Introduction

Recently, the performance of X-ray computed tomography (CT) and magnetic resonance imaging (MRI) method has been greatly improved, particularly in terms of their spatial resolution and technology for reconstructing the acquired images. Nuclear medicine imaging has been considered to be the most sensitive approach for diagnosing bone disorders such as bone metastases due to its ability to enable the early detection of abnormalities, namely, visualization of lesion sites before anatomical changes. For a long time, 99mTc-methylenediphosphonate (99mTc-MDP) and 99mTc-hydroxymethylenediphosphonate (99mTc-HMDP) have been widely used in bone imaging1,2,3,4,5. Because 99mTc has the convenient physical characteristics [moderate half-life (6.01ā€‰h) for clinical use, a generator-produced radionuclide, and appropriate gamma ray energy for imaging] and imaging methods using conventional gamma cameras are simple. 99mTc-MDP and 99mTc-HMDP are complexes of 99mTc with bisphosphonate analogs having high affinity for bone since the phosphate groups in the bisphosphonate can be coordinated with calcium in hydroxyapatite crystals in bone.

The use of [18F]NaF for bone imaging was initially reported by Blau et al. in 19626 and approved by the US Food and Drug Administration in 1972. [18F]NaF accumulates at a high level in bone because of chemisorption with the exchange of fluoride anions with the hydroxyl groups in hydroxyapatite [Ca10(PO4)6(OH)2]. However, [18F]NaF had not been widely used due to its limited availability and high cost, but it has recently been reevaluated. The images obtained using clinical positron emission tomography (PET) generally have high spatial resolution and PET/CT scanners have become widely available commercially. Although Even-Sapir et al. reported that [18F]NaF PET imaging is significantly more sensitive than 99mTc-MDP planar and 99mTc-MDP single photon emission computed tomography (SPECT) imaging7, the problems of limited availability and the high cost of cyclotrons have remained unresolved.

In recent years, 68Ga (T 1/2ā€‰=ā€‰68ā€‰min) has drawn substantial attention as a positron emission radionuclide for clinical PET because of its attractive radiophysical properties, such as reasonable half-life for clinical use; it has particularly been used as a generator-produced radionuclide. 68Ga-PET does not require an on-site cyclotron because 68Ga can be eluted from the generator on demand. Moreover, as the parent nuclide, 68Ge (T 1/2ā€‰=ā€‰271 days) has a long half-life, a generator could be used for a long period. Therefore, the demand for 68Ga-labeled compounds for the diagnosis of bone disorders, such as bone metastases, has increased. Some new radiogallium-labeled complexes for bone imaging have been developed in recent years8,9,10,11,12,13,14. Bisphosphonate analogs are used as carriers in these radiogallium-labeled complexes. For example, Fellner et al. reported that 68Ga-DOTA-conjugated bisphosphonate, 68Ga-BPAMD, showed high uptake in osteoblastic metastatic lesions in a first human PET study15. In addition, Suzuki et al. reported that 68Ga-NOTA-conjugated bisphosphonate, 68Ga-NOTA-BP, showed high bone affinity and rapid blood clearance in animal experiments10.

The acidic amino acid peptides (poly-glutamic and poly-aspartic acids) also have a high affinity for hydroxyapatite because side-chain carboxyl groups in the acidic amino acid peptides can be coordinated with calcium in hydroxyapatite, and could become carriers delivering drugs to bone16,17,18. Recently, 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) has been used as a chelating site, and Ga-DOTA-conjugated aspartic acid peptides [Ga-DOTA-(L-Asp)n], with varying peptide lengths (nā€‰=ā€‰2, 5, 8, 11, or 14), have been developed and evaluated using the easy-to-handle radioisotope 67Ga, which has a longer half-life (3.3 days), rather than 68Ga19. 67Ga-DOTA-(L-Asp)11 and 67Ga-DOTA-(L-Asp)14 show high affinity for hydroxyapatite, high accumulation in bone, and rapid blood clearance in biodistribution experiments in normal mice. Accordingly, the bone/blood ratios of 67Ga-DOTA-(L-Asp)11 and 67Ga-DOTA-(L-Asp)14 are comparable to those of 99mTc-HMDP and 67Ga-DOTA-Bn-SCN-HBP (Fig.Ā 1A), a Ga-DOTA-conjugated bisphosphonate, which was developed and evaluated in our previous study11. In these Ga-DOTA-conjugated aspartic acid peptide compounds, L-aspartic acid is used as the only component of the peptides. However, the peptides composed of D-amino acids could be more stable in vivo than the peptides built with L-amino acids because they are not readily recognized by the peptidases20. Thus, in this study, 67Ga-DOTA-(D-Asp)n (Fig.Ā 1B) of varying peptide lengths (nā€‰=ā€‰2, 5, 8, 11, or 14) were synthesized and evaluated. Moreover, to compare the different acidic amino acids as components of the carrier, 67Ga-DOTA-(L-Glu)14 (Fig.Ā 1C) and 67Ga-DOTA-(D-Glu)14 (Fig.Ā 1D) were synthesized and evaluated in vitro and in vivo.

