Steric guiding of metal ion binding to a purine residue by a non-coordinating amino group: Examplified by 9-[(2-phosphonomethoxy)ethyl]-2-aminopurine (PME2AP), an isomer of the antiviral nucleotide analogue 9-[(2-phosphonomethoxy)ethyl]adenine (PMEA), and by related compounds

Acid–base properties of an antivirally active acyclic nucleoside phosphonate: (S)-9-[3-hydroxy-2-(phosphonomethoxy)propyl]adenine (HPMPA)

Protonation equilibria for the parent compound of three highly potent antivirals have been studied by 1H NMR spectroscopy.

The exocyclic amino group of adenine in PtII and PdII complexes: a critical comparison of the X-ray crystallographic structural data and gas phase calculations

A detailed computational (DFT level of theory) study regarding the nature of the exocyclic amino group, N6H₂, of the model nucleobase 9-methyladenine (9MeA) and its protonated (9MeAH⁺) and deprotonated forms (9MeA_-H), free and metal-complexed, has been conducted. The metals are Pt^II and Pd^II, bonded to nitrogen-containing co-ligands (NH₃, dien, bpy), with N1, N6, and N7 being the metal-binding sites, individually or in different combinations. The results obtained from gas phase calculations are critically compared with X-ray crystallography data, whenever possible. In the majority of cases, there is good qualitative agreement between calculated and experimentally determined C6-N6 bond lengths, but calculated values always show a trend to larger values, by 0.02-0.08 Å. Both methods indicate, with few exceptions, a high degree of double-bond character of C6-N6, consistent with an essentially sp²-hybridized N6 atom. The shortest values for C6-N6 distances in X-ray crystal structures are around 1.30 Å. Exceptions refer to cases in which DFT calculations suggest the existence of a hydrogen bond with N6H₂ acting as a H bond acceptor, hence a situation with N6 having undergone a substantial hybridization shift toward sp³. Nevertheless, even in these cases the C6-N6 bond (1.392 Å) is still halfway between a typical C-N single bond (1.48 Å) and a typical C=N double bond (1.28 Å). This scenario is, however, not borne out by X-ray crystallographic results, and is attributed to the absence of counter anions and solvent molecules in the calculated structures.

Connectivity patterns and rotamer states of nucleobases determine acid-base properties of metalated purine quartets

Potentiometric pH titrations and pD dependent (1)H NMR spectroscopy have been applied to study the acidification of the exocyclic amino group of adenine (A) model nucleobases (N9 position blocked by alkyl groups) when carrying trans-a2Pt(II) (with a=NH3 or CH3NH2) entities both at N1 and N7 positions. As demonstrated, in trinuclear complexes containing central A-Pt-A units, it depends on the connectivity pattern of the adenine bases (N7/N7 or N1/N1) and their rotamer states (head-head or head-tail), how large the acidifying effect is. Specifically, a series of trinuclear complexes with (A-N7)-Pt-(N7-A) and (A-N1)-Pt-(N1-A) cross-linking patterns and terminal 9-alkylguanine ligands (9MeGH, 9EtGH) have been analyzed in this respect, and it is shown that, for example, the 9MeA ligands in trans-,trans-,trans-[Pt(NH3)2(N7-9MeA-N1)2{Pt(NH3)2(9EtGH-N7)}2](ClO4)6·6H2O (4a) and trans-,trans-,trans-[Pt(NH3)2(N7-9EtA-N1)2{Pt(CH3NH2)2(9-MeGH-N7)}2](ClO4)6·3H2O (4b) are more acidic, by ca. 1.3 units (first pKa), than the linkage isomer trans-,trans-,trans-[Pt(CH3NH2)2(N1-9MeA-N7)2{Pt(NH3)2(9MeGH-N7)}2](NO3)6·6.25H2O (1b). Overall, acidifications in these types of complexes amount to 7-9 units, bringing the pKa values of such adenine ligands in the best case close to the physiological pH range. Comparison with pKa values of related trinuclear Pt(II) complexes having different co-ligands at the Pt ions, confirms this picture and supports our earlier proposal that the close proximity of the exocyclic amino groups in a head-head arrangement of (A-N7)-Pt-(N7-A), and the stabilization of the resulting N6H(-)⋯H2N6 unit, is key to this difference.

