Trace Metal Impurities Effects on the Formation of [⁶⁴Cu]Cu-diacetyl-bis(N⁴-methylthiosemicarbazone) ([⁶⁴Cu]Cu-ATSM)

Abstract

[⁶⁴Cu]Cu-diacetyl-bis(N⁴-methylthiosemicarbazone) ([⁶⁴Cu]Cu-ATSM) is a radioactive hypoxia-targeting therapeutic agent being investigated in clinical trials for malignant brain tumors. For the quality management of [⁶⁴Cu]Cu-ATSM, understanding trace metal impurities’ effects on the chelate formation of ⁶⁴Cu and ATSM is important. In this study, we conducted coordination chemistry studies on metal–ATSM complexes. First, the effects of nonradioactive metal ions (Cu²⁺, Ni²⁺, Zn²⁺, and Fe²⁺) on the formation of [⁶⁴Cu]Cu-ATSM were evaluated. When the amount of Cu²⁺ or Ni²⁺ added was 1.2 mol or 288 mol, equivalent to ATSM, the labeling yield of [⁶⁴Cu]Cu-ATSM fell below 90%. Little effect was observed even when excess amounts of Zn²⁺ or Fe²⁺ were added to the ATSM. Second, these metals were reacted with ATSM, and chelate formation was measured using ultraviolet–visible (UV-Vis) absorption spectra. UV-Vis spectra showed a rapid formation of Cu²⁺ and the ATSM complex upon mixing. The rate of chelate formation by Ni²⁺ and ATSM was lower than that by Cu-ATSM. Zn²⁺ and Fe²⁺ showed much slower reactions with the ATSM than Ni²⁺. Trace amounts of Ni²⁺, Zn²⁺, and Fe²⁺ showed little effect on [⁶⁴Cu]Cu-ATSM’ quality, while the concentration of impurity Cu²⁺ must be controlled. These results can provide process management tools for radiopharmaceuticals.

1. Introduction

Hypoxia has been implicated in the poor prognosis of malignant tumors, such as recurrent brain tumors [1,2,3,4]. Extensive hypoxic areas are associated with a poor prognosis in patients, as malignant factors are promoted in hypoxic cancer cells [2,3,4,5]. Therefore, new tumor therapeutics targeting hypoxia have extensively been investigated to improve patient outcomes. We developed a promising radioactive hypoxia-targeting theranostic agent, [⁶⁴Cu]Cu-diacetyl-bis(N⁴-methylthiosemicarbazone) ([⁶⁴Cu]Cu-ATSM). Clinical studies on patients with recurrent brain tumors are underway.

Cu-ATSM (Figure 1) is labeled with several copper isotopes, such as ⁶⁰Cu, ⁶¹Cu, ⁶²Cu, and ⁶⁴Cu, for positron emission tomography (PET) imaging under hypoxia [6,7,8,9,10,11,12,13,14]. Cu-ATSM rapidly permeates cells, is reduced, and is trapped within cells under hypoxia [15,16,17,18,19]. In hypoxic cells, the levels of the biological reductant NAD(P)H and the activity of NAD(P)H-dependent reductive enzymes are upregulated, and highly reduced conditions lead to Cu-ATSM retention in the cells [16,20,21,22]. To develop hypoxia-targeting radiotherapeutics, ⁶⁴Cu was chosen to label ATSM. ⁶⁴Cu emits β+ for PET imaging (0.655 MeV, 17.8%), β^- (0.574 MeV, 40%), and particularly, Auger electrons bearing a high potential for killing cancer cells [21].

In our previous study [22], using a xenograft mouse model of glioblastoma, multiple doses of [⁶⁴Cu]Cu-ATSM significantly prolonged survival, but no toxicity was reported [22]. We stabilized the therapeutic dose of [⁶⁴Cu]Cu-ATSM by adding sodium L-ascorbate to our investigational drug formulation [23]. We also evaluated trace amounts of chemical impurities derived from the degradation of ATSM in our formulation of [⁶⁴Cu]Cu-ATSM [24]. We identified chemical impurities derived from the degradation of ATSM by using liquid chromatography with tandem mass spectrometry (LC-MS/MS). We also assessed their chemical hazards using quantitative structure–activity relationship (QSAR) applications and concluded that the potential risk posed by chemical impurities contained in the therapeutic dose of [⁶⁴Cu]Cu-ATSM is negligible.

