Gold(III)-induced oxidation of glycinet

Article abstract of DOI:10.1039/a902646k

NMR investigations of isotopically-labelled glycine show that Au^III induces deamination and subsequent decarboxylation of the amino acid with formation of glyoxylic acid, NH₄+, formic acid, CO₂ and metallic gold.

Full text of DOI:10.1039/a902646k

Gold(iii)-induced oxidation of glycine†
Juan Zou, Zijian Guo, John A. Parkinson, Yu Chen and Peter J. Sadler*
Department of Chemistry, University of Edinburgh, King’s Buildings, West Mains Road, Edinburgh, UK EH9 3JJ.
E-mail: P.J.Sadler@ed.ac.uk
Received (in Cambridge, UK) 1st April 1999, Accepted 11th June 1999
NMR investigations of isotopically-labelled glycine show
that AuIII induces deamination and subsequent decarboxyla-
tion of the amino acid with formation of glyoxylic acid,
acid 6 [1H: d 5.31(s)] with concomitant formation of NH4+ 5 [d
7.09 (d, 1JHN 73 Hz)] and Au(0). The identity of 6 was revealed
by further experiments. This species was stable over the pH
range 1.45–12.3 and a pH titration gave an associated pKa of
3.42. 13C–{1H} DEPT NMR data (CH group, 13C: d 89.4)
suggested assignment to the aldehyde group of glyoxylic acid
(lit.,13 pKa 3.46 at 298 K, I = 0). NMR spectra of glyoxylic acid
recorded at the same pH (1H: d 5.36; 13C: d 173.6 and 89.7)
further supported this assignment.
+
NH4 , formic acid, CO2 and metallic gold.
Several injectable 1+1 Au(i)–thiolate complexes and an orally-
active Au(i)–phosphine complex are used clinically for the
treatment of difficult cases of rheumatoid arthritis.1 Their
clinical application is limited, however, owing to severe host
toxicity such as kidney damage and blood disorders. The
molecular basis for these toxic side-effects of chrysotherapy is
poorly understood2 but recently Gleichmann et al.3 have
demonstrated that the oxidation of Au(i) to Au(iii) in vivo may
be responsible for some of the observed toxicity. In in-
flammatory situations, strong oxidants such as hypochlorite
(ClO2) and hydrogen peroxide (H2O2) are potentially available
in vivo and can oxidise Au(i)–thiolates and Au(i)–phosphines to
Au(iii).4 In view of this, and the antitumour activity of some
Au(iii) complexes,5 it is important to understand the chemistry
of Au(iii)–biomolecule interactions. Surprisingly, there are only
a few reports of reactions of Au(iii) with amino acids, and these
have focused on the sulfur-containing amino acids, namely
cyst(e)ine and methionine.6,7,8 Au(iii) can cleave the disulfide
bond of cystine to give the sulfonic acid6,7 and can oxidise the
sulfur of methionine stereospecifically to the sulfoxide.8 We
report here the first elucidation of the pathway of Au(iii)-
induced oxidation of the amino acid glycine. The identification
of reaction intermediates and final products was made possible
by the use of 13C and 15N isotopic labelling and multinuclear
NMR spectroscopy.‡ The proposed stepwise oxidation mecha-
nism provides a basis for obtaining further insights into the
mechanism of gold drug toxicity.
Both the Au(iii)– and Au(i)–imine intermediates are too
unstable to be observed by NMR during the reactions. The
+
NH4 produced reacts with [AuCl4]2 to give Au(iii)-ammine
adducts such as [AuCl3(NH3)] 7. This species was identified by
reaction of 1 with 15NH4Cl (1+2, pH 2.98). It gave rise to a
major HSQC cross-peak (1H/15N: d 5.93/212.9), which
1
exhibited H/2H isotope shift effects14–16 and chemical shifts
very similar to those of 7. Further substitution of Cl2 ligands by
NH3 is known to be difficult to achieve, and formation of
[Au(NH3)4]3+ requires the use of a large excess (super-saturated
+
solutions) of NH4 together with NH3 gas.17 Oxidative
decarboxylation of glyoxylic acid via further reaction with
Au(iii) gives rise to formic acid 8, carbon dioxide 9 and Au(0).
