Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2008 Aug 28.
Published in final edited form as: Inorg Chem. 2006 Mar 20;45(6):2688–2694. doi: 10.1021/ic051411p

Synthesis, Structures, and Electrochemistry of Au(III)-ethylenediamine Complexes and Interactions with Guanosine 5’-Monophosphate

Shourong Zhu a, Waldemar Gorski a, Douglas R Powell b, Judith A Walmsley a
PMCID: PMC2526558  NIHMSID: NIHMS63302  PMID: 16529492

Abstract

[Au(en)Cl2]Cl·2H2O has been synthesized and its structure has been solved for the first time by the single crystal X-ray diffraction method. The complex has square planar geometry about the Au(III) and the anionic Cl is located in the apical position and at a distance of 3.3033(10) Å compared to 2.2811(9) and 2.2836(11) Å for the coordinated Cl. [Au(en)Cl2]Cl·2H2O belongs to the space group Pbca with a = 11.5610(15) Å, b = 12.6399(17) Å, c = 13.2156(17) Å, α = β = γ = 90° and Z=8. Bond lengths of Au-N are 2.03 Å. [Au(en)Cl2]Cl·2H2O is less thermally stable than [Au(en)2]Cl3 due to the replacement of 2 Cl ligands by a second ethylenediamine ligand in the latter. Cyclic voltammetry shows that the formal potential of Au(III)/Au(0) becomes more negative in the series [AuCl4], [Au(en)Cl2]+, [Au(en)2]3+. 1H, 13C, and 31P NMR reveal that in aqueous solution [Au(en)Cl2]+ bonds to guanosine 5’-monophosphate, 5’-GMP, (1:1 mole ratio) via N7, although the stability is not very high. NMR data also indicate that N7-O6 or N7-phosphate 5’-GMP chelation, as found in some Au(III)-nucleotide complexes, is not present. The Au(III) complex undergoes hydrolysis at pH > 2.5−3.0 and, therefore N1 coordination to 5’-GMP is not observed. No direct coordination between 5’-GMP and [Au(en)2]Cl3 is observed.

Keywords: Gold(III), guanosine 5’-monophosphate, NMR, cyclic voltammetry, X-ray structure


Gold(III) complexes are isoelectronic and generally isostructural to platinum(II) complexes, and therefore it is anticipated that they will have similar activity to the Pt(II) antitumor drugs.1-5 However, compared to the corresponding Pt(II) complexes, Au(III) complexes have not been well explored chemically or electrochemically, possibly because few gold(III) complexes have been shown to be sufficiently stable in aqueous solution.6,7

Ethylenediamine (en) is a common chelating ligand that is widely used in transition metal complexes. Its gold complexes have been known since 1931,8-10 and the crystal structure of [Au(en)2Cl2]Cl was determined.11 This complex has a square bipyramidal geometry with 2 Cl at the apical positions. [Au(en)Cl2]Cl was reported to be isolated in 1967 by Brodersen et al.12 However, the crystal structure for this 1:1 complex is first reported herein, as well as the electrochemical properties and the reactivity with guanosine 5’-monophosphate (Figure 1) for both [Au(en)Cl2]Cl and [Au(en)2Cl2]Cl.

Figure 1.

Figure 1

Guanylic acid, H2(5’-GMP)

Interest in the reactions of nucleobases, nucleosides and nucleotides on gold surfaces has received considerable attention. Relatively few studies have been directed toward the isolation and characterization of these compounds with Au(III). Reaction of AuCl4 with a variety of nucleosides and nucleotides has shown that the bonding mode of the Au(III) is dependent upon the nucleoside or nucleotide. Based on solution studies, possible N7-O6 chelation is suggested for guanosine and possible N7-phosphate macrochelation for 5’-GMP.13,14 Mixed ligand complexes have been reported, such as [Au(L-N,N’)(5’-GMP)2], where L is N-(4-methylphenyl-2-pyridinecarboxamide)15 and the dimethylgold complex, [(CH3)2AuCl(guanosine)].16

Cytotoxicity studies of Au(III) complexes of polydentate amines, including ethylenediamine, have shown that they are as effective, or even more effective against certain tumor cell lines than cis-platin1. When tested against cis-platin resistant cell lines, the Au(III) complexes remained nearly as effective against non-resistant cell lines. It was also shown that the bis(ethyleneldiamine)Au(III) complex binds to calf thymus DNA non-covalently, most probably through electrostatic interactions.17

Our interest has been the cyclic polynuclear complexes of the d8 metal ions, especially those in which the metal ions are bridged by nucleotides. Complexes of [Pd(en)]2+ with purine nucleotides, in which the nucleotide uses two N donor sites (N7 and N1) to bridge between the Pd(II) centers, have been shown to be cyclic tetramers in aqueous solution.18,19 However, we have not found any evidence for cyclic species in the [Au(en)Cl2]+ system so far. Herein we report the synthesis of [Au(en)Cl2]Cl·2H2O and [Au(en)2]Cl3, their thermogravimetric and electrochemical properties, reactions with 5’-GMP in solution, and the crystal structure of [Au(en)Cl2]Cl·2H2O.

