Novel Hybrid Benzoazacrown Ligand as a Chelator for Copper and Lead Cations: What Difference Does Pyridine Make.

A synthetic procedure for the synthesis of azacrown ethers with a combination of pendant arms has been developed and the synthesized ligand, characterized by various techniques, was studied. The prepared benzoazacrown ether with hybrid pendant arms and its complexes with copper and lead cations were studied in terms of biomedical applications. Similarly to a fully acetate analog, the new one binds both cations with close stability constants, despite the decrease in both constants. The calculated geometry of the complexes correlate with the data from X-ray absorption and NMR spectroscopy. Coordination of both cations differs due to the difference between the ionic radii. However, these chelation modes provide effective shielding of cations in both cases, that was shown by the stability of their complexes in the biologically relevant media towards transchelation and transmetallation.

1. Introduction

Given the growing importance of nuclear medicine and the extending abilities of obtaining medically relevant radiometals, the design of functional polyazamacrocyclic ligands able to coordinate cations with suitable features has been attracting increased attention during the last decade. In the frame of systematic variation and the search for the best matching pairs between medical radiometals and chelators, we have synthesized those substituted with one type of arm and arm-free pyridine bisamide azacrown and benzoazacrown ethers [1,2,3,4,5,6]. We have shown that azacrown compounds with acetic arms are effective for binding bismuth [4,5], lead [7] and zinc [8] cations, regarding the nuclear medicine application of their radioisotopes. However, these ligands, including H3BA3A (Figure 1)—the closest analog of the ligand presented here, were unable to strongly chelate Cu2+. Given Cu2+ is a transition metal cation, preferring softer N-donors along with O-donor atoms, it can be effectively bound by aromatic nitrogen that is known to be less basic than aliphatic amines. Moreover, it has been recently shown that H4DOTA derivatives with a pyridine arm can significantly improve copper binding [9].

Figure 1

Ligands H3BA3A and H2BA2A1Py.

To the best of our knowledge, protonation and complexation constants of such hybrid ligands with the simultaneous presence of acetic and pyridyl pendant groups have not been described in the literature yet. There are examples of substitutions of all (two or three) acetates in azacrowns by pyridines [3,6,10,11] and it has been shown that pyridine, as a pendant arm, can significantly improve the chelation ability and kinetics of binding towards Cu2+ and Bi3+, including direct comparison with well-known H4DOTA [9,12,13].

In this work, we propose a new method for the synthesis of an azamacrocyclic ligand containing two types of additionally introduced coordinating acetic and pyridine groups (H2BA2A1Py, Figure 1). The Cu2+ and Pb2+ complexes with a novel ligand were studied by potentiometric methods, NMR methods, X-ray absorption spectroscopy and quantum-chemical calculations. In order to evaluate the features of radiolabeled complexes, radiolabeling with 64Cu and 210Pb isotopes has been analyzed.

2. Results

2.1. Synthesis

The synthesis consists of the following sequence: (1) introduction of one type of substituent in the bisamide azacrown compound; (2) reduction in amide groups; (3) introduction of another type of substituent into the obtained amino derivative (Scheme 1).

Scheme 1

Reagents and conditions: (a) 2-(chloromethyl)pyridine hydrochloride, K2CO3, MeCN, reflux; (b) (1) BH3 THF, 0 °C, Ar, (2) HCl, H2O, reflux, (3) KOH, H2O, 0 °C, Ar; (c) tert-butyl bromoacetate, K2CO3, KI, MeCN, reflux; (d) H2O, reflux.

Bisamide, of 15-membered benzoazacrown compound 1, was alkylated with chloromethylpyridine according to the method described earlier [6]. The resulting pyridyl derivative 2 was reduced using a borane–THF complex to azamacrocycle 3, containing one side chelating substituent. In this case, the amide groups were converted into amino groups, into which the second type of substituents was further introduced. The derivative 3 was alkylated with tert-butyl bromoacetate in acetonitrile in the presence of a base. At the last stage, tert-butyl ester 4 was hydrolyzed by boiling in water without the addition of any catalysts. The prepared ligand was characterized by 1H and 13C NMR, MS, elemental analysis (ESI) and potentiometric titration.

2.2. Ligand Protonation

Benzoazacrown ether H2BA2A1Py possesses several centers for proton coordination of a different nature including aliphatic and aromatic N atoms and carboxylate groups. For the determination of the protonation constants, the solution of the ligand was titrated potentiometrically with NaOH, using NaClO4 or KNO3 as background electrolytes. The obtained values of the protonation constants are given as logarithms in Table 1 and the corresponding pH distribution diagram of ligand protonated forms are shown in Figure 2.

Table 1

Protonation constants of BA2A1Py2− and the closest analog BA3A3− and stability constants of the studied complexes.

Cation	Complex a	BA2A1Py2−		BA3A3− [7,8]
		0.1 M NaClO4	0.1 M KNO3	0.1 M KNO3
H+	HL	9.0 (1)	9.3 (1)/9.3	9.9 (1)/9.9
	H2L	16.7 (1)	17.0 (1)/7.7	17.7 (1)/7.8
	H3L	21.2 (1)	20.8 (1)/3.8	21.2 (1)/3.5
	H4L	22.9 (1)	22.5 (1)/1.7	23.8 (2)/2.6
Cu2+	HLCu	21.2 (1)	20.0 (1)	20.4 (1)
	LCu	16.2 (2)	16.3 (1)	16.8 (1)
	LCuOH	18.8 (2)	18.6 (1)
Pb2+	HLPb	-	18.6 (1)	20.1 ± 0.1
	LPb	-	15.6 (1)	17.2 ± 0.1
	LPbOH	-	20.2 (2)	23.1 ± 0.1
	LPb(OH)2	-	22.4 (2)	26.3 ± 0.1

charges are omitted for clarity.

