Chemical and Biological Studies of Dichloro(2-((dimethylamino)methyl)phenyl)gold(III)

Article abstract of DOI:10.1021/ic950343b

Several new organogold(III) derivatives of the type [AuX₂(damp)] (damp = o-C₆H₄CH₂NMe₂) have been prepared [X = CN, SCN, dtc, or X₂ = tm; dtc = R₂NCS₂ (R = Me (dmtc) or Et (detc)); tm = SCH(CO₂)CH₂CO₂Na] together with [AuCl(tpca)(damp)]Cl (tpca = o-Ph₂PC₆H₄CO₂H), [Au(dtc)(damp)]Y (Y = Cl, BPh₄) and K[Au-(CN)₃(damp)]. The ¹³C NMR spectra of these and previous derivatives have been fully assigned. In [Au(dtc)₂-(damp)] and K[Au(CN)₃(damp)], the damp ligand is coordinated only through carbon, as shown by X-ray crystallography and/or NMR. [Au(detc)₂(damp)] has space group C2/c, with a = 29.884(4) ?, b = 13.446(2) ?, c = 12.401(2) ?, β = 99.45(3)°, V = 4915 ?³, Z = 8, and R = 0.057 for 1918 reflections. The damp and one detc ligand are monodentate, the other detc is bidentate; in solution, the complex shows dynamic behavior, with the detc ligands appearing equivalent. The crystal structure of [Au(dmtc)(damp)]BPh₄ [Pna2₁, a = 26.149(5) ?, b = 11.250(2) ?, c = 11.921(2) ?, V = 3507 ?³, Z = 4, R = 0.073, 1772 reflections] shows both ligands to be bidentate in the cation, but the two Au-S distances are nonequivaient. The crystal structure of [Au(tm)(damp)] has also been determined [P2₁/n, a = 18.267(7) ?, b = 9.618(3) ?, c = 18.938(4) ?, β = 113.45(3)°, V = 3053 ?³, Z = 8, R = 0.079, 1389 reflections]. The tm is bound through sulfur and the carboxyl group which allows five-membered ring formation. In all three structures, the trans-influence of the σ-bonded aryl group is apparent. [AuCl₂(damp)] has been tested in vitro against a range of microbial strains and several human tumor lines, where it displays differential cytotoxicity similar to that of cisplatin. Against the ZR-75-1 human tumor xenograft, both [AuCl₂(damp)] and cisplatin showed limited activity.

Full text of DOI:10.1021/ic950343b

Inorg. Chem. 1996, 35, 1659-1666
1659
Chemical and Biological Studies of Dichloro(2-((dimethylamino)methyl)phenyl)gold(III)
Richard V. Parish,* Brian P. Howe, Jonathan P. Wright, Ju1rgen Mack, and
Robin G. Pritchard
Department of Chemistry, UMIST, PO Box 88, Manchester M60 1QD, U.K.
Robert G. Buckley,* Amanda M. Elsome, and Simon P. Fricker
Johnson Matthey Technology Centre, Blount’s Court, Sonning Common, Reading RG4 9NH, U.K.
ReceiVed March 22, 1995X
Several new organogold(III) derivatives of the type [AuX2(damp)] (damp ) o-C6H4CH2NMe2) have been prepared
[X ) CN, SCN, dtc, or X2 ) tm; dtc ) R2NCS2 (R ) Me (dmtc) or Et (detc)); tm ) SCH(CO2)CH2CO2Na]
together with [AuCl(tpca)(damp)]Cl (tpca ) o-Ph2PC6H4CO2H), [Au(dtc)(damp)]Y (Y ) Cl, BPh4) and K[Au-
(CN)3(damp)]. The 13C NMR spectra of these and previous derivatives have been fully assigned. In [Au(dtc)2-
(damp)] and K[Au(CN)3(damp)], the damp ligand is coordinated only through carbon, as shown by X-ray
crystallography and/or NMR. [Au(detc)2(damp)] has space group C2/c, with a ) 29.884(4) Å, b ) 13.446(2) Å,
c ) 12.401(2) Å, â ) 99.45(3)o, V ) 4915 Å3, Z ) 8, and R ) 0.057 for 1918 reflections. The damp and one
detc ligand are monodentate, the other detc is bidentate; in solution, the complex shows dynamic behavior, with
the detc ligands appearing equivalent. The crystal structure of [Au(dmtc)(damp)]BPh4 [Pna21, a ) 26.149(5) Å,
b ) 11.250(2) Å, c ) 11.921(2) Å, V ) 3507 Å3, Z ) 4, R ) 0.073, 1772 reflections] shows both ligands to be
bidentate in the cation, but the two Au-S distances are nonequivalent. The crystal structure of [Au(tm)(damp)]
has also been determined [P21/n, a ) 18.267(7) Å, b ) 9.618(3) Å, c ) 18.938(4) Å, â ) 113.45(3)o, V ) 3053
Å3, Z ) 8, R ) 0.079, 1389 reflections]. The tm is bound through sulfur and the carboxyl group which allows
five-membered ring formation. In all three structures, the trans-influence of the σ-bonded aryl group is apparent.
[AuCl2(damp)] has been tested in Vitro against a range of microbial strains and several human tumor lines, where
it displays differential cytotoxicity similar to that of cisplatin. Against the ZR-75-1 human tumor xenograft, both
[AuCl2(damp)] and cisplatin showed limited activity.