Figure 1
figure 1

Chemical structures of (A) Ga-DOTA-Bn-SCN-HBP, (B) Ga-DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), (C) Ga-DOTA-(L-Glu)14, and (D) Ga-DOTA-(D-Glu)14.

Results

Preparation of 67Ga-DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), 67Ga-DOTA-(L-Glu)14, and 67Ga-DOTA-(D-Glu)14

67Ga-DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), 67Ga-DOTA-(L-Glu)14, and 67Ga-DOTA-(D-Glu)14 were prepared by complexing DOTA-(D-Asp)n, DOTA-(L-Glu)14, and DOTA-(D-Glu)14 with 67Ga, respectively. Radiochemical yields of 67Ga-DOTA-(D-Asp)2, 67Ga-DOTA-(D-Asp)5, 67Ga-DOTA-(D-Asp)8, 67Ga-DOTA-(D-Asp)11, 67Ga-DOTA-(D-Asp)14, 67Ga-DOTA-(L-Glu)14, and 67Ga-DOTA-(D-Glu)14 were 25%, 67%, 74%, 56%, 51%, 38%, and 68% respectively. After RP-HPLC purification, 67Ga-DOTA-(D-Asp)n, 67Ga-DOTA-(L-Glu)14, and 67Ga-DOTA-(D-Glu)14 had radiochemical purities of over 95%. The formation of 67Ga-DOTA-(D-Asp)n, 67Ga-DOTA-(L-Glu)14, and 67Ga-DOTA-(D-Glu)14 complexes were determined by examining the retention times in RP-HPLC analyses. The 67Ga-labeled complexes showed identical retention times as the corresponding nonradioactive complexes. The results indicated that the formation of 67Ga-labeled complexes were identical to those of nonradioactive Ga complexes, which were determined by MS.

Hydroxyapatite-binding assay

FigureĀ 2 shows the percentage of each 67Ga-DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), 67Ga-DOTA-(L-Glu)14, and 67Ga-DOTA-(D-Glu)14 bound to hydroxyapatite beads. Binding of each 67Ga-DOTA-(D-Asp)n to the beads increased with an increasing amount of hydroxyapatite, except for that of 67Ga-DOTA-(D-Asp)2. Binding of 67Ga-DOTA-(D-Asp)n to hydroxyapatite tended to increase with increasing length of amino acid chain. The binding affinities of 67Ga-DOTA-(L-Glu)14 and 67Ga-DOTA-(D-Glu)14 were comparable to that of 67Ga-DOTA-(D-Asp)14.

Figure 2
figure 2

Binding ratios of 67Ga-DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), 67Ga-DOTA-(L-Glu)14, and 67Ga-DOTA-(D-Glu)14 to hydroxyapatite beads. Data are shown as the meanā€‰Ā±ā€‰SD for four samples.

Biodistribution experiments

The biodistribution of 67Ga-DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), 67Ga-DOTA-(L-Glu)14, 67Ga-DOTA-(D-Glu)14, [18F]NaF, and 99mTc-MDP in normal mice is shown in TablesĀ 1ā€“9. Among these compounds, 67Ga-DOTA-(D-Asp)8, 67Ga-DOTA-(D-Asp)11, 67Ga-DOTA-(D-Asp)14, 67Ga-DOTA-(L-Glu)14, 67Ga-DOTA-(D-Glu)14, [18F]NaF, and 99mTc-MDP showed high accumulation and retention of radioactivity in bone. 67Ga-DOTA-(D-Asp)5 showed moderate accumulation of radioactivity in bone; however, the level of radioactivity decreased 3ā€‰h after injection. 67Ga-DOTA-(D-Asp)2 caused subtle accumulation of radioactivity in bone. Although there was little radioactivity in other tissues at 3ā€‰h after the injection of 67Ga-DOTA-(D-Asp)n, 99mTc-MDP, and [18F]NaF because of rapid excretion via the kidneys, the radioactivity in the kidneys after the injection of 67Ga-DOTA-(L-Glu)14 and 67Ga-DOTA-(D-Glu)14 was retained.