Figure 2

pH distribution diagrams for BA2A1Py.

The fully pyridine-substituted crowns described in the literature are featured by a decrease in overall basicity [3,6,10,11], but the value of the deviation of constants of the protonation steps appeared to be a function of a cumulative effect of the ligands’ structure, charge and intramolecular H-bond formation. It can be assumed that the first and the second protonation steps of BA2A1Py2− are associated with symmetric macrocyclic amino groups as in an arm-free azacrown ether [14] while the third one is presumably related to a middle pendant arm, as in the triacetate H3BA3A described earlier [8]. A diprotonated form prevails in the widest range from pH 4 up to pH 7.5 (Figure 2). Furthermore, the last detected protonation constant is lower than for the triacetic ligand, correlating with the total charge of triprotonated species for H3BA3A and H3BA2A1Py+. The latter is already 1+ charged compared to the neutral H3BA3A and consequently possesses less affinity for protons than H3BA3A. Moreover, compared to acetate H3BA3A, the presence of pyridyl as a pendant arm decreases the overall negative charge of fully deprotonated species, that is reflected in a slight decrease in the first protonation constant. Finally, overall, a less basic character of H2BA2A1Py is observed.

The proposed model for ligand protonation is confirmed by the 1H NMR spectra of the ligand in D2O at various pDs, corresponding to the maximum percentage of formation of certain protonated forms in solution (Figure 2). The proton resonances of the ligand at pD 11.2 correspond to the deprotonated form BA2A1Py (Figure 3). At pD 8.5, the monoprotonated species HBA2A1Py was formed and the macrocyclic proton signals in the spectrum are shifted to the downfield region, which is explained by the rapid proton exchange between the three amino groups of the ligand. Following acidification to pD 5.7, this leads to the formation of the diprotonated species H2BA2A1Py, causing the most pronounced resonance changes for macrocyclic protons H5, H6 and protons of acetate arms H8 (Table S1), indicating the localization of two H+ on the amino groups conjugated with acetate arms (Figure 3). At pD 2.6, the protons of the central arm H10 and the pyridine ring’s H12–15 exhibit a significant downfield shift (Figure 3, Table S1), which clearly indicates that the third proton is attached to the pyridine nitrogen atom, forming species H3BA2A1Py.

Figure 3

1H NMR spectra of ligand H and its complex with Pb2+ at various pD values in D2O.

2.3. Complexation Study

For the determination of complex formation constants, solutions of metal perchlorates, perchloric acid and ligands with different concentrations of the components were titrated with NaOH. A potentiometric titration of ligand in the presence of equimolar concentrations of Cu2+ and Pb2+ and a two-fold excess of Cu2+ were performed. The complex formation constants were calculated using Hyperquad software [15]. During the determination of the stability constant values, the protonation constants of the ligand and the ionic product of water were fixed, and the hydrolysis constants of the cations were considered as known, and no attempts were made to adjust their values. The time required for equilibration in the potentiometric titration experiments was within 5 min. The obtained values are presented in the Table 1 and the corresponding pH distribution diagram of cations’ forms are shown in Figure 4. Variation of the Cu2+/L ratio led to the same values of logβ for LCu and HLCu: 16.24 (4) and 19.88 (2), respectively. The formation of hydroxy species with limited solubility in water hampers equilibrium constant determination. That is why, due to the formation of an insoluble hydroxide of excess of Cu2+ (for the experiment with Cu2+/L = 2/1), titration points over pH 5.5 were excluded from the calculation of complex formation constants and, consequently, the logβ of LCuOH form was not calculated from this experiment.

Figure 4

pH distribution diagrams for complexes of Cu2+ (a) and Pb2+ (b) with BA2A1Py.

Both the cation formation of protonated, as well as hydroxyl complex species, is detected in the wide range of pH levels (Figure 4). Moreover, as a larger cation with a higher coordination number, Pb2+ forms species with one and two OH− groups. The lower overall basicity of H2BA2A1Py compared to H3BA3A induced a consequent decrease in the logβ of complexes with BA2A1Py, although to a different degree: only 0.5 log for Cu2+, while 1.6 log for Pb2+ (Table 1), showing a preference towards Cu2+. This fact proves our suggestion about an increase in affinity for Cu2+ of N-donor ligands and correlates with the described CuDOTAPy in vitro stability [9].

Noteworthy, the application of NaClO4 and KNO3 as background electrolytes allowed to compare the obtained data with the literature and ensure the lack of a crucial effect on the variation of background cations to the value of the logβ(ML), at least in the pH range and at the ionic strength used. The considered conditions fairly afford the assumed biologically relevant media. The obtained values demonstrate slight changes in the formation of protonated species H3BA2A1Py+, H4BA2A1Py2+ and HBA2A1PyCu+, apparently due to competition of Na+/K+ with H+ for association to carboxylic groups. However, this influence is negligible when protonation and coordination mainly occur via amino groups.

2.4. Complexes’ Structure Studies

In order to evaluate the structure of these complexes in the solution, we performed DFT calculations of the geometry of both complexes (Figure 5) and compared the interatomic distances with the coordination environment obtained, by the fitting of an X-ray absorption spectra of complexes (Table 2, Figure 6).

