Abstract
Manganese(II) is considered a valuable alternative to gadolinium(III) in developing contrast agents (CAs) for magnetic resonance imaging (MRI). However, due to the labile nature of common Mn(II) complexes, designing ligands that enable stable and inert complexation is essential before clinical application. Mn(1,4-Et4DO2A) bearing four ethyl groups, is found to have a log KMnL as high as 17.86, together with a pMn of 7.52 at pH 7.4. Kinetically, Mn(1,4-Et4DO2A) is approximately 20 times more inert than Mn(1,4-DO2A) in the presence of a 25-fold excess of Zn(II) at pH 6.0 and 37 °C, with the incorporation of a benzoic group to one pendant arm extending the half-life time (t1/2) to around 22 hours. Its r1 is measured as 2.34 and 2.20 mM−1 s−1 at 310 K and 1.5 and 3.0 T, respectively, representing an approximate 50% increase compared to Mn(1,4-DO2A). Mn(1,4-Et4DO2A) shows hepatic preference in mice, and its efficacy in diagnosing orthotopic hepatocellular carcinoma (HCC) is also confirmed.
Similar content being viewed by others
Introduction
In the past nearly 40 years, Gd(III)-based contrast agents (GBCAs) have been widely used in clinical magnetic resonance imaging (MRI) for enhanced diagnosis of soft tissue disorders1,2. Although there is general consensus on the high safety profile of GBCAs, several issues that have emerged over the past 20 years have raised concerns. The first, dating back to the early 2000s, is linked to the development of a severe condition known as nephrogenic systemic fibrosis (NSF) in patients with pre-existing kidney failure. The second issue is the more recent discovery of Gd(III) deposition in organs and tissues of patients who have undergone multiple contrast-enhanced MRI scans. In response to these concerns, regulatory agencies have recommended certain restrictions on the use of GBCAs, particularly for complexes with linear ligands. For these reasons, there is strong interest in finding viable alternatives to the widespread clinical use of GBCAs1,2. Mn(II) emerges as a promising candidate for developing safer and more biocompatible alternatives to GBCAs, thanks to its high-spin electronic configuration (S = 5/2) and fast water exchange. However, unlike GBCAs, Mn(II) complexes generally exhibit lower thermodynamic stability and kinetic inertness due to their low charge-to-radius ratio and the absence of ligand field stabilization energy3. Although Mn(II) is an essential element for the body, the release of Mn(II) from contrast agents (CAs) must be avoided for several reasons: First, Mn(II) release may lead to acute toxicity, as the dose of a CA is typically high for each scan to ensure effective contrast enhancement4. Second, free Mn(II) and its adducts with proteins, such as human serum albumin (HSA), have much higher longitudinal relaxivity than most small molecule Mn(II) complexes, potentially causing off-target signals5. Lastly, excessive exposure to free Mn(II) may result in clinical symptoms similar to those associated with Parkinson’s disease6.
Hexadentate open-chain EDTA is one of the most widely used ligands for constructing Mn(II) complexes as MRI contrast agents (CAs), and rigidifying the backbone of Mn(EDTA) significantly enhances its inertness. Examples include Mn(t-CDTA)7 and Mn(PhDTA)8 (Scheme 1), with half-lives (t1/2) of 12.3 and 19.1 h, respectively, compared to just 0.08 h for Mn(EDTA)9 (at pH 7.4, 25 °C and [Cu(II)] = 10-5 M). Mn(PyC3A)10 was developed by replacing one acetate arm of t-CDTA with a 2-methylpyridine group, resulting in 2-fold and 20-fold increases in inertness against Zn(II) challenge compared to Mn(t-CDTA) and Gd(DTPA), respectively. Mn(PyC3A) recently reached a phase II clinical trial as a potential alternative to GBCAs [https://www.cancer.gov/research/participate/clinical-trials-search/v?id=NCI-2024-03417&r=1]. More recently, Mn(BCN-DTA) was developed by cross-linking the cyclohexyl moiety of t-CDTA with a propylene bridge, showing a 7-fold increase in inertness compared to Mn(t-CDTA)11. Notably, the bispidine-derived ligands shown in Scheme 1 and supplementary Scheme S1 are well-known for their rigid bicyclic structures and their selectivity in chelating Mn(II) over Zn(II), giving rise to Mn(II) complexes with remarkably high stability and inertness, unmatched by other non-macrocyclic Mn(II) complexes4,12,13.
Left: Chemical structures of representative ligands used for Mn(II) complexation discussed in the text. Top right: Various chiral derivatives of the DOTA ligand. Bottom right: Design strategy for DO2A-based derivatives.
Macrocyclic ligands are potent to provide high stability and inertness when chelating metals. As shown in Scheme 1, Mn(NOTA) is a stable Mn(II) complex that received wide attention, but it suffers from low relaxivity due to the absence of a coordinated water molecule. In contrast, the hexadentate 1,4-DO2A ligand can accommodate Mn(II) while retaining a coordinated water molecule, thereby achieving reasonable relaxation rates14. Although Mn(1,4-DO2A) shares similar thermodynamic stability constants with Mn(NOTA) (log KMnL, 16.1 vs 16.3), Mn(1,4-DO2A) is more susceptible to challenges by exogenous metals, such as Zn(II) and Cu(II)15. The Mn(1,4-DO2AM) complex (supplementary Scheme S1) is a modified form of Mn(1,4-DO2A) containing amide arms, which exhibited more than a 11-fold improvement in inertness upon Zn(II) challenge when compared to Mn(1,4-DO2A)16. Despite Mn(3,9-PC2A) having a higher log KMnL (17.09), it was found to be less inert than Mn(1,4-DO2A) with half-lives (t1/2 at pH 7.4, 25 °C and [Cu(II)] = 10-5 M) of 48 and 21 h, respectively17. The replacement of the secondary amine in the macrocycle of Mn(3,9-PC2A) with an etheric oxygen atom gave Mn(3,9-OPC2A) (supplementary Scheme S1), which showed an outstanding t1/2 of 1625 h in a similar Cu(II) challenge experiment18. In another approach, an ethylamine pendant arm was attached to Mn(3,9-PC2A) to give Mn(HPC2A-EA), which was reported to be around 190 times more inert than Mn(PyC3A), while showing a pH-responsive relaxation behavior5. Therefore, rational design of macrocyclic ligands appears to be an ideal strategy for developing efficient and inert Mn(II) based contrast agents for MRI applications.
Numerous studies on Gd-DOTA-type complexes suggest that the inertness of macrocyclic complexes can be significantly enhanced by introducing chiral groups to the ring and side arms (Scheme 1). For example, Gd(DOTMA), modified with S-methyl groups on its arms, exhibited a preference for the twisted square antiprismatic (TSAP) geometry and faster water exchange kinetics compared to Gd-DOTA19. Further modification with four S-methyl groups symmetrically added to the macrocycle produced Gd(Me4DOTMA), which displayed a single square antiprismatic (SAP) stereoisomer and negligible stereoisomer conversion in aqueous solution20. Gd(Et4DOTA), containing four S-ethyl groups, demonstrated both significantly improved kinetic inertness and a faster water exchange rate compared to Gd(DOTA)21. Our recent research showed that attaching a single phenyl group to one of the arms of Gd(DOTA) significantly enhances its kinetic inertness22. Additionally, the incorporation of R-ethyl groups into the macrocyclic ring further stabilizes the complex, making it nearly SAP-pure; kinetic studies in 1 M HCl revealed that the complex exhibited negligible decomplexation even after almost 2 years at room temperature23.
Drawing inspiration from the enhanced inertness of Gd(DOTA)-type complexes using a chiral strategy, we aimed to rigidify Mn(1,4-DO2A) with chiral groups to create more inert complexes. As shown in Scheme 1, the macrocycle of 1,4-DO2A was symmetrically modified with four R-ethyl groups, yielding Mn(1,4-Et4DO2A). Furthermore, a modification with a p-benzoic acid group on one of the pendant arms was adopted to further rigidify Mn(1,4-DO2A) and Mn(1,4-Et4DO2A), giving Mn(L1) and Mn(L2) respectively. The Mn(II) complexes were characterized with UPLC-MS, crystallography, and density functional theory (DFT) calculations, and their thermodynamic stability and kinetic inertness were also compared. The relaxation properties of Mn(II) complexes discussed in the context were analyzed by using 1H NMRD and 17O NMR profiles. The in vivo MRI performance of Mn(II) complexes were examined using a mice model, and the potential of Mn(1,4-Et4DO2A) for diagnosing liver disease was further validated in a mouse model of orthotopic hepatocellular carcinoma (HCC).
Results and discussion
Synthesis and characterization of Mn(1,4-DO2A) derivatives
Mn(II) is smaller in size compared to Gd(III) (83 pm vs. 111 pm in ionic radii for coordination number 6 and 9, respectively) and typically has a coordination number of 6 or 7. The lower charge density of the cation makes Mn(II) complexes generally more labile than those of Gd(III). As a hexadentate ligand, the cyclic 1,4-DO2A can accommodate Mn(II) while retaining reasonable relaxivity, making it an ideal platform for developing Mn(II)-based contrast agents (CAs). In this study, structural modifications of 1,4-DO2A - including the addition of R-ethyl groups to the macrocycle and/or a p-benzoic acid moiety to the α-position of the arm - were explored.
Et4cyclen was used to construct chiral DO2A analogues, including 1,4-Et4DO2A and 1,7-Et4DO2A. As given in Scheme 2A and supplementary Scheme S2, the two neighboring secondary amines on the macrocycle of Et4cyclen were first protected with benzyl (Bn) groups, after which t-butyl acetate groups were attached to the remaining amines. The protecting groups were then removed via hydrogenation in a hydrogen atmosphere, followed by hydrolysis in HCl solution, yielding 1,4-Et4DO2A as the final product. The synthesis of 1,7-Et4DO2A began with Et4cyclen bearing four Bn groups (supplementary Scheme S3). Two of these Bn groups in the para position were selectively removed via hydrogenation in the presence of Pd/C. The resulting intermediate was then reacted with t-butyl bromoacetate and subsequently deprotected in a similar manner to 1,4-Et4DO2A, yielding 1,7-Et4DO2A.