Cu-64 was obtained by the ⁶⁴Ni(p,n)⁶⁴Cu reaction using a cyclotron. After the dissolution of the ⁶⁴Ni target with hydrochloric acid, no carrier added (n.c.a.) ⁶⁴Cu was obtained by chemical separation using an anionic exchange resin or cationic exchange resin [25,26,27,28]. Although the ⁶⁴Ni target was carefully purified before irradiation, minimal amounts (10⁰–10² ppb) of impurities such as Cu, Ni, Fe, and Zn remained in the irradiated sample solution. Most metal ions are present as divalent ions in solution and may react competitively with ATSM to form [⁶⁴Cu]Cu-ATSM chelates, thereby inhibiting [⁶⁴Cu]Cu-ATSM’s formation.

Our formulation of the [⁶⁴Cu]Cu-ATSM investigational drug contains 2.5 µg/mL of ATSM and up to 1.5 GBq/mL of ⁶⁴Cu [23]. The molar ratio of ATSM to n.c.a. ⁶⁴Cu in the therapeutic formulation was approximately 60. Therefore, sub-ppm-level contamination of metal ions may negatively affect the radiochemical yield of [⁶⁴Cu]Cu-ATSM. As of date, the effect of trace metal impurities on chelate formation by ⁶⁴Cu has been reported for some ligands by Ferreira et al. (DOTA-DBCB) [29], Boswell et al. (H₂CB-TE2A) [30], and Zeng et al. (EdF-DOTA) [31]; however, that for ATSM remains unclear. Therefore, control of trace metal impurities would be critical for the quality management of ⁶⁴Cu radiopharmaceutical agents. To obtain quantitative insights into the chelate formulations of ATSM and metal ions, we conducted studies on the coordination chemistry of metal–ATSM complexes in this study. First, we investigated the effect of non-radioactive metal ions on the formation of [⁶⁴Cu]Cu-ATSM. Copper, nickel, zinc, and iron were chosen for this experiment because they are possible contaminants in the target system of the cyclotron [32]. Next, these metals were made to react with ATSM, and the chelate formation rate was measured using ultraviolet–visible (UV-Vis) absorption spectra. These results can provide useful process management tools for this promising radiopharmaceutical for the treatment of recurrent brain tumors in future clinical studies.

2. Results

2.1. Experiment 1: Effect of Trace Metal Impurities on the Formation of [⁶⁴Cu]Cu-ATSM Complex

We investigated the radiochemical yield of [⁶⁴Cu]Cu-ATSM when metal solutions (Cu²⁺, Ni²⁺, Zn²⁺, and Fe²⁺) at nine different concentrations (0.0125, 0.025, 0.05, 0.1, 0.2, 0.4, 4, 10, and 100 ppm in 40.5 μL of the reaction mixture) were added during the synthesis. The concentration range was set at approximately 100 times the amount considered to be contaminated based on the literature [26,32]. The results are summarized in Figure 2 and

Table 1. Effect of coexisting transition metal ions (Cu²⁺ and Ni²⁺) on radiochemical yield of [⁶⁴Cu]Cu-ATSM. The contents of metal ions in the reaction mixture, shown as the unit of ppm and mol in parenthesis, are indicated as the sum of added metal ions and metal impurities derived from the target system.