Figure 5

Optimized geometries of complexes of BA2A1Py with Cu2+ (a) and Pb2+ (b).

Table 2

Structural parameters refined from fitting the EXAFS Cu K- and Pb L3- edges spectra for CuBA2A1Py and PbBA2A1Py complexes solutions at varied pHs.

Sample	N/O			C/N/O			C/N/O			C/N/O
	N	R, Å	σ, Å	N	R, Å	σ, Å	N	R, Å	σ, Å	N	R, Å	σ, Å
CuBA2A1Py_pH2.3	4.0	2.00	0.011	4.6	2.72	0.005	6.9	2.94	0.005	1.6	3.29	0.005
CuBA2A1Py_pH7.6	4.0	1.99	0.010	3.4	2.73	0.005	5.7	2.92	0.005	2.2	3.41	0.005
PbBA2A1Py_pH2.4	5.5	2.54	0.014	6.6	3.31	0.009	-
PbBA2A1Py_pH5.1	6.1	2.55	0.016	8.6	3.31	0.009

Figure 6

The fitted spectra for X-ray absorption of copper (a) and lead (b) complexes with H2BA2A1Py in the R- and k—spaces.

According to the geometries obtained in both complexes a macrocyclic cavity is opened, i.e., unfolded, and depending on the ionic radius, a cation is either encapsulated inside the cavity—Cu2+, or coordinated above—Pb2+ (Figure 5).

Cu2+ is held by two acetates from one side of the macrocycle, and the pyridyl arm from another side, forming distorted octahedral coordination. The shortest distances (~2 Å) to three macrocyclic amino groups and one acetate group, correspond to the equatorial plane, while another acetate and pyridine group can be attributed to axial positions at 2.30 and 2.76 Å, respectively. These ranges of distances are detected in the closest distance to the Cu2+ shells upon EXAFS fitting: 2, 2.7 Å. Taking into account that the nearly similar atomic scattering factors of C, N and O do not allow the atoms to be distinguished in the local surrounding of the Cu atom, the total high number (N) of neighbors obtained is fair. Noteworthy is the number of atoms in the second and third coordination shells of Cu2+, which looks to decrease by one upon a pH increase (Table 2). Apparently it is caused by conversion of CuLH (pH2.3) to CuL (pH7.6) with a slightly higher organized arrangement in the latter structure.

In the calculated structure, a larger Pb2+ cation is chelated above the cavity by macrocyclic amino groups, and all pendant arms with the interatomic distances 2.3–3.4 Å to N and O donor atoms. Atoms on these ranges are found in the experimental EXAFS spectra in both cases at pH 2.4 and 5.1.

In contrast to CuBA2A1Py, the number of closest neighbors increases upon conversion from the protonated form to the deprotonated form. Assumingly, it is provided by the formation of the protonated form due to the cation’s association with acetates, while in the PbBA2A1Py complex, both pendant arms and the macrocycle are in the vicinity of the cation (this suggestion correlates with the NMR results described below). The latter causes a higher number of atoms at 3.3 Å. Moreover, apparently symmetric environments of Cu2+ vs. Pb2+ affects the possibility to distinguish intermediate distances to neighboring atoms, that is not the case for Pb complexes.

2.5. NMR Spectroscopy of PbBA2A1Py

Furthermore, the structural features of the PbBA2A1Py complex in an aqueous solution were specified by NMR spectroscopy. The studies were carried out in D2O by comparing the 1H NMR spectra of the free ligand and the ligand in the presence of the Pb2+ cation. Changes in the 1H NMR spectrum observed, after the addition of the Pb2+ ion to the D2O solution of H2BA2A1Py, demonstrated geminal coupling of the methylene protons belonging to the macrocyclic fragment, as well as the pendant side arms. As a result, the eleven proton signals of the free ligand BA2A1Py2− transform to sixteen proton signals (Figure 3). This behavior is consistent with the presence of a single complex of PbBA2A1Py with a relatively rigid structure in the solution; fluxional interconversion at ambient temperatures was not observed (on the NMR time scale).

Additionally, no change in the number of peaks in the 13C NMR spectrum of the ligand H2BA2A1Py was found upon Pb2+ coordination, suggesting no chemically distinct isomers were formed (Figure S21). In the PbBA2A1Py complex spectrum, the resonances of the protons of the macrocycle, benzene, and pendant arms downfield change because of the polarization effect of the metal ion (Table S1). At the same time, some signals exhibit upfield shifts. This is probably due to the diamagnetic through-space shielding effect on the chemical shifts of the protons located under the surrounding carbonyl or pyridyl units. Moreover, the large upfield shift of the acetate group signal H8y indicates the shielding effect of the pyridine ring located close to them. The signal of the pyridine proton H15 has an insignificant downfield shift compared to the other pyridine signals, which can also confirm the close location of the acetate groups to the pyridine ring, since, in this case, the polarization effect of the metal cation is neutralized by the shielding effect of the carbonyl groups. The fact, that the methylene protons of the pyridyl substituent H10 appear as a singlet, can be explained by the location of this substituent in the plane of symmetry, which is perpendicular to the macrocycle and, therefore, the geminal protons H10 are magnetically equivalent. Considering that the diameter of the Pb2+ cation (2.40 Å) is larger than the size of the macrocyclic cavity (1.7–2.2 Å), it can be concluded that the Pb2+ cation is fixed by pyridyl and two acetate groups above the macrocyclic cavity, correlating with the calculated geometry, and the resulting complex PbBA2A1Py possesses Cs symmetry in an aqueous solution.