A a. t-Butyl bromoacetate, K2CO3, ACN; b. H2, Pd/C, EtOH; c. HCl, H2O. B a. Methyl 4-(1-bromo-2-methoxy-2-oxoethyl)benzoate, ACN; b. Ethyl bromoacetate, K2CO3, ACN; c. NaOH, MeOH, H2O; d. H2, Pd/C, EtOH. C a. H2, Pd/C, EtOH; b. Methyl 4-(1-bromo-2-methoxy-2-oxoethyl)benzoate, ACN; c. NaOH, MeOH, H2O.
The synthesis of 1,4-DO2A was based on a previous publication (supplementary Scheme S4)24. Then a p-benzoic acid moiety was introduced to one pendent arm of 1,4-DO2A in order to increase the rigidity of the Mn(II) complex, while introducing moderate hydrophilicity. As shown in Scheme 2B, cyclen-cis2Bn was selected for reaction with methyl 4-(1-bromo-2-methoxy-2-oxoethyl)benzoate and ethyl bromoacetate subsequently, followed by treatment with NaOH and subsequent hydrogenation in the presence of Pd/C. This yielded 1,4-DO2A functionalized with a p-benzoic acid group on one of the arms (L1). As a reference ligand, 1,7-DO2A was synthesized following De León-Rodríguez’s protocol (supplementary Scheme S5)25.
Due to the presence of chiral ethyl groups, attaching a p-benzoic acid group to either arm of 1,4-Et4DO2A can lead to isomerism. To increase the selectivity of the reaction taking advantage of steric hindrance, one Bn group of Et4cyclen-4Bn was replaced by a t-butyl acetate group in two steps (supplementary Scheme S6), and after removing the rest of Bn groups, the intermediate 6 was to react with methyl 4-(1-bromo-2-methoxy-2-oxoethyl)benzoate (Scheme 2C). After an additional step, the resulting L2 was analyzed using HPLC, where two adjacent peaks (L2-a and L2-b) were observed, accounting for 75.4% and 24.6% of the product, respectively (supplementary Scheme S7). To rationalize the preference for N2 or N4 substitution in the reaction from intermediate 6 to intermediate 7, density functional theory (DFT) simulations were employed to calculate the relative energy levels of the possible products (supplementary Scheme S8). Intermediate 7 with S-substitution on the arm attached to N2 (7a-S) was taken as the reference at the ground energy level, while its R-isomer counterpart (7a-R) was found to have an energy level 1.54 kcal mol−1 higher. In contrast, the N4-substituted intermediates (7b) were found to have much higher energy levels, with 7b-S and 7b-R measured at 4.99 and 12.72 kcal mol-1, respectively. As a result, the major product is expected to be 7a, regardless of whether it has S or R chirality, which corresponds to L2-a, as observed in the analytical HPLC trace (supplementary Scheme S7). This ligand was then isolated using a semi-preparative HPLC system for subsequent Mn(II) chelation.
Complexation was achieved by stirring a slight excess of ligand with MnCl2·4H2O in water at approximately pH 7.0. The formation of the complexes was subsequently confirmed by UPLC-MS (Supplementary Figs. S1–S6). Using the retention times of Mn(II) complexes in UPLC trials, their log P values were determined as indicators of hydrophobicity26. As shown in Supplementary Fig. S7 and Table S1, Mn(1,4-DO2A) was found to be more hydrophobic than Mn(1,7-DO2A), with log P values of 0.57 and 0.47, respectively. Mn(L1) exhibited similar hydrophobicity to Mn(1,4-DO2A), despite the attachment of a p-benzoic acid group (log P = 0.58). In contrast, Mn(1,4-Et4DO2A) and Mn(1,7-Et4DO2A) were more lipophilic, with log P values of around 1.00. Similar to Mn(Et4DO2A) analogues, two separate peaks were observed for Mn(L2), with both isomers showing slightly higher hydrophilicity than Mn(1,4-Et4DO2A), and log P values around 0.95. Despite similar log P values, Mn(L1) exhibited a two-fold higher solubility in pure water at 25 °C (0.54 g mL−1) compared to Mn(1,4-DO2A) (0.26 g mL−1), as shown in supplementary Table S1. Similarly, Mn(L2) demonstrated notably higher solubility (0.20 g mL−1) than Mn(1,4-Et4DO2A) (0.07 g mL−1). Furthermore, the trans-DO2A analogues were consistently found to be significantly more soluble in water than their cis-counterparts (supplementary Table S1).
Notably, single crystals corresponding to Mn(1,4-Et4DO2A) were obtained and characterized by X-ray diffraction, as shown in Fig. 1 and supplementary Table S2. The macrocycle of Mn(1,4-Et4DO2A) adopts a (δδδδ) conformation, while both pendant arms exhibit a Δ orientation in the crystal structure, as in twisted square antiprismatic (TSAP) isomers of DOTA-like chelates27. Furthermore, due to the positional variation of substitutions on the ring, the complexes can be classified as either “corner” or “side” isomers21,28,29,30. The conformation observed in the crystal structure exhibits the “side” characteristic. However, this does not explain the two neighboring peaks observed in UPLC analysis, where Mn(1,4-Et4DO2A) was detected. Interestingly, in the solid state, Mn(1,4-Et4DO2A) was found to be heptacoordinated, and the complex molecules were cross-linked via intermolecular coordination between the Mn(II) ion and the oxygen atom from the pendant arm of another molecule. This phenomenon was different from the Mn(1,4-DO2A) dimer reported by Antonio et al., where a doubly bridged bimetallic core was witnessed, with a Mn–Mn distance of 3.54 Å31. In Mn(1,4-Et4DO2A), the Mn – N distances are between 2.28 and 2.40 Å, a bit shorter than the Mn–N distances of Mn(1,4-DO2A), ranging from 2.31 to 2.45 Å. In contrast, the Mn–O bonds of Mn(1,4-Et4DO2A) were found longer than those of Mn(1,4-DO2A) (2.32 to 2.35 Å versus 2.16 to 2.27 Å), while the intermolecular Mn–O bridge was around 2.12 Å in length.
ORTEP plot (30% probability) of the crystal structure of C20H39ClMnN4O4 corresponding to the solid state structure of Mn(1,4-Et4DO2A). Hydrogen and chloride atoms were omitted for clarity. Two units of the coordination polymer are shown to illustrate the intermolecular connection.
Protonation constants and thermodynamic stability
In this work, pH-potentiometric titrations were adopted to determine the protonation constants of ligands and the thermodynamic stability constants of their Mn(II) complexes, where the ionic strength was maintained using 0.1 M KCl and the temperature was set at 25 °C. The titration and species distribution curves are shown in Supplementary Fig. S8, while the constants obtained by using Hyperquad201310 are listed in Table 1. The protonation constants of 1,4-DO2A and 1,7-DO2A were determined under the same conditions as the reference systems. These studies afforded overall protonation constants Σlog KiH of 26.5 and 28.2 for 1,4-DO2A and 1,7-DO2A, respectively, in excellent agreement with the reported values (Supplementary Table S3, i.e., 26.37 and 28.05, respectively)15. The incorporation of R-ethyl substituents leads to an increased basicity, with the overall protonation constants of 1,4-Et4DO2A and 1,7-Et4DO2A being Σlog KiH = 30.6 and 31.6, respectively. This basicity increase affects both the macrocyclic amines and acetate groups, as the first two protonation constants are related to amines and the remaining two to acetate groups32. For example, compared to 1,4-DO2A, 1,4-Et4DO2A displays higher values of log KH2 and log KH4, while a similar increase was also observed for log KH1 and log KH4 in 1,7-Et4DO2A compared with 1,7-DO2A.
The incorporation of p-benzoic acid to one of the arms rendered a fifth protonation constant for both L1 and L2. The overall protonation constants of amines (Σlog K2H) for both ligands remained consistent with those of the parent macrocycles 1,4-DO2A and 1,4-Et4DO2A, while the difference between log KH1 and log KH2 was enlarged. This may be related to a change in ligand conformation upon incorporation of the benzoate substituent. The overall basicity of the acetate arms of L1 (Σlog K3H + Σlog K4H = 7.9) was found to be higher than that of 1,4-DO2A (Σlog K3H + Σlog K4H = 5.6), but it remains identical for L2 compared with 1,4-Et4DO2A (Σlog K3H + Σlog K4H = 8.0).
The thermodynamic stability constants of Mn(1,4-DO2A) and Mn(1,7-DO2A) determined by potentiometric titration were found to be log KMnL = 15.5 and 14.7, respectively, in agreement with the values reported in the literature (supplementary Table S3). However, the thermodynamic stability of Mn(1,4-Et4DO2A) and Mn(1,7-Et4DO2A) is more than two orders of magnitude higher, with their log KMnL values being 17.9 and 17.1, respectively. It is interesting to note that the stability of Mn(1,4-Et4DO2A) is even higher than complexes with heptadentate ligands such as PCTA and DO3A (log KMnL = 16.83 and 16.55, respectively)33,34. Mn(L1) maintained similar stability to Mn(1,4-DO2A), while around one log unit decrease in stability constant was observed for Mn(L2) compared with Mn(1,4-Et4DO2A). This suggests that L2 may be too rigid to offer optimal accommodation of the Mn(II) ion inside the macrocyclic cavity. A similar drop in stability was reported previously for Gd(CYAAZTA) compared with the parent Gd(AAZTA) (log KGdL = 18.26 and 20.24, respectively), where CYAAZTA contains a rigid cyclohexyl ring35. This phenomenon can be further evidenced by a recent report in which Mn(t-CDTA) was rigidified with methylene (BCH-DTA), ethylene (BCO-DTA), or propylene (BCN-DTA) bridges, giving a decreasing order in stability following the trend Mn(t-CDTA) > Mn(BCN-DTA) > Mn(BCO-DTA) > Mn(BCH-DTA)11.