Introduction
these materials would be toxic. We reasoned that reduction of
the oxidation potential of gold(III) could be achieved by the
use of softer ligands, and we now have a large range of
complexes containing N-C- ligands, involving a σ-bonded
phenyl, naphthyl, or similar group with various nitrogen-
containing substituents. These materials are indeed more stable
in reducing, biological media.
In this paper we concentrate on derivatives of the 2-((dimeth-
ylamino)methyl)phenyl ligand (-C6H4CH2NMe2 ) damp),
which forms orthometalated C,N-complexes. Such complexes
have been known for about 10 years,6 but do not seem to have
been examined for biological activity. We have reprepared
known complexes of the type [AuX2(damp)] and further
characterized them spectroscopically; we have also obtained
some new materials, of which those where the ligands X are
cyanide or a dithiocarbamate are of special structural interest
because the amine group is detached from the coordination
sphere of the gold. This has been observed previously only
for two compounds, both of which contain two damp ligands:
[AuX(damp)2] (X ) Cl, CN).7 In later papers we shall describe
a range of complexes containing more novel N-C- ligands.8
Gold(III) is isoelectronic with platinum(II), and the complexes
of both metals normally adopt a square-planar configuration.
These similarities open the possibility that gold(III) complexes
might have biological activity parallel to cisplatin and its
homologues. Although [AuCl4]- is widely thought to be toxic,
it was used in the treatment of syphilis and alcoholism,1 and
dimethylgold(III) derivatives showed some degree of antitumor
activity.2 There have been many later studies in which gold
compounds have been tested for antitumor properties, although
most have involved gold(I); these studies have been reviewed
recently.3,4
To maintain the formal resemblance to cisplatin we have
chosen complexes having a single mononegative bidentate
ligand together with two monodentate anionic ligands; substitu-
tion reactions with a variety of other potentially bidentate ligands
of various charges have also been carried out. Previously we
described complexes [AuX2(L-L′)], in which L-L′ was pyri-
dine-2-carboxylate or an analogue.5 Unfortunately, in biological
media these complexes underwent reduction to metalic gold;
the implied oxidation of biological material might suggest that
X Abstract published in AdVance ACS Abstracts, February 1, 1996.
(1) Parish, R. V. Interdisc. Sci. ReV. 1992, 17, 221-228. Higby, G. J.
Gold Bull. 1982, 15, 130.
(2) National Cancer Institute, quoted by: Berners-Price, S. J.; Sadler, P.
J. In Frontiers in Bioinorganic Chemistry; Xavier, A. V., Ed.; VCH
Verlag: Weinheim, Germany, 1976, p 376.
(3) Ni Dubhghaill, O. M.; Sadler, P. J. In Metal Complexes in Cancer
Chemotherapy; Keppler, B. K., Ed.; VCH: Weinheim, Germany, 1993,
p 221.
(5) Dar, A.; Moss, K: Cottrill, S. M.; Parish, R. V.; McAuliffe, C. A;
Pritchard, R. G.; Beagley, B.; Sandbank, J. J. Chem. Soc., Dalton Trans
1992, 1907.
(6) (a) Vicente, J.; Chicote, M. T.; Bermu´dez, M. D. J. Organomet. Chem.
1984, 268, 191. (b) Vicente, J.; Chicote, M. T.; Bermu´dez, M. D.;
Jones, P. G.; Sheldrick, G. M. J. Chem. Res. Miniprint 1985; J. Chem.
Res., Synop. 1985, 954, 72.
(7) Vicente, J.; Bermu´dez, M. D.; Sa´nchez-Santana, M. J.; Paya, J. Inorg,
Chim. Acta 1990, 174, 53.
(4) Shaw, C. F. In Metal Compounds in Cancer Therapy; Fricker, S. P.,
Ed.; Chapman and Hall: London, 1994, p 46.
(8) Bonnardel, P. A.; Parish, R. V.; Wright, J. P. Manuscripts in
preparation.
0020-1669/96/1335-1659$12.00/0 © 1996 American Chemical Society

1660 Inorganic Chemistry, Vol. 35, No. 6, 1996
Parish et al.
Table 1. Selected Bond Lengths (Å) and Angles (deg)a
[Au(dmtc)(damp)]BPh4
[Au(detc)2(damp)]
2.04(2)
[Au(tm)(damp)]b
2.16(5), 2.04(5)
Au-C
Au-N
Au-Sd
Au-Sd
2.03(2)
2.23(2)
2.340(9)
2.179(9)
Au-C
Au-Sc
Au-Sc,d
Au-S
Au-C
Au-S
Au-Od
Au-N
2.325(5)
2.391(5)
2.311(5)
2.23(2), 2.23(2)
2.09(3), 2.12(3)
1.85(5), 2.00(5)
C-Au-S
S-Au-Sc
S-Au-N
N-Au-C
C-Au-S
N-Au-S
S-C-S
100.0(6)
C-Au-S
S-Au-S
S-Au-S
S-Au-C
C-Au-S
S-Au-S
S-C-S
95.2(3)
74.7(2)
98.8(2)
C-Au-S
S-Au-O
O-Au-N
N-Au-C
C-Au-O
S-Au-N
99(2), 98(1)
85(1), 84(1)
95(2), 97(2)
80(2), 98(1)
172(2), 178(1)
174(2), 177(2)
78.6(3)
97.7(6)
81.5(8)
171.3(6)
177.6(6)
111(2)c
91.3(5)
169.9(5)
173.3(2)
120(2),c 127(1)f
a Data are listed clockwise from the Au-C bond (see Figures 1, 3, and 4). b Two independent molecules. c For bidentate dtc. d Trans to C.
e Trans to N. f For monodentate dtc.