Methods and Materials

Materials

NaAuCl4·2H2O, ethylenediamine, D2O, and dmso-d6 were purchased from Aldrich Chemical Co. and used as received. H2(5’-GMP) and Na2(5’-GMP) were purchased from Sigma and used as received. All solutions other than those for NMR measurements were prepared using deionized water that was purified with a Barnstead NANOpure cartridge system (18 MΩ).

Syntheses

[Au(en)Cl2]Cl·2H2O

NaAuCl4·2H2O (0.218 g; 0.548 mmol) was dissolved in 20 mL of absolute ethanol at room temperature. To this solution was added 37 μL of ethylenediamine (0.55 mmol). An orange-brown precipitate formed immediately. After stirring the mixture for 20 min, the precipitate disappeared and the solution became slightly turbid. After filtration, the solution was concentrated to 5 mL and left to stand at room temperature. Pale yellow crystals of [Au(en)Cl2]Cl·2H2O formed gradually. Yield: 0.15 g. (68 %). Elemental analysis for [Au(en)Cl2]Cl·2H2O: Found %: C, 6.11; H, 2.97; N, 6.94, Cl, 26.71; calc %: C, 6.01; H, 3.03; N, 7.01; Cl, 26.63. 1H NMR: 3.30 ppm in D2O. The complex darkens in color at 105 °C; decomposes at 195 °C.

[Au(en)2]Cl3·0.5HCl

NaAuCl4·2H2O (0.20 g; 0.50 mmol) was dissolved in 5 mL of deionized H2O. To this solution was added 1.0 mL of 1.0 M aqueous solution of en. An orange precipitate formed at once and gradually redissolved (pH 4.3). A trace amount of undissolved precipitate was removed by filtration and 30 mL of ethanol was added to the filtrate. The very pale yellow precipitate which formed immediately was filtered off and air dried. Yield: 0.16 g (72 %). Elemental analysis for[Au(en)2]Cl3·0.5HCl: Found %: C, 11.45; H, 3.72; N, 12.98, Cl, 27.87; calc %: C, 10.87; H, 3.76; N, 12.69; Cl, 28.09. 1H NMR: 3.27 ppm in D2O. The complex darkens in color at 170 °C; decomposes at 195 °C.

Nucleotide Stock Solutions

Na2(5’-GMP) or H2(5’-GMP) was dissolved in H2O or D2O and the concentration of the nucleotide was determined by UV spectroscopy (λmax 252 nm; ε = 1.37 × 104 M−1cm−1; pH 7). The required amount of Au(III) compound was added to the 5’-GMP solution.

Methods

Thermogravimetric Analysis

Measurements were made on a Shimadzu TGA-50 analyzer under N2 atmosphere with a flow rate of 20 mL min−1 in a Pt cell. Samples weighed 3−7 mg. The temperature increased at 20 °C min−1 from room temperature to 900 °C.

NMR

Spectra were obtained in D2O on a Varian Inova 500 MHz in D2O. The temperature dependent 1H studies were referenced to internal tetramethylammonium ion at 3.185 ppm and the spectra at 26 °C were referenced to the HOD line at 4.800 ppm (The TMA+ resonance is very near that of ethylenediamine and can overlap at times). 31P spectra were referenced to external 85 % H3PO4 and 13C spectra to DSS. pD = pH (meter reading) + 0.40.

Electrochemical Measurements

A CHI 832 workstation (CH Instruments, Inc.) was used to collect electrochemical data. Experiments were performed at room temperature (20 ± 1 °C) in a conventional three-electrode system with 3.0 mm-diameter glassy carbon (GC) disk working electrode (Bioanalytical Systems, Inc.), a Pt wire as the auxiliary electrode, and a Ag/AgCl/3M NaCl (BAS) reference electrode. Prior to use, the glassy carbon electrodes were wet polished on an Alpha A polishing cloth (Mark V Lab) with successively smaller particles (0.3 and 0.05 μm diameter) of alumina. The slurry that accumulated on the electrode surface was removed by ultrasonication for 30 s in methanol and deionized water.

The concentration of the Au complexes was 1.0 mM and the background electrolyte was 0.40 M NaCl. The experiments were repeated at least three times and the means of measurements are presented with the standard deviations or relative standard deviations.

Bulk electrolysis of stirred solutions of Au complexes was performed using a GC rod (∼0.6 cm2) as a working electrode. The reductive electrolysis was continued for ∼4 h until the current dropped to ∼1 % of its initial value. The progress of electrolysis was also followed by using UV-visible spectrometry.