The conditional stability of the complexes at physiological pH can be assessed by calculating the pMn values at pH 7.4, and it is a more reliable parameter for assessing the thermodynamic integrity of complexes in biological applications. As shown in Table 1, the pMn values of the investigated complexes are consistent with their corresponding stability constants log KMnL, with the highest value of pMn being determined for Mn(1,4-Et4DO2A) at 7.5, close to that of Mn(NOTA)36 and Mn(EDTA)8 (supplementary Table S3, pMn = 7.75 and 7.83, respectively). It can be seen that the values of pMn are little influenced by the chiral modifications to the macrocycle and pendent arms, with values ranging from 7.2 to 7.5 following the sequence Mn(L2) = Mn(1,4-DO2A) < Mn(L1) < Mn(1,4-Et4DO2A). However, the concentration of free Mn(II) at equilibrium (pH 7.4) for Mn(1,4-Et4DO2A) is expected to be approximately an order of magnitude higher (i.e., lower pMn value) compared to complexes derived from pyclen. For instance, hexadentate Mn(3,9-PC2A)17 and Mn(3,9-OPC2A)18 exhibited pMn values of 8.64 and 8.69, respectively (Supplementary Table S3). Furthermore, the incorporation of an additional donor atom significantly enhanced the pMn of Mn(1,4-DO2A) from 7.20 to 8.66, as demonstrated by Mn(DO3A)34. Collectively, these observations suggest that modifications to the Mn(1,4-DO2A) macrocycle and its pendant arms with additional groups (i.e., R-Et, p-benzoic acid) preserve a largely similar coordination environment, thus leading to only limited variations in their affinity for Mn(II).
Transmetallation study against Zn(II)
Dissociation of Mn(II) complexes may occur through several distinct mechanisms, notably including spontaneous, proton-assisted, and metal-assisted pathways. The latter mechanism frequently involves transmetallation reactions with competing metal ions such as Zn(II) or Cu(II)37. The dissociation reactions of Mn(1,4-DO2A) and Mn(1,7-DO2A) follow the acid-catalyzed mechanism, with negligible contribution from the spontaneous dissociation pathway15. By contrast, Mn(3,6-PC2A) and Mn(3,9-PC2A) dissociate through the proton-assisted pathway, which is inhibited by the presence of Cu(II)17. Therefore, to directly compare the inertness of our Mn(II) complexes with reported systems, we employed a widely adopted Zn(II) transmetallation assay developed by Caravan et al. In this assay, the complex was incubated in an aqueous medium containing 25 mM Zn(II) at pH 6.0 and 37 °C10. As shown in Table 2 and Supplementary Fig. S9, Mn(1,4-DO2A) and Mn(1,7-DO2A) are characterized by similar half-lives (t1/2) of 35 and 37 min respectively, which represents a two-fold increase with respect to Mn(PyC3A)10 and Gd(EOB-DTPA) (supplementary Table S4, t1/2 = 17 and 15 min respectively), showing the intrinsic advantage of Mn(DO2A) complexes in resisting transmetallation. The t1/2 value for Mn(3,6-PC2A) was reported as 35 min17, a duration comparable to that of Mn(1,4-DO2A). Remarkably, the attachment of an ethylamine group to Mn(3,9-PC2A) extended its t1/2 dramatically from 21 min to 54.4 h while preserving a coordinated water molecule, as evidenced by Mn(HPC2A-EA) (supplementary Table S4)5. This exceptional inertness is surpassed only by the most stable Mn(II)-bispidine derivatives4,12,13. In this study, the R-ethyl modification on Mn(1,4-DO2A) resulted in a 20-fold improvement in the inertness of the complex, as the half-life of Mn(1,4-Et4DO2A) was found to be 717 min, which is close to that of Mn(3,9-OPC2A) (i.e., 664 min)18. In contrast, the enhancement of inertness was less significant for Mn(1,7-Et4DO2A) (t1/2 = 352 min) in spite of similar stability constant to that of Mn(1,4-Et4DO2A) (Fig. 2 and Table 2). Thus, the introduction of chiral substituents enhances the kinetic inertness of the complexes by providing structural rigidification, which likely hinders dynamic processes including isomer interconversion by ring inversion and arm rotation. Additional examples of rigidified ligands that result in enhanced complex inertness are provided by t-CDTA7, PhDTA8, CHXOCTAPA38 and Et4DOTA21. It is interesting to note that the decoration of a p-benzoic acid moiety on one of the side arms resulted in higher inertness than Mn(1,4-DO2A) and Mn(1,4-Et4DO2A) as witnessed by Mn(L1) and Mn(L2), with their half-lives determined at 60 and 1305 min, respectively (Table 2 and Supplementary Fig. S9). This finding is consistent with our previous observations for Gd(DOTA) analogues, where addition of phenyl groups to the side arm contributed to a higher inertness as investigated in an acidic condition22,23.
Transmetallation kinetics of Mn(1,7-Et4DO2A), Mn(1,4-Et4DO2A), and Mn(L2) against Zn(II) monitored by transverse relaxation rate (r2) measurements as a function of time at 37 °C and 1.4 T.
1H NMRD and 17O NMR measurements
About 10 years ago, an in-depth study on the 1H and 17O NMR relaxometric properties of Mn(II) complexes with a series of structurally related cyclen-based ligands was conducted. The results revealed the presence of equilibria in aqueous solution between species with different hydration states (q = 0 and q = 1), and consequently, different coordination numbers (CN = 6 or 7)14. Also in this study, we conducted a multinuclear NMR relaxometric analysis to evaluate how structural variations of the chelators influence the physicochemical parameters that determine the magnetic properties of the complexes. The relaxivity values (r1) - a measure of the efficiency of a paramagnetic complex in increasing the longitudinal relaxation rate of the solvent in a 1 mM solution - of the new Mn(II) complexes discussed in this work range between 1.3 (Mn(1,7-Et4DO2A)) and 3.1 mM−1 s−1 (Mn(1,4-Et4DO2A)), at 20 MHz and 298 K. Since all the complexes exhibit relatively similar molecular weights, these differences once again reflect the presence of equilibria in solution between q = 0 and q = 1 species, with varying relative populations. Mn(1,7-Et4DO2A) is a typical example of a complex without a water molecule in its coordination sphere, as indicated by its r1 value, which is very similar to that of the parent complex Mn(1,7-DO2A). Consequently, its relaxivity is determined solely by the contribution of the outer sphere (OS). On the other hand, the r1 value for Mn(1,4-Et4DO2A) closely matches those of Mn(EDTA)9 and Mn(3,9-OPC2A)18 (Supplementary Table S5, r1 = 3.2 and 3.1 mM−1 s−1, respectively, at 20 MHz and 298 K), which is characteristic of a complex with q = 1. Structural modifications on the coordination skeleton of open-chain Mn(II) complexes vary the r1 to limited extents, e.g., the r1 values of Mn(t-CDTA)7, Mn(PhDTA)8 and Mn(PyC3A)10 with q = 1 range from 3.3 to 3.7 mM−1 s−1, while the monohydrated Mn(PC2A) analogues and derivatives exhibited similar relaxivities as well, as shown for Mn(3,9-PC2A)17, Mn(PC2A-EA)5 and Mn(3,9-OPC2A)18 in Supplementary Table S5. Mn(1,4-DO2A) displays a r1 of 2.1 mM−1 s−1 due to the insufficient inner sphere (IS) contribution, with q < 1, and a similar r1 of 2.5 mM−1 s−1 was determined for Mn(1,4-DO2AM)16. Up to now, exceptions were only reported for the Mn(II)-bispidine chelates (supplementary Scheme S1, CN = 6 or 8, and q = 1), with relaxivities as high as 4.4 mM−1 s−1 at 20 MHz and 298 K4,13. This suggests that the improved relaxation rate of Mn(1,4-Et4DO2A) can be attributed to the enhanced availability of water molecules to the coordination of Mn(II).
For a pure OS complex, we measured NMRD profiles at three different temperatures and proceed with data analysis based on Freed’s theory. The best-fit procedure provides the electronic relaxation parameters (Δ2 and τV), the diffusion coefficient (D), and the distance of closest approach (a) between bulk water molecules and the metal center. For Mn(1,7-Et4DO2A), the NMRD profiles and calculated parameters are shown in Supplementary Fig. S10 and Table S6. However, when the complexes have q > 0, the best approach is to combine the measurement of NMRD profiles with 17O NMR data, particularly the temperature dependence of the transverse relaxation rate (R2) and the shift (Δω) of the 17O signal from isotopically enriched bulk water molecules. In this case, r1 receives a significant contribution from the inner sphere relaxation mechanism, which depends on additional parameters such as the rotational correlation time (τR), the number q of coordinated water molecule(s), their distance (rMnH) from the Mn(II) ion, and their residence lifetime in the coordination site (τM = 1/kex). A global fit of the 1H data (using the Solomon-Bloembergen-Morgan equations) and 17O NMR data (using the Swift-Connick equations) allows for an accurate estimation of these parameters, providing a detailed evaluation of the correlation between structural aspects and relaxometric properties (Table 3). All the results from the relaxometric measurements are presented in Supplementary Figs. S10 and S11, while Fig. 3A, B displays only the NMRD profiles and 17O NMR data for a few representative complexes.
A 1H NMRD profiles of Mn(1,4-Et4DO2A) (□) ([Mn(II)] = 1.19 mM, pH = 6.8), Mn(L1) (○) ([Mn(II)] = 7.14 mM, pH = 7.3) and Mn(L2) (◇) ([Mn(II)] = 6.10 mM, pH = 7.0) at 298 K; B Reduced transverse relaxation rates (1/T2r) and chemical shifts (Δωr) as a function of temperature for Mn(1,4-Et4DO2A) (□) ([Mn(II)] = 1.19 mM, pH = 6.8) and Mn(L2) (◇) ([Mn(II)] = 6.10 mM, pH = 7.0).
The q value varies significantly according to the different structural modifications. This is also evident when comparing the NMRD profiles, which have the same shape but differ in amplitude, as the IS contribution to relaxivity is directly proportional to q (Fig. 3A and Supplementary Fig. S10). Substitutions on the side arm of Mn(1,4-Et4DO2A) likely hinder the access of water molecules to the metal center, as observed in Mn(L2) (q = 0.5). In contrast, minor changes in hydration state are noted for the achiral complexes, with q values determined at around 0.9 for both Mn(1,4-DO2A) and Mn(L1). However, two Mn(1,4-DO2A) derivatives, namely Mn(1,4-DO2AMBz) and Mn(1,4-BzDO2AM), were reported to preserve one coordinated water molecule39. In Mn(1,4-DO2AMBz), two benzyl groups were linked to the ends of the pendant arms of the Mn(1,4-DO2A) framework. Conversely, for Mn(1,4-BzDO2AM), the benzyl groups were attached to the secondary amines of the macrocycle. Therefore, similar modifications to Mn(1,4-Et4DO2A) that avoid substitution at the α-site of the arm might prevent the loss of accessible water molecules.