Crystallographic Studies. All measurements were performed at
room temperature using graphite monochromated Mo KR radiation (λ
) 0.710 69 Å).
We have also carried out a range of biological tests to evaluate
[AuCl2(damp)] for antimicrobial and antitumor activity. In-
cluded in these tests, as comparisons, are cisplatin and also
[AuCl2(ppy)] (ppy ) 2-pyridylphenyl); the latter is a gold(III)
complex with similar structural features.9
[Au(dmtc)(damp)]BPh4 was examined using a Rigaken AFC6S
diffractometer, and [Au(detc)2(damp)] and [Au(tm)(damp)] with an
Enraf-Nonius CAD-4. ω scans were used on the first two samples
while [Au(detc)2(damp)] was scanned by the ω/2θ technique.
Crystal data and experimental details are given in the Supporting
Information. The structures were solved by Patterson methods and
refined by full-matrix least-squares using the TEXSAN suite of
programs.10 The gold, sulfur, and some terminal carbon atoms were
subjected to anisotropic refinement in [Au(detc)2(damp)], while only
the gold and sulfur atoms were refined anisotropically in [Au(dmtc)-
(damp)]BPh4 and [Au(tm)(damp)]. Hydrogen atoms were placed in
chemically reasonable positions in all three cases, except for the
carboxylic acid proton in [Au(tm)(damp)], which was ignored. Ad-
ditionally, the phenyl ring in [Au(dmtc)(damp)]BPh4 was constrained
to be a regular hexagon of side 1.40 Å. The aromatic rings in the
other compounds were accurately planar (maximum deviation from the
least-squares plane: 0.03(2) and 0.14(8) Å for the detc and tm
derivatives respectively).
Experimental Section
The complex [AuCl2(damp)] was prepared as previously described6a
by reaction of [HgCl(damp)] and Me4N[AuCl4]. The derivatives [AuX2-
(damp)] (X ) Br, I, CN) were prepared by metathesis, again by
literature methods.6a Analogous procedures were used to make
complexes with X ) SCN (from KSCN), dmtc, or detc (from NaS2-
CNR2, R ) Me, Et) or with X2 ) mercaptosuccinate (from the
corresponding silver salt). The complex [AuCl(tpca)(damp)]Cl (tpca
) Ph2PC6H4CO2H) was prepared analogously to the corresponding PPh3
complex.6a The complex [AuCl2(ppy)] was prepared according to the
literature method.9 Analytical data for all materials prepared are given
in the Supporting Information.
(Mercaptosuccinato)[2-((dimethylamino)methyl)phenyl]gold-
(III). Silver mercaptosuccinate (0.35 g, 1.0 mmol) was added in one
portion to a solution of [AuCl2(damp)] (0.20 g, 0.50 mmol) in acetone
(30 cm3). The mixture was stirred in the dark for 1 h, evaporated to
dryness, and extracted with dichloromethane. Addition of hexane (30
cm3) to the extract and refrigeration afforded small yellow cubes of
the complex after 24 h; crystallization continued for 7 d affording a
total of 0.17 g, 70%).
The default TEXSAN weighting scheme of 1/σ6(Fo) was used for
all three studies.
Selected bond lengths and angles are given in Table 1; final atomic
coordinates are deposited as Supporting Information and additional
material is available from the Cambridge Crystallographic Data Center
(atomic coordinates, thermal parameters, and the remaining bond
distances and angles).
Pharmacological Studies. Microbial Strains. In Vitro antimicro-
bial activity was assessed using a panel of bacteria and fungi
representing microorganisms of clinical importance. The bacterial
strains used were Staphylococcus aureus NCTC 6571, Enterococcus
faecalis NCTC 8727, Klebsiella pneumoniae NCTC 9633, Escherichia
coli NCTC 10418, and Pseudomonas aeruginosa NCTC 10662. The
fungi consisted of Candida albicans NCFC 3153, Cryptococcus
neoformans NCFC 3003, Aspergillus fumigatus NCFC 2140, and a
strain each of Candida albicans and Aspergillus niger which were
obtained from the University of Surrey.
Potassium Tricyano[2-((dimethylamino)methyl)phenyl]aurate-
(III). The complex [AuCl2(damp)] (0.20 g, 0.5 mmol) was dissolved
in acetone (30 cm3), and potassium cyanide (0.19 g, 3.0 mmol) was
added. This suspension was stirred for 6 h and filtered. Removal of
solvent gave an oil which solidified in Vacuo after 30 min at 35 °C.
The solid was washed with hexane. Yield: 0.078 g, 35%. The product
was found to be highly soluble in water.
NMR measurements were made on Brucker AC250 (1H, 13C) and
AC200 (31P) spectrometers, the latter using 85% H3PO4 as reference.
Concentrations were 0.05-0.2 mol dm-3 13C spectra were measured
.
at a flip angle of 28.2° and data point resolution of 0.137 Hz. 13C
DEPT90° spectra gave a negative signal for the CH2 group and positive
signals for CH and CH3 groups; quaternary carbon atoms did not
respond. For the 13C-1H shift-correlation (COSY) spectra the following
pulse sequence was used:
Cell Lines. The CHO fibroblast11 and the human tumor cell lines
SW403, SW620, and SW1116 colon carcinoma,12 HT29/219 colorectal
carcinoma,13 ZR-75-1 breast carcinoma,14 HT1376 bladder carcinoma,15
and the SK-OV-3 ovarian carcinoma16 were obtained from the ECACC.