X-Ray Structure Determination

. A colorless prism-shaped crystal of dimensions 0.54 × 0.36 × 0.20 mm was selected for structural analysis. Intensity data were collected using a Bruker D8 instrument with an APEX detector and graphite-monochromated Mo Kα radiation.20 A total of 11,038 reflections were measured in the range 2.84 ≤ θ ≤ 26.00° using ω scan frames. The intensity data were corrected by a face-indexed, analytical absorption correction 21 and an empirical absorption correction22 giving transmittances in the range of 0.0421 − 0.1419. These data were merged to form a set of 1897 independent data with Rint = 0.0337 and a completion of 99.8%. The space group Pbca was determined by systematic absences and confirmed by refinement. The structure was solved by direct methods and refined by full-matrix least squares on F2.21 The final difference map had maxima and minima of 1.766 and −1.140 e/Å3, respectively.

Results and Discussion

Structure of [Au(en)Cl2]Cl·2H2O

The crystal data and structure refinement are given in Table 1 and the ORTEP view and crystal packing of [Au(en)Cl2]Cl·2H2O are shown in Figures 2 and 3. Selected bond distances [Å] and angles [°] are listed in Tables 2 and 3 and hydrogen bond distances are given in Table 4. The Au(III) ion is four coordinated in a square planar geometry as expected. The Au-N bond lengths are 2.029(4) and 2.030(3) Å, which is roughly the same as those found in [Au(phen)Cl ]Cl2 (2.03∼2.06 Å)6 and macrocyclic polyamine complexes,23 but slight longer than in Au(III)-peptide complexes (1.98∼2.02 Å)24,25 and a pyridine-2-carboxamido complex.26 They are slightly shorter than in an EDTA complex (2.06∼2.11 Å).27 The Au(III)-Cl bond lengths for the coordinated Cl are 2.2811(9) and 2.2836(11) Å, which is roughly the same as found in the literature.6,11,27 The root mean square deviation from planarity for the 5 atoms comprising the squa re plane (N(1), N(4), Au, Cl(1), Cl(2) is 0.0177.

Table 1.

Crystallographic data for [Au(en)Cl2]Cl · 2H2O.

chemical formula C2 H12 Au Cl3 N2 O2
formula weight 399.45
space group (No. 61) Pbca
a 11.5610(15) Å
b 12.6399(17) Å
c 13.2156(17) Å
V 1931.2(4) Å3
Z 8
T −173(2) °C
λ 0.71073 Å
Dcalcd 2.748 g cm−3
μ 160.19 cm−1
R(F, obsd)* 0.0221
wR(F2, all)* 0.0620
*

R = Σ ∥Fo| - |Fc∥ / Σ |Fo|, wR = {Σ [w(Fo2Fc2)2] / Σ [w(Fo2)2]}1/2

Figure 2.

Figure 2

ORTEP drawing of [Au(en)Cl2]Cl·2H2O with the atomic numbering scheme.

Figure 3.

Figure 3

Packing diagram of[Au(en)Cl2]Cl·2H2O. Inline graphic Cl; Inline graphicN; Inline graphicAu; Inline graphicO

Table 2.

Selected Bond lengths [Å] and angles [o] for [Au(en)Cl2]Cl·2H2O

Au(1)-N(1) 2.029(4) Au(1)-N(4) 2.030(3)
Au(1)-Cl(1) 2.2811(9) Au(1)-Cl(2) 2.2836(11)
Au(1)-Cl(3) 3.3033(10)
N(1)-Au(1)-Cl(1) 89.90(11) N(4)-Au(1)-Cl(1) 173.89(10)
N(1)-Au(1)-Cl(2) 175.65(11) N(4)-Au(1)-Cl(2) 91.27(10)
Cl(1)-Au(1)-Cl(2) 94.32(4) N(1)-Au(1)-N(4) 84.47(15)
Cl(1)-Au(1)-Cl(3) 87.63(3) Cl(2)-Au(1)-Cl(3) 85.30(3)

Table 3.

Torsion angles [o] for [Au(en)Cl2]Cl·2H2O

N(4)-Au(1)-N(1)-C(2 −12.3(3)
Cl(1)-Au(1)-N(1)-C(2) 170.1(2)
Cl(2)-Au(1)-N(1)-C(2) −24.0(16)
Au(1)-N(1)-C(2)-C(3) 38.8(4)
N(1)-C(2)-C(3)-N(4) −53.7(4)
C(2)-C(3)-N(4)-Au(1) 42.5(4)
N(1)-Au(1)-N(4)-C(3) −16.7(3)
Cl(1)-Au(1)-N(4)-C(3) 6.4(13)
Cl(2)-Au(1)-N(4)-C(3) 162.4(3)

Table 4.