Table 1 Biodistribution of radioactivity after i.v. injection of 67Ga-DOTA-(D-Asp)2 in micea.
Table 2 Biodistribution of radioactivity after i.v. injection of 67Ga-DOTA-(D-Asp)5 in micea.
Table 3 Biodistribution of radioactivity after i.v. injection of 67Ga-DOTA-(D-Asp)8 in micea.
Table 4 Biodistribution of radioactivity after i.v. injection of 67Ga-DOTA-(D-Asp)11 in micea.
Table 5 Biodistribution of radioactivity after i.v. injection of 67Ga-DOTA-(D-Asp)14 in micea.
Table 6 Biodistribution of radioactivity after i.v. injection of 67Ga-DOTA-(L-Glu)14 in micea.
Table 7 Biodistribution of radioactivity after i.v. injection of 67Ga-DOTA-(D-Glu)14 in micea.
Table 8 Biodistribution of radioactivity after i.v. injection of [18F]NaF in micea.
Table 9 Biodistribution of radioactivity after i.v. injection of 99mTc-MDP in micea.

Urine Analyses

The results of urine analysis using RP-HPLC after injection of 67Ga-DOTA-(L-Asp)14 and 67Ga-DOTA-(D-Asp)14 showed that a part of these complexes metabolized to more hydrophilic complexes; some radioactivity was eluted earlier than the intact complex. The ratio of the intact complex after injection of 67Ga-DOTA-(D-Asp)14 (85.8ā€‰Ā±ā€‰17.4%) was significantly higher than that of 67Ga-DOTA-(L-Asp)14 (55.0ā€‰Ā±ā€‰13.9%).

Discussion

It has been shown that the bisphosphonate structure is very useful as a carrier of physiologically active molecules or compounds with medicinal properties. This is particularly true for bone lesions because of the high affinity of bisphosphonate for hydroxyapatite, which is plentiful in bone but not in soft tissues21,22. Stable radiometal complex-conjugated bisphosphonate compounds have been designed as bone-seeking radiopharmaceuticals; they have been synthesized and evaluated for the diagnosis and therapy of bone metastases11,23,24,25,26,27,28,29,30. The available data show that bisphosphonate is an excellent carrier of radioisotopes to bone lesions. Our recent study has shown that L-aspartic acid peptides could also work as carriers of radioisotopes to bone lesions; L-aspartic acid peptides have high affinity for hydroxyapatite19,31. Thus, we assumed that D-aspartic acid peptides might be even better carriers. They should have a similar degree of affinity for hydroxyapatite but higher stability in vivo than the L-aspartic acid compounds.

In the hydroxyapatite-binding assay, 67Ga-DOTA-(D-Asp)n with a longer amino acid chain showed higher affinity for hydroxyapatite than the short-chain compounds. The binding patterns of 67Ga-DOTA-(D-Asp)n were almost the same as those of 67Ga-DOTA-(L-Asp)n 19. A previous study reported that the dissociation constants and the maximal binding rates of Fmoc-peptide compounds for hydroxyapatite show no significant differences among Fmoc-(L-Asp)n, Fmoc-(D-Asp)n, and Fmoc-(L-Glu)n (nā€‰=ā€‰2, 4, 6, 8, 10)32. This is consistent with the results of hydroxyapatite binding assay in our study. We found that aspartic acid peptides had the same degree of affinity for hydroxyapatite regardless of their optical isomeric form. Moreover, there were no differences between the affinities of aspartic acid peptides and glutamic acid peptides for hydroxyapatite.