Cu2+			Ni2+
Metal Ion in ppm (mol)	Radiochemical Yield (%)	Ratio (1)	Metal Ion in ppm (mol)	Radiochemical Yield (%)	Ratio
0.073 (1.9 × 10−8)	98.8	0.19	0.0125 (3.4 × 10−9)	99.0	0.036
0.086 (2.2 × 10−8)	97.4	0.22	0.025 (6.9 × 10−9)	98.8	0.072
0.11 (2.9 × 10−8)	96.6	0.28	0.05 (1.4 × 10−8)	97.7	0.14
0.16 (4.1 × 10−8)	94.8	0.41	0.1 (2.8 × 10−8)	98.2	0.29
0.26 (6.7 × 10−8)	91.0	0.66	0.2 (5.5 × 10−8)	98.4	0.58
0.46 (1.2 × 10−7)	88.8	1.2	0.4 (1.1 × 10−7)	98.2	1.2
4.1 (1.0 × 10−6)	25.0	10	4 (1.1 × 10−6)	98.3	12
10 (2.6 × 10−6)	8.3	26	10 (2.8 × 10−6)	94.9	29
100 (2.6 × 10−5)	1.0	255	100 (2.8 × 10−5)	85.9	288

⁽¹⁾ Metal ion (mol)/ATSM (mol).

and

Table 2. Effect of coexisting transition metal ions (Zn²⁺ and Fe²⁺) on radiochemical yield of [⁶⁴Cu]Cu-ATSM. The contents of metal ions in the reaction mixture, shown as the unit of ppm and mol in parenthesis, are indicated as the sum of added metal ions and metal impurities derived from the target system.

Zn2+			Fe2+
Metal Ion in ppm (mol)	Radiochemical Yield (%)	Ratio	Metal Ion in ppm (mol)	Radiochemical Yield (%)	Ratio
0.064 (1.6 × 10−8)	93.5	0.16	0.030 (8.8 × 10−9)	98.5	0.088
0.077 (1.9 × 10−8)	94.9	0.19	0.043 (1.2 × 10−8)	96.4	0.12
0.10 (2.5 × 10−8)	93.7	0.25	0.068 (2.0 × 10−8)	96.7	0.20
0.15 (3.8 × 10−8)	96.1	0.38	0.12 (3.4 × 10−8)	94.6	0.34
0.25 (6.3 × 10−8)	95.7	0.62	0.22 (6.3 × 10−8)	96.6	0.63
0.45 (1.1 × 10−7)	94.1	1.1	0.42 (1.2 × 10−7)	95.4	1.2
4.1 (1.0 × 10−6)	98.7	10	4.0 (1.2 × 10−6)	96.4	12
10 (2.5 × 10−6)	98.0	25	10 (2.9 × 10−6)	96.4	29
100 (2.5 × 10−5)	96.1	248	100 (2.9 × 10−5)	96.4	290

. Figure 2 shows the plots of the radiochemical yield of [⁶⁴Cu]Cu-ATSM against the concentration of metal ions in the solutions. The top axis shows the molar ratio of the metal ions to ATSM. Table 1 and Table 2 show the concentration of metal ions, radiochemical yield, and molar ratio of M²⁺ to ATSM (i.e., M²⁺/ATSM). As shown in Figure 2a, Cu²⁺ inhibited the formation of the [⁶⁴Cu]Cu-ATSM complex at lower concentrations compared to other metal ions. When 1.2 mol equivalent of Cu²⁺ or more was added to ATSM, the radiochemical yield of [⁶⁴Cu]Cu-ATSM was reduced to lower than 90%, which is regarded as a quality standard for radiopharmaceuticals.

Figure 2b shows the concentration-dependent effects of Ni²⁺ on [⁶⁴Cu]Cu-ATSM complex formation. The radiochemical yield of [⁶⁴Cu]Cu-ATSM fell below 90% upon adding 288 equivalent Ni²⁺ ions to ATSM. Thus, the effect of Ni²⁺ on [⁶⁴Cu]Cu-ATSM formation was 240 times weaker than that of Cu²⁺. Little effects of Zn²⁺ and Fe²⁺ were observed, even with the addition of 248 mol or 290 mol equivalents of ATSM (Figure 2c and 2d, respectively, and Table 2).