Hydrogen bonds for [Au(en)Cl2]Cl·2H2O [Å and o]

D-H...A* d(D-H) d(H...A) d(D...A) <(DHA)
N(1)-H(1A)...O(2S) 0.93(6) 1.88(6) 2.791(5) 164(4)
N(1)-H(1B)...Cl(3)#1 0.80(5) 2.39(5) 3.180(4) 170(5)
N(4)-H(4A)...O(2S)#2 0.81(5) 2.04(5) 2.807(5) 157(5)
N(4)-H(4B)...O(1S) 0.82(5) 2.11(5) 2.925(5) 173(5)
O(1S)-H(1S1)...Cl(3) 0.81(5) 2.35(5) 3.135(3) 164(5)
O(1S)-H(1S2)...Cl(3)#3 0.81(6) 2.45(6) 3.243(3) 166(5)
O(2S)-H(2S1)...Cl(3) 0.66(5) 2.60(5) 3.218(3) 157(6)
O(2S)-H(2S2)...O(1S)#4 0.79(5) 2.04(5) 2.813(4) 168(5)

Symmetry transformations used to generate equivalent atoms:

#1

-x+1/2, y+1/2, z

#2

x, -y+1/2, z+1/2

#3

-x, -y, -z+1

#4

x+1/2, y, -z+1/2

*

D is H-bond donor; A is H-bond acceptor

The Au(III)- Cl(3) (ionic) distance is 3.3033(10) Å with the Cl-Au vector ∼90° relative to the Au(N2Cl2) square plane. Although much longer than the Au-Cl covalent bond distance and a little longer than the 3.103(4) Å found by Minacheva11 for the square bipyramidal [Au(en)2Cl2]Cl compound, it is short enough to suggest a possible weak interaction between the Cl and the Au(III).The bonding of a fifth ligand, and sometimes a sixth, to a primarily square planar complex is well known.28 An unusual compound, Au(2,2'-bpy)Cl3·2.25H2O, has been reported by Lippert et al, in which two square planar [Au(bpy)Cl2]+ ions are connected by a long Au-Cl-Au bridge (av. Au-Cl of 3.218(3) Ǻ).29 Since the bpy ligands in the two units are said to be parallel to each other, it is likely that stacking holds the units sufficiently close to allow bridging.

The N(1)-Au-N(4) bond angle is 84.47(15)°, while the Cl(1)-Au-Cl(2) bond angle is 94.32(4) °. The difference between these is expected because of the greater non-bonding to non-bonding electron pair repulsions between the two coordinated Cl. The same type of variation between the Na-Au-Nb and Cla-Au-Clb bond angles was reported for [Au(phen)Cl2]Cl,6 although both bond angles were smaller than those we report for [Au(en)Cl2]Cl·2H2O. The Au-N, Au-Cl bond lengths and Cl-Au-Cl and N-Au-N angles are comparable with Au2(trien)Cl5 complex, where trien is triethylenetetraamine.30

The cations and water molecules crystallize in layers in the b direction. A layer of chlorides separates every two layers of the cations and waters. There is extensive H-bonding between the NH2, Cl(3), and the water of hydration which serves to stabilize the structure (Table 4). As shown in Figure 3, two H2O molecules are associated with each other through a H-bond. One of the pair (H2O(1S)) also forms hydrogen bonds to two Cl(3) and to the N of one en. The other H2O(2S) hydrogen bonds to one Cl(3) and to en in 2 different cations. There is additional H-bonding between NH2 and Cl(3).

Thermal analysis

As shown in Figure 4, [Au(en)Cl2]Cl·2H2O loses its water of hydration beginning at 77 °C and all water is lost by 127 °C. The 9% weight loss agrees with the loss of 2 moles of H2O. Scheme 1 gives the temperature ranges of the decompositions, the experimental mass losses, and the products.

Figure 4.

Figure 4

TGA curves of (a) [Au(en)Cl2]Cl·2H2O; (b) [Au(en)2Cl2]Cl· 0.5HCl, run in an N2 atm with flow rate of 20 mL/min and temperature increase 20 °C/min.

Scheme 1.

Scheme 1

[Au(en)2]Cl3·0.5HCl behaves in a similar manner. After loss of 0.5 mole HCl (4%), it loses 2 Cl and 2 ethylenediamine (42%) between 220−390 °C and the remaining AuCl decomposes between 390−600 °C. The ethylenediamine and Cl are lost at a higher temperature in Au(en)2Cl3·0.5HCl than in [Au(en)Cl2]Cl·2H2O, probably due to the stronger coordination of the ethylenediamine ligand compared to the Cl ligand.