In in vivo studies, it is known that the peptides that composed of D-amino acids are more stable than the L-amino acid peptides20. A study examining the Fmoc compounds reported that, after a single i.v. administration, the plasma concentration of Fmoc-(L-Asp)6 decreased more rapidly than the concentration of Fmoc-(D-Asp)6. Degradation products did not appear in the plasma after the injection of Fmoc-(D-Asp)6, but Fmoc-(L-Asp)4 and Fmoc-(L-Asp)2 were detected in plasma after the injection of Fmoc-(L-Asp)6 32. Therefore, we had expected to observe increased accumulation in bone after the injection of 67Ga-DOTA-(D-Asp)n, caused by their superior in vivo stability. In urine analyses, 67Ga-DOTA-(L-Asp)14 metabolized to more hydrophilic complexes, which should be 67Ga-DOTA conjugated with shorter aspartic acid peptides, because of the cleavage of an amide bond in the peptide. These compounds were diluted before the full-length compound during the RP-HPLC using an ODS column. This indicates that 67Ga-DOTA-(D-Asp)14 is more stable than 67Ga-DOTA-(L-Asp)14. Since 67Ga-DOTA conjugated with shorter aspartic acid peptides should show lower accumulation in bone than 67Ga-DOTA conjugated with long aspartic acid peptides, we expected that 67Ga-DOTA-(D-Asp)n, which has higher stability, would show higher accumulation in bone than 67Ga-DOTA-(L-Asp)n. However, against our expectations, the accumulation of radioactivity in bone was comparable for 67Ga-DOTA-(L-Asp)n and 67Ga-DOTA-(D-Asp)n. Not only 67Ga-DOTA-(D-Asp)n but also 67Ga-DOTA-(L-Asp)n immediately accumulated in bone or was excreted into urine via the kidneys with little degradation; both molecule types showed extremely rapid clearance from the blood. There was no difference between the biodistributions of 67Ga-DOTA-(L-Asp)n and 67Ga-DOTA-(D-Asp)n.

To compare the biodistributions of 67Ga-DOTA-conjugated acidic amino acid peptides with the biodistributions of other typical bone-seeking compounds, biodistribution experiments of 99mTc-MDP and [18F]NaF were performed. 67Ga-DOTA-(D-Asp)11, 67Ga-DOTA-(D-Asp)14, 99mTc-MDP, and [18F]NaF showed excellent biodistribution as bone imaging agents, such as high bone accumulation and low radioactivity in non-target tissues. Among these agents, as [18F]NaF showed the highest bone uptake, [18F]NaF may have the most preferable biodistribution as a bone imaging agent. However, the bone/non-target tissue radioactivity ratios of 99mTc-MDP and 67Ga-DOTA-(D-Asp)n (nā€‰=ā€‰11 or 14) are sufficient for bone imaging, and 99mTc and 68Ga have some convenient physical properties as radionuclides. Thus, 99mTc-MDP and 68Ga-DOTA-(D-Asp)n (nā€‰=ā€‰11 or 14) should be useful in a clinical context.

The 67Ga-DOTA-conjugated L-glutamic acid peptide, 67Ga-DOTA-(L-Glu)14, and the 67Ga-DOTA-conjugated D-glutamic acid peptide, 67Ga-DOTA-(D-Glu)14, also showed rapid clearance from the blood and high accumulation in bone, similarly to 67Ga-DOTA-(L-Asp)14 and 67Ga-DOTA-(D-Asp)14. Generally, radiometal-labeled peptides tend to show a high accumulation of radioactivity in the kidneys. It has been reported that the accumulation of radioactivity in the kidneys after the injection of 111In-labeled peptides is affected by their molecular charges33,34. As the renal brush border membrane is negatively charged, a repulsive force could arise between this membrane and negatively charged compounds. Such repulsive force could inhibit the reabsorption of these compounds into renal proximal tubular cells. The introduction of negative charges into radiometal-labeled peptides has also been studied to develop a method of decreasing the accumulation of radioactivity in the kidneys35. The extremely low accumulation of radioactivity in the kidneys after the injection of 67Ga-DOTA-(L-Asp)14 and 67Ga-DOTA-(D-Asp)14 may have been caused by their negative charges. We had expected that 67Ga-DOTA-(L-Glu)14 and 67Ga-DOTA-(D-Glu)14, being negatively charged like 67Ga-DOTA-(L-Asp)14 and 67Ga-DOTA-(D-Asp)14, would also cause low accumulation of radioactivity in the kidneys. However, contrary to our expectations, high accumulation and retention or slower clearance of radioactivity in the kidneys were observed after the injection of 67Ga-DOTA-(L-Glu)14 or 67Ga-DOTA-(D-Glu)14. The mechanism behind these phenomena are unclear, but we must conclude that the glutamic acid peptides are not appropriate as carriers to the bone in the nuclear medicine imaging because of their association with high radioactivity in the kidneys.