2.2. Experiment 2: Chelate Formation of ATSM with Metal Ions

In Experiment 2, we examined chelate formation when metal ions (Cu²⁺, Ni²⁺, Zn²⁺, and Fe²⁺) were added to the ATSM with UV-Vis spectra over time. Figure 3 shows spectral changes in the ATSM after adding transition metal ions. Cu²⁺ chelated ATSM within 5 min. Ni²⁺ also formed chelates with ATSM; however, the reaction rate was slower than that with Cu²⁺. In the case of Fe²⁺, the ATSM peak at 337 nm gradually decreased, and a weak shoulder appeared at around 400 nm, indicating a slower reaction rate of Fe²⁺ than that of Ni²⁺. The spectrum of Zn²⁺ remained unchanged after 24 h, indicating a prolonged reaction rate.

The reactions for Cu²⁺ and Ni²⁺ were studied semi-quantitatively; the concentrations of Cu- and Ni-ATSM were estimated using the molar extinction coefficients, ε, which were determined at 7680 (Cu-ATSM at 477 nm) and 14,300 cm M^–1 (Ni-ATSM at 402 nm) in this study (Figure S1). Although the ε values in DMSO have not been reported, the values in DMF (Cu-ATSM) [33] and pyridine (Ni-ATSM) [34] are similar to our findings. Figure 4 shows the time-dependent increase in the M-ATSM concentration; the vertical axis represents the concentration of M-ATSM relative to the complete complex concentration. The concentration of Cu-ATSM reached 100% immediately after 5 min, suggesting the completion of the chelate formation. In contrast, the complexation yield of Ni²⁺ with ATSM gradually increased from 5.8% at 5 min to over 50% at 2 h, indicating that the formation of Ni-ATSM is slower and weaker than that of Cu-ATSM.

2.3. Experiment 3: Chelate Formation of Zn-ATSM in the Presence of Brønsted Base

In Experiment 2, the UV-Vis spectrum of ATSM showed that only Zn²⁺ did not react under this formulation condition. To understand the underlying reason, we hypothesized that deprotonation of ATSM would be necessary for coordination with Zn²⁺, and, therefore, we checked whether the use of a Brønsted base changes the reaction rate of Zn-ATSM. Figure 5 shows the changes in the UV-Vis spectrum of ATSM after mixing with ZnCl₂ and sodium methoxide as a Brønsted base. In contrast to the Zn-ATSM spectrum shown in Figure 3c, Zn-ATSM began to form 15 min after the initiation of the reaction. The ATSM peak slightly decreased for 20 h, but that of Zn-ATSM did not change from 15 min to 20 h.

2.4. Experiment 4: Differences in the Formation Constants of Cu-ATSM and Ni-ATSM

To confirm the difference of the formation constants between Cu²⁺ and Ni²⁺ with ATSM, 150 μM CuCl₂, 150 μM NiCl₂, and 150 μM, ATSM were mixed in a total volume of 600 µL, and the UV-Vis spectra change was obtained (Figure 6). Compared to Figure 3a,b, the Cu-ATSM spectrum was observed within 5 min of the start of the reaction, and there was no significant change in the spectrum after 2 h, suggesting that the formation constant of Cu-ATSM was much larger than that of Ni-ATSM. The results showed that upon mixing Cu²⁺ and Ni²⁺, the chelate formation rates of ATSM and Cu²⁺ were much higher than those of Ni²⁺, and the formation of Cu-ATSM was thermodynamically predominant.

3. Discussion

[⁶⁴Cu]Cu-ATSM is a promising therapeutic radiopharmaceutical for clinical trials. The ⁶⁴Cu used in [⁶⁴Cu]Cu-ATSM was prepared using a cyclotron and contained trace metal impurities, mainly Cu²⁺, Ni²⁺, Zn²⁺, and Fe²⁺, derived from cyclotron target systems and production lines. Therefore, to obtain insights into the effects of trace metal impurities on the chelate formation of ⁶⁴Cu and ATSM ligands, we conducted coordination chemistry studies on metal–ATSM complexes.