Electrochemistry of Gold Complexes

The survey cyclic voltammograms recorded at glassy carbon electrodes in aqueous solutions of gold complexes (1.0 mM in 0.40 M NaCl) displayed a series of distinct current peaks A, B, B1, and C (Figure 5). The original reduction peak A shifted toward more negative potentials in the order Ep(AuCl4, 0.410 V) > Ep(Au(en)Cl2+, 0.140 V) > Ep(Au(en)23+, −0.290 V). This indicated that a bidentate ethylenediamine ligand stabilized the Au(III) oxidation state against reduction, as previously reported for chelating N donor ligands.31 The position of the peaks B, B1, and C, which developed on consecutive voltammetric scans, was essentially not influenced by the composition of the gold complex. The gold complexes displayed good electrochemical stability as indicated by their peak currents and peak potentials, which were stable (± 10%) at least for 4 h.

Figure 5.

Figure 5

Survey cyclic voltammograms recorded at a glassy carbon electrode in 1.0 mM solutions of Au(III) complexes: (a) Au(en)Cl2+, (b) Au(en)23+, (c) AuCl4 . The vertical arrows indicate a direction of change in peak currents upon repetitive scanning of the electrode potential. Background electrolyte, 0.40 M NaCl. Scan rate, 100 mV s−1.

The origin of the current peaks can be conveniently discussed by using the cyclic voltammogram for the Au(en)Cl2+ complex (Figure 5a) as an example. On the first cathodic (negative-going) potential scan, the current peak A at 0.14 V was formed due to the reduction of the gold complex. The reduction process involved three electrons per molecule (n = 3) as indicated by the coulometric analysis of the exhaustive electrolysis at 0.0 V. The Au(III) → Au(0) reduction was also confirmed by the presence of gold deposit on the electrode surface after the electrolysis. The voltammetric and chronoamperometric analyses showed no intermediate Au(I) complexes, which is compatible with the fact that the ethylenediamine is not a π-acceptor ligand that could stabilize the intermediate oxidation state of gold. Such stabilization has been demonstrated in the case of the gold complexed with the π-acceptor phosphine ligands.32

On the anodic (positive-going) backward scan, a symmetrical current peak B was formed at ∼1.00 V. This peak was present only when the potential on the forward scan was scanned beyond the peak A. Thus, the peak B was ascribed to anodic dissolution Au(0) → Au(III) of the gold presence of high chloride concentration in the solution, the anodic stripping of Au(0) deposit was deposit. In the driven by the complexation of Au(III) ions with Cl ions. A small peak B1 that appeared at ∼0.80 V was probably due to the competing process of gold oxide formation. The current after the peak B did not decay to zero, which suggested that the formation of gold oxide precluded a total dissolution of the Au(0) deposit from the electrode surface.

The scanning of the potential again toward more negative values yielded a broad reduction peak C at ∼0.40 V, which was not present during the first cathodic scan. The peak C can be ascribed to two parallel processes, which involved the reduction of gold oxide and AuCl4 complex that were formed at the electrode/solution interface during the preceding anodic scan. The supportive argument for the latter process was that the reduction of the AuCl4 reference complex yielded a current peak (Figure 5c, peak A) at the same potential as that of the peak C. The foregoing discussion can be summarized by the following overall peak assignments

Peak A (0.14 V, cathodic): Au(en)Cl2+ +3eAu0 + en + 2Cl
Peak B (1.0 V, anodic): Au0 + 4Cl → AuCl4 + 3e
Peak B1 (0.80 V, anodic): 2Au0 + 3H2O → Au2O3 + 6H+ + 6e
Peak C (0.40 V, cathodic): Au2O3 + 3H2O + 6e → 2Au0 + 6OH
AuCl4 + 3e → Au0 + 4Cl

Such an interpretation of peaks B, B1, and C is in agreement with the independence of the peak potentials upon the composition of the starting gold complex.

In order to avoid complications from excessive deposition/dissolution processes, the AuCl4, [Au(en)Cl2]+, and [Au(en)2]3+ complexes were studied in narrower potential windows that included only their original reduction peak A (Figure S1, Supporting Information). The analysis of scan rate (ν) dependence of the peak currents (Ip) and peak potentials (Ep) revealed that the Ip increased linearly with ν1/2 (R2=0.997−0.994) and Ep was directly proportional to log(ν) (R2=0.998−0.991) while shifting toward higher overpotentials. This indicated that molecules of gold complexes were transported by diffusion to the electrode surface, where they underwent electrochemically irreversible reduction. The voltammetry and chronoamperometry revealed that the substitution of Cl ions with bulkier ethylenediamine ligands resulted in lowering of the diffusion coefficient (Table 1S, Supporting Information).

