Characterization of Lanthanide(III) DOTP Complexes: Thermodynamics, Protonation, and Coordination to Alkali Metal Ions

Several solution properties of complexes formed between the trivalent lanthanide ions (Ln^III) and the macrocyclic ligand DOTP^8-, including stability constants, protonation equilibria, and interactions of the LnDOTP^5- complexes with alkali metal ions, have been examined by spectrophotometry, potentiometry, osmometry, and ¹H, ³¹P, and ²³Na NMR spectroscopy. Spectrophotometric competition experiments between DOTP and arsenazo III for complexation with the Ln^III ions at pH 4 indicate that the thermodynamic stability constants (log K_ML) of LnDOTP^5- range from 27.6 to 29.6 from La^III to Lu^III. The value for LaDOTP^5- obtained by colorimetry (27.6) was supported by a competition experiment between DOTP and EDTA monitored by ¹H NMR (27.1) and by a potentiometric competition titration between DTPA and DOTP (27.4). Potentiometric titrations of several LnDOTP^5- complexes indicated that four protonation steps occur between pH 10 and 2; the protonation constants determined by potentiometry were consistent with ³¹P shift titrations of the LnDOTP^5- complexes. Dissection of the ³¹P shifts of the heavy LnDOTP^5- complexes (Tb → Tm) into contact and pseudocontact contributions showed that the latter dominated at all pH values. The smaller ³¹P shifts observed at lower pH for TmDOTP^5- were partially due to relaxation of the chelate structure which occurred upon protonation. The ³¹P shifts of other LnDOTP^5- complexes (Ln = Pr, Nd, Eu) showed a different pH-dependent behavior, with a change in chemical shift direction occurring after two protonation steps. This behavior was traced to a pH-dependent alteration of the contact shift at the phosphorus nuclei as these complexes were protonated. ²³Na NMR studies of the interactions of TmDOTP^5- with alkali and ammonium cations showed that Et₄N⁺ and Me₄N⁺ did not compete effectively with Na⁺ for the binding sites on TmDOTP^5-, while K⁺ and NH₄⁺ competed more effectively and Cs⁺ and Li⁺ less effectively. A ²³Na shift of more than 400 ppm was observed at low Na⁺/TmDOTP^5- ratios and high pH, indicating that Na⁺ was bound near the 4-fold symmetry axis of TmDOTP^5- under these conditions. Osmolality measurements of chelate samples containing various amounts of Na⁺ indicated that at high Na⁺/TmDOTP^5- ratios at least three Na⁺ ions were bound to TmDOTP^5-. These ions showed a significantly smaller ²³Na-bound shift, indicating they must bind to the chelate at sites further away from the 4-fold symmetry axis. Fully bound ²³Na shifts and relaxation rate enhancements and binding constants for all Na_xH_yTmDOTP species were obtained by fitting the observed ²³Na shift and relaxation data and the osmometric data, using a spreadsheet approach. This model successfully explained the ²³Na shift and osmolality observed for the commercial reagent Na₄HTmDOTP·3NaOAc (at 80 mM at pH 7.4).