Synthesis of DOTA-pyridine chelates for ⁶⁴Cu coordination and radiolabeling of αMSH peptide

Abstract

Background

⁶⁴Cu is one of the few radioisotopes that can be used for both imaging and therapy, enabling theranostics with identical chemical composition. Development of stable chelators is essential to harness the potential of this isotope, challenged by the presence of endogenous copper chelators. Pyridyl type chelators show good coordination ability with copper, prompting the present study of a series of chelates DOTA-xPy (x = 1–4) that sequentially substitute carboxyl moieties with pyridyl moieties on a DOTA backbone.

Results

We found that the presence of pyridyl groups significantly increases ⁶⁴Cu labeling conversion yield, with DOTA-2Py, −3Py and -4Py quantitatively complexing ⁶⁴Cu at room temperature within 5 min (1 × 10^− 4 M). [⁶⁴Cu]Cu-DOTA-xPy (x = 2–4) exhibited good stability in human serum up to 24 h. When challenged with 1000 eq. of NOTA, no transmetallation was observed for all three ⁶⁴Cu complexes. DOTA-xPy (x = 1–3) were conjugated to a cyclized α-melanocyte-stimulating hormone (αMSH) peptide by using one of the pendant carboxyl groups as a bifunctional handle. [⁶⁴Cu]Cu-DOTA-xPy-αMSH retained good serum stability (> 96% in 24 h) and showed high binding affinity (Ki = 2.1–3.7 nM) towards the melanocortin 1 receptor.

Conclusion

DOTA-xPy (x = 1–3) are promising chelators for ⁶⁴Cu. Further in vivo evaluation is necessary to assess the full potential of these chelators as a tool to enable further theranostic radiopharmaceutical development.

Background

Personalized medicine is predicated on a philosophy of tailoring disease diagnosis, monitoring and therapy for individual patients in order to ensure maximal medical benefit during disease management (Vogenberg et al. 2010). Theranostics is an approach in which a single compound can be used to both diagnose and treat disease, and is an important tool in the move toward personalized medicine (Yordanova et al. 2017). Quantitative target visualization of a therapeutic radiopharmaceutical can help predict if a patient will benefit from a therapy and avoid unnecessary or ineffective treatments. In many instances, modern radiotheranostics include isotopes used for therapy, such as yttrium (⁹⁰Y), lutetium-177 (¹⁷⁷Lu) and actinium-225 (²²⁵Ac), and pairs them with compounds containing a different, imageable isotope for imaging, such as gallium-68 (⁶⁸Ga), fluorine-18 (¹⁸F) and indium-111 (¹¹¹In) (Langbein et al. 2019; Yordanova et al. 2017). One key assumption is that the imaging tracer has identical biodistribution as the therapeutic radiopharmaceutical. This assumption is not always true considering differences in molar activity (activity per mole of compound), in vivo stability and inherent biological sensitivity to even the slightest perturbations in pharmaceutical structure (Di 2015). The above matched-pair philosophy is driven primarily by the fact that not many isotopes are capable of both imaging and therapy. Copper-64 (⁶⁴Cu, t_1/2 = 12.7 h) decays via electron capture (branching ratio [BR] = 44%), β⁻ emission (39%, 0.573 MeV), and β⁺ emission (17%, E_β,avg = 0.655 MeV). Its half-life makes the radionuclide a versatile isotope for imaging, therapy or theranostics using small molecules, peptides, antibodies, and nanoparticle targeting and carrier platforms (Cai and Anderson 2014; Gutfilen et al. 2018). While the unique decay profile permits ⁶⁴Cu, in principle, to be a dual PET-imaging and targeted therapy isotope, the high positron branching ratio precludes the clinical use of ⁶⁴Cu for radiotherapy. The pure β⁻ emitter, ⁶⁷Cu (t_1/2 = 61.8 h) is available by high gamma irradiation and shows promise as an element-matched theranostic pair for ⁶⁴Cu PET imaging and ⁶⁷Cu targeted therapy (Smith et al. 2012). As one of the only copper isotopes that can be produced via proton irradiation on lower energy medical cyclotrons, ⁶⁴Cu is more readily available over other isotopes, making it more ideal for use in imaging (McCarthy et al. 1997).

The development of chelators for Cu is essential in order to harness its theranostic power. Cu is the third most abundant metal in the human body, and plays an important role in many biological functions, such as redox electron transfer, structure shaping, and catalysis (Gutfilen et al. 2018). As a result, there are many Cu chelating proteins in vivo and it is a challenge to develop thermodynamically and kinetically stable Cu chelators. DOTA (1,4,7,10-tetraazacyclododecane- 1,4,7,10-tetraacetic acid) (De León-Rodríguez and Kovacs 2008), TETA (1,4,8,11-tetraazacyclote- tradecane-1,4,8,11-tetraacetic acid) (Wadas and Anderson 2007) and their derivatives are commonly used, but they are unstable in vivo (Fig. 1) (Smith 2004). Over time, NOTA (1,4,7-triazacyclononane-1,4,7-triacetic acid) (Cooper et al. 2012), the cross-bridged cyclam (CB-TE2A) (Sprague et al. 2007), and sarcophagine-based chelators (Di Bartolo et al. 2006) have been developed and show improved stability (Anderson and Ferdani 2009; Cai and Anderson 2014).

Fig. 1

Examples of commonly used ⁶⁴Cu chelators

The development of bifunctional chelators capable of complexing radiometals at ambient temperature, in short reaction times, and sub-micromolar concentrations is of increasing interest. Chelators containing pyridyl moieties, including pyridine, bipyridine and terpyridine, have been previously used to complex Cu (Czerwińska et al. 2017; Lee et al. 2011; Leussing and Hansen 1957), and as Cu(II) is a d9 metal of borderline hardness, it favors ligation with softer bases including amines and imines. We hypothesized for this study that the addition of pyridine moieties in place of the carboxylic acid prostheses on DOTA (referred to herein as DOTA-xPy, x = 1–4; compounds 4–8 in Scheme 1) would maintain a higher affinity for the Cu(II) ion.