This was confirmed by two separate reactions of 1 with
glyoxylic acid to give formic acid 8 (1H: d 8.20, pH 1.52) and
with 2-[13C]Gly 2 to give [13C]8 (1H: d 8.20, 1JCH 218 Hz, 13C:
d 166.4). Therefore formic acid is formed from the aldehyde
Reactions of [AuIIICl4]2 1 with [15N]glycine [15N] 2 (1+2,
starting pH 2.44)§ were monitored by 1D 1H NMR (data
acquired at intervals over a 12 h period) and by 2D [1H, 15N]
HSQC-TOCSY NMR spectroscopy (acquired after 8 h of
reaction).9 On the basis of these results, a mechanism for the
reaction of Au(iii) with glycine can be proposed (Scheme 1).
Initial coordination of Gly to Au(iii) via the amino group gives
[AuCl3(Gly-N)] 3 [d(1H): CH2 3.85; d(1H/15N): 15NH2
6.88/2.6], which can undergo chelation to form [AuCl2(Gly-
N,O)] 4 [d(1H): CH2 3.75; d(1H/15N): 15NH2 6.75/29.2]. The
identification of the O,N-chelate 4 is based on chemical shift
arguments compared with analogous Pt(ii)–am(m)ine sys-
tems.10 For example, d(15N) of S,N-chelated Met in [Pt(Hdien-
N,N)(15N-Met-S,N)] is ca. 10 ppm to lower frequency compared
to monodentate Met in [Pt(dien)(15N-Met-N)].11 In our case,
d(15N) for 15NH2 of coordinated Gly differs by 12 ppm between
the monodentate and O,N-chelate Au complexes.
Once formed, the chelate 4 appears to be unreactive. A two-
electron transfer from Gly to Au(iii) in 3 gives rise to a Au(i)–
imine intermediate. Imines with a proton on N are seldom stable
and are known to undergo fast hydrolysis to the corresponding
aldehyde and NH4+.12 An imine ligand on Au(i) would also be
readily displaced by Cl2 and would hydrolyse to give glyoxylic
† Reactions of [AuIIICl4]2 1 with [15N]glycine as monitored by 1D 1H NMR
and 2D [1H, 15N] HSQC-TOCSY NMR can be viewed in electronic form,
see: http://www.rsc.org/suppdata/cc/1999/1359/.
Scheme 1
Chem. Commun., 1999, 1359–1360
1359

View Article Online
CVCP for an ORS Award to Y. C., Professor A. M. Sargeson
(Australian National University) and Dr R. M. Hartshorn
(University of Canterbury, New Zealand) for helpful discus-
sion, and Johnson Matthey plc for loan of gold salts.
Notes and references
‡ NMR spectra were recorded at 298 K on a Brüker DMX 500 NMR
spectrometer (1H 500.13 MHz, 15N 50.7 MHz), using the procedures
described previously.24 The 1D 13C-{1H} and 2D [1H, 13C] HETCOR NMR
spectra were acquired using standard procedures.
§ NMR samples were prepared in 90% H2O–10% D2O (600 ml). Fresh
solutions of Na[AuCl4]·2H2O (30 mM) and Gly (60 mM) were prepared
separately and then mixed in the following molar ratios (Au+Gly) at 298 K:
1+1, 2+1, 3+1, 5+1, 8+1, 10+1, 1+2 and 1+4. Fresh solutions of dl-Ala and
[AuCl4]2 were mixed in a 2+1 molar ratio at 298 K. The concentration of
[AuCl4]2 was 15 mM for all reactions. For reactions at neutral pH, the pH
values of the solutions of [AuCl4]2 and Gly were adjusted separately prior
to mixing, and remeasured immediately after mixing.