1H: D0-90-D0 D0-D3-90 BB
(10) TEXSAN. TEXRAY Structure Analysis Package. Molecular Structure
Corporation 3200A Research Forest Drive, The Woodlands, TX, 1985.
(11) Puck, T. T.; Cieciura, J.; Robinson, R. J. Exp. Med. 1958, 108, 945.
(12) Liebovitz, A.; Stinson, J. C.; McCombs, W. B., III; McCoy, C. E.;
Mazur, K. C.; Mabry, N. D. Cancer Res. 1976, 36, 4562.
(13) Fogh, J.; Trempe, G. Human tumor Cells in Vitro Fogh J., Ed.; Plenum
Press: New York, 1975, p 115.
(14) Engel, L. W.; Young, N. A.; Tralka, T. S.; Lippman, M. E.; O’Brien,
S. J.; Joyce, M. J. Cancer Res. 1978, 38, 3352.
(15) Rasheed, S.; Gardner, M. B.; Rongey, R. W.; Nelson-Rees, W. A.;
Arnstein, P. J. Natl. Cancer Inst. 1977, 58, 881.
13C: D1 180 90-D4 FID
This allowed the 1H signals to be paired with the appropriate carbon
signals. In H decoupling experiments, each of the aromatic-proton
doublets was irradiated in turn, allowing identification of pairs of
adjacent protons. Finally, irradiation at the CH2 frequency gave a strong
NOE on the signal at δ 7.15 and a weak, negative effect at δ 7.75; the
former signal was therefore attributed to H3.
1
(16) Fogh, J.; Wright, W. C.; Loveless, J. D. J. Natl. Cancer Inst. 1977,
58, 209.
(9) Constable, E. C.; Leese, T. A. J. Organomet. Chem. 1989, 363, 419.

Dichloro(2-((dimethylamino)methyl)phenyl)gold(III)
Inorganic Chemistry, Vol. 35, No. 6, 1996 1661
Antimicrobial Testing. Minimum inhibitory concentration (MIC)
values for each test organism were determined using the plate dilution
technique17 to assess the antimicrobial activity of [AuCl2(damp)] and
[AuCl2(ppy)]. Iso-sensitest agar and 2% purified agar (Unipath Ltd.)
supplemented with 10% yeast nitrogen base18 were used for the bacterial
and fungal MIC determinations respectively. An inoculum of 106
organisms mL-1 was applied with a multipoint inoculating device, and
the plates were examined for microbial growth after incubation for either
18 h at 37 °C (bacteria) or 48 h at 30 °C (fungi). Ciprofloxacin and
amphotericin B were included as controls.
Mammalian-Cell Toxicity. Comparative cytotoxicity against mam-
malian cells was assessed using the Chinese Hamster Ovary (CHO)
fibroblast cell line.19 The cells were seeded onto 96-well microtitre
plates at a concentration of 5 × 104 cells mL-1, in 200 µL of medium.
After a 24-h preincubation period, the cells were treated with test
Figure 1. Structure of the cation in [Au(dmtc)(damp)]BPh4. Bond
lengths and angles are given in Table 1 and Figure 2.
compound for 2 h at concentrations of 0.1, 1.0, 10.0, and 100 µg mL-1
.
The solution of the compound was then replaced with fresh medium
and the cells were incubated for a further 48 h. Cell growth was assayed
using sulforhodamine B as previously described.20 Cell survival (%)
was calculated relative to untreated control cells.
The complex [AuCl(tpca)(damp)]Cl seems to be completely
similar to its PPh3 analogue,6a but its conductivity is low (31.2
S mol-1 at 10-3 mol dm-3 in MeNO2); the most probable
explanation for this is association between the uncoordinated
carboxylic acid group and the chloride ion: -CO2H‚‚‚Cl-. A
less likely possibility is coordination of the second chloride with
displacement and protonation of the amine group to form a
zwitterionic complex; given the general reluctance of the amine
group to be displaced, this seems unlikely. In both phosphine
complexes, IR data indicate, as previously deduced,6a that the
phosphine is trans to the nitrogen of damp [ν(Au-Cl) 315 and
305 cm-1 for PPh3 and tpca, respectively].
The complex K[Au(CN)3(damp)] is difficult to prepare
reproducibly. It is a 1:1 electrolyte in acetonitrile (156 S mol-1
at 10-3 mol dm-3), and NMR data (see below) indicate that
the gold atom is four-coordinate and that the damp-amine group
is not coordinated.
Unexpectedly, a similar detachment of the amine group occurs
in [Au(detc)2(damp)], as shown by X-ray crystallography (see
below). In gold-damp complexes, this mode of bonding has
only been previously observed in the bis(damp) compounds
[AuX(damp)2] (X ) Cl, CN).7 The NMR spectra show that
this compound is fluxional in solution, with the damp ligands
being alternately mono- and bidentate.