When the solution of the complex of the ligand with the Pb2+ cation is alkalized to pD 10.5, in accordance with the data of potentiometric titration, a hydroxo- complex PbBA2A1PyOH is formed. The observed 1H NMR spectrum almost completely coincides with the spectrum of the complex PbBA2A1Py (Figure 3), which means that the ligand geometry does not change. Most likely, the Pb2+ cation in the complex coordinates the water molecule, which in an alkaline medium turns into an OH− group. Upon acidification of the solution of the complex to pD 2.4, in the 1H NMR spectrum, the signals of the macrocyclic protons H5, H6, and H7, as well as the protons of the acetate arms H8, exhibit a slight downfield shift relative to the spectrum of the complex PbBA2A1Py, while the overall shape of the spectrum does not change (Figure 3, Table S1). It can be concluded that, upon protonation, H+ adds to the side amino groups of the macrocycle, while the Cs symmetry of the complex PbHBA2A1Py is retained. The latter can be caused by coordination of cations via acetates without participation of the protonated macrocycle. A similar structure of the protonated form of the complex is also characteristic of H4DOTA protonated intermediate complexes [16,17].

2.6. Ligand Lipophilicity

As one of the important features for molecules to be used as part of radiopharmaceuticals, we evaluated the lipophilicity of the molecule as the partition coefficient of BA2A1Py labelled by 210Pb between n-octanol and water. Due to the aromatic pendant arm, the obtained value logP = −2.11 is higher than for BA3A3− (logP = −2.26). Of note, benzene causes a slight decrease in polarity of benzoazacrowns, compared to the smaller dipyridyl-acetic cyclononane derivative—NO1A2Py logP = −2.43 [12].

2.7. Radiolabeling of Complexes

Radiolabeled complexes were synthesized with 64Cu and 210Pb. To simulate the conditions of preparation of a highly active radiopharmaceutical we added the carrier of CuCl2 to reach the concentration of cations of 10 nM, as soon as 1 kBq of 64Cu provided only a picomolar concentration of cations, while for 210Pb, a 10 nM content of Pb2+ was achieved with applied activity. The low chemical content of the cations with sufficient radioactivity, due to the short half-life of medically relevant radionuclides, requires the concentration of ligands for a >95% radiolabeling yield to be determined. Moreover, the presence of additives as pH-adjusting or buffering agents, e.g., acetates, hydrochloric acid, and corresponding impurities, even in highly purified commercially available sources, in the amounts exceeding the radiometal’s content, also impacts the thermodynamic equilibrium. Finally, in highly diluted solutions with low contents of both reactants, i.e., radiometal and ligand, kinetic hindrance for the complexation can also take place.

In the considered systems, a high radiochemical yield of the complexation reaction above 95% was achieved at the H2BA2A1Py concentration of 50 μM (Table 3) for both [210Pb]Pb2+ and [64Cu]Cu2+. Competitive chelation of radiometals by an excess of acetate was initially simulated according to the complexation constants in applied concentrations (Figure S25). It was shown that, even such a significant prevalence of the concentration of acetate can hamper the binding by ligands only at low pH values. It was directly confirmed with regard to the same values of radiolabeling yield obtained for some concentration points with and without NaOAc (Table 3).

Table 3

Labeling efficiency (%) of complexes at different ligand concentrations.

c(H2BA2A1Py), μM	PbL, %, 0.15 M Ac (No Ac)	CuL, %, 0.15 M Ac
1000	100 ± 7	97 ± 9
500	100 ± 9	92 ± 8
200	99 ± 9	–
100	100 ± 8 (99%)	99 ± 2
80	99 ± 7	–
50	95 ± 9 (99%)	96 ± 3
20	73 ± 7	–
10	62 ± 3	34 ± 2

It is important to note that in the labeling experiments a solution of the complex was applied on TLC plates immediately after the mixing of the reactants at room temperature. These solutions were additionally analyzed for radiolabeling efficiency after 30 min and 24 h. The same value of radiolabeling yield, irrespective of the duration of synthesis, indicated the rapid kinetics of cation binding (less than a minute).

The formation of radiolabeled complexes, as well as the chemical identity of labeled and stable complexes, was confirmed by TLC and HPLC analysis. The retention time of an LPb stable complex and [210Pb]LPb with or without the addition of NaOAc prepared at pH 5–6.5 were the same (tR = 5.2 ± 0.2 min, Figure S26) and proved the identity of the stable complex in a concentration of 2 mM and radiolabeled with c(Pb2+) = 10 nM. Additionally, this experiment demonstrated not only the lack of competition of acetate with BA2A1Py for the chelation of radiometals, but also dismissed the possibility of the formation of ternary complexes between radiometal, acetate and ligand that could take place regarding a high content of acetate anion in the radiolabeling process (compared to the ligand and cation). Similarly, in applied TLC conditions, Rf of free [210Pb]Pb2+ or [64Cu]Cu2+ in the presence of 0.15 M acetate, diverged significantly from ligand-bound species. Additionally, Rf and the labeling yield for chelates was the same for radiolabeled complexes prepared with or without 0.15 M AcONa, indicating the formation of the same species irrelative to the presence of acetate.

2.8. Stability of Radiolabeled Complexes in Biologically Relevant Media

Taking into account the changes in the stability constants with the cations considered, it is expedient to compare the stabilities of radiolabeled complexes of BA2A1Py with Cu2+ and Pb2+ in the competing medium, such as serum proteins, because BA3A3− complexes of Cu2+ and Pb2+ have shown that only the latter was stable enough in this challenging environment, despite the close values of logβ [7,8].