Fig. 1 Concentration vs. time profiles for species observed during the
reaction of [AuCl4]2 1 (15 mM) with [15N]glycine ([15N]2) (30 mM) at pH
2.87, 298 K. Concentrations were obtained from peak integrals using TSP
as a standard. The profile for free Gly (ca. 70% unreacted after 10 h) is not
shown.
¶ The same products were observed for the reactions conducted under N2
and in the absence of light.
1 C. F. Shaw III, Perspect. Biol. Med., 1979, 2, 287; P. J. Sadler, Struct.
Bonding (Berlin), 1976, 29, 171; K. C. Dash and H. Schmidbaur, Met.
Ions Biol. Syst., 1983, 14, 180; P. J. Sadler and R. E. Sue, Metal-Based
Drugs, 1994, 1, 107.
2 C. F. Shaw III, Comments Inorg. Chem., 1989, 8, 233; S. L. Best and
P. J. Sadler, Gold Bull., 1996, 29, 87.
3 E. Gleichmann, D. Schuhmann and M. Kubickamuranyi, Eur. J.
Pharmacol., 1990, 183, 78.
4 C. F. Shaw III, S. Schraa, E. Gleichmann, Y. P. Grover, L. Dunemann
and A. Jagarlamudi, Metal-Based Drugs, 1994, 1, 351; K. Takahashi, P.
Griem, C. Goebel, J. Gonzalez and E. Gleichmann, Metal-Based Drugs,
1994, 1, 483.
5 R. V. Parish, B. P. Howe, J. P. Wright, J. Mack, R. G. Pritchard, R. G.
Buckley, A. M. Elsome and S. P. Fricker, Inorg. Chem., 1996, 35, 1659;
P. Calamai, S. Carotti, A. Guerri, T. Mazzei, L. Messori, E. Mini, P.
Orioli and G. P. Speroni, Anti-Cancer Drug Des., 1998, 13, 67.
6 C. F. Shaw III, M. P. Cancro, P. L. Witkiewicz and J. E. Eldridge, Inorg.
Chem., 1980, 19, 3198; P. L. Witkiewicz and C. F. Shaw III, J. Chem.
Soc., Chem. Commun., 1981, 1111.
7 D. H. Brown and W. E. Smith, Am. Chem. Soc. Symp. Ser., 1983, 209,
401; W. E. Smith and J. Reglinski, Perspect. Bioinorg. Chem., 1991, 1,
183.
8 E. Bordignon, L. Cattalini, G. Natile and A. Scatturi, J. Chem. Soc.,
Chem. Commun., 1973, 878; G. Natile, E. Bordignon and L. Cattalini,
Inorg. Chem., 1976, 15, 246; A. A. Isab and P. J. Sadler, Biochim.
Biophys. Acta, 1979, 492, 322.
9 To view supplementary data in electronic form see: http://www.rsc.org/
suppdata/cc/1999/1359.
10 S. J. Berners-Price and P. J. Sadler, Coord. Chem. Rev., 1996, 151, 1,
and references therein.
11 Y. Chen, Z. Guo, P. del S. Murdoch, E. Zang and P. J. Sadler, J. Chem.
Soc., Dalton Trans., 1998, 1503.
12 Advanced Organic Chemistry: Reactions, Mechanisms, and Structure,
ed. J. March, John Wiley & Sons, New York, 1985, p. 784.
13 Critical Stability Constants, ed. A. E. Martell and R. M. Smith, Plenum
Press, New York, 1977, p. 65.
14 T. C. Stringfellow, G. Wu and R. E. Wasylishen, J. Phys. Chem. B,
1997, 101, 9651.
15 R. B. Calvert and J. R. Shapley, J. Am. Chem. Soc., 1978, 100, 7726.
16 F. A. L. Anet and A. H. Dekmezian, J. Am. Chem. Soc., 1979, 101,
5449.
17 L. H. Skibsted and J. Bjerrum, Acta Chem. Scand., Ser. A, 1974, 28,
740.
group of glyoxylic acid. Formic acid can be further oxidised by
Au(iii) to give Au(0) and carbon dioxide,18 but this step was not
observed in the current study using 2-[13C]Gly, probably due to
the slow rate of the reaction at low pH.18 The variations in
concentration of the species observed by NMR during the
reaction of 1 with [15N]2 with time are shown in Fig 1.