All the Mn(II) complexes demonstrate a fast exchange of the coordinated H2O, as evidenced by the increase of R2 with decreasing temperature (Fig. 3B and Supplementary Fig. S11). The large kex and small ΔHM# values indicate that the energy barrier for the formation of the transition state is low and that the 6- and 7-coordinate states are close in energy. The rotational correlation time (τR) of the Mn(II) complexes shows a strong correlation with their molecular weights (Supplementary Fig. S12), which provides a good indication of the validity and reliability of the data analysis. It is also noteworthy that there is a pronounced variation in the Δ2 parameter (the mean square zero-field splitting energy) across the different complexes, which can be attributed to differences in stereochemical rigidity and the symmetry imposed by the substituents.
DFT calculations
For the chiral Mn(DO2A) analogues, the presence of two distinct peaks in UPLC-MS traces indicates the formation of different isomers. However, there is only one stereoisomer present in the crystal of Mn(1,4-Et4DO2A). Furthermore, there are obvious differences in the stability and inertness of these Mn(DO2A) derivatives after structural modifications, which likely affect the metal coordination environment. Therefore, DFT calculations with the wB97XD/Def2-TZVP model were used to obtain structural information on the investigated Mn(II) complexes in solution. DFT calculations indicate that the observed isomerism of chiral Mn(DO2A) derivatives is related to the presence of “corner” and “side” conformations of the macrocyclic unit, as described previously for chiral Gd(DOTA) derivatives21. The calculated “corner” and “side” isomers of Mn(1,4-Et4DO2A) are shown in Fig. 4. Of note, we have included in our models two explicit second-sphere water molecules to obtain a better description of the interaction between the metal ion and the coordinated water molecule40. In the absence of explicit second-sphere water molecules, geometry optimizations lead to the departure of the water molecule from the first coordination sphere. The mixed explicit/continuum solvent model indicates that the coordination of a water molecule is favorable, with the relative energies of the q = 0 form with respect to the q = 1 one being ΔEZPE = +3.05 and +4.75 kJ mol-1 for the “corner” and “side” isomers, respectively. The Mn-N distances of the calculated “side” isomer (2.34–2.45 Å) are in reasonable agreement with the crystal data (2.28–2.40 Å), but its pendant arms show slightly different orientations, similar to the behavior of Mn(1,4-DO2A)15 and Mn(L2) (Supplementary Fig. S13). The Mn-O distances involving carboxylate groups are rather different, a situation that is observed in the solid state structure as well. On the other hand, the structure calculated for the “corner” isomer has the pendant arms twisted towards opposite orientations and a long Mn-Ow length at 2.48 Å. The long Mn-Ow bond is compatible with the fast water exchange observed by 17O NMR measurements14. The ethyl substituents of the macrocyclic unit occupy equatorial positions in both the “corner” and “side” isomers. The introduction of a p-benzoic acid group to the pendant arm to give Mn(L2) induces some changes in the calculated Mn-N distances, whose average values remain nevertheless very similar to those of Mn(1,4-Et4DO2A) (~2.38 Å), while the Mn-OAc distances get slightly longer (Fig. 4). The average Mn-N and Mn-O bond distances remain very similar in the calculated structures of Mn(1,4-DO2A) and Mn(1,4-Et4DO2A), indicating that the introduction of the ethyl substituents does not introduce significant hindrance for the coordination to the metal ion.
Stereoisomers of A Mn(1,4-Et4DO2A) and B Mn(L2) with S substitution. The “corner” and “side” ring conformations were considered for isomerism and shown in left and right, respectively.
Cytotoxicity
The in vitro cytotoxicity of Mn(II) complexes was evaluated against human hepatic stellate cells (LX-2). Pre-grown LX-2 cells were treated with varying concentrations of Mn(II) complexes (0.05, 0.1, 0.3, 0.5, and 1.0 mM), and after 24 h of incubation at 37 °C, cell viability was assessed using the Cell Counting Kit-8 (CCK-8), with absorbance measured at 450 nm. As shown in Fig. 5, free Mn(II) exhibited significant toxicity, reducing cell viability to 38% at 0.3 mM, with increasing Mn(II) concentrations leading to further cell suppression. However, all the investigated Mn(II) complexes were notably less toxic, with cell viabilities remaining comparable to the control group treated with saline, even at concentrations as high as 1 mM. This suggests minimal leakage of Mn(II) from the complexes during incubation.
The cells were treated with increasing concentration of the complexes (0.05, 0.1, 0.3, 0.5, and 1 mM), and free Mn(II) was included for comparison. The experiments were performed in triplicate.
In vivo MRI of mice
The in vivo MRI performance of the investigated Mn(II) complexes was evaluated in normal BALB/c mice using a 3.0 T clinical MRI scanner. T1-weighted MR images are shown in Fig. 6 and Supplementary Fig. S14. Mn(1,4-DO2A) was predominantly excreted through the renal route, while the addition of R-ethyl groups significantly contributed to a hepatic preference. After intravenous administration of Mn(1,4-Et4DO2A), the signal intensity in the hepatic region increased by ~55% relative to the baseline, reaching 76% by 18 min post-injection, followed by rapid excretion into the small intestine via the gallbladder. In contrast, the attachment of a p-carboxyphenyl group noticeably reduced hepatocellular uptake, as observed with Mn(L2), where the relative enhancement peaked at only 37% after 6 min post-injection, likely due to its rapid renal excretion. The achiral counterpart Mn(L1) was primarily excreted renally, as shown in Supplementary Fig. S14.
T1-weighted MRI of normal BALB/c mice treated with Mn(1,4-DO2A), Mn(1,4-Et4DO2A) and Mn(L2) (0.05 mmol kg−1) respectively. A Coronal MR images showing liver and kidney panels recorded before and after administration of (a) Mn(1,4-DO2A), (b) Mn(1,4-Et4DO2A), and (c) Mn(L2), respectively; B Relative signal enhancements in the liver and kidney regions from 1 to 30 min post-injection. The data were presented as mean ± SD, n = 3.
Due to its exceptional liver contrast enhancement, Mn(1,4-Et4DO2A) shows promise for diagnosing hepatic disorders such as hepatocellular carcinoma (HCC). Compared to normal hepatocytes, HCC cells exhibit little to no expression of transporters like OATPs, which affects the uptake of contrast agents (CAs) and enables differentiation of HCC from healthy tissue via MRI26,41,42,43. An orthotopic HCC mouse model was subjected to MR scanning 1 and 2 weeks after inoculation of H22 cells into the liver, with images captured before and after administration of Mn(1,4-Et4DO2A), as shown in Fig. 7 and Supplementary Fig. S15. In the 1-week model, the tumor region was not visible in either coronal or transverse T1-weighted pre-scan images but appeared as a hyperintense region in the T2-weighted pre-scan, which was used to identify the tumor area prior to CA administration. Following the infusion of Mn(1,4-Et4DO2A) via tail vein at a dose of 0.05 mmol kg−1, the liver-to-tumor contrast-to-noise ratio (CNR) progressively increased, peaking at around 24 min post-injection with a CNR of 43.8. In images taken at 12 min post-injection, the 1-week tumor was visually measured to be ~3.0 × 2.3 mm in the coronal plane and 2.5 × 2.5 mm in the transverse plane. The increase in tumor size led to a more pronounced contrast difference between normal hepatocytes and the tumor, as observed in the 2-week mouse model in Supplementary Fig. S15. Due to its larger size, the central region of the tumor was visible even before contrast enhancement, with a liver-to-tumor CNR of 15.5 (Fig. 7B). The pre-contrast T2-weighted scan was able to localize both the tumor center and its peripheral invasion into normal tissue, indicated by hyperintensity. Following the injection of Mn(1,4-Et4DO2A), the CNR rapidly increased to 37.7 at 1-min post-injection and further to 56.8 by 6 min, maintaining this level for the next 20 min. At 12 min post-injection, the 2-week tumor was measured to be ~6.5 × 3.5 mm in the coronal plane and 5.0 × 2.5 mm in the transverse plane, nearly double the size of the 1-week tumor (Supplementary Fig. S15).
MRI for diagnosing orthotopic hepatocellular carcinoma (HCC) using Mn(1,4-Et4DO2A). A Coronal and transverse MRI images of the liver in a mouse bearing HCC, taken 1-week post-implantation. T2-weighted scans demonstrated hyperintensity in the tumor tissue in both panels. The images were acquired 12 min after injection. B Comparison of the contrast-to-noise ratio (CNR) between liver and tumor tissues in mice at 1- and 2-weeks post-tumor seeding, respectively. Data were reported as mean ± SD (n = 3, **P < 0.004, ***P = 0.0004, ****P < 0.0001, unpaired t test).
Biodistribution
In MRI experiments, anesthesia can delay the elimination of contrast agents (CAs) in mice due to hypothermia44,45. Therefore, to assess accurately the pharmacokinetics of CAs, a biodistribution study in mice without anesthesia is essential. By measuring Mn(II) levels in various organs and tissues, the actual pharmacokinetic profile can be revealed. Considering the inertness, relaxivity, and excretion rates, Mn(1,4-Et4DO2A) and Mn(L2) were evaluated as potential CAs by quantifying Mn(II) levels in different tissues at 5 min, 30 min, and 24 h post-injection using ICP-MS (Fig. 8). At 5 min post-injection, the Mn(1,4-Et4DO2A) levels in the liver were nearly twice those of Mn(L2) (9.20 vs. 5.03 μg per g tissue, accounting for 17.67% and 8.55% of the injected dose, respectively) as shown in supplementary Table S7 and Table S8. In contrast, the levels of Mn(L2) in the intestine were relatively higher compared to Mn(1,4- Et4DO2A), with 19.72 vs. 12.52 μg per g tissue, respectively. Given the low hepatic signal enhancement of Mn(L2), it is suggested that both Mn(1,4-Et4DO2A) and Mn(L2) are primarily eliminated via the hepatic pathway, with Mn(L2) exhibiting a higher excretion rate. At the same time, their concentrations in the kidney were comparable, measuring 8.70 and 9.61 μg per g tissue for Mn(1,4-Et4DO2A) and Mn(L2), respectively, at this time point. By 30 min post-injection, the hepatic levels of Mn(II) dropped to ~3 μg per g tissue for both complexes, while the concentration of Mn(1,4-Et4DO2A) in the intestine doubled from its 5-min level, reaching 24.43 μg per g tissue due to hepatic excretion. In contrast, Mn(L2) showed a lower intestinal concentration of 13.90 μg per g tissue. After 24 h, both complexes were nearly fully eliminated from the body, with Mn(II) levels in major organs and tissues comparable to those in the saline-treated group, except for a slightly elevated level of Mn(L2) in the intestine.