In this study, no differences in the biodistributions between L-aspartic acid [67Ga-DOTA-(L-Asp)n] and D-aspartic acid [67Ga-DOTA-(D-Asp)n] compounds were observed, presumably because of their extremely rapid blood clearance. Recently, we have proposed a new concept of using a bifunctional peptide containing an aspartic acid peptide linker as a carrier to bone metastases and an RGD peptide, which has high affinity for Ī±vĪ²3 integrin, as a carrier to primary cancer31. In this compound, L-aspartic acid is used as a composite component of the aspartic acid peptide linker. A D-aspartic acid peptide linker may be effective in the new approach. Higher stability of the D-aspartic acid peptide linker should be effective for higher accumulation in target tissues because the blood clearance of bifunctional peptide does not occur as rapidly as that of 67Ga-DOTA-(D-Asp)n. Further studies are needed to examine the effectiveness of a D-aspartic acid peptide linker in the drug design concept.

Methods

Materials

Electrospray ionization mass (ESI-MS) analyses were performed with a LCQ (Thermo Fisher Scientific, Waltham, MA, USA). Matrix assisted laser desorption/ionization-time of flight mass (MALDI-TOF-MS) analyses were performed with ABI 4800 plus (AB SCIEX, Foster, CA, USA). [67Ga]GaCl3 was supplied by Nihon Medi-Physics Co., Ltd. (Tokyo, Japan). [18F]NaF was prepared in Fukui University and transported to Kanazawa University. [99mTc]Pertechnetate (99mTcO4 āˆ’) was eluted in saline solution from generators (Nihon Medi-Physics Co., Ltd). 99mTc-MDP was prepared by the addition of 99mTcO4 āˆ’ solution into the mixture of MDP (Wako Pure Chemical Industries, Ltd., Osaka, Japan), tin(II) chloride, and ascorbic acid solution. 1,4,7,10-Tetraazacyclododecane-1,4,7-tris(t-butyl acetate) (DOTA-tris) was purchased from Macrocyclics (Dallas, TX, USA). 9-Fluorenylmethoxycarbonyl (Fmoc)-D-Asp(OtBu)-Wang resin, Fmoc-D-Asp(OtBu), and Fmoc-L-Glu(OtBu) were purchased from Merck KGaA (Darmstadt, Germany). Fmoc-L-Glu(OtBu)-Wang resin and 2-chlorotrityl chloride resin were purchased from Watanabe chemical Industries, LTD. (Hiroshima, Japan). Fmoc-D-Glu(OtBu) was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). Other reagents were of reagent grade and used as received.

Synthesis of DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14)

The protected peptidyl resin was manually constructed by an Fmoc-based solid-phase methodology using a method described previously19. After the construction of the peptide chain on the resin, the Fmoc protecting group was removed using 20% piperidine in dimethylformamide (DMF), and a mixture containing two equivalents of DOTA-tris, 1,3-diisopropylcarbodiimide (DIPCDI), and 1-hydroxybenzotriazole hydrate (HOBt) in dimethylformamide (DMF) was added and allowed to react for 2ā€‰h. For the cleavage of peptides from the resin and deprotection, 0.5ā€‰mL of thioanisole and 5ā€‰mL of trifluoroacetic acid (TFA) were added to the completely protected peptide resin at 0ā€‰Ā°C. After stirring at room temperature for 2ā€‰h, the resin was removed by filtration, and ether was added to the filtrate at 0ā€‰Ā°C to precipitate crude peptide. The crude products were purified by reversed-phase (RP)-HPLC using a Hydrosphere 5C18 column (10ā€‰Ć—ā€‰150 mm; YMC, Kyoto, Japan) at a flow rate of 4ā€‰mL/min with an isocratic mobile phase of water containing 0.1% TFA [in the case of DOTA-(D-Asp)2] or using a Cosmosil 5C18-AR 300 column (10ā€‰Ć—ā€‰150 mm; Nacalai Tesque, Kyoto, Japan) at a flow rate of 4ā€‰mL/min with a gradient mobile phase from water containing 0.1% TFA to 20% methanol in water containing 0.1% TFA for 20ā€‰min [in the case of DOTA-(D-Asp)n (nā€‰=ā€‰5, 8, 11, or 14)]. UV Chromatograms (220ā€‰nm) were obtained. The fraction containing DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14) was determined by mass spectrometry and collected. The solvent removal from the fraction was performed by freeze-drying to provide DOTA-(D-Asp)n as white powder.