Experiment 1 indicates that trace amounts of Cu²⁺ and Ni²⁺ affect the radiochemical yield of [⁶⁴Cu]Cu-ATSM. In contrast, Fe²⁺ and Zn²⁺ did not affect radiochemical yield, even when the concentration of these metals reached up to 100 ppm. This result is consistent with those of Experiment 2 (vide infra). Defined as a specification for the investigational drug formulation of [⁶⁴Cu]Cu-ATSM, the radiochemical yield of the chelate should be not less than 90%. Considering this criterion, the maximum metal ion concentrations for [⁶⁴Cu]Cu-ATSM formation were 0.26 ppm (260 ppb) for Cu²⁺, 10 ppm for Ni²⁺, and 100 ppm for Fe²⁺ and Zn²⁺. These values were defined as the highest experimental concentrations at which radiochemical yield could be maintained above 90% in the presence of the impurities.

The metal impurities in the irradiated ⁶⁴Ni target were estimated using previously reported values [26,32,35,36]. The concentration of metal impurities can differ among laboratories because it depends on the cyclotron irradiation, mainly on the target and subsequent separation process. The ⁶⁴Cu source was produced at our institute, QST, and we referred to the values of Ohya et al. [32]. The reported concentrations in 10 mL solutions were Cu: 36 ± 22, Ni: 12 ± 27, Fe: 64 ± 48, and Zn: 157 ± 42 ppb. Assuming that the impurities came from only 10 mL of the solution recovered from the target and that all impurities were in the final product, the estimated concentrations of the metal impurities in our products were well below the recommended levels mentioned above, indicating that the metal impurities from the target did not affect the radiochemical yields in our formulation. In this study, the concentration of Cu²⁺ impurities in the reaction mixture derived from the target system was estimated to be 60.8 ppb in the reaction mixture. This value was approximately 20% of the allowed value (260 ppb). In the production process of ⁶⁴Cu, the ⁶⁴Ni fraction is usually recovered to regenerate as the target; as a result, the purity of the ⁶⁴Ni target increases, and control of the copper impurities in the target is essential. According to the values of metal impurities reported by other institutes, the copper content reaches the permissible value in one case; thus, in such a case, the removal of Cu²⁺ from the ⁶⁴Ni target using a chelate resin, for instance, is required [37,38].

The chelate formation reaction was examined qualitatively from a kinetic point of view, with contrasting results obtained for Cu²⁺ and Ni²⁺. Cu²⁺ was made to quantitatively react with ATSM within 5 min, whereas Ni²⁺ showed a slow reaction. The formation of Ni-ATSM complexes was almost complete after 24 h, but the amount of formed Ni-ATSM was only 86% of that of the theoretical amount from the quantitative reaction. This observation suggests that the system was in chemical equilibrium, and Ni²⁺ did not react quantitatively with ATSM. Thus, the reaction of metal ions with the ATSM is more favorable with Cu²⁺ than with Ni²⁺. The low inhibition of Ni²⁺ and ATSM can be explained by both thermodynamic and kinetic consequences.

The reaction of Zn²⁺ with ATSM showed little spectral change, suggesting a slow reaction rate for Zn²⁺. One reason for the slow reaction may be that deprotonation of ATSM is required to coordinate with Zn²⁺. The reaction proceeded more rapidly when we added two equivalents of sodium methoxide, a strong Brønsted base (Figure 5). This finding suggests that a kinetic factor is involved, whereby Zn²⁺ does not inhibit the formation of the Cu-ATSM complex. Therefore, deprotonating agents may inhibit the ions involved in [⁶⁴Cu]Cu-ATSM formation, and the concentration of deprotonating agents in the [⁶⁴Cu]Cu-ATSM formulation must be strictly controlled.

The kinetic experiments indicated that the reaction rates are different between Cu²⁺ and Ni²⁺. This phenomenon is frequently observed in the formation of metal complexes in aqueous solutions. The exchange reaction rate of the solvent molecule dominated most of the formation reaction rates, and no ligand could react faster than the solvent molecule. A significant difference in the solvent exchange reaction in aqueous solutions was observed between Cu²⁺ and Ni²⁺; the exchange rate constant, k_ex, of Cu²⁺ was 10⁵ times larger than that of Ni²⁺ in water [39]. We expected a similar relationship with DMSO concentration.