In the nine-fold excess of fetal bovine serum, both complexes of [64Cu]CuBA2A1Py and [210Pb]PbBA2A1Py have shown a lack of radionuclide loss during 4 h and 2 days, respectively (Figure 7). Comparing the obtained data with earlier published results on complexes with BA3A3− [7,8] it is evident that, despite close complexation constants for CuBA3A and CuBA2A1Py, the stability of the latter radiolabeled chelate is higher. This showed up even after 1 h of incubation and was well pronounced after 4 h. Wherein, both complexes of Pb2+ are stable after 2 days of incubation despite more than one order of difference of logβ (Table 1).

Figure 7

Stability of Cu2+ (a) and Pb2+ (b) complexes with BA2A1Py2− compared to complexes with BA3A3− in the presence of serum proteins. Data on M-BA3A complexes are taken from earlier published papers [7,8].

Additionally, the stability of complexes was evaluated in the presence of biologically relevant cations such as Fe3+, Zn2+, Mg2+, Ca2+ and Cu2+ in aqueous solutions and in an isotonic solution of 0.15 M NaCl. The intactness of complexes after incubation for 24 h has been shown by TLC and gamma spectrometry. No loss of radionuclide, in both cases, was detected, meaning the stability of the complexes towards not only transchelation, but also transmetallation.

High stability of both considered complexes with H2BA2A1Py despite different values of logβ and coordination modes of cations in both complexes should be explained according to individual peculiarities. According to all the obtained results, it is evident that the copper complexes’ stability was mainly governed by the presence of a softer donor atom, while lead complexes’ stability cannot be affected by such substitution even though a complexation constant decreased by a 1.6 order of magnitude. Moreover, shielding of cations from surrounding agents due to pendant arms seems to play an important role for in vitro stability, irrespective of coordination inside or above the macrocycle.

3. Discussion

Earlier we have developed a series of azacrown ethers with an aryl substituent in the macrocycle. It was shown that 15-azacrown-5 and 18-azacrown-6 with acetate pendant arms, are appropriate for various medically relevant radiometals [4,5,7,8], but are inappropriate for Cu2+. Additionally, the complete substitution of acetate by pyridyl induced a decrease in the water-solubility of the ligands [3]. In this regard, and based on the recent results of DOTAPy complexes with Cu2+ [9], in this work we proposed an original and effective method for the synthesis of a hybrid benzoazacrown compound bearing acetate and pyridine pendant arms, providing a water-soluble character and stronger chelation of the copper cation.

Protonation and complexation constants allowed us to fully describe the ligand and Pb2+ complex species by NMR spectroscopy. The protonation model is close to the acetate analog H3BA3A with macrocyclic symmetric amines, to be the most basic, and the third proton to be associated with the middle pendant arm [8]. Additionally, it is demonstrated that in the acidified solution in the complex, an PbHBA2A1Py cation is held by acetates lacking coordination by the macrocyclic amino groups, similar to the complexation process described for H4DOTA [16,17]. Wherein, obviously, due to the larger size of the azacrown cavity providing faster deprotonation of macrocyclic amines, H2BA2A1Py forms complexes almost immediately, that is confirmed by radiolabeling experiments. These experiments were performed in the pH range where ML complexes should be formed, and according to the calculated structure, supported by EXAFS measurements, Cu2+ is coordinated inside the macrocycle BA2A1Py capped by pendant arms from both sides, whereas, Pb2+ is located above the benzoazacrown ether cavity chelated by all pendant groups from one side, respectively.

Finally, radiolabeled complexes of BA2A1Py and BA3A3− with 210Pb turned out to be stable towards transchelation and transmetallation. In the case of 64Cu-radiolabeled complexes with BA3A and BA2A1Py, only the latter has demonstrated stability in a competing medium, opening perspectives of such a kind of variation of pendant arms for Cu2+ chelation. Moreover, regarding the tuning of the water-soluble character of the ligand, the partition coefficient between octanol/water showed a slight decrease in hydrophilicity in H2BA2A1Py, compared to H3BA3A.

4. Materials and Methods

All commercially available reagents were used without further purification. The progress of reactions was followed with TLC using aluminum oxide (Merck, Darmstadt, Germany, 60 F254, neutral). Bisamide crown compound 1 [18] and pyridyl derivative 2 [6] were synthesized following the literature procedures.

1H and 13C NMR spectra were recorded at 25 °C on Bruker Avance 500, Varian Inova 400 MHz spectrometers (Bremen, Germany). Chemical shifts for 1H and 13C are reported in parts per million (δ) relative to deuterated solvent as an internal reference (CDCl3 1H δ = 7.27 and 13C δ = 77.00; D2O δ = 4.75). Spin–spin coupling constants for the proton spectra (J) are given in hertz (Hz). Spectral assignments were based, in part, on the two-dimensional NMR experiments (COSY, NOESY, HSQC and HMBC). Electrospray ionization mass spectrometry (ESI-MS) analyses were performed using a Finnigan LCQ Advantage (Thermo Fisher Scientific, Waltham, MA, USA) mass spectrometer equipped with an octopole iontrap mass-analyzer, an MS Surveyor pump, a Surveyor auto sampler, a Schmidlin-Lab nitrogen generator (Neuheim, Switzerland) and MestReNova (14.2.0-26256, Santiago de Compostela, Spain, 2020) for data collection and processing. Direct infusion of the sample solution was used. Positive electrospray ionization was achieved using an ionization voltage at 4.5 kV at a temperature of 200 °C. Electrospray full scan spectra in the range m/z 100–2000 was obtained by infusion at 0.05 mL·min–1 of 50 μM acetonitrile solutions of compounds 3, and 4 and aqueous solutions of the free ligand H2BA2A1Py, and in the presence of 2 eq. of Cu(ClO4)2 and Pb(ClO4)2. Melting points were determined on a Mel-temp II apparatus in open capillary tubes.