DOTA-(D-Asp)2 MS (ESI): m/z 635 (Mā€‰+ā€‰H)+, Yield: 30.4%

DOTA-(D-Asp)5 MS (ESI): m/z 980 (Mā€‰+ā€‰H)+, Yield: 39.8%

DOTA-(D-Asp)8 MS (ESI): m/z 1325 (Mā€‰+ā€‰H)+, Yield: 11.7%

DOTA-(D-Asp)11 MS (ESI): m/z 1670 (Mā€‰+ā€‰H)+, Yield: 12.1%

DOTA-(D-Asp)14 MS (MALDI): m/z 2015 (Mā€‰+ā€‰H)+, Yield: 13.6%

Synthesis of DOTA-(L-Glu)14

A resin-binding protected peptide was constructed by same procedure as mentioned above using Fmoc-L-Glu(OtBu)-Wang resin, Fmoc-L-Glu(OtBu), and tris-DOTA. For the cleavage of peptides from the resin and the deprotection, 0.5ā€‰mL of thioanisole and 5ā€‰mL of TFA were added to the fully protected peptide resin at 0ā€‰Ā°C. After stirring at room temperature for 2ā€‰h, the crude product was purified by RP-HPLC at a flow rate of 4ā€‰mL/min with a gradient mobile phase from water containing 0.1% TFA to 20% methanol in water containing 0.1% TFA for 20ā€‰min. The solvent removal from the fraction was performed by freeze-drying to provide DOTA-(L-Glu)14 and as white powder.

DOTA-(L-Glu)14 MS (ESI): m/z 2212 (Mā€‰+ā€‰H)+, Yield: 14.6%

Synthesis of DOTA-(D-Glu)14

Fmoc-D-Glu(OtBu) (4 molar equivalents to resin) was dissolved in dichloromethane. 2-Chlorotrityl chloride resin and N,N-diisopropylethylamine (DIEA, 3.5 equiv.) were added. The reaction mixture was rotated for 1ā€‰h, and 1ā€‰mL of methanol was added to react further for 30ā€‰min at room temperature. Construction, cleavage, deprotection, and purification of the peptide were performed by the same procedure as mentioned above. DOTA-(D-Glu)14 was obtained as white powder.

DOTA-(D-Glu)14 MS (ESI): m/z 2212 (Mā€‰+ā€‰H)+, Yield: 2.1%

Preparation of Ga-DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), Ga-DOTA-(L-Glu)14, and Ga-DOTA-(D-Glu)14

Ga-DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), Ga-DOTA-(L-Glu)14, and Ga-DOTA-(D-Glu)14 were synthesized using a method described previously19.

Ga-DOTA-(D-Asp)2 MS (ESI): m/z 701 (M)+

Ga-DOTA-(D-Asp)5 MS (ESI): m/z 1046 (M)+

Ga-DOTA-(D-Asp)8 MS (ESI): m/z 1391 (M)+

Ga-DOTA-(D-Asp)11 MS (ESI): m/z 1736 (M)+

Ga-DOTA-(D-Asp)14 MS (MALDI): m/z 2081 (M)+

Ga-DOTA-(L-Glu)14 MS (ESI): m/z 2278 (M)+

Ga-DOTA-(D-Glu)14 MS (ESI): m/z 2278 (M)+

Preparation of 67Ga-DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), 67Ga-DOTA-(L-Glu)14, and 67Ga-DOTA-(D-Glu)14

Approximately 50ā€‰Ī¼g of DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), DOTA-(L-Glu)14 or DOTA-(D-Glu)14 was dissolved in 75ā€‰Ī¼L of 0.2ā€‰M ammonium acetate buffer (pH 5.0), and 25ā€‰Ī¼L of 67GaCl3 solution (1.85 MBq) in 0.01ā€‰M HCl was added and allowed to react at 80ā€‰Ā°C for 8ā€‰min. 67Ga-labeled peptides were purified by RP-HPLC performed using a Hydrosphere 5C18 column (4.6ā€‰Ć—ā€‰250 mm; YMC) at a flow rate of 1ā€‰mL/min with an isocratic mobile phase of water containing 0.1% TFA [in the case of 67Ga-DOTA-(D-Asp)2] or using a Cosmosil 5C18-AR 300 column (4.6ā€‰Ć—ā€‰150 mm) at a flow rate of 1ā€‰mL/min with a gradient mobile phase from water containing 0.1% TFA to 20% methanol in water containing 0.1% TFA for 20ā€‰min [in the case of 67Ga-DOTA-(D-Asp)n (nā€‰=ā€‰5, 8, 11, or 14), 67Ga-DOTA-(L-Glu)14, and 67Ga-DOTA-(D-Glu)14].