Primary in Witro Tumor Screen. Primary screening for antitumor
activity was conducted against an in Vitro panel of seven human tumor
lines.21 The cells were seeded onto 96-well microtitre plates at
concentrations of 5 × 104 to 1 × 105 cells mL-1. After a 24-hour
pre-incubation period, the cells were treated with test compound for 2
h at concentrations of 0-200 µg mL-1
. The compound was then
replaced with fresh medium and the cells were incubated for a further
72 h. Cell growth was assayed using sulforhodamine B. Cell survival
(%) was calculated relative to untreated control cells. Dose/survival
curves were constructed from these data, and the IC50 (concentration
of compound giving 50% survival) was calculated.
Xenograft Study. The ZR-75-1 tumor was implanted subcutane-
ously in the flanks of nude mice. The tumor-bearing animals were
then randomized into treatment groups of six mice and a control group
of ten mice. Test compounds were administered intraperitoneally in
0.5% carboxymethylcellulose in 0.9% saline. For the experiment using
[AuCl2(damp)] the doses were given on days 0, 7, 14, and 21, and
tumor dimensions were taken and animals weighed on days 0, 7, 14,
21, and 28. Cisplatin was also evaluated in this model using a similar
procedure but with slight differences in the dosing/tumor measurement
intervals, as indicated in Table 8. The tumor volume was calculated
from the tumor dimensions using the formula: length × width × depth
× π/6. Results are expressed as relative tumor volumes (RTV) where
RTV ) 100(mean volume of tumors at assessment time)/(mean volume
of tumors at the start of the experiment). Results were analyzed using
Student’s t test.
Crystallographic Studies
The xenograft study was carried out in the Cancer Research
Campaign Laboratories at the Charing Cross Hospital, London.
[Au(dmtc)(damp)]BPh4. The mono(dmtc) cation has the
expected structure in which gold(III) is square-planar coordi-
nated to two chelated ligands (Figure 1 shows the cation, and
Figure 2a gives a schematic of the important bond lengths and
angles). The two Au-S bond lengths are appreciably different,
with that trans to C being the longer as expected on trans-
influence grounds. This asymmetry extends to the C-S bonds
of the dmtc ligand, the longer bond being to the sulfur trans to
N. The structure therefore approximates to that of a chelated
Me2NC(dS)S- ligand, in which the “anionic” sulfur atom is
coordinated trans to N.
The Au-C and Au-N bond lengths are similar to, but slightly
longer than, those reported for [AuCl(R)(damp)] (R ) C6H5,
C6F5)22 and [Au(X-Y)(damp)] (X-Y ) oxin, H2NC6H4S).23
[Au(detc)2(damp)]. The dimensions and structure of the bis-
(detc) derivative is shown in Figures 2 and 3. The gold atom
appears to be approximately octahedrally coordinated: the
damp-C atom, one detc-S atom of one detc and two of the other
Results
Several complexes of the types [AuX2(damp)] and [Au(X-X)-
(damp)] (X, X-X ) mono- and bidentate mononegative ligands)
have been obtained by procedures analogous to those reported
previously,6 of which those with X ) CN, SCN, dmtc, and detc,
[AuCl(tpca)(damp)]Cl (tpca ) o-Ph2PC6H4CO2H); [Au(X-X)-
(damp)]Y (X-X ) dmtc, detc, tm; Y ) Cl, BPh4), and K[Au-
(CN)3(damp)] are new. Analytical, infrared, and conductivity
data are entirely consistent with the expected formulations
(Supporting Information).
(17) Washington, J. A.; Barry, A. L. Dilution Test Procedures. In Manual
of Clinical Microbiology; 2nd ed.; Lennette, E. H., Spaulding, E. H.,
Traunt, J. P., Eds.; American Soceity for Microbiology: Washington,
DC, 1974, p 410.
(18) Shadomy, S. Appl. Microbiol. 1969, 17, 871.
(19) Fricker, S. P. Toxicol. In Vitro 1994, 8, 821.
(20) Higgins, J. D. III; Neely, L.; Fricker, S. J. Inorg. Biochem. 1993, 49,
149.
(21) Fricker, S. P. Screening for New Metal Anti-tumor Agents. In Metal
Ions in Biology and Medicine; Collery, P., Poirier, L. A., Manfait,
M., Etienne, J-C., Eds.; John Libbey Eurotext: Paris, 1990, p 452.
(22) Vicente, J.; Chicote, M. T.; Bermu´dez, M. D.; Sanchez-Santana, M.
J.; Jones, P. G. J. Organomet. Chem. 1988, 354, 381. Vicente, J.;
Bermu´dez, M. D.; Escribano, J; Carrillo, M. P.; Jones, P. G. J. Chem.
Soc., Dalton Trans, 1990, 3083.
(23) Vicente, J.; Chicote, M. T.; Bermu´dez, M. D.; Jones, P. G.; Fittschen,
C.; Sheldrick, G. M. J. Chem. Soc., Dalton Trans. 1986, 2361.

1662 Inorganic Chemistry, Vol. 35, No. 6, 1996
Parish et al.
Figure 4. Molecular structure of [Au(tm)(damp)].
by the inability to have three bidentate ligands present without
increasing the coordination number of the metal atom above
four. The gold atom is only 0.0153 Å out of the coordination
plane.
One other example of mixed coordination for dtc ligands in
a gold(III) complex is known: [Au(detc)3].24e In this case, one
detc ligand is symmetrical and bidentate (Au-S ) 2.333 Å),
the other two are monodentate (Au-S ) 2.350 Å).