Compound 3

4.1. Synthesis and Analysis of Compounds

Bisamide crown compound 2 (309 mg, 0.95 mmol) was dissolved in 1 M BH3·THF solution (12 mL) at 0 °C and stirred under an inert atmosphere overnight. Excess BH3∙THF was destroyed by adding of MeOH (10 mL). The solvent was evaporated under vacuum. Then, 1 M HCl (10 mL) was added to the residue and the reaction mixture was refluxed for 8 h. The solution was washed with CH2Cl2, and then the pH was adjusted to 9–10 by adding NaOH. The product was extracted with CH2Cl2. The solvent was evaporated under vacuum to give a brown oil (206 mg, yield 61%). 1H NMR (CDCl3, 400 MHz, 25 °C): 2.73 (t, 4H, H(7), J = 4.3, 5.1), 2.84–2.89 (m, 8H, H(5,6)), 3.79 (s, 2H, H(8)), 4.10 (t, 4H, H(4), J = 4.3, 4.7), 6.97 (s, 4H, H(1,2)), 7.08 (t, 1H, H(12), J = 5.9, 6.3), 7.48 (t, 1H, H(11), J = 7.0, 7.8), 7.52 (d, 1H, H(10), J = 7.4), 8.45 (d, 1H, H(13), J = 4.3). 13C NMR (CDCl3, 400 MHz, 25 °C): 47.16 (C-7), 48.82 (C-5), 54.97 (C-6), 60.60 (C-8), 69.22 (C-4), 115. 39 (C-2), 122.02 (C-1,12), 122.92 (C-10), 136.96 (C-11), 148.94 (C-13), 149.34 (C-3), 160.12 (C-9). MS (ESI): m/z [M + H]+ calcd for C20H28N4O2 + H+: 357.43; found: 357.3. Anal. Calcd for C20H28N4O2·3H2O: C, 58.52; H, 8.35; N, 13.65%. Found: C, 58.40; H, 8.51; N, 13.58%.

Compound 4

A solution of tert-butyl bromoacetate (363 mg, 1.86 mmol) in MeCN (5 mL) was added to a mixture of 3 (332 mg, 0.93 mmol), K2CO3 (513 mg, 3.72 mmol) and KI (308 mg, 1.86 mmol) in MeCN (15 mL) and refluxed for 15 h. The solvent was evaporated under vacuum and the residue was dissolved in H2O and extracted with CHCl3. The crude product obtained by concentration of the combined organic extracts was purified by column chromatography (Al2O3, EtOAc/MeOH 10:1). The product was obtained as a brown oil (250 mg, yield 46%). 1H NMR (CDCl3, 400 MHz, 25 °C): 1.41 (s, 18H, H(11)), 2.77 (t, 4H, H(7), J = 5.9, 7.0), 3.02 (t, 4H, H(6), J = 6.6, 7.4), 3.13 (t, 4H, H(5), J = 3.9), 3.27 (s, 4H, H(8)), 3.79 (s, 2H, H(12)), 4.06 (t, 4H, H(4), J = 3.9), 6.82–6.88 (m, 4H, H(1, 2)), 7.12 (t, 1H, H(16), J = 5.9, 6.3), 7.46 (d, 1H, H(14), J = 7.8), 7.60 (t, 1H, H(15), J = 7.4), 8.48 (d, 1H, H(17), J = 5.1). 13C NMR (CDCl3, 400 MHz, 25 °C): 28.10 (C-11), 52.91 (C-7), 52.94 (C-6), 53.43 (C-5), 56.93 (C-8), 61.69 (C-12), 68.33 (C-4), 80.77 (C-10), 112.20 (C-2), 120.63 (C-1), 121.73 (C-16), 122.82 (C-14), 136.28 (C-15), 148.56 (C-3), 148.80 (C-17), 160.16 (C-13), 170.91 (C-9). MS (ESI): m/z [M + H]+ calcd for C32H48N4O6 + H+: 585.37; found: 585.5; m/z [M + Na]+ calcd for C32H48N4O6 + Na+: 607.35; found: 607.48. Anal. Calcd for C32H48N4O6·2H2O: C, 61.91; H, 8.44; N, 9.03%. Found: C, 61.85; H, 8.53; N, 9.08%.