Biodistribution of A Mn(1,4-Et4DO2A) and B Mn(L2) in normal BALB/c mice. The complexes were administered via tail vein injection at a dose of 0.05 mmol kg−1. The concentration levels in various organs and tissues, including the intestine, kidney, liver, heart, spleen, lungs, brain, and blood, were analyzed using ICP-MS at 5 min, 30 min, and 24 h post-injection. The data were presented as mean ± SD, n = 3.
Conclusions
In this study, strategic structural modifications were introduced to Mn(1,4-DO2A) to enhance its thermodynamic stability, kinetic inertness, and effectiveness as a relaxation probe. Among the hexadentate Mn(II) complexes tested, Mn(1,4-Et4DO2A), with four chiral ethyl groups, demonstrated the highest stability, and its inertness was nearly doubled by incorporating a rigid phenyl analog into one of the side arms. Mn(1,4-Et4DO2A) also showed a 50% increase in r1 relaxivity compared to Mn(1,4-DO2A) at 25 °C and 20 MHz. However, further modifications to its pendant arm hindered water coordination to the Mn(II) ion, reducing relaxivity to a notable degree. In addition, Mn(1,4-Et4DO2A) exhibited significant liver uptake, and its effectiveness in diagnosing liver tumor was confirmed in an orthotopic HCC mouse model. Overall, Mn(1,4-Et4DO2A) stands out for its high stability and inertness, making it a promising alternative to gadolinium-based contrast agents for MRI diagnostics, including HCC.
Methods
General information
All chemicals were purchased from commercial suppliers, including Energy Chemical (Shanghai, China), Macklin (Shanghai, China), and Aladdin (Shanghai, China) and used as received. The 1H and 13C NMR were performed using a QUANTUM-I-400MHz spectrometer (Q.One Instruments, Wuhan, China), where deuterated solvents including CDCl3, D2O, and DMSO-d6 (Macklin, Shanghai, China) were used. The standard solutions of HCl (1 M) and KOH (2 M) were acquired from BOLINDA Technology Co., Ltd. (Shenzhen, China) and then used for titration study after dilution with ultra-pure water (Milli-Q, 18.2 MΩ cm−1) if necessary. The Mn(II) complexes were characterized by an Agilent ultraperformance liquid chromatography (UPLC) equipped with both diode array (DAD) and electro spray ionization (ESI)-mass spectrometry (MS) detectors (1290 Infinity II + 6135MS). The quantification of Mn(II) was conducted on an Agilent 7850 inductively coupled plasma mass spectrometry (ICP − MS), in which a calibration curve was plotted by infusing the Mn(II) standards (Guobiao Testing & Certification Co., Ltd., Beijing, China) before samples in 2% nitric acid, and 1 ppb of Tb(III) was adopted as the inner standard.
Liquid chromatography
(A) The above mentioned UPLC-MS system was equipped with a Waters C18 column (2.5 μm, 2.1 × 100 mm), and 10 mM ammonium acetate solution and pure ACN were adopted as mobile phases A and B, respectively. The instrument method started with 95% A and 5% B at a flow of 0.4 mL min−1 for 1 min, then the portion of B increased to 50% in 5 min and to 95% in another 1 min. The column was further rinsed for 1 min, and the portion of B was dropped to 5% in 1 min and kept steady for another 1 min before the next injection. (B) A Waters alliance e2695 RP-HPLC system with the aid of a photodiode array (PDA) detector was used for common analytical works, such as monitoring reactions. Initially, a Waters C18 column (5 μm, 4.6 × 150 mm) was rinsed with 0.05% TFA solution (phase A, 90%) and pure ACN (phase B, 10%) at a flow rate of 1 mL min−1, afterwards the content of B increased to 100% in 10 min and recovered to 10% in the next 2 min. The column was rinsed for a further 3 min before the next injection. (C) The separation of certain compounds was conducted using a Waters semi-preparative HPLC system equipped with a PDA detector and a Waters C18 column (5 μm, 19 × 250 mm). Water containing 0.05% TFA and pure ACN were adopted as phases A and B, respectively, and the portion of ACN increased linearly from 10% to 50% in 20 min, followed by a recovery back to 10% in another 2 min. The column was rinsed for a further 3 min at a constant flow rate of 7 mL min−1 before next injection.
Intermediate 1
Et4cyclen-cis2Bn (1.0 g, 2.1 mmol) and K2CO3 (0.6 g, 4.3 mmol) were mixed with 20 mL of ACN, and t-butyl bromoacetate (1.0 g, 5.2 mmol) was then added to the solution. The mixture was further stirred overnight at room temperature, and the solution was concentrated after filtration. The residue was dissolved in 30 mL of EA and extracted with HCl solution (1 M, 30 mL × 3), and then the aqueous solution was neutralized with K2CO3 and extracted with DCM subsequently (30 mL × 3). The combined organic phase was dried over anhydrous Na2SO4 and concentrated to give intermediate 1 (1.2 g, 85% yield). 1H NMR (400 MHz, CDCl3, δ ppm): 0.88 (t, J = 7.28 Hz, 3H), 0.99 (m, 9H), 1.13 (m, 1H), 1.30 (m, 3H), 1.44 (s, 9H), 1.51 (s, 9H), 1.67 (m, 3H), 1.87 (m, 3H), 2.12 (m, 2H), 2.31 (dd, J1 = 9.04 Hz, J2 = 12.44 Hz, 1H), 2.80 (m, 2H), 3.01 (m, 5H), 3.14 (m, 3H), 3.23 (d, J = 13.92 Hz, 1H), 3.30 (m, 2H), 3.68 (d, J = 13.80 Hz, 1H), 3.91 (d, J = 13.88 Hz, 1H), 7.27 (m, 10H). 13C NMR (100 MHz, CDCl3, δ ppm): 11.53, 11.90, 23.04, 23.14, 23.53, 28.13, 28.18, 47.96, 48.06, 49.85, 52.26, 52.64, 55.66, 56.29, 80.30, 80.42, 126.53, 126.56, 128.05, 128.11, 128.68, 128.72, 140.75, 141.06, 171.91, 172.10. ESI-MS m/z: [M + H]+ calcd for C42H69N4O4 693.5, found 693.6.
Intermediate 2
Intermediate 1 (1.0 g, 1.4 mmol) was dissolved in 30 mL of EtOH together with 0.3 g Pd/C (10%), and the mixture was stirred overnight at 70 °C under a hydrogen atmosphere. After filtration, the solvent was evaporated under reduced pressure to give intermediate 2 (0.6 g, 81% yield). 1H NMR (400 MHz, CDCl3, δ ppm): 0.90 (m, 12H), 1.13 (s, 1H), 1.24 (m, 1H), 1.44 (s, 18H), 1.63 (m, 5H), 2.57 (m, 5H), 2.88 (s, 4H), 3.36 (m, 4H). 13C NMR (100 MHz, CDCl3, δ ppm): 10.46, 10.63, 11.70, 12.43, 19.78, 20.78, 23.52, 25.33, 28.09, 28.13, 29.70, 43.97, 44.02, 50.43, 52.57, 54.78, 60.90, 81.18, 171.41, 171.53. ESI-MS m/z: [M + H]+ calcd for C28H57N4O4 513.4, found 513.4.
1,4-Et4DO2A
Intermediate 2 (1.0 g, 2.0 mmol) was dissolved in 20 mL of HCl solution (6 M) and stirred overnight at 60 °C, and after drying the solvent, the residue was harvested as 1,4-Et4DO2A with equivalent yield. 1H NMR (400 MHz, D2O, δ ppm): 1.07 (m, 12H), 1.57 (m, 4H), 1.80 (m, 1H), 1.96 (m, 3H), 2.65 (t, J = 13.16 Hz, 0.5H), 2.88 (m, 0.5H), 3.22 (m, 10H), 3.35 (d, J = 14.48 Hz, 1H), 3.45 (d, J = 16.48 Hz, 1H), 3.57 (d, J = 17.00 Hz, 2H), 3.80 (d, J = 16.44 Hz, 1H). 13C NMR (100 MHz, D2O, δ ppm): 9.05, 9.15, 9.28, 9.37, 10.24, 10.34, 10.91, 18.30, 18.73, 21.07, 23.95, 43.42, 45.65, 50.48, 51.37, 52.41, 54.20, 54.59, 55.36, 56.92, 62.05, 171.60, 176.28. ESI-MS m/z: [M + H]+ calcd for C20H41N4O4 401.3, found 401.2.
Intermediate 3
Cyclen-cis2Bn (2.0 g, 5.6 mmol) was dissolved in 30 mL of ACN together with 4-(1-bromo-2-methoxy-2-oxoethyl)benzoate (1.9 g, 6.7 mmol), and this mixture was stirred overnight at room temperature under nitrogen atmosphere. The mixture was concentrated and dissolved in 30 mL of EA, and then HCl solution (1 M, 30 mL × 3) was used to extract the product from the EA phase. The aqueous solution was neutralized with K2CO3 and subsequently extracted with DCM (30 mL × 3), and the intermediate 3 was harvested after evaporating the solvent (2.8 g, 87% yield). 1H NMR (400 MHz, CDCl3, δ ppm): 2.71 (s, 4H), 2.90 (br, 4H), 3.05 (s, 4H), 3.16 (br, 2H), 3.33 (d, J = 15.68 Hz, 2H), 3.77 (d, J = 3.76 Hz, 2H), 3.81 (s, 3H), 3.96 (s, 3H), 4.71 (s, 1H), 6.87 (s, 2H), 7.30 (m, 4H), 7.41 (m, 3H), 7.46 (m, 2H), 7.53 (d, J = 8.08 Hz, 2H), 8.13 (d, J = 8.12 Hz, 2H). 13C NMR (100 MHz, CDCl3, δ ppm): 48.61, 49.09, 50.05, 51.14, 52.23, 62.85, 127.62, 128.07, 128.51, 128.66, 129.49, 129.70, 129.81, 130.08, 130.68, 134.30, 138.39, 166.30, 171.57. ESI-MS m/z: [M + H]+ calcd for C33H43N4O4 559.3, found 559.1.