[Au(tm)(damp)]. In this compound the point of interest is
the coordination of the thiomalate ligand (Figure 4). The soft
nature of gold(III) dictates choice of the thiolate group as one
donor; it is, as expected on trans-influence grounds, trans to
N. The carboxylate group bound to the carbon atom R to the
thiol group also coordinates, giving a five-membered chelate
ring; coordination of the other carboxylate would give a six-
membered ring. The Au-S bond distance, 2.23(2) Å, is slightly
shorter than for the dithiocarbamates, mainly due to the fact
that it is trans to the nitrogen donor. In the related 2-amino-
benzenethiolate derivative, the Au-S bond is 2.267 Å (trans
to nitrogen).23 [Au(tm)(damp)] appears to be the first example
of a structure in which a gold atom is bonded to four different
donor atoms.
Figure 2. Schematic of principal bond lengths and bond angles for
(a) [Au(dmtc)(damp)]BPh4, (b) [Au(detc)2(damp)], (c) [Au(tm)(damp)]
(average of two independent molecules).
The [Au(tm)(damp)] molecules are linked in strands by
hydrogen bonding between the free carboxylate group and the
carbonyl oxygen atom of the coordinated carboxyl group.
Spectroscopic Studies. Detailed H/13C, COSY, selective
1
Figure 3. Molecular structure of [Au(detc)2(damp)].
decoupling, and NOE studies were made of [AuCl2(damp)], in
both DMSO-d6 and (CD3)2CO, in order to determine fully the
assignments for each proton and carbon atom. Eventually, the
only ambiguity remaining was for C1 and C2, the two carbon
atoms not bound to hydrogen. It was then assumed that the
the atom bound to the gold atom would have the higher chemical
shift. This assignment is supported by the data for [HgCl-
(damp)] given below, where C1 can be unambiguously identified
by its coupling to the mercury atom, and by data for a range of
other aryl mercury(II) and gold(III) derivatives.8,25-27 The final
assignments are summarized diagrammatically in Figure 5.
These results slightly amend those reported previously.6 It was
then assumed that the order of chemical shifts was the same
for each derivative, and all the available data are summarized
in Table 2.
define a conventional square plane. The damp-N atom and the
second S atom of one detc group lie above and below the AuCS3
square plane (the Au-S and Au-N vectors form angles of 63.1°
and 71.2° with the plane, and S-Au-N ) 151.9(4)°). How-
ever, these bonds are considerably elongated: Au-N ) 3.01-
(2) Å and Au-S ) 3.196 Å. While the damp and detc ligands
are not constrained to have these atoms in these positions, the
interactions with gold must be very weak. The damp ligand
and one detc must therefore be considered to be essentially
monodentate. The other detc is bidentate and more symmetrical
than that in [Au(dmtc)(damp)]BPh4. The three Au-S bond
distances are comparable to those in other gold(III) chelated-
dithiocarbamate derivatives.24 The adoption of this configura-
tion, rather than that with two monodentate detc groups, is
presumably driven by the relatively soft nature of gold(III),
which favors coordination of sulfur rather than nitrogen, and
For the damp ligand, substantial solvent shifts are observed.
However, comparing data for the same solvent, large positive
coordination chemical shifts are seen for C1, as expected.
Although only a limited range of data is available, the chemical
(24) (a) Noordick, J. H.; Buerskens, P. J. J. Cryst. Mol. Struct. 1971, 1,
339. (b) Buerskens, P. T.; Blaauw, H. J. A.; Cras, J. A.; Steggerda, J.
J. Inorg. Chem. 1968, 7, 805. (c) Buerskens, P. T.; Cras, J. A.; van
der Linden, J. G. M. Inorg. Chem. 1970, 9, 475. (d) Cras, J. A.;
Noordik, J. H.; Buerskens, P. T.; Verhoeven, A. M. J. Cryst. Mol
Struct. 1971, 1, 155. (e) Noordik, J. H. Cryst. Struct. Commun. 1973,
2, 81. (f) Uso´n, R.; Laguna, A.; Laguna, M; Castilla, M. L.; Jones, P.
G.; Mayer-Ba¨se, K. J. Organomet. Chem. 1987, 336, 453.
(25) Rowland, K. E.; Thomas, R. D. Magn. Reson. Chem. 1976, 21, 437.
(26) van der Ploeg, A. F. M. J.; van der Kolk, C. E. M.; van Koten, G. J.
Organomet. Chem. 1981, 212, 283.
(27) Parish, R. V.; Mack, J.; Hargreaves, L.; Wright, J. P.; Buckley, R.
G.; Elsome, A. M.; Fricker, S. P.; Theobald, B. R. C. J. Chem. Soc.,
Dalton Trans. 1996, 69.