Compound H2BA2A1Py

Ditert-butyl ester 4 (250 mg, 0.43 mmol) was dissolved in H2O (10 mL) and refluxed for 18 h. The solvent was evaporated under vacuum. The product was obtained as a dark yellow solid (199 mg, yield 98%). Mp.165–168 °C. 1H NMR (D2O, 500 MHz, 25 °C): 2.99 (br.s, 4H, H(7)), 3.61 (br.s, 4H, H(6)), 3.64 (br.s, 4H, H(5)), 3.68 (br.s, 2H, H(10)), 3.77 (s, 4H, H(8)), 4.31 (br.s, 4H, H(4)), 6.99 (br.s, 4H, H(1,2)), 7.29 (t, 1H, H(14), J = 5.3, 7.2), 7.41 (d, 1H, H(12), J = 6.0), 7.75 (t, 1H, H(13), J = 7.3), 8.35 (d, 1H, H(15), J = 4.6). 13C NMR (D2O, 500 MHz, 25 °C): 49.10 (C-7, C-10), 53.37 (C-5, C-6), 55.90 (C-8), 61.50 (C-4), 112.93 (C-2), 122.27 (C-1), 123.69 (C-14), 124.94 (C-12), 139.07 (C-13), 146.52 (C-3), 148.07 (C-15), 154.72 (C-11), 169.69 (C-9). MS (ESI): m/z [M + H]+ calcd for C24H32N4O6 + H+: 473.24; found: 473.3; m/z [M + Na]+ calcd for C24H32N4O6 + Na+: 495.22; found: 495.19.; m/z [M + K]+ calcd for C24H32N4O6 + K+: 511.2; found: 511.2. Anal. Calcd for C24H32N4O6·2,5H2O: C, 55.69; H, 7.21; N, 10.83%. Found: C, 55.68; H, 7.12; N, 10.55%.

4.2. Potentiometric Titration

Potentiometric titration was performed using an 848 Titrino Plus autotitrator (Metrohm, Herisau, Switzerland) equipped with a 5 mL autoburette and Metrohm combined pH glass electrode (model 60262100). A constant temperature of 25.0 ± 0.1 °C in the titration vessel was maintained with a thermostat throughout the experiments. Nitrogen flow through the titration vessel was used in all experiments. A combined glass electrode was calibrated by titration of a preliminarily standardized solution of HClO4 by NaOH solution with a known concentration. The calculation of the standard electrode potential, E0, and the slope of electrode function, was performed using the GLEE program (glass electrode evaluation) according to the Gran method [19]. The ionic product of water, pKw = 13.77, was taken for 0.1 M KNO3. For determination of the ligand’s protonation constants, potentiometric titration of a 0.001 M H2BA2A1Py solution in the presence of 0.01 M HClO4 was carried and the protonation constants were calculated with the HyperQuad2008 program. To determine the stability constants of Pb2+ and Cu2+ complexes with H2BA2A1Py, titration of a 0.001 M ligand solution was carried out in the presence of an equimolar amount of a Pb2+ or Cu2+, or two-fold excess of Cu2+, and a 0.01 M HClO4. To maintain constant ionic strength during an experiment a solution of 0.1 M KNO3/NaClO4 was used as a background electrolyte. The stability constants of complexes were calculated with the HyperQuad2008 program using the hydrolysis constants, the ionic product of water, the preliminarily determined ligand’s protonation constants and Eo value [15].

4.3. NMR Study of Ligand H2BA2A1Py Protonation and Complex Formation with Pb2+ Ion

Ligand H samples were prepared in D2O at various pD values for NMR studies of protonation. The pD values were selected in accordance with the data of potentiometric titration to obtain spectra of various protonated forms (L2−, HL−, H2L, H3L+) by adding small volumes of concentrated solutions of NaOH and HClO4. Samples of the Pb2+ complex for the NMR measurements was prepared by dissolving 1 eq. ligand and 2 eq. Pb(ClO4)2 in D2O, followed by adjustment of the desired pD with small volumes of concentrated NaOH and HClO4. The accurate pD measurements in D2O were obtained by a direct reading in a D2O solution using a combined glass/AgCl electrode after appropriate calibration procedures by usage of standard buffers.

4.4. Determination of logP

An aqueous solution of the corresponding ligand 0.5 mM, 0.05 M NaOAc and 4 kBq of 210Pb at pH5.5 were mixed with an equal volume of n-octanol (Sigma, Roedermark, Germany) (500 μL/500 μL) and resulting samples were thoroughly shaken for 20 min at room temperature. After centrifugation, organic and aqueous phases were separated and measured on a gamma-spectrometer. The logP was calculated under the following conditions of the corresponding ligand, 0.05 M NaOAc and 4 kBq of 210Pb at pH5.5 were mixed with an equal volume of n-octanol (Sigma, Roedermark, Germany) (500 μL/500 μL) and resulting of samples were thoroughly shaken for 20 min at room temperature. After centrifugation, organic and aqueous phases were separated and measured on gamma spectrometer. The logP was calculated as follows where A(oct)—count rate of 210Pb in the octanol’s phase, A(aq)—count rate of 210Pb in the aqueous phase.

4.5. High Performance Liquid Chromatography (HPLC)

Analytical HPLC was performed on a Waters system (Milford, MA, USA) equipped with a Waters 1525 binary HPLC pump, Waters C-18 reversed-phase column (4.6 × 75 mm), Waters 2489 UV-Vis detector and Waters Fraction collector III, using following modes:

For the copper complex: isocratic total flow 1 mL/min, A—0.01%TFA in mQ water—0.85 mL/min, B—CH3OH—0.15 mL/min

For the lead complex: isocratic total flow 1 mL/min, A—mQ water—0.9 mL/min, B—CH3CN—0.1 mL/min

Retention times of non-radioactive complexes PbBA2A1Py (2 mM) and CuBA2A1Py (1.7 mM) prepared at pH 5–7 were 5.2 ± 0.4 min and 5.3 ± 0.2 min, respectively. Detection of stable complexes was performed via a UV/Vis absorption at 267 nm (Figure S26). Retention time of radiolabeled [210Pb]PbBA2A1Py complex was the same: activity peaked on fractions of 5 and 5.5 mL, as it was shown by measuring the activity of fractions collected every 0.5 min, using a gamma spectrometer (Figure S26).