Intermediate 4
Intermediate 3 (2.0 g, 3.5 mmol) was dissolved in 30 mL of ACN together with K2CO3 (0.5 g, 3.5 mmol), and ethyl bromoacetate (0.7 g, 4.0 mmol) was added to this solution while constant stirring. The mixture was stirred for 4 h at room temperature, and the solution was dried under reduced pressure after filtration. The residue was then dissolved in 30 mL of EA and extracted with HCl solution (1 M, 30 mL × 3), and the combined aqueous solution was treated with DCM (30 mL × 3) after neutralization with K2CO3. The collected organic phase was dried over anhydrous Na2SO4 and concentrated to give intermediate 4 (1.9 g, 86% yield). 1H NMR (400 MHz, CDCl3, δ ppm): 1.24 (m, 3H), 2.63 (m, 11H), 2.85 (m, 7H), 3.21 (s, 2H), 3.49 (m, 5H), 3.75 (s, 3H), 3.95 (s, 3H), 4.13 (m, 2H), 4.63 (s, 1H), 7.29 (m, 10H), 7.39 (m, 2H), 7.52 (d, J = 5.80 Hz, 2H), 8.01 (d, J = 5.04 Hz, 2H). 13C NMR (100 MHz, CDCl3, δ ppm): 14.34, 49.71, 51.53, 52.17, 52.39, 52.56, 52.78, 55.06, 59.93, 60.09, 68.20, 126.78, 128.07, 128.12, 128.92, 129.04, 129.09, 129.47, 129.52, 139.71, 139.91, 142.69, 166.93, 171.68, 172.33. ESI-MS m/z: [M + H]+ calcd for C37H49N4O6 645.4, found 645.2.
L1. Intermediate 4 (1.0 g, 1.5 mmol) was dissolved in 30 mL of MeOH, and NaOH (0.2 g, 6 mmol) dissolved in 1 mL of water was added to this solution while constant stirring. The mixture was further stirred overnight at 50 °C, and the solvent was evaporated to give crude product as a sodium salt. Afterwards, Pd/C (0.3 g, 10%) was added to this product together with 30 mL of ethanol, and the mixture was heated to 60 °C overnight under a hydrogen atmosphere. After evaporating the solvent, the residue was purified by using a semi-preparative HPLC system to give L1 (0.4 g, 63% yield). 1H NMR (400 MHz, D2O, δ ppm): 2.34 (s, 1H), 2.47 (s, 1H), 3.01 (m, 14H), 3.42 (s, 1H), 3.62 (s, 1H), 7.17 (d, J = 8.20 Hz, 2H), 7.75 (d, J = 7.76 Hz, 2H). 13C NMR (100 MHz, D2O, δ ppm): 41.30, 42.58, 43.22, 45.15, 46.15, 51.18, 52.27, 53.76, 64.88, 130.08, 137.32, 169.33, 171.66, 173.72. ESI-MS m/z: [M + H]+ calcd for C19H29N4O6 409.2, found 409.3.
Intermediate 6
Intermediate 5 (2.0 g, 3.0 mmol) was dissolved in 30 mL of EtOH together with 0.3 g of Pd/C (10%), and the mixture was stirred overnight at 80 °C under a hydrogen atmosphere. Intermediate 6 was harvested after filtration and drying the solvent (1.0 g, 83% yield). 1H NMR (400 MHz, CDCl3, δ ppm): 0.88 (m, 12H), 1.20 (m, 3H), 1.44 (s, 9H), 1.55 (m, 4H), 2.16 (t, J = 11.20 Hz, 1H), 2.43 (m, 5H), 2.65 (m, 3H), 2.80 (d, J = 9.96 Hz, 1H), 2.89 (d, J = 15.84 Hz, 1H), 3.29 (d, J = 16.56 Hz, 1H), 3.34 (br, 1H). 13C NMR (100 MHz, CDCl3, δ ppm): 10.12, 10.34, 10.46, 12.36, 19.00, 24.65, 25.05, 25.24, 25.32, 28.16, 44.85, 44.88, 49.29, 51.57, 53.52, 54.58, 56.57, 57.59, 80.75, 171.60. ESI-MS m/z: [M + H]+ calcd for C22H47N4O2 399.4, found 399.3.
L2. Intermediate 6 (1.0 g, 2.5 mmol) and methyl 4-(1-bromo-2-methoxy-2-oxoethyl)benzoate (0.9 g, 3.0 mmol) were dissolved in 30 mL of ACN, and this mixture was heated to 50 °C for 2 days under a nitrogen atmosphere. Intermediate 7 was harvested as a crude product after drying the solvent. Then the residue was dissolved in 30 mL of MeOH, and NaOH (0.4 g, 10.0 mmol) in 2 mL of water was added as follows. After stirring at 50 °C overnight while constant stirring, the given product was then purified by using a reverse phase HPLC system to harvest L2 (0.5 g, 38% yield). 1H NMR (400 MHz, D2O, δ ppm): -0.02 (m, 1H), 0.35 (t, J = 7.24 Hz, 3H), 0.68 (m, 2H), 0.82 (t, J = 7.32 Hz, 3H), 0.88 (t, J = 7.48 Hz, 3H), 0.99 (t, J = 7.40 Hz, 3H), 1.28 (m, 1H), 1.49 (m, 4H), 1.90 (m, 1H), 2.46 (t, J = 12.68 Hz, 1H), 2.83 (m, 7H), 2.99 (m, 2H), 3.31 (m, 1H), 3.86 (m, 1H), 4.02 (d, J = 18.04 Hz, 1H), 4.19 (d, J = 18.16 Hz, 1H), 4.82 (s, 1H), 7.27 (d, J = 7.88 Hz, 2H), 7.90 (d, J = 8.40 Hz, 2H). 13C NMR (100 MHz, D2O, δ ppm): 9.73, 10.11, 10.19, 10.29, 19.41, 20.67, 21.01, 21.99, 37.64, 38.17, 42.31, 50.70, 51.67, 55.12, 57.45, 57.60, 62.87, 64.28, 129.52, 130.12, 130.20, 138.90, 169.60, 169.78, 174.54. ESI-MS m/z: [M + H]+ calcd for C27H45N4O6 521.3, found 521.2.
Complexation of ligands with Mn(II)
The solutions of the ligands (i.e., DO2A, Et4DO2A, L1, and L2) were standardized using a relaxometric method, by which the ligand was dissolved in Tris buffered saline (TBS, pH 7.0) and mixed with increasing amounts of Mn(II) until reaching the point when r2 increased sharply26,46. Generally, the ligand (1.05 mmol) was dissolved in pure water (15 mL) and its pH was adjusted to around 7.0 with a NaOH solution (0.1 M). MnCl2·4H2O (0.2 g, 1.0 mmol) in 5 mL of water was slowly added to the mixture while constant stirring, and the mixture was allowed to stir overnight under a nitrogen atmosphere at room temperature while maintaining the pH at around 7.0. The complexation was monitored using a 1.4 T magnet (Huan Tong Nuclear Magnetic, China), and the complexes were characterized by UPLC-MS. The final aqueous solution was freeze-dried to give a solid, and its Mn(II) content was determined using ICP-MS.
DFT calculations
DFT calculations were performed with Gaussian 09 (revision D.01) software package47. TPSSh/Def2TZVP level was applied by using the atom-centered density matrix propagation (ADMP) molecular dynamics model40. Water (H2O) was chosen as the solvent, and the polarizable continuum model (PCM) was used as the solvent model. wB97XD/Def2-TZVP model was used for the optimization of Mn(II) complexes.
Crystallography
Single crystals of C20H39ClMnN4O4 were grown from aqueous media (CCDC NO. 2412450). A suitable crystal was selected and mounted on a Bruker D8 VENTURE dual wavelength Mo/Cu diffractometer. The crystal was kept at 222.00 K during data collection. Using Olex248, the structure was solved with the SHELXT49 structure solution program using intrinsic phasing and refined with the SHELXL50 refinement package using least squares minimization.
log P estimation
The log P values of Mn(II) complexes in this research were estimated using liquid chromatography according to previous reports22,23,26. Standards including DMSO, DMF, pyridine, phenol, and nitrobenzene were analyzed using the same UPLC method as shown in the section of methods for liquid chromatography, and their corresponding retention time (TR) was recorded. A calibration curve was built between log TR of standard chemicals and their log P, then, the log P of Mn(II) complexes was calculated based on the equation with their known TR.
pH-potentiometric titrations
The pH-potentiometric titrations were performed on a Metrohm Eco titrator at 25 °C for the determination of protonation constants of ligands and the thermodynamic stabilities of their Mn(II) complexes. The Mn(II) stock solution was prepared by dissolving MnCl2·4H2O in ultra-pure water and adjusted with 1 M of HCl to pH 6, and its concentration was further calibrated by using ICP-MS. The contents of ligands were determined by plotting 1/T2 along with successive titration of Mn(II) of known concentration. The ligand (around 3 mM) was dissolved in 0.1 M of KCl solution, and the solution was adjusted to be around pH 1.7 with a total volume of 10 mL. This solution was then titrated with 0.5 M of KOH solution with a volume of 10 μL for each step until the pH reached 12.0, while constantly bubbling nitrogen. The recorded volume-pH pairs were fitted using Hyperquad201310 to produce the protonation constants. For the titration of stability constants, slightly excessive ligand and Mn(II) (1.05:1 in molar ratio) were dissolved in 0.1 M of KCl solution, and similarly, the titration was performed between pH 1.7 to 12.0 with the addition of NaOH (0.5 M) in a nitrogen atmosphere. Hyperquad2013 was used to produce stability constants and species distribution curves. All the titrations were duplicated in triplicate.