Dichloro(2-((dimethylamino)methyl)phenyl)gold(III)
Inorganic Chemistry, Vol. 35, No. 6, 1996 1663
Table 2. 13C Chemical Shifts in Gold(III)-Damp Complexes
compound
solvent
C1
C2
C3
C4
C5
C6
CH2
CH3
[AuCl2(damp)]
[AuCl2(damp)]
CDCl3
148.2
148.9
151.1
152.9
149.1
152.8
153.1
149.2
145.0
144.5
146.9
128.4
128.4
143.9
146.9
145.6
151.3
147.3
152.2
152.4
146.9
135.0
140.9
136.5
138.5
138.5
123.5
124.2
128.5
127.7
125.1
128.7
129.2
132.6
124.1
125.9
129.4
128.4
128.4
129.2
129.4
132.8
132.6
128.1
132.0
133.1
136.1
128.7
129.2
132.6
127.7
127.5
128.2
127.8
128.7
131.4
127.8
131.7
131.9
133.4
128.4
127.8
131.1
126.4
126.4
131.3
131.1
133.6
139.1
129.1
132.9
131.9
136.2
131.1
133.3
139.7
127.5
127.5
76.0
76.0
76.3
78.2
74.0
77.1
77.7
73.6
71.6
66.4
70.1
64.0
63.9
54.0
53.9
54.4
57.5
53.4
56.6
56.9
51.6
51.3
45.4
48.6
45.0
44.7
(CD3)2CO
CD2Cl2
(CD3)2SO
CDCl3
(CD3)2SO
(CD3)2SO
CDCl3
CDCl3
CDCl3
(CD3)2SO
CDCl3
CDCl3
[AuBr2(damp)]
[Au(CN)2(damp)]a
[Au(detc)(damp)]Clb
[Au(detc)(damp)]BPh4
c
d
[Au(dmtc)(damp)]BPh4
[AuCl(PPh3)(damp)]Cle
[Au(tm)(damp)]f
[Au(detc)2(damp)]g
K[Au(CN)3(damp)]h
Hdampi
Hdampi,j
a CN signals: 144.2, 109.8. b detc signals: 195.5 (SSCN); 48.4, 47.1 (NCH2CH3); 12.7, 12.4 (NCH2CH3). c detc signals: 197.8 (SSCN); 52.2,
50.8 (NCH2CH3); 16.5, 16.2 (NCH2CH3). d dmtc signals: 196.9 (SSCN); 46.1, 44.8 (NCH3). e PPh3 signals at 129.6, 129.4, 128.3, 127.2, 125.7,
125.0. f tm signals: 183.0, 172.0 (C ) O); 103.0 (C-S); 51.3 (CH2). g detc signals: 200.5 (SSCN); 46.9 (NCH2CH3); 12.3 (NCH2CH3). h Solution
containing [Au(CN)2(damp) + 2KCN; CN signals: 129.7, 118.0. i Numbering corresponds to that for the gold complexes. j From ref 26.
Scheme 1
chelated damp complexes, but similar to those for free N,N-
dimethylbenzylamine (Table 2). There is also a small shift to
lower frequency for C2. Only a single set of signals is seen for
the dtc groups [1H: δ(CH2) 3.78, δ(CH3) 1.15], even though
the X-ray data show one to be bidentate and the other
monodentate. The apparent equivalence is best explained by
fluxionality, the two dtc ligands being alternately mono- and
bidentate. This switching of coordination is fast on the NMR
timescale, and it is presumably the sulfur atom trans to the
C-bonded damp ligand which is labile (Scheme 1).
In an attempt to freeze the fluxionality, the spectrum of [Au-
(detc)(damp)] was remeasured at -40 °C. No additional signals
were seen, but the ethyl group peaks diminished in relative
intensity, suggesting that they were broadened.
The 13C spectrum of [Au(CN)2(damp)] contains two signals
additional to those for the damp ligand; their low intensity
Figure 5. NMR chemical shift data for [AuCl2(damp)] in (a) CDCl3
and (b) (CD3)2CO. Shifts for H are given in italics.
suggests that both are for carbon atoms not bonded to hydrogen,
and this is confirmed by the DEPT spectrum. A unique signal
at 109.8 ppm is close to that reported for [Au(CN)4]- (105.2
ppm in D2O),28 and can confidently be assigned to a bound
cyanide ligand. Of the three other quaternary carbon signals,
two are assigned to C1 and C2 of the damp ligand by comparison
with the dithiocarbamate complexes; the signal at 144.2 ppm
must then correspond to the second cyanide ligand. It is
significantly more shielded than the other, which suggests that
it is cis to the aryl group. The fourth coordination position is
occupied by the amine group, as shown by the CH2 and CH3
chemical shifts.
The spectrum of K[Au(CN)3(damp)] could not be obtained
directly, owing to the difficulty of synthesis. However, on
addition of 2 equiv of KCN to a DMSO-d6 solution of [Au-
(CN)2(damp)], extra peaks are seen in the 13C spectrum,
indicating the presence of a second damp-containing species.
In particular, the CH2NMe2 chemical shifts show that the amine
1
shift appears to decrease as the trans-influence of the ligand
opposite increases. Significant coordination shifts are seen for
the adjacent carbon atoms, C2 and C6, also positive. The signals
for the CH2 and CH3 carbon atoms show substantial positive
coordination shifts in cases where the ligand is known to chelate.
In other cases, these shifts are close to those of the free ligand.
This a useful diagnostic criterion of chelation.
In [Au(dtc)(damp)]+ (dtc ) dmtc, detc), in which both ligands
are chelated, the signals for the two alkyl groups of the dtc
ligands are nonequivalent, reflecting the absence of a plane of
symmetry between them. [1H data, dtc ) detc: δ(CH2) 3.96,
3.89; δ(CH3) 1.54, 1.50 for the chloride in CDCl3; δ(CH2) 4.81,
4.77; δ(CH3) 1.29, 1.26 for the BPh4 salt in DMSO-d6.]
For the bis(dithiocarbamate) compound [Au(detc)2(damp)],
the chemical shifts of the methylene and methyl groups of the
damp ligand confirm that the amine group is not coordinated
in solution. The shifts are quite different from those of the
(28) Pesek, J. J.; Mason, W. R. Inorg. Chem. 1979, 18, 924.