Scheme 1

Synthesis of cyclen-xPy (x = 1–4, compounds 1–4) and DOTA pyridine derivatives DOTA-xPy (x = 1–3, compounds 5–7).

Furthermore, we present the application of novel DOTA-pyridine derivatives and the labeling of a peptide targeting the Melanocortin 1 receptor (MC1R) (referred to herein as DOTA-xPy-αMSH (x = 1–3)). MC1R is overexpressed in primary and metastatic melanoma (López et al. 2007; Tafreshi et al. 2019). With low normal tissue expression, this receptor is an attractive target for molecular imaging and treatment of late-stage melanoma, which has a low long-term survival rate and no curative treatment options available (Andtbacka et al. 2015; Hertzman Johansson and Egyhazi 2014; Hinrichs and Rosenberg 2014; Postow et al. 2015). We have extensively studied analogues of α-melanocyte-stimulating hormone (αMSH) for MC1R-targeted imaging and therapy. We have developed ⁶⁸Ga or ¹⁸F labeled cyclic αMSH (CCZ01048, CCZ01064, CCZ01096) for PET imaging in preclinical models of mouse or human melanoma (Zhang et al. 2020; Zhang et al. 2019; Zhang et al. 2018; Zhang et al. 2017). In all cases, tracers achieved good tumor visualization in PET images with excellent tumor-to-normal tissue ratios. Moreover, we recently evaluated ²²⁵Ac labeled αMSH derivatives, which also exhibited excellent tumor-to-normal tissue ratios (Ramogida et al. 2019; Yang et al. 2020). Knowing the potential of tumor accumulation and internalization, and low uptake in normal tissues, we were interested in labeling an αMSH peptide with ⁶⁴Cu to assess the unique theranostic power of ⁶⁴Cu. [⁶⁴Cu]Cu-αMSH peptide will allow for real-time monitoring of physiological processes during therapy studies, and when coupled with the longer half-life of ⁶⁴Cu over other PET isotopes (i.e. ⁶⁸Ga and ¹⁸F), it will enable later time points for imaging and biodistribution studies, and a more accurate assessment of the longer-term biological fate of both peptide and radionuclide.

Results and discussions

The chelator synthesis started with the cyclen-xPy (x = 1–4, compounds 1–4, Scheme 1). All derivatives were synthesized using a single reaction followed by purification by HPLC. Each compound was then reacted with tert-butyl bromoacetate and deprotected to yield the final product DOTA-xPy (x = 1–3, compounds 5–7) (Subat and König 2001; Veiga et al. 2013). Compounds cyclen-1Py (Aime et al. 1999; Aoki et al. 2004; Subat et al. 2007), cyclen-2Py (El Hajj et al. 2009), cyclen-4Py (Natrajan et al. 2010) and DOTA-1Py (Aime et al. 1999) were previously reported, while compounds cyclen-3Py, DOTA-2Py and DOTA-3Py are new compounds to the best of our knowledge. DOTA-1Py and DOTA-3Py were obtained as a single compound. DOTA-2Py could potentially maintain both cis- and trans- stereoisomer configurations. Because ¹H NMR of cyclen-2Py indicated a cis- derivative, DOTA-2Py was assumed to be cis-. The complexation of ^natCu to DOTA-xPy (x = 1–3) was confirmed by IR and UV-vis spectra (Fig. S1 and S2 in Supporting Information). In the UV-vis spectra, we observed increased absorbance between 200 and 400 nm, with a new shoulder peak emerging at around 290 nm, suggesting metal complexation formation. A new peak at 720–740 nm and the blue color of the product indicated the formation of Cu(II). Also, for all three chelates, UV-vis spectra showed saturation at 1.2 eq. Cu(II), suggesting the complex has 1:1 Cu(II):chelate ratio. Infrared (IR) analysis showed the decrease of the stretching absorption at 1700 cm^− 1 for C=O and 1200 cm^− 1 for C-N (Spreckelmeyer et al. 2017), indicating that both COO and pyridyl moieties bind to the metal. In attempting to interpret the coordination geometry, we considered both the relative frequency and intensities of both asymmetric (ν_as, ~ 1600 cm^− 1) and symmetric (ν_s, ~ 1400 cm^− 1) COO stretching vibrations for the free and bound chelates (Berestova et al. 2017). For the DOTA-1Py and -2Py chelates, we observed the evolution of dual maxima upon metal binding, suggesting there exist multiple binding configurations for the COO moieties around the metal. That said, the spectroscopic data did not allow for the elucidation of a definitive orientation of binding motifs around the metal. Future studies will focus on a full structural analysis to aid in determining the coordination behavior for these chelates. The labeled peptide [⁶⁴Cu]Cu-DOTA-2Py-αMSH showed a double peak on HPLC gamma trace under isocratic conditions, which could indicate the formation of stereoisomers upon metal coordination.

⁶⁴Cu was produced by proton irradiation of enriched ⁶⁴Ni solid metal targets using TRIUMF’s TR13 cyclotron. After acid dissolution and purification by cation exchange resin (AG1-X8), ⁶⁴Cu was obtained in dilute HCl. ⁶⁴Cu labeling of DOTA, DOTA-1Py, DOTA-2Py, DOTA-3Py, and cylen-4Py were examined at various ligand concentrations. The results (Fig. 2 and Fig. 3) indicated that the subsequent addition of pyridine moiety served to enhance ⁶⁴Cu labeling conversion in general. While DOTA showed little complexation with ⁶⁴Cu at ambient temperature even at ligand concentrations of 1 × 10^− 3 M (100 nmol ligand, 3 MBq ⁶⁴Cu), having one, two or three pyridyls significantly improved radiochemical conversion (RCC), with DOTA-3Py demonstrating efficient complexation of ⁶⁴Cu (RCC close to 95%) at ligand concentrations of 1 × 10^− 5 M (1 nmol ligand, 3 MBq ⁶⁴Cu) and at ambient temperature. The differences in ⁶⁴Cu labeling performance between DOTA-2Py and -4Py were not significant.

