Matthias Bastian

Stability and structure of binary and ternary metal ion complexes of orotidinate 5'-monophosphate (OMP3−) in aqueous solution

Abstract The stability constants of the 1:1 complexes formed between Mg2 +, Ca2 +, Sr2 +, Ba2 +, Mn2 +, Co2 +, Ni2 +, Cu2 +, Zn 2+ or Cd 2+ and orotidinate 5′-monophosphate (OMP3-) were determined by potentio-metric pH titrations in aqueous solution (I =0.1 M, NaNO3; 25°C). In addition to the stability constants of these M(OMP)− complexes, for several cases also the corresponding acidity constants for the release of the proton from the H(N-3) site were calculated; i.e., the formation of M(OMP-H)2- complexes was quantified. On the basis of recent measurements for simple phosphate monoesters [R-MP2-; R is a noncoordinating residue; S.S. Massoud and H. Sigel, Inorg. Chem., 27, 1447-1453 (1988)], evidence is provided that the somewhat increased stability of all the mentioned M(OMP)− complexes is mainly the result of a charge effect of the carboxylate group (in position 6 of OMP3-) and not of a direct participation in complex formation; i.e., there are no indications for the formation of significant amounts of...

On the metal ion binding properties of orotidine

The stability constants of the 1:1 complexes formed between orotidinate (Or−) and Cu2+ or Zn2+ were determined by potentiometric pH titrations in solvent mixtures (vol./vol.) consisting of 70% 1,4- dioxane and 30% water at 25 °C and I=0.1 M (NaNO3). Orotidine is first deprotonated at the carboxylic acid residue in position 6 (pKHH(Or)=2.36) and next at the H(N-3) site (pKHOr=10.67). The latter site plays no role in the physiological pH range, i.e. the obvious site for metal ion coordination is the carboxylate group at position 6. For comparison we have studied therefore also the stabilities of the 1:1 complexes formed between Cu2+ or Zn2+ and the simple carboxylates (CA−): HCOO−, CH3COO−, CH2ClCOO−, CHCl2COO− and CHF2COO−. Plots of log KMM(CA) versus pKHH(CA) indicated that the stability of the M(Or)+ complexes is largely governed by the basicity of the corresponding carboxylate groups, but there are also indications that in the mentioned solvent mixture some (macro)chelate formation may occur, i.e. the carboxylate-coordinated metal ion is possibly interacting with a further site of the Or− ligand (probably the ‘ether’ oxygen of the ribosyl residue which is accessible in the less favored anti conformation). The upper limit for the formation degree of such a chelate is 45%. It should be pointed out that in aqueous solution (I=0.1 M, NaNO3; 25 °C) the basicity of the carboxylate group is strongly reduced (pKHH(Or)=0.5±0.3) and consequently the metal ion affinity of this group is also expected to be considerably smaller; indeed, estimations of the stability of the orotidinate complexes of Cu2+, Zn2+ or Mg2+ confirm this expectation. Some relevant points following from the present results regarding biological systems are indicated.

Acid-base and metal ion-binding properties of 2′-deoxycytidine 5′-monophosphate (dCMP2−) alone and coordinated to cis-diammine-platinum(II). Formation of mixed metal ion nucleotide complexes