Interaction with 5’-GMP

The AuCl4 ion and other Au(III) chloro complexes are known to undergo hydrolysis in water at relatively low pH.7,13 [Au(en)Cl2]Cl hydrolyzed in aqueous solution above pD ∼2.5 as shown by changes in the 1H NMR of the ethylenediamine ligand (Figure S2, Supporting Information). At pD 1.4, the ethylenediamine chemical shift was 3.28 ppm, which is assigned to the [Au(en)Cl2+] complex. At pD 3.4, a new line appeared at 3.30 ppm and the intensity of the 3.28 ppm resonance was greatly reduced. The NMR line at 3.0 ppm is likely to be an hydroxo complex, such as Au(en)(OH)Cl+ or Au(en)(OH)2+.6 At the even higher pD of 5.8, three distinct species were present (3.22, 3.27, 3.30 ppm) and a brown precipitate was observed, which precluded continuing above pD 5. This precipitate contained no ethylenediamine or 5’-GMP and was assumed to be an oxide or hydroxide of Au(III). A line at 3.22 ppm was broad and was probably a soluble polymeric Au(III) hydoxo or oxo species.

The 1H NMR spectra of 10 mM aqueous solutions of [Au(en)Cl2]Cl containing different amounts of 5’-GMP in acidic solution are shown in Figure 6. There are clearly two 5’-GMP H8 signals and two H1’ signals, one each from unreacted H2(5’-GMP) and from a Au(en)-5’-GMP complex. The upfield of the two H8 (8.84 ppm) and the downfield of the two H1’(6.13 ppm) resonances are assigned to Au(en)-5’-GMP, while the other H8 (8.92 ppm) and H1’ (6.09 ppm) signals are those of unreacted 5’-GMP. The unreacted 5’-GMP resonance assignments were made on the basis of a pD-dependent experiment (Figure 7) which clearly showed the expected upfield shift of the H8 line of unreacted 5’-GMP with increasing pD.18 Even in the 2 mM 5’-GMP solution where there was a five-fold excess of [Au(en)Cl2]Cl, a considerable amount of unreacted H2(5’-GMP) was still present. This indicates that either the stability constant of the Au(en)-5’-GMP complex is rather small or the acidity of the solution is such that the N7 of 5-’GMP is partially protonated, or possibly both. Integration of the NMR spectra in Figure 6 shows that the mole ratio of reacted 5’-GMP to reacted [Au(en)Cl2]Cl remained at 1:1, irrespective of the amounts of the two components in the solution. Although it appears (Figure 6) that the concentration of Au(en)-5’-GMP complex is decreasing with increasing 5'-GMP concentration, it is simply the large excess of 5'-GMP that gives that appearance in the spectra.

Figure 6.

Figure 6

1H NMR of 10 mM [Au(en)Cl2]Cl with various concentrations of H2(5’-GMP) in D2O, pD 2.5. U = unreacted 5’-GMP; R = Au(en)-GMP complex.

Figure 7.

Figure 7

1H NMR of 10 mM Au(en)-5’GMP in D2O at various pD. U = unreacted 5’-GMP; R = Au(en)-GMP complex.

The pD study of a solution of 20 mM Au(en)Cl3 containing 20 mM 5’-GMP was carried out over the pD range 2.3−4.6 (Figure 7). Some precipitate of hydrolyzed Au(III) species was observed at pD 4.3 and above. The chemical shift of the H8 of the Au(en)-GMP complex remained constant at 8.83 pm over the pD range studied, but that of unreacted 5’-GMP gradually shifted upfield with increasing pD until it reached 8.2 ppm, the chemical shift at which N7 is completely deprotonated. It is well known that metal-N7 purine bonding causes a large downfield shift of the H8 resonance. Therefore, the fact that the H8 resonance in the Au(en)-GMP complex remained constant at 8.83 ppm proves that the Au(III) is bonding to the N7 of the 5’-GMP. The 31P NMR of solutions under the same conditions and pD 3.6 also showed two resonances, 0.88 ppm (unreacted 5’-GMP) and 0.53 ppm (reacted 5’-GMP). Although this indicated a slightly different environment for the bonded 5’-GMP, it showed that a possible N7-phosphate macrochelate structure was not being formed because the 31P resonance for Au(III)-OPO3R bonding would be located considerably downfield, as observed in a Co(III)-5’-GMP complex at 12 ppm.33