Fig. 2

Radio-HPLC chromatogram of ⁶⁴Cu labeling of (a) DOTA-1Py (retention time, t_R = 12.5 min), (b) DOTA-2Py (t_R = 12.7 min), (c) DOTA-3Py (t_R = 14.5 min), and (d) cyclen-4Py (t_R = 16.7 min) at ligand concentration 1 × 10^− 3 M (100 nmol ligand, 3 MBq ⁶⁴Cu) and ambient temperature

Fig. 3

⁶⁴Cu labeling conversion at various ligand concentrations for DOTA, DOTA-1Py, DOTA-2Py, DOTA-3Py, and cyclen-4Py (n = 3) at ambient temperature and 5 min reaction time

We then studied the reaction kinetics for labeling DOTA-2Py, −3Py and -4Py at 1 × 10^− 4 M (10 nmol ligand, 3 MBq ⁶⁴Cu, DOTA-1Py not studied because the labeling was less efficient). Under the same conditions as described above, the reactions were monitored at 1, 5, 10, 20 and 30 min by radioTLC. All three chelators demonstrated near quantitative metal incorporation within 5 min at ambient temperature (Fig. 4).

Fig. 4

⁶⁴Cu labeling conversion of DOTA-2Py, DOTA-3Py and cyclen-4Py at 1, 5, 10, 20 and 30 min performed with ligand concentration of 1 × 10^− 4 M (10 nmol ligand, 3 MBq ⁶⁴Cu) and ambient temperature (n = 3)

The stability of [⁶⁴Cu]Cu-DOTA-2Py, −3Py and -4Py were examined in phosphate buffer and human serum at 1 h and 24 h at 37 °C. All three ligands were largely stable (> 94%) in both phosphate buffer and human serum in 24 h (Table 1).

Table 1 Stability of [⁶⁴Cu]Cu-DOTA-xPy (x = 2–4) and [⁶⁴Cu]Cu-DOTA-xPy-MSH (x = 1–3) in phosphate buffer and human serum at 37 °C

NOTA has been the widely used chelate for labeling with ⁶⁴Cu as it forms kinetically stable complexes (Cooper et al. 2012). For comparison, we examined the kinetic inertness of [⁶⁴Cu]Cu-DOTA-2Py, −3Py and -4Py using NOTA as the transchelation challenge reagent. After the formation of [⁶⁴Cu]Cu-DOTA-2Py, −3Py and -4Py at ligand concentration of 1 × 10^− 4 M (10 nmol ligand, 3 MBq ⁶⁴Cu), 1000 eq. of p-SCN-Bn-NOTA were added and the mixture was monitored by radio-HPLC over 4 h. During this time, no transchelation was observed for all three ⁶⁴Cu complexes, indicating the kinetic inertness of [⁶⁴Cu]Cu-DOTA-2Py, −3Py and -4Py. As control experiments, p-SCN-Bn-NOTA was labeled with ⁶⁴Cu, and 1000 eq. of DOTA-2Py, −3Py or -4Py were added and no transchelation was observed for up to 4 h. Interestingly, when ⁶⁴Cu was added to an equimolar mixture (1 × 10^− 4 M ligand) of DOTA-xPy (x = 2–4) and p-SCN-Bn-NOTA, [⁶⁴Cu]Cu-DOTA-2Py, −3Py and -4Py were formed with RCYs of 20%, 19% and 28%, while [⁶⁴Cu]Cu-p-SCN-Bn-NOTA formed at RCYs of 78%, 79% and 70%, respectively, indicating DOTA-xPy (2–4) chelators may have slower kinetics compared to NOTA.

With radiolabeling efficiency, stability and kinetic inertness of the Cu-complexes established, we attached DOTA-1Py, −2Py, −3Py to an αMSH peptide through a carboxylic moiety of each chelator (Fig. 5). Although DOTA-1Py-αMSH was expected to complex ⁶⁴Cu less efficiently, it was included to study the trend with increasing pyridyls. The peptide was synthesized via the step-wise addition of amino acids followed by on-resin cyclization. The final Fmoc protecting group was removed by mixing with HOBt, HBTU and isopropyl ethylamine and agitated at ambient temperature overnight. The deprotection and purification by HPLC was done by standard procedures (Zhang et al. 2017) and characterized by mass spectrometry.

Fig. 5

Chemical structure of DOTA-3Py-αMSH

Radiolabeling was performed under similar conditions as discussed for the chelators themselves. [⁶⁴Cu]CuCl₂ was mixed with DOTA-xPy-αMSH in MES buffer (0.4 M, pH 6.2) and allowed to sit at ambient temperature. At 1 × 10^− 5 M peptide concentration (1 nmol), labeling with 3 MBq of ⁶⁴Cu was quantitative after 15 min for DOTA-2Py-αMSH and DOTA-3Py-αMSH and 66% for DOTA-1Py-αMSH (Fig. 6). The resulting [⁶⁴Cu]Cu-DOTA-xPy-αMSH (x = 1–3) complexes were stable in both phosphate buffer and human serum at 37 °C for 24 h (radiochemical purity, RCP > 96%, Table 1).