Reactions of 1 with [15N]2 at higher molar ratios gave rise to
similar NMR spectra but higher product yields. For example, in
the presence of a 10-fold excess of 1, all [15N]2 reacted within
1
14 h. The H NMR spectrum of a solution containing 1 and
[15N]2 in a 1+2 molar ratio, pH 7.14, recorded after 14 h showed
peaks for the CH2 signal of [15N]2 at d 3.56, together with
singlets at d 5.07 (assigned to the CH of glyoxylate) and at d
8.46 (assigned to formate). The course of the reaction between
1 and 2 at physiological pH§ therefore appears to be similar to
that at low pH. However, rapid proton exchange at pH 719 made
it impossible to carry out a detailed 1H,15N NMR study.
The mechanism in Scheme 1 predicts that the products from
+
the reaction§ of dl-alanine with [AuCl4]2 should be NH4 ,
1
pyruvic acid and acetylaldehyde. This was verified by H and
13C NMR spectroscopy, although the reaction was slower than
that of Gly.
A parallel mechanism has recently been proposed for Fe(iii)-
assisted oxidative cleavage of a C–N bond.20 Fe(iii)-bound
bis(2-pyridylmethyl)aminoacetate (BPG) is transformed into
Fe(iii)–bis(2-pyridylmethyl)amine via an Fe(iii)–imine inter-
mediate which undergoes hydrolysis with release of glyoxylic
acid. Photodecomposition of complexes such as [CoIII(Gly3)]
and [CoIII(Ala3)] have been reported to give rise to dipeptide
amides, together with CO2 and aldehydes.21 Photoredox
transformation of an Fe(iii) complex with N-(phosphonome-
thyl)glycine has also been reported, and when irradiated with
UV light the Gly derivative decomposes to give formaldehyde,
ammonia and a new species containing phosphorus.22 The
reactions with Au(iii) studied here were not activated by
light.¶
Gold(iii)-induced oxidation of Gly and related amino acids
may be important in relation to the severe toxicity of gold drugs.
It is possible that Au(iii) can also deaminate the amino terminus
of peptides and proteins. For example, during the reactions of
Au(iii) with tripeptides GGH and GGG, the formation of
colloidal gold is observed together with other unidentified
species.23 Au(iii) modification of peptides is likely to influence
MHC peptide presentation and T cell recognition systems,
allowing gold to have a major influence on the immune
system.
We thank the Wellcome Trust (Wellcome Travelling Re-
search Fellowship to J. Z), the Biotechnology and Biological
Sciences Research Council, and the Engineering and Physical
Science Research Council for their support for this work, the
18 B. S. Maritz and R. van Eldik, J. Inorg. Nucl. Chem., 1976, 38, 1749.
19 NMR of Proteins and Nucleic Acids, ed. K. Wüthrich, John Wiley &
Sons, New York, 1986, p. 24.
20 M.-C. Rodriguez, F. Lambert, I. Morgenstern-Badarau, M. Cesario, J.
Guilhem, B. Keita and L. Nadjo, Inorg. Chem., 1997, 36, 3525.
21 V. Balzani, V. Carassiti, V. Moggi and N. Sabbatini, Inorg. Chem.,
1965, 4, 1247.
22 M. J. Sima, Chem. Pap.—Chem. Zvesti., 1997, 51, 258.
23 S. L. Best, Ph. D. thesis, University of London, 1996.
24 N. Kratochwil, Z. Guo, P. del S. Murdoch, J. A. Parkinson, P. J.
Bednarski and P. J. Sadler, J. Am. Chem. Soc., 1998, 120, 8253; Y.
Chen, Z. Guo, J. A. Parkinson and P. J. Sadler, J. Chem. Soc., Dalton
Trans., 1998, 3577.
Communication 9/02646K
1360
Chem. Commun., 1999, 1359–1360