Under the conditions of Experiment 1, the maximally allowed Ni²⁺ concentration of 10 ppm corresponded to 170 μM, and the molar ratio Ni²⁺/ATSM was 29. This indicated that Ni²⁺ did not inhibit [⁶⁴Cu]Cu-ATSM formation up to 29 equivalents to ATSM. This observation suggests the low chelate formation ability of Ni²⁺, although the coexisting glycine, added as an auxiliary complexing agent to maintain Cu²⁺ in the solution, can weaken the coordination of ATSM to some extent through the competition of ATSM with glycine. This coexisting glycine also competed with [⁶⁴Cu]Cu-ATSM formation because the molar ratio of total Cu, the sum of cold Cu and ⁶⁴Cu, to ATSM was only 0.68 at 91% radiochemical yield.

The effect of metal ions on the formation of [⁴⁴Sc]Sc-DOTATATE (Figure S2) has been reported [40]. Two nmol/mL of Fe^2+/3+ and Zn²⁺ and 10 nmol/mL of Cu²⁺ significantly reduced the labeling yield, while Ca²⁺ and Al³⁺ did not affect it up to 2 μmol/mL. The formation of [⁴⁴Sc]Sc-DOTATATE was more sensitive to metal ion contamination, in contrast to our [⁶⁴Cu]Cu-ATSM. This high inhibition of DOTATATE results from the fact that the DO3A moiety of DOTATATE can coordinate to a wide range of M²⁺ and M³⁺ ions with high formation constants, reducing the selectivity of DOTATATE.

A limitation of this study is that the formation constant of the ATSM complex was not determined. However, these values can be referred to for the same coordination atom. The formation constant, K_f, for Ni²⁺ and Cu²⁺ complexes of 1,10-diaza-4,7-dithiadecane (Figure S3) has been reported; the log K_f value is 7.41 for Ni²⁺ and 10.70 for Cu²⁺ [41]. These values indicated that the ligand formed a more stable (approximately 1000 times) complex with Cu²⁺ than with Ni²⁺. The values of ATSM can be larger than 1,10-dizaza-4,7-dithiodecane because of the rigid framework of the ligand, and the same stability relationship holds for ATSM. The Irving–Williams series can interpret this stability relationship. The Irving–Williams series shows that the order of stability of the divalent transition metal complex of a given ligand is Mn²⁺ < Fe²⁺ < Co²⁺ < Ni²⁺ < Cu²⁺ > Zn²⁺ [42]. The hard and soft acid and base theory [43] can also explain the large formation constant of Cu-ATSM. Fe²⁺, Ni²⁺, Cu²⁺, and Zn²⁺ are Lewis acids that are intermediates between hard and soft acids, and Cu²⁺ has the highest electronegativity and polarizability among these ions. This indicates that Cu²⁺ is the softest acid in the series and has the highest affinity for soft Lewis bases containing sulfur atoms. Thus, ATSM has the highest affinity for Cu²⁺, and the inhibition of the formation of [⁶⁴Cu]Cu-ATSM by Ni²⁺, Fe²⁺, and Zn²⁺ can be negligible in future manufacturing practices.

4. Materials and Methods

4.1. Reagents and Materials

All reagents and solvents from commercial sources (Fujifilm Wako Pure Chemical, Osaka, Japan) were of high quality and were used without further purification. ATSM and Cu-ATSM were purchased from ABX Advanced Biochemical Laboratories (Radeberg, Germany). Japan Pharmacopoeia grade distilled water (Otsuka Pharmaceutical Factory, Tokyo, Japan) was used for preparation of aqueous solutions. Spectrochemical-analysis-grade dimethyl sulfoxide (DMSO) was purchased from Fujifilm Wako Pure Chemical Co., Ltd., Osaka, Japan. The Ni²⁺ complex of ATSM was prepared following a previously reported method [44].

4.2. Determination of Radiochemical Yield

The radiochemical yield of [⁶⁴Cu]Cu-ATSM was determined by radio-TLC. One microliter of the sample solution was spotted onto an HPTLC silica gel 60 F₂₅₄ glass plate (Merck Millipore, Burlington, MA, USA) and developed using methanol. Radioactivity on the TLC plates was measured using a radio-TLC system (Raytest PET miniGITA Star; Elysia s.a., Liège, Belgium).