Fig. 6

Radio-HPLC chromatograms of (a) [⁶⁴Cu]Cu-DOTA-1Py-αMSH (RCY = 66%), (b) [⁶⁴Cu]Cu-DOTA-2Py-αMSH, and (c) [⁶⁴Cu]Cu-DOTA-3Py-αMSH

Modifications of peptides may adversely impact target binding affinity (Di 2015). We studied the binding affinity of ^natCu-DOTA-xPy-αMSH (x = 1–3) and found that binding of all three peptides towards MC1R remained in the in low nanomolar level (Table 2), although lower than Ga-DOTA-αMSH complex (K_i = 0.31 ± 0.06 for CCZ01048) (Zhang et al. 2017).

Table 2 Inhibition constants (K_i) of ^natCu-DOTA-xPy-αMSH (x = 1–3) towards MC1R (n = 3)

Conclusion

We have synthesized a new set of chelators DOTA-1Py, −2Py, −3Py and -4Py and demonstrated that substitution of the carboxylic groups with one or two pyridine groups on a DOTA backbone can significantly improve the binding affinity towards ⁶⁴Cu. We covalently attached the chelators to an αMSH peptide and demonstrated that DOTA-xPy-αMSH (x = 1–3) can be quantitatively labeled by ⁶⁴Cu at ambient temperature. Moreover, all three ^natCu-DOTA-xPy-αMSH (x = 1–3) retained good binding affinity towards MC1R. Metabolic stability and biodistribution evaluation of [⁶⁴Cu]Cu-DOTA-xPy-αMSH (x = 1–3) is underway.

Material and methods

General

All reactions were carried out with commercial solvents and reagents that were used as received. Concentration and removal of trace solvents was done via a Büchi rotary evaporator using dry ice/acetone condenser, and vacuum applied from an aspirator or Büchi V-500 pump. Nuclear magnetic resonance (NMR) spectra were recorded using D₂O or DMSO-d₆ or MeOD-d₄. Signal positions are given in parts per million from tetramethylsilane (δ = 0) and were measured relative to the signal of the solvent. ¹H NMR spectral data are tabulated in the order: multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; quint, quintet; m, multiplet). NMR spectra were recorded on a Bruker 400 (400 MHz) or a Bruker 300 (300 MHz) spectrometer at 25 °C unless otherwise noted. MS spectra was recorded on a Waters ZQ equipped with ESCI ion source (low resolution), a Kratos Concept IIHQ (high resolution) or a Bruker Autoflex MALDI-TOF. High performance liquid chromatography (HPLC) of radiolabeled samples was performed on an Agilent 1200 series instrument equipped with a diode array detector and Raytest GABI Star scintillation detector. Radioactivity was measured on a Capintec 55TR dose calibrator.

Chelator synthesis

Compounds cyclen-1Py (1) (Aime et al. 1999; Aoki et al. 2004; Subat et al. 2007), cyclen-2Py (2) (El Hajj et al. 2009), cyclen-4Py (4) (Natrajan et al. 2010) and DOTA-1Py (5) (Aime et al. 1999) are previously reported and we synthesized with a modified procedure. Compound 2′ was co-produced with 2.

Synthesis of cyclen-1Py (1), cyclen-2Py (2, 2′), cyclen-3Py (3), and cyclen-4Py (4): To a suspension of cyclen (172.3 mg, 1 mmol) and K₂CO₃ (552 mg, 4 mmol) in anhydrous acetonitrile (25 mL), a solution of 2-(bromomethyl)pyridine hydrobromide (329 mg, 1.3 mmol) in 10 mL anhydrous acetonitrile was added dropwise over a period of 1 h at 0 °C and the reaction mixture was allowed to stand at room temperature overnight. After the reaction was complete, the precipitate was filtered off and the solvent was removed by evaporation. The crude products were purified on reverse phase semi-preparative HPLC using Phenomenex Luna C18 (250 mm × 100 mm) column at 3 mL/min with the following method: A: H₂O with 0.1% TFA, B: CH₃CN with 0.1% TFA; 0–20 min, 100% A isocratic; 20-30 min, 100%–85% A, 0%–15% B; 30-45 min, 85% A, 15% B. The fraction at 6.3 min, 15.5 min and 33.0 min were collected and lyophilized to yield 1 as a colorless solid (40 mg, 15%), 2 (2′) as a yellowish solid (63 mg, 18%) and 3 as a yellowish solid (76 mg, 17%), respectively.

1-(Pyridin-2-ylmethyl)-1,4,7,10-tetraazacyclododecane (cyclen-1Py, 1): ¹H NMR (300 MHz, D₂O) δ 8.75 (dd, J = 5.9, 1.6 Hz, 1H), 8.55 (td, J = 8.0, 1.6 Hz, 1H), 8.09–7.93 (m, 2H), 4.16 (s, 2H), 3.27 (t, J = 5.2 Hz, 4H), 3.20–3.10 (m, 4H), 3.01–2.87 (m, 8H). ¹³C NMR (126 MHz, MeOD) δ 158.56, 149.53, 140.87, 125.18, 125.09, 58.38, 51.06, 46.07, 44.19, 44.01. HR-MS: calcd. For C₂₄H₂₆N₅ ([M + H]⁺): 264.2188; found: 264.2112.

1,4-Di(pyridin-2-ylmethyl)-1,4,7,10-tetraazacyclododecane and 1,2-Di(pyridin-2-ylmethyl)-1,4,7,10-tetraazacyclododecane (cyclen-2Py, 2 and 2′): ¹H NMR (300 MHz, D₂O) δ 8.65–8.55 (m, 2H), 8.30 (td, J = 7.9, 1.7 Hz, 2H), 7.88–7.71 (m, 4H), 4.08 (s, 4H), 3.43 (s, 4H), 3.32 (dd, J = 6.7, 4.3 Hz, 4H), 3.08–2.92 (m, 8H). ¹³C NMR (126 MHz, MeOD) δ 161.92 (q, J = 36.5 Hz, CF₃CO₂H) 154.44, 148.71, 141.43, 126.23, 125.61, 117.59 (q, J = 291.1 Hz, CF₃CO₂H) 57.24, 51.90, 51.51, 44.51, 43.80. ESI-MS calcd. For C₂₀H₃₁N₆ ([M + H]⁺): 355.3; found: 355.3.

