at 37 • C in an atmosphere containing 5% CO 2 for 2 h. After incubation, each well was washed with 1 mL of PBS (pH of 7.4) and the cells were lysed with 1 M NaOH (0.2 mL × 2). The radioactivity of the cells was measured using a γ counter (Wallac 2470 Wizard; PerkinElmer, Waltham, MA, USA). The protein concentration was measured using a bicinchoninic acid protein assay kit (Thermo Fisher Scientific, Waltham, MA, USA).

Stability in Murine Plasma
[ 99m Tc]Tc-BOD (the mixtures of two isomers; 740 kBq) was added to freshly prepared murine plasma (200 µL) collected from ddY mice (male, 5 weeks old). After incubating the solution at 37 • C for 6 h, acetonitrile (200 µL) was added. Subsequently, the solution was centrifuged (4,000 g, 5 min) and the supernatant was analyzed using RP-HPLC. The analytical method of RP-HPLC was the same as that described in the radiolabeling section.

In Vivo Biodistribution
Male ddY mice (5 weeks old) were purchased from Japan SLC. Inc. (Shizuoka, Japan). [ 99m Tc]Tc-BOD (the mixtures of two isomers) (480 kBq, 200 µL in saline supplemented with 0.1% DMSO) was injected into ddY mice via the tail vein. After injection, the mice were sacrificed at 0.5, 1, 3, or 6 h post-injection (n = 5, per time-point). The blood was collected, organs of interest (blood, spleen, pancreas, stomach, intestine, kidney, liver, heart, and lung) were dissected and weighed, and the radioactivity in each organ or tissue was measured using a γ counter (PerkinElmer). The results are expressed as a percentage of the injected dose per gram (%ID/g) for the selected organs, except the stomach whose result is expressed as a percentage of the injected dose (%ID).

Statistics
The student's t-test was used to assess the significance of differences. Differences at the 95% confidence level (p < 0.05) were considered significant.

Conclusions
In this study, we developed [ 99m Tc]Tc-BOD and [ 99m Tc]Tc-MBOD as novel BODIPY-based probes radiolabeled with 99m Tc. [ 99m Tc]Tc-BOD exhibited higher accumulation in foam cells and suitable properties for in vivo imaging. Thus, the [ 99m Tc]Tc-BOD may have potential as an imaging probe to detect lipid droplets.