4.3. Experiment 1: Effect of Trace Metal Impurities on the Formation of [⁶⁴Cu]Cu-ATSM Complex

The experiments were performed on a scale of 1/400 of the clinical formulation, and the concentrations of the reagents, including ⁶⁴Cu and ATSM, were kept constant.

Cyclotron-produced ⁶⁴CuCl₂ (29.25 MBq; 3.20 pmol) was dissolved in 20 μL of 0.125 M aqueous glycine solution, and 0.5 μL of 0.5 mM ATSM in DMSO was added and mixed well. Then, 20 μL of each metal aqueous solution (Cu²⁺, Ni²⁺, Zn²⁺, and Fe²⁺) at nine different concentrations (0.0125, 0.025, 0.05, 0.1, 0.2, 0.4, 4, 10, and 100 ppm in 40.5 μL of the reaction mixture) were added. The mixed solutions were incubated at room temperature for 10 min. The radiochemical yield of [⁶⁴Cu]Cu-ATSM was determined by radio-TLC. The effect of adding these metal solutions was evaluated by plotting the radiochemical yield against the concentration of metal ions as the sum of the added metal ions and metal impurities derived from the target system.

4.4. Experiment 2: Chelate Formation of ATSM with Metal Ions

Briefly, 300 μM of metal solutions (Cu²⁺, Ni²⁺, Zn²⁺, and Fe²⁺) were prepared by dissolving their chlorides into DMSO, and 300 µL of each solution was mixed with 300 µL of 300 μM ATSM in DMSO. UV-Vis spectra were measured at 5, 15, 30, 1, 2, 5, and 24 h after mixing. Spectra were measured using a NanoDrop One Microdrop UV-Vis Spectrometer (ND-ONE-W; Thermo Fisher Scientific, Waltham, MA, USA) with 1 μL of each mixed solution. The molar absorption coefficient (ε) of Cu-ATSM and Ni-ATSM was determined on 100 μM DMSO solutions of Cu-ATSM and Ni-ATSM. The concentrations of the Cu-ATSM and Ni-ATSM in each reaction mixture were analyzed based on the obtained ε values.

4.5. Experiment 3: Chelate Formation of Zn-ATSM in the Presence of Brønsted Base

The spectrum of the reaction between ATSM and Zn²⁺ remained almost unchanged for 24 h in Experiment 2. One possible reason for the slow reaction is that deprotonation of ATSM is required to coordinate with Zn²⁺. Therefore, we confirmed whether the reaction rate of Zn-ATSM is changed by using the Brønsted base. A total of 600 µL of DMSO solution containing 200 µL of a ZnCl₂ (300 µM), 200 µL of ATSM (300 µM), and 200 µL of sodium methoxide (600 µM) was stirred at room temperature. The UV-Vis spectra were measured at 15 min, 30 min, 1 h, and 20 h after mixing. The spectra were measured using the same method as in Experiment 2.

4.6. Experiment 4: Differences in the Formation Constants of M-ATSM in the Reaction of ATSM with Cu²⁺ and Ni²⁺ Mixtures

Specifically, DMSO solutions of 150 µM CuCl₂ (200 µL) and 150 µM NiCl₂ (200 µL) were mixed, and 200 µL solution of ATSM in DMSO (150 µM) was added at room temperature. Changes in UV-Vis spectra were measured at 5 min, 1 h, and 2 h after mixing. The spectra were measured using the same method as in Experiment 2.

5. Conclusions

This study showed that 1.2 mol Cu²⁺ or 288 mol Ni²⁺, equivalent to ATSM, decreased the labeling yield of [⁶⁴Cu]Cu-ATSM below 90%. Zn²⁺ and Fe²⁺ had little effect on the labeling of [⁶⁴Cu]Cu-ATSM, even with the addition of 248 or 290 mol equivalents of ATSM. Based on these data, we conclude that trace amounts of Ni²⁺, Zn²⁺, and Fe²⁺ have little effect on the quality of [⁶⁴Cu]Cu-ATSM; however, the concentration of Cu²⁺ must be controlled. These results can provide process management tools for radiopharmaceuticals.