1,4,7-Tri(pyridin-2-ylmethyl)-1,4,7,10-tetraazacyclododecane (cyclen-3Py, 3): ¹H NMR (500 MHz, MeOD) δ 8.63 (dd, J = 5.4, 1.6 Hz, 2H), 7.93 (td, J = 7.8, 1.7 Hz, 2H), 7.81 (td, J = 7.8, 1.8 Hz, 1H), 7.62 (d, J = 7.9 Hz, 2H), 7.57 (dd, J = 5.1, 1.6 Hz, 1 H), 7.52–7.45 (m, 2H), 7.39 (d, J = 7.8 Hz, 1H), 7.23 (dd, J = 7.5, 4.9 Hz, 1H), 4.54 (s, 2H), 4.06 (s, 4H), 3.63 (t, J = 5.3 Hz, 4H), 3.42 (t, J = 5.1 Hz, 4H), 3.23 (t, J = 5.3 Hz, 4H), 3.09–2.96 (m, 4H). ¹³C NMR (126 MHz, MeOD) δ 162.02 (CF₃CO₂H), 161.73 (CF₃CO₂H), 161.44, 156.25, 150.49, 149.82, 148.33, 141.39, 139.42, 126.33, 125.64, 125.14, 118.65 (CF₃CO₂H), 116.35 (CF₃CO₂H), 58.18, 57.11, 52.65, 50.12, 43.73. HR-MS: calcd. For C₂₆H₃₆N₇ ([M + H]⁺): 446.3032; found: 446.2989.

Preparation of cyclen-4Py: A mixture of cyclen (35 mg, 0.2 mmol) and K₂CO₃ (414 mg, 3 mmol) in dry acetonitrile (15 mL) was placed in an ice bath and then a solution of 2-(bromomethyl)pyridine (190 mg, 0.75 mmol) in 10 mL dry CH₃CN was added dropwise over a period of 1 h. The resulting mixture was stirred for additional 10 h. The precipitate was filtered, and the filtrate concentrated in vacuo to give the crude product which was further purified by HPLC using the same method described above. ¹H NMR (300 MHz, D₂O) δ 8.57 (d, J = 5.3 Hz, 4H), 8.00 (td, J = 7.8 Hz, 1.1 Hz, 4H), 7.80 (d, J = 7.5 Hz, 4H), 7.66 (t, J = 6.3 Hz, 4H), 4.36 (s, 8H), 3.45 (s, 16H). ESI-MS: calcd. For C₃₂H₄₁N₈ ([M + H]⁺): 537.3; found: 537.3.

Preparation of DOTA-1Py: A solution of tert-butyl bromoacetate (59 mg, 0.3 mmol) in dry CH₃CN (5 mL) was added dropwise to a mixture of cyclen-1py (26 mg, 0.1 mmol), K₂CO3 (83 mg, 0.6 mmol) and KI (10 mg, 0.06 mmol) in dry CH₃CN (10 mL). The resulting mixture was stirred at room temperature for 1 h and then refluxed overnight. After cooling to room temperature, the mixture was filtered, and the residue washed with dry CH₃CN. The filtrate was collected and evaporated to yield the title compound as a white solid. The solid was dissolved in 10 mL of mixed solvent containing TFA (9.5 mL), TIPS (0.25 mL) and H₂O (0.25 mL). The reaction mixture was stirred at room temperature for 2 h. The solvent was removed by air flow and the residue was dissolved in CH₃CN/H₂O (1/1). The crude product was purified on reverse phase semi-preparative HPLC using Phenomenex Luna C18 (250 mm × 100 mm) column at 3 mL/min with the following method: A: H₂O with 0.1% TFA, B: CH₃CN with 0.1% TFA; 0-5 min, 100% A isocratic; 5–25 min, 100%–60% A, 0%–40% B. The fraction at 6.7 min was collected and lyophilized to give the title compound as a colorless oil (32 mg, 74%). ¹H NMR (600 MHz, D₂O) δ 8.69 (ddd, J = 5.9, 1.6, 0.6 Hz, 1H), 8.46 (td, J = 7.9, 1.6 Hz, 1H), 8.03 (d, J = 8.0 Hz, 1H), 7.94 (ddd, J = 7.3, 5.9, 1.3 Hz, 1H), 4.09 (s, 2H), 3.89 (d, J = 15.8 Hz, 2H), 3.68 (d, J = 16.4 Hz, 4H), 3.47–3.44 (m, 4H), 3.35 (s, 2H), 3.31–3.28 (m, 2H), 3.15 (s, 2H), 3.06–3.02 (m, 2H), 2.92–2.86 (m, 4H). ¹³C NMR (151 MHz, D₂O) δ 175.48, 169.03, 162.83 (q, J = 35.6 Hz, CF₃CO₂H), 150.13, 147.53, 143.16, 129.61, 128.39, 126.93, 116.26 (q, J = 291.7 Hz, CF₃CO₂H), 55.34, 53.60, 52.87, 52.05, 50.36, 48.15, 48.01. ESI-MS: calcd. For C₂₀H₃₂N₅O₆ ([M + H⁺]): 438.2; found: 438.2.