Solution properties of metal ion complexes formed with the antiviral and cytostatic nucleotide analogue 9-[2-(phosphonomethoxy)ethyl]-2-amino-6-dimethylaminopurine (PME2A6DMAP)

The acidity constants of protonated 9-[2-(phosphonomethoxy)ethyl]-2-amino-6-dimethylaminopurine (H3(PME2A6DMAP)+) are considered, and the stability constants of the M(H;PME2A6DMAP)+ and M(PME2A6DMAP) complexes (M2+ = Mg2+, Ca2+, Sr2+, Ba2+, Mn2+, Co2+, Ni2+, Cu2+, Zn2+, or Cd2+) were measured by potentiometric pH titrations in aqueous solution (25 °C; I = 0.1 mol/L, NaNO3). In the M(H;PME2A6DMAP)+ species, H+ and M2+ (mainly outersphere) are at the phosphonate group; this is relevant for phosphoryl-diester bridges in nucleic acids because, in the present system, there is no indication for a M2+–purine binding. This contrasts, for example, with the complexes formed by 9-[2-(phosphonomethoxy)ethyl]adenine, M(H;PMEA)+, where M2+ is mainly situated at the adenine residue. Application of log [Formula: see text] vs. [Formula: see text] plots for simple phosph(on)ate ligands, R–PO32− (R being a residue that does not affect M2+ binding), proves that all M(PME2A6DMAP) complexes have larger stabilities than what would be expected for a M2+–phosphonate coordination. Comparisons with M(PME–R) complexes, where R is a noncoordinating residue of the (phosphonomethoxy)ethane chain, allow one to conclude that the increased stability is due to the formation of five-membered chelates involving the ether–oxygen of the –CH2–O–CH2–PO32− residue: the percentages of formation of these M(PME2A6DMAP)cl/O chelates, which occur in intramolecular equilibria, vary between 20% (Sr2+, Ba2+) and 50% (Zn2+, Cd2+), up to a maximum of 67% (Cu2+). Any M2+ interaction with N3 or N7 of the purine moiety, as in the parent M(PMEA) complexes, is suppressed by the (C2)NH2 and (C6)N(CH3)2 substituents. This observation, together with the previously determined stacking properties, offers an explanation why PME2A6DMAP2– has remarkable therapeutic effects.

Extent of intramolecular π-stacks in aqueous solution in mixed-ligand copper(II) complexes formed by heteroaromatic amines and several 2-aminopurine derivatives of the antivirally active nucleotide analog 9-[2-(phosphonomethoxy)ethyl]adenine (PMEA)

The acidity constants of twofold protonated, antivirally active, acyclic nucleoside phosphonates (ANPs), H(2)(PE)(±), where PE(2-)=9-[2-(phosphonomethoxy)ethyl]adenine (PMEA(2-)), 2-amino-9-[2-(phosphonomethoxy)ethyl]purine (PME2AP(2-)), 2,6-diamino-9-[2-(phosphonomethoxy)ethyl]purine (PMEDAP(2-)), or 2-amino-6-(dimethylamino)-9-[2-(phosphonomethoxy)ethyl]purine (PME(2A6DMAP)(2-)), as well as the stability constants of the corresponding ternary Cu(Arm)(H;PE)(+) and Cu(Arm)(PE) complexes, where Arm=2,2'-bipyridine (bpy) or 1,10-phenanthroline (phen), are compared. The constants for the systems containing PE(2-)=PMEDAP(2-) and PME(2A6DMAP)(2-) have been determined now by potentiometric pH titrations in aqueous solution at I=0.1M (NaNO(3)) and 25°; the corresponding results for the other ANPs were taken from our earlier work. The basicity of the terminal phosphonate group is very similar for all the ANP(2-) species, whereas the addition of a second amino substituent at the pyrimidine ring of the purine moiety significantly increases the basicity of the N(1) site. Detailed stability-constant comparisons reveal that, in the monoprotonated ternary Cu(Arm)(H;PE)(+) complexes, the proton is at the phosphonate group, that the ether O-atom of the -CH(2)-O-CH(2)-P(O)(2)(-)(OH) residue participates, next to the P(O)(2)(-)(OH) group, to some extent in Cu(Arm)(2+) coordination, and that π-π stacking between the aromatic rings of Cu(Arm)(2+) and the purine moiety is rather important, especially for the H·PMEDAP(-) and H·PME(2A6DMAP)(-) ligands. There are indications that ternary Cu(Arm)(2+)-bridged stacks as well as unbridged (binary) stacks are formed. The ternary Cu(Arm)(PE) complexes are considerably more stable than the corresponding Cu(Arm)(R-PO(3)) species, where R-PO(3)(2-) represents a phosph(on)ate ligand with a group R that is unable to participate in any kind of intramolecular interaction within the complexes. The observed stability enhancements are mainly attributed to intramolecular-stack formation in the Cu(Arm)(PE) complexes and also, to a smaller extent, to the formation of five-membered chelates involving the ether O-atom present in the -CH(2)-O-CH(2)-PO(3)(2-) residue of the PE(2-) species. The quantitative analysis of the intramolecular equilibria involving three structurally different Cu(Arm)(PE) isomers shows that, e.g., ca. 1.5% of the Cu(phen)(PMEDAP) system exist with Cu(phen)(2+) solely coordinated to the phosphonate group, 4.5% as a five-membered chelate involving the ether O-atom of the -CH(2)-O-CH(2)-PO(3)(2-) residue, and 94% with an intramolecular π-π stack between the purine moiety of PMEDAP(2-) and the aromatic rings of phen. Comparison of the various formation degrees of the species formed reveals that, in the Cu(phen)(PE) complexes, intramolecular-stack formation is more pronounced than in the Cu(bpy)(PE) species. Within a given Cu(Arm)(2+) series the stacking intensity increases in the order PME2AP(2-) <PMEA(2-) <PMEDAP(2-) <PME(2A6DMAP)(2-). One could speculate that the reduced stacking intensity of PME2AP(2-), together with a different H-bonding pattern, could well lead to a different orientation of the 2-aminopurine moiety (compared to the adenine residue) in the active site of nucleic acid polymerases and thus be responsible for the reduced antiviral activity of PME2AP compared with that of PMEA and the other ANPs containing a 6-amino substituent.