Abstract The acidity constants of diprotonated 2′-deoxycytidine 5′-monophosphate, i.e. H 2 (dCMP) ± , were determined by potentiometric pH titration in aqueous solution (25 °C; I = 0.1 M, NaNO 3 ) and compared with the previously determined (S.S. Massoud and H. Sigel, Inorg. Chem., 27 (1988) 1447–1453) corresponding constants of diprotonated cytidine 5′-monophosphate, i.e. H 2 (CMP) ± . The absence of the 2′-hydroxy group makes dCMP 2− slightly more basic, compared with CMP 2− . The stability constants of the M(H·dCMP) + and M(dCMP) complexes of Mg 2+ , Cu 2+ and Zn 2+ were determined and those for the corresponding CMP complexes reevaluated. It is concluded that in the M(H·dCMP) + and M(H·CMP) + species the metal ion is mainly located at N-3 and the proton at the phosphate group. On the basis of recent measurements with simple phosphate monoesters and phosphonate derivatives, i.e. R-PO 3 2− with R being a non-coordinating residue (H. Sigel et al., Helv. Chim. Acta, 75 (1992) 2634–2656), it is shown that the stability of all the M(dCMP) and M(CMP) complexes is solely determined by the basicity of the phosphate group. Coordination of two H(dCMP) − ions via N-3 to cis -(NH 3 ) 2 Pt 2+ gives H 2 [ cis -(NH 3 ) 2 Pt(dCMP) 2 ], abbreviated as H 2 (Pt(dC) 2 ), the synthesis of which is described and the acidity constants of which were determined. Pt 2+ bound to the N-3 sites apparently has only a small effect on the basicity of the two phosphate groups in Pt(dC) 2 2− . In addition, also via pitentiometric pH titrations, the stability constants of the M(H·Pt(dC) 2 ) + and M(Pt(dC) 2 ) complexes with Mg 2+ , Cu 2+ and Zn 2+ were determined. Based on the previously determined (see the above Ref.) linear log K M(R−PO 3 ) M versus p K H(R-PO 3 ) H relationships it is shown that the metal ion-binding properties of the phosphate groups in the mentioned platinum(II) complex are still remarkable, allowing thus the formation of mixed metal ion complexes. In fact, the effect of Pt 2+ at the N-3 sites on the binding properties of the phosphate groups is relatively small; to a first approximation, though there are some minor additional effects, one may conclude that also in these cases the complex stabilities are mainly determined by the basicity of the phosphate groups.

Ternary complexes in solution1 with hydrogen phosphate and methyl phosphate as ligands

Abstract The stability constants of the 1:1 complexes formed between Cu(Arm) 2+ , where Arm = 2,2′-bipyridyl or 1,10-phenanthroline, and methyl phosphate, CH 3 OPO 3 2− , or hydrogen phosphate, HOPO 3 2− , were determined by potentiometric pH titration in aqueous solution (25°C; l = 0.1 M, NaNO 3 ). On the basis of previously established log K versus p K a straight-line plots (D. Chen et al., J. Chem. Soc., Dalton Trans. (1993) 1537–1546) for the complexes of simple phosphate monoesters and phosphonate derivatives, R-PO 3 2− , where R is a non-coordinating residue, it is shown that the stabilities of the Cu(Arm) (CH 3 OPO 3 ) complexes are solely determined by the basicity of the -PO 3 2− residue. In contrast, the Cu(Arm) (HOPO 3 ) complexes are slightly more stable (on average by 0.15 log unit) than expected on the basicity of HPO 4 2− ; this is possibly due to a more effective solvation including hydrogen bonding, an interaction not possible with coordinated CH 3 OPO 3 2− species. Regarding biological systems the observation that HOPO 3 2− is somewhat favored over R-PO 3 2− species in metal ion interactions is meaningful.

Ternary complexes in solution (part 551) with phosphonates as ligands. Various intramolecular equilibria in mixed-ligand complexes containing the antiviral 9-(2-phosphonomethoxyethyl)adenine, an adenosine monophosphate analogue

The stability constants of the mixed-ligand complexes formed between Cu(arm)2+, where arm = 2,2′-bipyridyl (bipy) or 1,10-phenanthroline (phen), and the dianions of phosphonomethoxyethane (PME2–) or 9-(2-phosphonomethoxyethyl)adenine (PMEA2–) were determined by potentiometric pH titration in aqueous solution at 25 °C and l= 0.1 mol dm–3(NaNO3). The stability of the binary (arm)(PMEA)2– stacks was estimated and the experimental conditions for the titrations were carefully selected such that self-association of the adenine derivative PMEA and of its complexes was negligible, i.e. it was made certain that the properties of the monomeric Cu(arm)(PMEA) complexes were studied. The ternary Cu(arm)(PMEA) complexes are considerably more stable than the corresponding Cu(arm)(R-PO3) complexes, where R-PO32– represents a phosphonate (or a phosphate monoester) with a group R that is unable to participate in any kind of interaction within the complexes as, for example, methylphosphonate or ethylphosphonate. This increased stability is attributed to intramolecular stack formation in the Cu(arm)(PMEA) complexes and also to the formation of five-membered chelates involving the ether oxygen present in the –O–CH2–PO32– residue of PMEA2–. The latter interaction is separately quantified by studying the ternary Cu(arm)(PME) complexes which can form the five-membered chelates but where no intramolecular ligand–ligand stacking is possible. Application of these results allows a quantitative analysis of the intramolecular equilibria involving three structurally different Cu(arm)(PMEA) species, e.g. of the Cu(bipy)(PMEA) system about 3% exist with the metal ion solely co-ordinated to the phosphonate group, 10% as a five-memebered chelate involving the -O–CH2–PO32– residue of PMEA2–, and 87% with an intramolecular stack between the adenine moiety of PMEA2– and the aromatic rings of bipy. In addition, the Cu(arm)(PMEA) complexes may be protonated leading to Cu(arm)(H·PMEA) species for which it is concluded that the proton is mainly located at the phosphonate group. However, of this species two isomers still coexist, one where Cu(arm)2+ forms a stack with the adenine residue of H(PMEA)– and another one where Cu(arm)2+ co-ordinates in an adenosine-type fashion to the nucleic base moiety of H(PMEA)–; the percentages of the formation degree of these isomeric species have been estimated. Finally, the properties of adenosine 5′-monophosphate (AMP2–) and of its PMEA2– analogue are compared in their ternary Cu(arm)(AMP) and Cu(arm)(PMEA) systems. The co-ordinating properties of the ether oxygen, which are crucial for the antiviral properties of PMEA, are discussed.