Comparison of kinetic inertness
Caravan’s assay was adopted to evaluate the kinetic inertness of the Mn(II) complexes10. The working buffer was prepared by dissolving 25 mM of ZnCl2 in MES solution (50 mM, 0.1 M KCl), the pH of which was adjusted to be 6.0 with HCl solution (6 M). Then 1 mM of Mn(II) complexes were dissolved in the working buffer and incubated at 37 °C, and the relaxation time (T2) of each sample was recorded after certain time of incubation. The pseudo-first-order rate constants (kd) and half-life (t1/2) were estimated by fitting incubation time and relaxation rate (1/T2) according to the following equations \({{{{\rm{X}}}}}_{{{{\rm{t}}}}}=({{{{\rm{X}}}}}_{0}-{{{{\rm{X}}}}}_{{{{\rm{e}}}}}){{{{\rm{e}}}}}^{-{{{{\rm{k}}}}}_{{{{\rm{d}}}}}{t}}+{{{{\rm{X}}}}}_{{{{\rm{e}}}}}\) and t1/2 = ln2/kd, where X0, Xt, and Xe are the relaxation rates (1/T2) at start, time t, and equilibrium of the reaction, respectively.
1H NMRD analysis
Relaxivities were obtained from the observed longitudinal relaxation rates of solvent protons (T1) by subtracting the diamagnetic contribution, which was determined from the relaxation rates of acidified water. The exact Mn(II) concentration required to determine the relaxivities was obtained by the BMS method51,52. The magnetic-field dependence of the longitudinal relaxation rate (r1) of solvent protons (1H NMRD profiles) was measured in aqueous solution by using a variable field relaxometer equipped with an HTS-110 3 T Metrology Cryogen-free Superconducting Magnet (Mede, Italy) with frequency ranging from 20 to 120 MHz (0.47 to 3.00 T). The measurements were performed using the standard inversion recovery sequence (20 experiments, 2 scans) with a typical 90° pulse width of 3.5 μs, and the reproducibility of the data was within ±0.5%. The temperature was controlled with a Stelar VTC-91 heater airflow. Additional points in the 0.01–10 MHz frequency range were collected on a Fast-Field Cycling (FFC) Stelar SmarTracer Relaxometer. 1H NMRD profiles were collected at 283, 298, and 310 K. All the experiments were repeated three times with a reproducibility of the data within ± 0.5%.
17O NMR
Variable temperature 17O NMR data were collected on a Bruker Avance III spectrometer (11.7 T) equipped with a 5 mm double resonance Z-gradient broadband probe and Bruker BVT-3000 unit for temperature control. The samples were prepared in a 3 mm NMR tube by mixing 188 μL of a∼20 mM complex solution at physiological pH, 22 μL of D2O with 10% of tert-butanol, and 10 μL of H217O (Cambridge Isotope, 2% isotope enrichment). The transverse relaxation rates were calculated from the signal full width at half-height. The bulk magnetic susceptibility contribution was subtracted from the 17O NMR shift data using the 1H NMR shifts of the tert-butanol signal as the internal reference.
Cytotoxicity
The human hepatic stellate cell line LX-2 was incubated against Mn(II) complexes to examine the corresponding cytotoxicity. The LX-2 cells were seeded in a 96-well plate at a density of 5000 cells per well in 100 μL of Dulbecco’s modified Eagle’s medium (DMEM), and the cells were further incubated for 24 h at 37 °C, 5% CO2, and saturated humidity before further operation. The Mn(II) complexes were dissolved in 100 μL of DMEM to concentrations including 0.05, 0.1, 0.3, 0.5, and 1 mM, respectively, and then applied to cells after removing the medium, followed with 24 h of incubation at standard conditions. Afterwards, 10 μL of cell counting kit-8 (CCK-8) solution was added to the wells filled with cells, and the absorptions at 450 nm were recorded using a microplate reader after 4 h of incubation. The cell viability was acquired as the percentage of the control group in absence of Mn(II) complex. The experiment was repeated in triplicate.
Mouse model of orthotopic HCC
All animal experiments were carried out according to the Institutional Ethical Guidelines on Animal Care and were approved by the Institute of Animal Care and Use Committee at Wenzhou Institute, University of Chinese Academy of Sciences. Male BALB/c mice (25 ± 3 g, Charles River, China) were acclimatized for 1 week before operation. The mice were fasted for 12 h but with supply of water, then avertin (0.6 mL per 20 g) was used for anesthesia, followed with disinfection with iodophor. H22 cells suspension was mixed with basement-membrane matrix (1 × 107 cells mL−1), then 10 μL of this mixture was injected to the right lobe of the liver through a right abdominal transverse incision. The incision was sutured properly after sending the liver back to the abdominal cavity, and antibiotics were given to the mice in the next few days to avoid infection. The mice were allowed to recover in a warmed-up cage and used for MRI study after 1 more week.
In vivo MRI of mice
Male BALB/c mice (25 ± 3 g) were anesthetized with avertin (0.6 mL per 20 g mice weight) and then fixed in a mice specific coil under a 3.0 T clinical magnet (Ingenia elition, Philips). Background scanning was conducted before administration of complexes by using a T1-weighted sequence for both coronal and transverse planes, and the mice were further scanned for around 30 min after intravenous infusion of the complex in phosphate-buffered saline (PBS, pH 7.4) at a dose of 0.05 mmol kg−1. For the mice bearing liver tumor, both T1 and T2-weighted sequences were used for the pre-scanning. The parameters of T1-weighted sequence were set as TE = 9.7 ms, TR = 191.1 ms, FA = 50°, FOV = 50 × 50 mm, matrix size = 200 × 135, 12 slices and slice thickness = 1.5 mm with 0.15 mm interval. The parameters of T2-weighted sequence were set as TE = 120 ms, TR = 4000 ms, FA = 90°, FOV = 50 × 50 mm, matrix size = 200 × 168, 12 slices and slice thickness = 1.5 mm with 0.15 mm interval.
MRI data analysis
MicroDicom viewer was used to handle the acquired MR images, where the coronal and transverse images of the same panel pre- and post-injection were compared by the signal intensity (SI) from the drawn regions of interest (ROI) on organs or tissues. The CAs induced signal changes in liver and kidney were described with relative enhancement (RE), calculated as (SIpost – SIpre)/SIpre × 100%, where SIpre and SIpost were SI before and after infusion of CAs. The contrast between liver and the embedded tumor was described with liver to tumor contrast to noise ratio (CNR), calculated as (SIliver – SItumor)/SDair, where SDair was the standard deviation (SD) of the SI in the ROI adjacent to the mice.
Biodistribution
Male BALB/c mice (25 ± 3 g, Charles River, China) were fixed in a mice holder without anesthesia and then infused with Mn(1,4-Et4DO2A) and Mn(L2) via tail vein, respectively, at a dose of 0.05 mmol kg−1. By 5 min, 30 min, and 24 h postinjection, the mice were euthanized, respectively, and organs and tissues, including heart, liver, spleen, lung, kidney, intestine, brain, and blood, were collected. Around 0.1 g of the above organs and tissues was mixed with 0.4 mL of conc. HNO3, and the mixtures were heated to 70 °C overnight, followed by dilution with water and then filtration through a 0.22 μm membrane. The given solutions were then delivered to ICP-MS (Agilent 7850) for quantification of Mn(II).
Data availability
The data underlying this study are available in the published article and the Supplementary Information. 1H and 13C NMR spectra are provided as Supplementary Data 1. The X-ray crystallographic coordinates of structures reported in this Article have been deposited at the Cambridge Crystallographic Data Centre (CCDC), under deposition number CCDC 2412450, which can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. The CIF file of the crystallographic structure of Mn(1,4-Et4DO2A) is also available as Supplementary Data 2. Correspondence and requests for materials should be addressed to C. P., M. B., or L. D..
References
Caravan, P. Divalent manganese complexes as potential replacements for gadolinium-based contrast agents. Invest. Radiol. 59, 187–196 (2024).
Wahsner, J., Gale, E. M., Rodriguez-Rodriguez, A. & Caravan, P. Chemistry of MRI contrast agents: current challenges and new frontiers. Chem. Rev. 119, 957–1057 (2019).
Daksh, S., Kaul, A., Deep, S. & Datta, A. Current advancement in the development of manganese complexes as magnetic resonance imaging probes. J. Inorg. Biochem. 237, 112018 (2022).
Ndiaye, D. et al. Unprecedented kinetic inertness for a Mn2+‐bispidine chelate: a novel structural entry for Mn2+‐based imaging agents. Angew. Chem. Int. Edit. 59, 11958–11963 (2020).
Botár, R. et al. Stable and inert Mn(II)-based and pH-responsive contrast agents. J. Am. Chem. Soc. 142, 1662–1666 (2020).
Harischandra, D. S. et al. Manganese-induced neurotoxicity: new insights into the triad of protein misfolding, mitochondrial impairment, and neuroinflammation. Front. Neurosci. 13, 654–654 (2019).
Molnár, E. et al. Remarkable differences and similarities between the isomeric Mn(II)- cis - and trans- 1,2-diaminocyclohexane- N, N, N ′, N ′-tetraacetate complexes. Inorg. Chim. Acta 472, 254–263 (2018).
Pota, K. et al. Taking the next step toward Inert Mn2+ complexes of open-chain ligands: the case of the rigid PhDTA ligand. New J. Chem. 42, 8001–8011 (2018).
Kalman, F. K. & Tircso, G. Kinetic inertness of the Mn2+ complexes formed with AAZTA and some open-chain EDTA derivatives. Inorg. Chem. 51, 10065–10067 (2012).
Gale, E. M., Atanasova, I. P., Blasi, F., Ay, I. & Caravan, P. A manganese alternative to gadolinium for MRI contrast. J. Am. Chem. Soc. 137, 15548–15557 (2015).
Martinelli, J. et al. Improving the stability and kinetic inertness of Mn(II) complexes by increasing the bridge length in bicyclic CDTA‐like ligands. Chem. Eur. J. 30, e202400570 (2024).