Preparation of [18F]NaF

No-carrier-addedĀ [18F]fluoride was produced via theĀ 18O(p,n)18F reaction fromā€‰>ā€‰98% enrichedĀ [18O]water (Taiyo Nippon Sanso Corporation, Tokyo, Japan) on an RDS eclipse RD/HP medical cyclotron (Siemens, Knoxville, TN, USA). [18F]NaF was prepared by elutingĀ [18F]fluoride trapped on an anion exchange column (QMA Plus Light; Waters Corporation, Milford, MA, USA) with saline after washing the anion exchange column with water.

Hydroxyapatite-binding assays

Hydroxyapatite-binding assays were performed in accordance with previously described procedures19,25. In brief, hydroxyapatite beads (Bio-Gel; Bio-Rad, Hercules, CA, USA) were suspended in Tris/HCl-buffered saline (50ā€‰mM, pH 7.4) at 2.5ā€‰mg/mL, 10ā€‰mg/mL, and 25ā€‰mg/mL. For the solutions of 67Ga-DOTA-(D-Asp)n (nā€‰=ā€‰2, 5, 8, 11, or 14), 67Ga-DOTA-(L-Glu)14, or 67Ga-DOTA-(D-Glu)14, ligand concentrations were adjusted to 19.5ā€‰ĀµM by adding DOTA-(D-Asp)n, DOTA-(L-Glu)14, or DOTA-(D-Glu)14. Two hundred microliters of each of 67Ga-DOTA-(D-Asp)n, 67Ga-DOTA-(L-Glu)14, or 67Ga-DOTA-(D-Glu)14 solution was added to 200ā€‰Ī¼L of the hydroxyapatite suspension, and samples were gently shaken for 1ā€‰h at room temperature. After centrifugation at 10,000ā€‰g for 5ā€‰min, a part of the radioactivity in the supernatants was measured using a gamma counter (AccuFLEX Ī³ ARC-7010, Hitachi, Ltd., Tokyo, Japan). Control experiments were performed according to the same procedure without hydroxyapatite beads, which showedā€‰<ā€‰0.1% adsorption of radioactivity to vials. The ratios of binding were determined as follows:

Hydroxyapatite binding (%)ā€‰=ā€‰(1 āˆ’ [sample supernatant radioactivity]/[control supernatant radioactivity])ā€‰Ć—ā€‰100

Biodistribution experiments

Experiments with animals were conducted in strict accordance with the Guidelines for the Care and Use of Laboratory Animals of Kanazawa University. The animal experimental protocols used were approved by the Committee on Animal Experimentation of Kanazawa University (Permit Number: AP-132633). Biodistribution experiments were performed after intravenous administration of each diluted tracer solution (37ā€“740ā€‰kBq/100ā€‰Ī¼L) to 6-week-old male ddY mice (27ā€“32ā€‰g, Japan SLC, Inc., Hamamatsu, Japan). Four or five mice at each time point after the administration of each compound were sacrificed by decapitation at 10, 60, and 180ā€‰min post-injection. Tissues of interest were taken and weighed. Complete left femurs were isolated as representative bone samples, radioactivity was determined using gamma counters (AccuFLEX Ī³ ARC-8001 in the case of 18F, Hitachi, Ltd.), and background counts and physical decay were corrected during counting.

Urine Analyses

67Ga-DOTA-(L-Asp)14 was prepared according to a method described previously19. 67Ga-DOTA-(L-Asp)14 or 67Ga-DOTA-(D-Asp)14 solution (370ā€‰kBq / 200ā€‰Ī¼L) was intravenously injected to 6-week-old male ddY mice. At 1ā€‰h post-injection, mice were sacrificed and their urea samples were taken from the bladders. After ultrafiltration (Microcon-30, Merck KGaA), the filtrate samples were analyzed by RP-HPLC at a flow rate of 1ā€‰mL/min with a gradient mobile phase from water containing 0.1% TFA to 20% methanol in water containing 0.1% TFA for 20ā€‰min.