The self-association of flavin mononucleotide (FMN2−) as determined by 1H NMR shift measurements

The concentration dependence of the (1)H NMR chemical upfield shifts of the protons H6, H9, H7alpha, and H8alpha of the 7,8-dimethylisoalloxazine residue of flavin mononucleotide (FMN(2-)) has been measured and the self-stacking tendency of FMN(2-) was quantified with the isodesmic model of indefinite non-cooperative self-association. The stacking tendency of FMN(2-) is considerable and described in the concentration range of 0.0025-0.1 M with the indicated model by K = 27 +/- 15 M(-1) (25 degrees C; I = 0.1-0.3 M). This result is compared with related ones from the literature. The caveats regarding the self-stacking properties of FMN(2-) and their dependence on the concentration are discussed.

Acid-base and metal ion-binding properties of flavin mononucleotide (FMN2−). Is a ‘dielectric’ effect responsible for the increased complex stability?

Abstract Due to contradictions in the literature we have redetermined the acid-base properties of riboflavin (RiFl; vitamin B2), i.e. 7,8-dimethyl-10-ribityl-isoalloxazine, and of flavin mononucleotide (FMN2−), also known as riboflavin 5′-phosphate, via potentiometric pH titrations (I = 0.1 M, NaNO3; 25 °C). In contrast to various claims, the isoalloxazine ring cannot be protonated at pH > 1, a result in agreement with an early study (pKa = −0.2; L. Michaelis, M.P. Schubert and C.V. Smythe, J. Biol. Chem., 116 (1936) 587–607); deprotonation of the ring system occurs in both compounds with p K a ⋍ 10 . The pKa value of ∼ 0.7 determined for the deprotonation of H2(FMN) must be attributed to the release of the first proton from the fully protonated phosphate group; its second proton is released with pKa = 6.18 in agreement with the acidity constants of various other monoprotonated monophosphate esters. The stability constants of the 1:1 complexes formed between Mg2+, Ca2+, Sr2+, Ba2+, Mn2+, Co2+, Ni2+, Cu2+, Zn2+ or Cd2+ (M2+) and FMN2− were determined by potentiometric pH titrations in aqueous solution (I = 0.1 M, NaNO3; 25 °C). The log stability constants of all these M(FMN) complexes are about 0.2 log units higher than expected from the basicity of the phosphate group. This slight stability increase cannot be attributed to the formation of a seven-membered chelate involving the ribit-hydroxy group at C-4′ as the stability constants for the M2+ 1:1 complexes of glycerol 1-phosphate (G1P2−) demonstrate: G1P2− contains the same structural unit which would also allow in this case the formation of the mentioned seven-membered chelate; however, the stability of the M(G1P) complexes is solely determined by the basicity of the phosphate group. Hence, in agreement with earlier conclusions (J. Bidwell, J. Thomas and J. Stuehr, J. Am. Chem. Soc., 108 (1986) 820–825) regarding Ni(FMN) one must conclude that the slight stability increase of the M(FMN) complexes has to be attributed to the isoalloxazine ring. The equality of the stability increase of the complexes for all the mentioned ten metal ions precludes its attribution to an interaction with an N site and makes a specific interaction with an O site also somewhat unlikely. In addition, carbonyl oxygens appear as not very favorable for the formation of macrochelates by a further interaction with already phosphate-coordinated metal ions. Therefore, we propose that the slight but significant stability increase originates from M(FMN) species (with a formation degree of about 30%) in which the hydrophobic flavin residue is close to the metal ion, thereby lowering the ‘effective’ dielectric constant in the microenvironment of the metal ion and thus indirectly promoting the −PO32−/M2+ interaction.