Preparation of DOTA-2Py: cyclen-2py (39 mg, 0.2 mmol) was dissolved in dry acetonitrile (10 mL), K₂CO₃ (55 mg, 0.4 mmol) and KI (7 mg, 0.04 mmol) was added to the solution. Then a solution of tert-butyl bromoacetate (35 mg, 0.1 mmol) in dry CH₃CN (5 mL) was added dropwise. The mixture was stirred at room temperature for 1 h and then refluxed overnight. After cooling to room temperature, the mixture was filtered, and the filtrate was evaporated to yield the title compound as a yellowish solid. The solid was dissolved in 10 mL of mixed solvent containing TFA (9.5 mL), TIPS (0.25 mL) and H₂O (0.25 mL). The reaction mixture was stirred at room temperature for 2 h. The solvent was removed by air flow and the residue was dissolved in CH₃CN/H₂O (1/1) for further HPLC purification using the same method described above. The fraction from 14 to 14.5 min was collected and lyophilized to give the title compound as colorless oil (36 mg, 77%). ¹H NMR (500 MHz, MeOD) δ 8.44 (dd, J = 5.3, 1.7 Hz, 2H), 7.79 (td, J = 7.7, 1.7 Hz, 2H), 7.55 (s, 2H), 7.35–7.31 (m, 2H), 4.36–4.08 (m, 4H), 3.73 (s, 4H), 3.53–2.95 (m, 18H, including MeOD). ¹³C NMR (75 MHz, DMSO) δ 170.77, 158.86 (CF₃COOH), 158.43 (CF₃COOH), 153.45, 149.07, 137.97, 124.36, 123.76, 119.20 (CF₃COOH), 115.25 (CF₃COOH), 57.12, 53.93, 50.00, 49.75, 49.44. HR-MS: calcd. For C₂₄H₃₅N₆O₄ ([M + H]⁺): 471.2720; found: 471.2558.

Preparation of DOTA-3Py: To a suspension of cyclen-3py (45 mg, 0.1 mmol), K₂CO₃ (28 mg, 0.2 mmol) and KI (3 mg, 0.02 mmol) in dry acetonitrile (15 mL), a solution of tert-butyl bromoacetate (20 mg, 0.1 mmol) in dry CH₃CN (5 mL) was added. The resulting mixture was refluxed overnight. The precipitated solids were removed by filtration and the filtrate was concentrated to give the title compound as a yellowish solid. The solid was dissolved in 10 mL of mixed solvent containing TFA (9.5 mL), TIPS (0.25 mL) and H₂O (0.25 mL). The reaction mixture was stirred at room temperature for 2 h. The solvent was removed by air flow and the residue was dissolved in CH₃CN/H₂O (1/1) for further HPLC purification using the same method described above. The fraction at 15.2 min was collected and lyophilized to give the title compound as colorless oil (41 mg, 81%). ¹H NMR (600 MHz, D₂O) δ 8.60 (d, J = 5.7 Hz, 2H), 8.50 (d, J = 4.9 Hz, 1H), 8.37–8.30 (m, 2H), 8.02 (s, 2H), 7.83 (td, J = 7.8, 1.6 Hz, 1H), 7.73 (t, J = 7.1 Hz, 2H), 7.45–7.39 (m, 1H), 7.32 (s, 1H), 4.32 (s, 2H), 4.16–3.91 (m, 4H), 3.83 (s, 2H), 3.73–3.42 (m, 8H), 3.30–2.94 (m, 8H). ¹³C NMR (151 MHz, D₂O) δ 178.19, 162.97 (q, J = 34.7 Hz, CF₃COOH), 156.69, 156.64, 147.95, 147.90, 147.82, 147.74, 147.64, 138.80, 138.78, 138.57, 124.37, 124.23, 124.20, 123.94, 123.65, 123.41, 116,3 (q, J = 276.3 Hz, CF₃COOH) 59.88, 59.68, 59.15, 58.40, 53.02, 52.09, 51.86, 51.46, 51.25, 51.00, 49.18. HR-MS: calcd. For C₂₈H₃₈N₇O₂ ([M + H]⁺): 504.3087; found: 504.2943.

IR and UV-vis titration of DOTA-xPy (x = 1–3)

A PerkinElmer Fourier transfer-infrared spectrometer (FT-IR) was used to obtain the spectra of three chelators and their respective metal complexes with Cu²⁺ metal ions, at 298 K, covering wavenumbers 650–4000 cm^− 1. The six samples had been dried to be solid samples as TFA salts. Samples were re-dissolved in ACN and dried three times to remove extra TFA. Approximately 2–5 mg of each sample was used to cover the 2 mm diameter crystal on the attenuated total reflectance (ATR) top plate. UV-vis absorbance was recorded on an Agilent Cary100 UV-visible spectrophotometer. 5 mg of each chelator was dissolved in water to make a solution of 0.26–29 mM. 0.2 eq. of Cu(II) aqueous solution (trace metal basis) was added for each titration. UV-vis spectra at 200–800 nm was recorded using a 1 cm pathlength quartz cell.

Production of ⁶⁴Cu

Isotopically enriched ⁶⁴Ni metal powder (50 mg) was dissolved in HCl. Aqueous ammonia was added to adjust the pH to ~ 9. ⁶⁴Ni is then electroplated onto a rhodium backing disk (35 mm diameter, 1 mm thickness) using a small custom plating apparatus overnight. The plated nickel layer was approximately 8 mm diameter × 110 μm thickness. The target was irradiated for 1 h at 10 μA proton current 13 MeV and left in the cyclotron for 2 h to allow for the decay of the short-lived isotopes. The major co-produced isotopes were ⁶¹Co (2.9%) and ¹⁰³Pb (2.0%) after cool-down. The ⁶⁴Ni layer of the target was dissolved in hot concentrated HCl. After evaporation to dryness, the residue was re-dissolved in 6 M HCl and loaded onto a 1 cm AG1-X8 solid extraction column pre-conditioned with 6 M HCl. The column was washed with 6 M HCl (10 mL) and ⁶⁴Cu was eluted with milli-Q water. Radionuclide purity (> 99%) was confirmed by gamma spectroscopy.

RadioTLC

RadioTLC was carried out using BioScan system 200 Image Scanner. iTLC-SA plates used as stationary phase, developed using 0.2 M EDTA (pH 5). Under these conditions uncomplexed ⁶⁴Cu^II migrates with the solvent front (R_f = 1), while the ⁶⁴Cu-complex remains at the baseline (R_f = 0).