4.6. DFT Calculations

The calculations were performed at the DFT level of theory [20,21] with PBE0 hybrid functional [22,23] and dispersion correction D4 [24]. The atomic orbitals were modelled with the def2-tzvp basis set [25] (and the corresponding effective core potential for the Pb2+ ion [26]). All the calculations were performed in the Orca5 package [27].

4.7. Radiochemical Separation of 210Pb

Radiochemical separation of 210Pb using Sr-resin (Triskem Int. France) is described in the literature [28]. Briefly, a solution of 226Ra dissolved in 0.1 M HNO3 was loaded on a Sr-resin chromatographic column, a solution of 0.1 M HNO3 was passed through a column until the radium was completely rinsed, then 210Pb was washed off with 0.05 M (NH4)3cit. The separation of 210Pb from 226Ra was monitored by using gamma spectrometry (with an HPGe-detector GR3818 Canberra Ind.) at 46.5 keV (210Pb) and 186.2 keV (226Ra) peaks. The resulting solution was evaporated and redissolved in 0.1 M HCl.

4.8. Production and Radiochemical Separation of 64Cu

Nickel target was irradiated using the cyclotron at the Skobeltsyn Institute of Nuclear Physics, Moscow State University. The 64Cu was produced according to the 64Ni(p,n)64Cu reaction [29] from a nickel metal target (0.5 g) of natural isotopic composition using an 8-MeV proton beam at 1 μA for 3.5 h. Separation was performed on Cu Resin (Triskem), as previously described [30]. The resulting carrier-free radionuclide solutions in each experiment was added along with the solution of stable Cu(ClO4)2.

4.9. Radiolabeling Studies

Radionuclide stock solutions for labeling were prepared in 0.1 M HCl (PanReac AppliChem, Darmstadt, Germany, HPLC grade diluted in deionised water): (a) 210Pb stock solution 45 kBq/mL; (b) 64Cu stock solution 10 kBq/mL, samples 2.5 kBq/mL or 2.7 × 10−13 M.

The complex was labeled at room temperature in the 0.15 M NaOAc at pH5–6. For radiolabeling experiments, 4 kBq of 210Pb and 1 kBq of 64Cu per sample were used, leading to a final concentration of cation to be 20 nM and 0.3 pM, respectively. In the case of the copper complex, a stable Cu2+ carrier was added to reach the concentration 10 nM of cations. Thin layer chromatography (TLC) was used to control the bound fraction of the radionuclide. TLC experiments were performed using cellulose on aluminum plates (Sigma) as a stationary phase and 10 mM NaOH dissolved in 0.9% NaCl as an eluent for the 210Pb-labeled complex and silica gel on aluminum plates (Sigma) and 10% AcONH4:CH3OH (1:1) for 64Cu-labeled complex. Analysis of the distribution of radioactivity on the plates was carried out using gamma spectroscopy.

To select the optimal concentration of ligand, the 210Pb solution was mixed with 0.15 M NaOAc and various concentrations of ligand; the degree of complex formation was determined by the TLC method described above, followed by gamma spectrometry of the TLC plates. To study the labeling as a function of pH, solutions of the complex were prepared in the same way with the addition of 0.5 M HCl for pH 2.5 and 0.1 M NaOH for pH 10.5.

4.10. Stability of Complexes in Biologically Relevant Media

The resulting complex with a radiochemical purity of >95% was added to solutions of microelements (5 mM Ca2+ and Mg2+, 0.1 mM Fe3+, Zn2+ and Cu2+) and isotonic solution of 0.15 M NaCl. The bound radionuclide fraction was measured by TLC and gamma spectroscopy after 24 h of incubation at 37 °C.

For the serum stability experiment, labeled compounds with a higher activity content, 40 kBq 210Pb and 10 kBq 64Cu, were synthesized. The obtained complexes were mixed with a nine-fold excess of fetal bovine serum (HyClone, Logan, UT, USA) and incubated at 37 °C, then an aliquot of this solution was taken at fixed time points and the protein fraction was precipitated with an excess of ethanol, cooled to 2–4 °C, centrifuged and the supernatant was separated. The content of radioactivity in the supernatant was established using gamma spectrometry. The experiment, without an addition of ligand, was carried out in the same conditions.

4.11. EXAFS Measurement and Data Treatment

The Cu K- and Pb L3- edges X-ray spectra for samples of corresponding aqueous solutions containing 0.005 M of complex, at pH 2.3 and 7.6 for Cu2+ and pH2.4 and 5.1 for Pb2+, were collected at the beamline “Structural Materials Science” [31] using the equipment of “Kurchatov Synchrotron Radiation Source” (Moscow, Russia). The storage ring with electron beam energy of 2.5 GeV and a current of 80–100 mA was used as the source of radiation. All the spectra were collected in the transmission mode using a Si (111) channelcut monochromator. EXAFS data (χ) were analyzed using the IFEFFIT data analysis package [32]. EXAFS data reduction used standard procedures for the pre-edge subtraction and spline background removal. The Fourier transformation (FT) of the k2 weighted EXAFS functions χ was calculated over the range of photoelectron wave numbers k = 2–11.0 Å−1 (for both complexes). The structural parameters, including interatomic distances (Ri), coordination numbers (Ni) and Debye–Waller factors (σ2), were found by the non-linear fit of theoretical spectra (eq.) to experimental ones.

The theoretical data were simulated using the photoelectron mean free path λ(k), amplitude F, and phase shift F calculated ab-initio using the program FEFF6 [33].