Stability and structure of the Mg2+, Ca2+ and Cu2+ complexes of orotidinate 5′-monophosphate (OMP)3− in various aqueous 1,4-dioxane mixtures

Abstract The stability constant of the 1:1 complexes formed between Mg2+, Ca2+ or Cu2+ and orotidinate 5′- monophosphate (OMP3−) were determined by potentiometric pH titrations in water containing 30 or 50% (vol./vol.) 1,4-dioxane (I=0.1 M, NaNO3; 25 °C). In addition to the stability constants of these M(OMP)− complexes, for Mg(OMP)− and Ca(OMP)− the acidity constants for the release of the proton from the H(N-3) site were also calculated, i.e. the formation of the corresponding M(OMP- H)2− complexes was quantified. All the corresponding equilibrium constants for aqueous solution are known from a previous study (M. Bastian and H. Sigel, J. Coord. Chem., 23 (1991) 137) and in the discussions these data are also taken into account. On the basis of recent measurements in aqueous dioxane mixtures with simple phosphate monoesters (M.C.F. Magalhāes and H. Sigel, J. Indian Chem. Soc., in press), evidence is provided that the somewhat increased stability of the Cu(OMP)− complex in the various solvents is mainly the result of a charge effect of the carboxylate group (in position 6 of OMP3−) and not of a direct participation in complex formation, i.e. there are no indications for the formation of significant amounts of macrochelates involving the phosphate and the carboxylate groups. This result is in agreement with the dominating syn conformation of OMP3− in which the 5′- phosphate and 6-carboxylate groups are pointing away from each other. However, for the Cu2+ complex of salicyl phosphate (SaP3−) evidence is given by evaluating previously published equilibrium constants (R. W. Hay, A. K. Basak, M. P. Pujari and A. Perotti, J. Chem. Soc., Dalton Trans., (1986) 2029) that a simultaneous coordination of a phosphate and a carboxylate group to the same metal ion is possible; it is estimated that the eight-membered chelate of Cu(SaP)− reaches a formation degree of about 65 (±25)%.

Ternary Complexes in Solution+ with Phosphonates as Ligands. Intramolecular Equilibria in the Mixed Ligand Cu2+ Complexes Formed by 2,2′-Bipyridyl or 1,10-Phenanthroline and the Dianion of Phosphonylmethoxyethane in Water-Dioxane Mixtures

The stability constants of the mixed ligand complexes formed by Cu2+, 2,2′-bipyridyl or 1,10-phenanthroline (= Arm), and the dianion of phosphonylmethoxyethane (PME2-), ethyl phosphonate (EtP2-), methyl phosphonate (MeP2-), or D-ribose 5′-monophosphate (RibMP2-) (= R–PO32-) were determined by potentiometric pH titrations in water containing 30 or 50% (v/v) 1,4-dioxane (I = 0.1 M, NaNO3; 25°C). The corresponding results regarding water as solvent were taken from our earlier work. Previous measurements with simple phosphate monoesters, together with the present results for RibMP2-, were used to establish log versus straight line plots. With the aid of the equilibrium constants determined for the MeP2- and EtP2- systems it is shown that simple phosphonates, i.e., those without an additional binding site, fit also on the same straight lines. Therefore, it could be demonstrated with these reference lines that the Cu(Arm)(PME) complexes in all solvents have a higher stability than expected for a sole phosphonate Cu2+ coordination. This increased stability is attributed to the formation of 5-membered chelates involving the ether oxygen present in the – CH2– O – CH2–PO32- residue of PME2-. The formation degree of the 5-membered chelates in the Cu(Arm)(PME) systems varies only between about 65 and 85% in the three mentioned solvents, despite the fact that the stabilities of the Cu(Arm)(PME) complexes increase by more than 1.8 log units by going from water to 50% dioxane-water. It is concluded that (i) such 5-membered chelates will also be formed in mixed ligand complexes of other metal ions in solvents with a reduced polarity, and (ii), more importantly, that the same interactions will also occur with the parent compound of PME2-, i.e. the dianion of 9-(2-phosphonylmethoxyethyl)adenine (PMEA2-), a compound which shows antiviral properties and for which the ether oxygen is important.