Labeling of DOTA, DOTA-1Py, DOTA-2Py, DOTA-3Py and cyclen-4py (cyclen-4Py) with ⁶⁴Cu

To a solution of ligand (10 μL, 1 × 10^− 3 M) containing MES buffer (83 μL, 0.4 M, pH = 6.2) in an Eppendorf tube was added [⁶⁴Cu]CuCl₂ (7 μL 3 MBq). Final concentration of the ligand was 1 × 10^− 4 M. The resulting mixture was kept at room temperature for 30 min. After the reaction, radiochemical purity of the radio-ligands was confirmed by both radio-HPLC and radio-iTLC. HPLC method: Phenomenex Luna C18 column (150 mm × 4.6 mm), 1 mL/min. A: H₂O with 0.1% TFA, B: CH₃CN with 0.1% TFA, 0–5 min: 100% A isocratic; 5–25 min: 100%–60% A, 0–40% B. [⁶⁴Cu]Cu-DOTA-1Py, t_R = 12.5 min; [⁶⁴Cu]Cu-DOTA-2Py, t_R = 12.7 min; [⁶⁴Cu]Cu-DOTA-3Py, t_R = 14.5 min; [⁶⁴Cu]Cu-cyclen-4Py, t_R = 16.7 min.

Concentration dependence ⁶⁴Cu labeling

To aqueous solution of ligands (10 μL, 1 × 10^− 2 M, 1 × 10^− 3 M, 1 × 10^− 4 M, 1 × 10^− 5 M, 1 × 10^− 6 M, 1 × 10^− 7 M) in MES buffer (83 μL, 0.4 M, pH = 6.2) was added [⁶⁴Cu]CuCl₂ (3 MBq, 7 μL of 0.01 M HCl). Final concentrations of the ligand were 1 × 10^− 3 M, 1 × 10^− 4 M, 1 × 10^− 5 M, 1 × 10^− 6 M, 1 × 10^− 7 M, 1 × 10^− 8 M. The mixtures were kept at room temperature for 15 min and then analyzed by radio-HPLC and radio-iTLC using the conditions described above.

Kinetics of radiolabeling studies

To an aqueous solution of ligand (10 μL, 1 × 10^− 3 M) in MES buffer (83 μL, 0.4 M, pH = 6.2) was added [⁶⁴Cu]CuCl₂ (3 MBq, 7 μL in 0.01 M HCl). The reaction mixture was kept at room temperature. Aliquots of the mixtures were analyzed by radio-HPLC and radio-iTLC using the conditions described above to determine the radiolabeling yields.

Ligand competition radiolabeling

To equimolar aqueous mixtures of new ligand (10 μL, 1 × 10^− 3 M DOTA-xPy (x = 1–4)) and commercially available chelator p-SCN-Bn-NOTA (10 μL, 1 × 10^− 3 M NOTA-SCN) in MES buffer (73 μL, 0.4 M, pH = 6.2) was added [⁶⁴Cu]CuCl₂ (3 MBq, 7 μL in 0.01 M HCl). Final concentrations of new ligand and NOTA-SCN were 1 × 10^− 4 M in the reaction mixtures. The mixtures were kept at room temperature for 15 min and then analyzed by radio-HPLC using the same method described above to determine the labeling ratio of radiolabeled new ligand and [⁶⁴Cu]Cu-NOTA-SCN (t_R = 24.2 min).

Ligand challenge studies

To a solution of ligand (10 μL, 1 × 10^− 3 M) containing MES buffer (83 μL, 0.4 M, pH = 6.2) in an Eppendorf tube was added [⁶⁴Cu]CuCl₂ (7 μL 3 MBq). Final concentration of the ligand was 1 × 10^− 4 M. After 10 min, full complexation of the ligand was checked by radio-HPLC and radio-iTLC. NOTA-SCN (1000-fold excess) dissolved in water (50 μL) was subsequently added to the reaction solution, and stirred for 4 h at ambient temperature. The mixture was assayed by radio-HPLC to determine the decomposition. As an opposing experiment, an aqueous solution of NOTA-SCN (10 μL, 1 × 10^− 3 M) and [⁶⁴Cu]CuCl₂ (3 MBq, 7 μL in 0.01 M HCl) were subsequently added into MES buffer (83 μL, 0.4 M, pH = 6.2). After full complexation, DOTA-xPy (x = 1–4, 1000-fold excess) dissolved in water (50 μL) was added to the complex solution and the mixture was stirred for 4 h. The mixture was analyzed by radio-HPLC to determine the decomposition.

In vitro stability of radio-ligands

Pre-formed ⁶⁴Cu-labeled complexes (100 μL) were added to phosphate buffer (300 μL, pH 7.4) or human serum (300 μL). After 1 h or 24 h of incubation at 37 °C, the buffer samples were analyzed by semi-preparative HPLC. The serum samples were firstly treated with ethanol to precipitate and separate proteins (twice the volume of the mixture) and centrifuged for 1 min (13,200 rpm). The filtrate was decanted and further filtered (VWR sterile syringe filter, 0.2 μm PES), and subsequently analyzed by radio-iTLC using the conditions outlined above.