The NMR results strongly support the formation of a 1:1 complex of a monomeric Au(III)(en) complex with monodentate N7-GMP bonding and a fourth ligand of Cl, OH, or H2O. Alternatively, an N7-O6 chelate structure,proposed, but not confirmed by several authors,13,14 could be formed. In order to determine the presence of this type of bonding, the 13C NMR spectrum of 11 mM Au(en)Cl3-GMP (1:1 mole ratio) in D2O at pD 2.8 was run and compared with that of 5’-GMP under the same conditions ((Table 5) and Figure S3, Supporting Information). As was observed for the 1H spectrum of Au(en)Cl3-GMP (1:1 mole ratio), lines were present for both complexed and unreacted H2(5’-GMP). In all cases the resonances of the complexed species were located downfield of the corresponding resonances of the unreacted 5’-GMP. The only 13C resonances that exhibited significant differences in the chemical shifts between the complexed and unreacted species were C8 (Δδ = 2.9 ppm), C5 (Δδ = 1.6 ppm), and C1’ (Δδ = 1.3 ppm). The relatively large Δδ values for the C8 and C5 carbons flanking the N7 position is just what is expected for Au(III)-N7 GMP bonding, while the small Δδ of 0.1 ppm for C6 suggests that the C$#xFF1D;O is not involved in the bonding. Therefore, these data eliminate the N7-O6 chelate structure as a possible Au(III) bonding mode at pD 2.8. The Δδ for C1’ was ascribed to a conformational change of the ribose in the complexed species. These data also confirm that bonding of the Au(en) moiety is to the N7 of the 5'-GMP.

Table 5.

13C NMR Chemical Shifts of 5'-GMP and Au(en)-5′-GMP in D2O at pD 2.8

5′-GMP/ppma Au(en)-5′-GMP complex/ppmb,c
C6 156.49 156.60, 156.38
C2 155.20 155.27, 155.25
C4 150.36 150.34
C8 136.31 139.21, 136.26
C5 110.95 112.65, 110.69
C1' 89.01d 90.31, 89.14
C4' 84.19e 84.40e, 84.19e
C2' 74.60 74.92, 74.63
C3' 69.82 69.78, 74.63
C5' 63.84f 64.00, 63.66
en (react) 51.87, 50.90
en (unreact) 51.13
a

15 mM

b

11 mM

c

the more downfield line is 5′-GMP in Au(en)-5′-GMP and the upfield line is unreacted 5′-GMP

d

doublet, J = 1.0 Hz

e

split by 31P, J = 8.7 Hz

f

split by 31P, J = 5.0 Hz

The thermal stability of a solution which was 10 mM in [Au(en)Cl2]Cl and 15 mM in H2(5’-GMP) (pD 2.4), was investigated over the temperature range 2−50 °C. The complex decomposed slightly as the temperature was raised, decreasing from 33 % [Au(en)Cl2]Cl reacted at 2 °C to 27 % reacted at 50 °C. This implies that the coordination of 5’-GMP to [Au(en)Cl2] +is an exothermic process.

As one might expect, there is no significant reaction between Au(en)2Cl3 and Na2(5’-GMP). The only 1H NMR signals of 5’-GMP observed were those typical of unreacted nucleotide. This is due to the fact that the two coordinated bidentate en ligands bind to the Au(III) more strongly than 5’-GMP. Although no direct coordination of 5’-GMP was observed in Au(en)2Cl3-GMP system, H-bonding between 5’-GMP and the ethylenediamine in the Au(en)2Cl3 complex, as reported for other coordinatively saturated complexes,33,34 cannot be ruled out.

Supplementary Material

1si20050817_12
2si20050818_09
3si20050818_09
4si20050818_09
5si20051230_02

Acknowledgements

The authors thank the National Institutes of Health (NIGMS/SCORE program, grant SO6-08194) and The University of Texas at San Antonio for research support.

Abbrev

en

ethylenediamine (1,2-diaminoethane)

5’-GMP

guanosine 5’-monophosphate

Footnotes

Supporting Information Available Crystallographic data file in CIF format, CV as a function of scan rate, diffusion coefficients, 1H NMR of [Au(en)Cl2]Cl as a function of pD, 13C NMR available. This material is available free of charge via the Internet at http://pubs.acs.org.