Complex formation of cadmium with sugar residues, nucleobases, phosphates, nucleotides, and nucleic acids

Cadmium(II), commonly classified as a relatively soft metal ion, prefers indeed aromatic-nitrogen sites (e.g., N7 of purines) over oxygen sites (like sugar-hydroxyl groups). However, matters are not that simple, though it is true that the affinity of Cd(2+) towards ribose-hydroxyl groups is very small; yet, a correct orientation brought about by a suitable primary binding site and a reduced solvent polarity, as it is expected to occur in a folded nucleic acid, may facilitate metal ion-hydroxyl group binding very effectively. Cd(2+) prefers the guanine(N7) over the adenine(N7), mainly because of the steric hindrance of the (C6)NH(2) group in the adenine residue. This Cd(2+)-(N7) interaction in a guanine moiety leads to a significant acidification of the (N1)H meaning that the deprotonation reaction occurs now in the physiological pH range. N3 of the cytosine residue, together with the neighboring (C2)O, is also a remarkable Cd(2+) binding site, though replacement of (C2)O by (C2)S enhances the affinity towards Cd(2+) dramatically, giving in addition rise to the deprotonation of the (C4)NH(2) group. The phosphodiester bridge is only a weak binding site but the affinity increases further from the mono- to the di- and the triphosphate. The same also holds for the corresponding nucleotides. Complex stability of the pyrimidine-nucleotides is solely determined by the coordination tendency of the phosphate group(s), whereas in the case of purine-nucleotides macrochelate formation takes place by the interaction of the phosphate-coordinated Cd(2+) with N7. The extents of the formation degrees of these chelates are summarized and the effect of a non-bridging sulfur atom in a thiophosphate group (versus a normal phosphate group) is considered. Mixed ligand complexes containing a nucleotide and a further mono- or bidentate ligand are covered and it is concluded that in these species N7 is released from the coordination sphere of Cd(2+). In the case that the other ligand contains an aromatic residue (e.g., 2,2'-bipyridine or the indole ring of tryptophanate) intramolecular stack formation takes place. With buffers like Tris or Bistris mixed ligand complexes are formed. Cd(2+) coordination to dinucleotides and to dinucleoside monophosphates provides some insights regarding the interaction between Cd(2+) and nucleic acids. Cd(2+) binding to oligonucleotides follows the principles of coordination to its units. The available crystal studies reveal that N7 of purines is the prominent binding site followed by phosphate oxygens and other heteroatoms in nucleic acids. Due to its high thiophilicity, Cd(2+) is regularly used in so-called thiorescue experiments, which lead to the identification of a direct involvement of divalent metal ions in ribozyme catalysis.