Preparation of DOTA-1Py-αMSH, DOTA-2Py-αMSH and DOTA-3Py-αMSH

Peptide synthesis for Fmoc-Pip-Nle-CycMSH_hex-resin (Fmoc-αMSH-resin) was performed as described in previously published procedures (Zhang et al. 2017). Subsequently, Fmoc was removed. (2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU, 5 eq.), hydroxybenzotriazole (HOBt, 5 eq.) and DOTA-xPy (x = 1–3, 5 eq.) were dissolved in DMF (minimum), then added to the resin (1 eq.) and initiated by the addition of N,N-diisopropylethylamine (15 eq.). Coupling was carried out for ~ 16 h at room temperature with shaking. The resin was washed extensively with DMF and DCM and solvent removed in a flow of N₂. The peptide was cleaved and deprotected by soaking the resin with a mixture of TFA/TIPS/water/phenol (90/2.5/2.5/5) and shaking at room temperature for 3 h. After ether precipitation and filtration, the filtrate was collected and dried by N₂. The residue was dissolved in CH₃CN and water and purified by semi-preparative HPLC using Phenomenex Luna C18 (250 mm × 100 mm) column at 3 mL/min with the following method: A: H₂O with 0.1% TFA, B: CH₃CN with 0.1% TFA; 0–5 min, 90% A, 10% B; 5-25 min, 90%–10% A, 10%–90% B. DOTA-1Py-αMSH, t_R = 14.9 min; DOTA-2Py-αMSH t_R = 15.1 min; DOTA-3Py-αMSH t_R = 15.4 min.

MALDI-TOF MS: DOTA-1Py-αMSH: calcd. 1541.8 for [C₇₅H₁₀₉N₂₂O₁₄]⁺ ([M + H]⁺), found: 1542.0. DOTA-2Py-αMSH: calcd. 1574.9 for [C₇₉H₁₁₂N₂₃O₁₂]⁺ ([M + H]⁺), found: 1574.9. DOTA-1Py-αMSH: calcd. 1607.9 for [C₈₃H₁₁₅N₂₄O₁₀]⁺ ([M + H]⁺), found: 1607.9.

Preparation of non-radioactive complexes (^natCu-DOTA-Py-MSH, ^natCu-DOTA-2Py-MSH and ^natCu-DOTA-3Py-MSH)

A solution of peptide (200 μL, 1 × 10^− 3 M) and CuSO₄ (100 μL, 1 × 10^− 2 M) in an Eppendorf tube was kept at room temperature for 2 h and subsequently purified by C-18 Sep-Pak (Waters). Non-radioactive title complexes were confirmed by ESI-MS: ^natCu-DOTA-1Py-αMSH: calcd. 801.9 for [C₇₅H₁₀₇⁶³CuN₂₂O₁₄]²⁺ ([M + 2H]⁺), found: 801.3. ^natCu-DOTA-2Py-αMSH: calcd. 818.4 for [C₇₉H₁₁₂⁶³CuN₂₃O₁₂]⁺ ([M + H]²⁺), found: 818.0. ^natCu-DOTA-1Py-αMSH: calcd. 834.9 for [C₈₃H₁₁₅⁶³CuN₂₄O₁₀]⁺ (M²⁺), found: 834.4. HPLC retention times were recorded using method: Phenomenex Luna C18 column (150 mm × 4.6 mm), 1 mL/min. A: H₂O with 0.1% TFA, B: CH₃CN with 0.1% TFA, 0–5 min: 100% A isocratic; 5–25 min: 100%–0% A, 0–100% B. ^natCu-DOTA-1Py-αMSH, t_R = 15.6 min; ^natCu-DOTA-2Py-αMSH, t_R = 16.3 min; ^natCu-DOTA-3Py-αMSH, t_R = 15.4 min.

⁶⁴Cu labeling of bioconjugates DOTA-1Py-αMSH, DOTA-2Py-COOH-αMSH and DOTA-3Py-αMSH

To a solution of bioconjugate (10 μL, 1 × 10^− 4 M) in MES buffer (83 μL, 0.4 M, pH = 6.2) was added ⁶⁴CuCl₂ (3 MBq, 7 μL). Final concentration of the bioconjugate was 1 × 10^− 5 M. The resulting mixture was kept at room temperature for 15 min. After the reaction, radiochemical purity of the radiolabeled bioconjugate was confirmed by both radio-HPLC and radio-iTLC. HPLC method: Phenomenex Luna C18 column (150 mm × 4.6 mm), 1 mL/min. A: H₂O with 0.1% TFA, B: CH₃CN with 0.1% TFA, 0–5 min: 100% A isocratic; 5–25 min: 100%–0% A, 0–100% B. [⁶⁴Cu]Cu-DOTA-1Py-αMSH, t_R = 15.9 min; [⁶⁴Cu]Cu-DOTA-2Py-αMSH, t_R = 16.7 min; [⁶⁴Cu]Cu-DOTA-3Py-αMSH, t_R = 15.7 min.

In vitro stability of radiolabeled peptides

Pre-formed ⁶⁴Cu-labeled bioconjugate (100 μL) was mixed in phosphate buffer (300 μL, pH 7.4) or human serum (300 μL). After 1 h or 4 h of incubation at 37 °C, the buffer samples were analyzed by semi-preparative HPLC. The serum samples were firstly treated with ethanol to precipitate and separate proteins (twice the volume of the mixture) and centrifuged for 1 min (13,200 rpm). The filtrate was decanted and further filtered (VWR sterile syringe filter, 0.2 μm PES), and subsequently analyzed by radio-iTLC using the conditions outlined above.

Binding affinity of ^natCu-DOTA-xPy towards MC1R

K_i was measured using the same method previously reported (Zhang et al. 2017). Briefly, 5 × 10⁵ B16F10 cells/well were seeded in complete growth medium in a 24 well Poly-D-Lysine plate (Corning BioCoat™, Ref No. 354414) and grown to 85–90% confluency at standard conditions. The growth medium was replaced with reaction medium (RPMI, 2 mg/ml BSA, 20 mM HEPES) and the non-radioactive ligand added at a final concentration of 0.5 pM to 5 μM, along with 0.1 nM [¹²⁵I]I-αMSH. The cells were incubated for 1 h at 25 °C with mild shaking. The supernatant was removed and cells were washed twice with ice-cold PBS. Cells were harvested with trypsin and counted on WIZARD 2480 gamma counter. The measurements were repeated in three separate experiments and reported as mean ± standard deviation.