Cieslik, P. et al. Exceptional Manganese (II) Stability and Manganese(II)/Zinc(II) selectivity with rigid polydentate ligands. Angew. Chem. Int. Edit. 61, e202115580 (2022).
Ndiaye, D. et al. Mn2+ bispidine complex combining exceptional stability, inertness, and MRI efficiency. J. Am. Chem. Soc. 144, 22212–22220 (2022).
Rolla, G. A., Platas-Iglesias, C., Botta, M., Tei, L. & Helm, L. 1H and 17O NMR relaxometric and computational study on macrocyclic Mn(II) complexes. Inorg. Chem. 52, 3268–3279 (2013).
Garda, Z. et al. Physico-chemical properties of Mn(II) complexes formed with cis- and trans-DO2A: thermodynamic, electrochemical and kinetic studies. J. Inorg. Biochem. 163, 206–213 (2016).
Forgács, A. et al. A bisamide derivative of [Mn(1,4‐DO2A)] – solution thermodynamic, kinetic, and NMR relaxometric studies. Eur. J. Inorg. Chem. 2016, 1165–1174 (2016).
Garda, Z. et al. Complexation of Mn(II) by rigid pyclen diacetates: equilibrium, kinetic, relaxometric, density functional theory, and superoxide dismutase activity studies. Inorg. Chem. 60, 1133–1148 (2021).
Csupász, T. et al. A new oxygen containing pyclen-type ligand as a manganese(II) Binder for MRI and 52Mn PET applications: equilibrium, kinetic, relaxometric, structural, and radiochemical studies. Molecules 27, 371 (2022).
Aime, S. et al. Properties, solution state behavior, and crystal structures of chelates of DOTMA. Inorg. Chem. 50, 7955–7965 (2011).
Opina, A. C. et al. Analysis of the isomer ratios of polymethylated-DOTA complexes and the implications on protein structural studies. Dalton Trans. 45, 4673–4687 (2016).
Dai, L. et al. Chiral DOTA chelators as an improved platform for biomedical imaging and therapy applications. Nat. Commun. 9, 857 (2018).
Xu, W. et al. Rational design of Gd-DOTA-type contrast agents for hepatobiliary magnetic resonance imaging. J. Med. Chem. 66, 8993–9005 (2023).
Xu, W. et al. Chiral Gd-DOTA as a versatile platform for hepatobiliary and tumor targeting MRI contrast agents. J. Med. Chem. 66, 14669–14682 (2023).
Bellouard, F. et al. cis-diprotected cyclams and cyclens: a new route to symmetrically or asymmetrically 1,4-disubstituted tetraazamacrocycles and to asymmetrically tetrasubstituted derivatives. J. Chem. Soc. Perkin Trans. 1, 3499–3505 (1999).
De León-Rodríguez, L. M., Kovacs, Z., Esqueda-Oliva, A. C. & Miranda-Olvera, A. D. Highly regioselective N-trans symmetrical diprotection of cyclen. Tetrahedron Lett. 47, 6937–6940 (2006).
Wang, J. et al. Manganese-based contrast agents for magnetic resonance imaging of liver tumors: structure-activity relationships and lead candidate evaluation. J. Med. Chem. 61, 8811–8824 (2018).
Clough, T. J., Jiang, L., Wong, K. L. & Long, N. J. Ligand design strategies to increase stability of gadolinium-based magnetic resonance imaging contrast agents. Nat. Commun. 10, 1420 (2019).
Payne, K. M. & Woods, M. Isomerism in Benzyl-DOTA derived bifunctional chelators: implications for molecular imaging. Bioconjug. Chem. 26, 338–344 (2015).
Rust, L., Payne, K. M., Carniato, F., Botta, M. & Woods, M. Differences in the relaxometric properties of regioisomeric Benzyl-DOTA bifunctional chelators: implications for molecular imaging. Bioconjug. Chem. 30, 1530–1538 (2019).
Tircso, G., Webber, B. C., Kucera, B. E., Young, V. G. & Woods, M. Analysis of the conformational behavior and stability of the SAP and TSAP isomers of lanthanide(III) NB-DOTA-type chelates. Inorg. Chem. 50, 7966–7979 (2011).
Bianchi, A. et al. Thermodynamic and structural aspects of manganese(II) complexes with polyaminopolycarboxylic ligands based upon 1,4,7,10-Tetraazacyclododecane (cyclen). Crystal structure of dimeric [MnL]2·2CH3OH containing the new ligand 1,4,7,10-Tetraazacyclododecane-1,4-diacetate. J. Chem. Soc. Dalton Trans. 6, 917–922 (2001).
Huskens, J. et al. Alkaline earth metal and lanthanide(III) complexes of ligands based upon 1,4,7,10-tetraazacyclododecane-1,7-bis(acetic acid). Inorg. Chem. 36, 1495–1503 (1997).
Uzal-Varela, R. et al. Thermodynamic stability of Mn(II) complexes with aminocarboxylate ligands analyzed using structural descriptors. Inorg. Chem. 61, 14173–14186 (2022).
Garda, Z. et al. Effect of the nature of donor atoms on the thermodynamic, kinetic and relaxation properties of Mn(II) complexes formed with some trisubstituted 12-membered macrocyclic ligands. Front. Chem. 6, 232 (2018).
Vágner, A. et al. High kinetic inertness of A bis-hydrated Gd-complex with a constrained AAZTA-like ligand. Chem. Commun. 52, 11235–11238 (2016).
Drahoš, B. et al. Mn2+ complexes of 1-Oxa-4,7-diazacyclononane based ligands with acetic, phosphonic and phosphinic acid pendant arms: stability and relaxation studies. Dalton Trans 40, 10131–10146 (2011).
Drahoš, B. et al. Dissociation kinetics of Mn2+ complexes of NOTA and DOTA. Dalton Trans. 40, 1945–1951 (2011).
Tircso, G. et al. Approaching the kinetic inertness of macrocyclic gadolinium(III)-based MRI contrast agents with highly rigid open-chain derivatives. Chemistry 22, 896–901 (2016).
Forgács, A. et al. Optimising the relaxivities of Mn2+ complexes by targeting human serum albumin (HSA). Dalton Trans. 46, 8494–8504 (2017).
Patinec, V. et al. Hyperfine coupling constants on inner-sphere water molecules of a triazacyclononane-based Mn(II) complex and related systems relevant as MRI contrast agents. Inorg. Chem. 52, 11173–11184 (2013).
Haimerl, M. et al. Quantification of contrast agent uptake in the hepatobiliary phase helps to differentiate hepatocellular carcinoma grade. Sci. Rep. 11, 22991 (2021).
Baek, A. R. et al. Effect of structural fine-tuning on chelate stability and liver uptake of anionic MRI contrast agents. J. Med. Chem. 65, 6313–6324 (2022).
Baek, A. R. et al. Gadolinium complex of 1,4,7,10-tetraazacyclododecane-1,4,7-trisacetic Acid (DO3A)-ethoxybenzyl (EOB) conjugate as a new macrocyclic hepatobiliary MRI contrast agent. J. Med. Chem. 60, 4861–4868 (2017).
Kiryu, S., Inoue, Y., Watanabe, M. & Ohtomo, K. Effect of isoflurane anesthesia and hypothermia on the hepatic kinetics of Gd-EOB-DTPA: evaluation using MRI of conscious mice. J. Magn. Reson. Imaging 34, 354–360 (2011).
Murase, K., Assanai, P., Takata, H., Saito, S. & Nishiura, M. A simple and inexpensive system for controlling body temperature in small animal experiments using MRI and the effect of body temperature on the hepatic kinetics of Gd-EOB-DTPA. J. Magn. Reson. Imaging 31, 1744–1751 (2013).
Chen, K. et al. Mn(II) complex of lipophilic group-modified ethylenediaminetetraacetic acid (EDTA) as a new hepatobiliary MRI contrast agent. J. Med. Chem. 64, 9182–9192 (2021).
Frisch, M. J. et al. Fox Gaussian 09, Revision A.02. Gaussian, Inc., Wallingford CT (2016).
Dolomanov, O. V., Bourhis, L. J., Gildea, R. J., Howard, J. A. K. & Puschmann, H. OLEX2: a complete structure solution, refinement and analysis program. J. Appl. Crystallogr. 42, 339–341 (2009).
Sheldrick, G. M. SHELXT– integrated space-group and crystal-structure determination. Acta Crystallogr. A 71, 3–8 (2015).
Sheldrick, G. M. Crystal structure refinement with SHELXL. Acta Crystallogr. C 71, 3–8 (2015).
Evans, D. F. The determination of the paramagnetic susceptibility of substances in solution by nuclear magnetic resonance. J. Chem. Soc. 0, 2003–2005 (1959).
Corsi, D. M., Platas-Iglesias, C., van Bekkum, H. & Peters, J. A. Determination of Paramagnetic Lanthanide(III) Concentrations from Bulk Magnetic Susceptibility Shifts in NMR Spectra. Magn. Reson. Chem. 39, 723–726 (2001).
Acknowledgements
The National Natural Science Foundation of China (22075281 and 22105201), Zhenjiang Provincial Natural Science of Foundation of China (LZ21B010001), University of Chinese Academy of Science (WIUCASQD2020008), and Wenzhou high-level innovation team (Development and application team of functional liver cancer-on-a-chip) financially supported this work. This manuscript is part of the project NODES, which has received funding from the MUR-M4C2 1.5 of PNRR with grant agreement no. ECS00000036 (M.B.).
Author information
Authors and Affiliations
Contributions
W.X., C.P., M.B., and L.D. designed the experiments. W.X., Z.C., F.C., Y.L., X.Y., X.X., J.X., G.M., Y.D., and Y.J. performed experiments. W.X., F.C., C.P., M.B., and L.D. co-wrote the paper. X.R., Z.Y., F.Y., C.P., M.B., and L.D. supervised the project. All authors have given approval to the final version of the article.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Communications Chemistry thanks the anonymous reviewers for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Xu, W., Cai, Z., Carniato, F. et al. Design of Mn(1,4-DO2A) derivatives as stable and inert contrast agents for magnetic resonance imaging. Commun Chem 8, 215 (2025). https://doi.org/10.1038/s42004-025-01615-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s42004-025-01615-x