References

  • 1.Messori L, Marcon G. Met. Ions Biol. Syst. 2004;42:385–424. [PubMed] [Google Scholar]
  • 2.Messori L, Marcon G, Orioli P. Bioinorg. Chem. Appl. 2003;1:177–187. doi: 10.1155/S1565363303000141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Tiekink ERT. Gold Bull. (London) 2003;36:117–124. [Google Scholar]
  • 4.Zhang CX, Lippard SJ. Curr. Opin. Chem. Biol. 2003;7:481–489. doi: 10.1016/s1367-5931(03)00081-4. [DOI] [PubMed] [Google Scholar]
  • 5.Marcon G, Carotti S, Coronnello M, Messori L, Mini E, Orioli P, Mazzei T, Cinellu MA, Minghetti G. J. Med. Chem. 2002;45:1672–1677. doi: 10.1021/jm010997w. [DOI] [PubMed] [Google Scholar]
  • 6.Abbate F, Orioli P, Bruni B, Marcon G, Messori L. Inorg. Chim. Acta. 2000;311:1–5. [Google Scholar]
  • 7.Fry FH, Hamilton GA, Turkevich J. Inorg. Chem. 1966;5:1943–1946. [Google Scholar]
  • 8.Gibson CS, Colles WM. J. Chem. Soc. 1931:2407–2416. [Google Scholar]
  • 9.Block BP, Bailar JC., Jr. J. Am. Chem. Soc. 1951;73:4722–4725. [Google Scholar]
  • 10.Makotchenko EV. Russ. J. Coord. Chem. (Trans. of Koord. Khim.) 2003;29:720–723. [Google Scholar]
  • 11.Minacheva LK, Gladkaya AS, Sakharova VG, Don GM, Shchelokov RN, Porai-Koshits MA. Zh. Neorg. Khim. 1988;33:683–687. [Google Scholar]
  • 12.Brodersen K, Kahlert T. Z. Anorg. Allg. Chem. 1967;355:323–327. [Google Scholar]
  • 13.Gibson DW, Beer M, Barrnett RJ. Biochemistry. 1971;10:3669–3679. doi: 10.1021/bi00796a006. [DOI] [PubMed] [Google Scholar]
  • 14.Chatterji D, Nandi US, Podder SK. Biopolymers. 1977;16:1863–1878. doi: 10.1002/bip.1977.360160904. [DOI] [PubMed] [Google Scholar]
  • 15.Yang T, Zhang J-Y, Tu C, Lin J, Liu Q, Guo Z-J. Wuji Huaxue Xuebao. 2003;19:45–48. [Google Scholar]
  • 16.Mizuno Y, Komiya S. Inorg. Chim. Acta. 1986;125:L13–L15. [Google Scholar]
  • 17.Marcon G, O'Connell T, Orioli P, Messori L. Metal-Based Drugs. 2000;7:253–256. doi: 10.1155/MBD.2000.253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Zhu S, Matilla A, Tercero JM, Vijayaragavan V, Walmsley JA. Inorg. Chim. Acta. 2004;357:411–420. [Google Scholar]
  • 19.Wirth W, Blotevogel-Baltronat J, Kleinkes U, Sheldrick WS. Inorg. Chim. Acta. 2002;339:14–26. [Google Scholar]
  • 20.SMART Software Reference Manual (1998) and SAINT Software Reference Manual (1998). Bruker-AXS, 5465 E. Cheryl Pkwy, Madison, WI 53719−1173 USA.
  • 21.Sheldrick GM. 2000. International Tables for Crystallography, Vol C. Kluwer; Boston: 1965.
  • 22.Sheldrick GM. SADABS. University of Göttingen; Germany: 2002. [Google Scholar]
  • 23.Suh MP, Kim IS, Shim BY, Hong D, Yoon T-S. Inorg. Chem. 1996;35:3595–3598. [Google Scholar]
  • 24.Wienken M, Lippert B, Zangrando E, Randaccio L. Inorg. Chem. 1992;31:1983–1985. [Google Scholar]
  • 25.Best SL, Chattopadhyay TK, Djuran MI, Palmer RA, Sadler PJ, Sovago I, Varnagy K. J. Chem. Soc., Dalton Trans. 1997:2587–2596. [Google Scholar]
  • 26.Hill DT, Burns K, Titus DD, Girard GR, Reiff WM, Mascavage LM. Inorg. Chim. Acta. 2003;346:1–6. [Google Scholar]
  • 27.Cornejo AA, Castineiras A, Yanovsky AI, Nolan KB. Inorg. Chim. Acta. 2003;349:91–96. [Google Scholar]
  • 28.Marangoni G, Pitteri B, Bertolasi V, Ferretti V, Gilli G. J. Chem. Soc., Dalton Trans. 1987:2235–2240. [Google Scholar]
  • 29.Micklitz W, Lippert B, Müller g., Mikulcik P, Riede J. Inorg. Chim. Acta. 1989;165:57–64. [Google Scholar]
  • 30.Messori L, Abbate F, Orioli P, Tempi C, Marcon G. Chem. Commun. 2002:612–613. doi: 10.1039/b111395j. [DOI] [PubMed] [Google Scholar]
  • 31.Messori L, Abbate F, Marcon G, Orioli P, Fontani M, Mini E, Mazzei T, Carotti S, O'Connell T, Zanello P. J. Med. Chem. 2000;43:3541–3548. doi: 10.1021/jm990492u. [DOI] [PubMed] [Google Scholar]
  • 32.Koelle U, Laguna A. Inorg. Chim. Acta. 1999;290:44–50. [Google Scholar]
  • 33.Fletcher TM, Walmsley JA. J. Inorg. Biochem. 1997;68:239–249. [Google Scholar]
  • 34.Farmer PJ, Cave JR, Fletcher TM, Rhubottom JA, Jr., Walmsley JA. Inorg. Chem. 1991;30:3414–3420. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1si20050817_12
2si20050818_09
3si20050818_09
4si20050818_09
5si20051230_02

RESOURCES