1664 Inorganic Chemistry, Vol. 35, No. 6, 1996
Table 3. NMR Parameters for [HgCl(damp)] in CDCl3
Parish et al.
CH3
a
C1
C2
C3
C4
C5
C6
CH2
δ(13C)
149.2
144.0
(144.5)
51.6
127.5
(127.5)
...
128.8
(128.7)
31
129.1
(129.1)
168
137.3
(137.3)
141.5
64.7
(64.7)
102
44.6
(44.6)
10.6
J(199Hg)/Hz
2471
(212)
(178)
(101)
(11.0)
a Numbers in parentheses are from ref 26.
Table 4. Antibacterial MIC Values (µg/mL)
compound
S. aureus
E. faecalis
K. pneumoniae
E. coli
P. aeruginosa
[AuCl2(damp)]
[AuCl2(ppy)]
ciprofloxacin
1.0-2.5
1.0-2.5
<0.25
1.0-2.5
2.5-10
1.0-2.5
10-25
25-100
<0.25
10-25
25-100
<0.25
10-25
25-100
<0.25
Table 5. Antifungal MIC Values (µg mL-1
)
compound
C. albicans NCFC 3153
C. albicans
C. neoformans
A. niger
A. fumigatus
[AuCl2(damp)]
[AuCl2(ppy)]
amphotericin B
10-25
25-100
<0.25
25-100
25-100
<0.25
2.5-100
1.0-2.5
<0.25
10-25
25-100
1.25-1.0
2.5-10
25-100
0.25-1.0
Table 6. Comparative Cytotoxicity against CHO Cells
(% Survival)
group is not coordinated, and the C2 shift is lower even than
that for [Au(dtc)2(damp)] (Table 2); both observations show that
the damp ligand is monodentate. The signals for the coordinated
cyanide ligands are distinct, but have shifted toward each other
(129.7, 118.0 ppm). This is consistent with an increase in
shielding by the aryl group, which is now free to rotate out of
the coordination plane. These observations are consistent with
the presence of [Au(CN)3(damp)]-. However, no signal is seen
for a free CN- ion (variously reported at 164.1-167.4 in
D2O),25,29 and it is possible that rapid exchange is occurring
with one of the coordinated groups. Such an exchange would
be expected to result in a broadening of this peak, with
consequent reduction in intensity, as is seen in reactions of [Au-
(OAc)2(damp)] with aqueous acetate solutions;27 this would
explain the similarity in the intensities of the two cyanide signals
which would otherwise be expected to be in 2:1 ratio.
cytotoxicity for concn ca
compound
c ) 100
c ) 10
c ) 1.0
c ) 0.1
[AuCl2(damp)]
[PtCl2(NH3)2]
ciprofloxacin
amphotericin B
2.2
26.2
90.0
2.4
69.6
89.0
101.2
72.7
84.3
92.9
99.9
88.2
91.3
96.9
96.7
102.3
a c in µg mL-1
.
Pharmacological Tests. Antimicrobial Activity. The abil-
ity of the compounds to inhibit microbial growth was determined
by measuring their minimum inhibitory concentration values
(MIC). The MIC was taken as the range between the highest
concentration of compound which allowed growth and the
lowest which inhibited growth.31 The results (Tables 4 and 5,
for bacteria and fungi, respectively) show that both [AuCl2-
(damp)] and [AuCl2(ppy)] have moderate, broad-spectrum anti-
microbial activity; there is slightly greater specificity against
the Gram positive organisms S. aureus and E. faecalis. The
complex [AuCl2(damp)] was a little more potent than [AuCl2-
(ppy)], but neither was as active as the controls, ciprofloxacin
(a quinolone antibiotic) and amphotericin B (a polyene anti-
fungal drug).
Mammalian-Cell Toxicity. The cytotoxicity of [AuCl2-
(damp)] was assessed against CHO cells to obtain an indication
of its toxicity against a mammalian host32 (Table 6). Cisplatin,
ciprofloxacin and amphotericin B were included, as examples
of a cytotoxic metal complex, an antibacterial agent, and an
antifungal agent respectively. The complex [AuCl2(damp)] has
a similar cytotoxicity to cisplatin and is more cytotoxic than
either ciprofloxacin or amphotericin B, with cytotoxic concen-
trations similar to the MIC values.
When a large excess of cyanide ion is added to solutions of
[Au(CN)2(damp)] or [Au(OAc)2(damp)], very complex 13C
spectra are observed, suggesting extensive decomposition.
The 13C spectrum of the intermediate [HgCl(damp)] was also
re-examined (Table 3). It was possible to see couplings to 199Hg
1
(16.8%, I ) /2) which had not been reported previously:24 in
particular there is a large coupling to C1 (1J ) 2471 Hz), which
is within the range of similar reported 1J values.25,30 The very
small coupling for the signal at δ 128.8 indicates that this must
be C4 (4J ) 31 Hz); this changes the previous assignment,26
but the new assignment conforms to the order of coupling
constants seen for bis(arylmercury(II)) derivatives:25,30 1J . 3J
6
4
> J > J. 199Hg-coupling was also observed to the carbon
atoms and the protons of the methylene group and to the carbon
of the methyl groups. The first of these is similar to the value
of 3J observed for the methyl groups in bis(o-tolylmercury(II)).25
The chemical shifts of these groups are similar to those of
compounds in which the damp ligand is monodentate, [Au(dtc)2-
(damp)] and K[Au(CN)3(damp)], and to those of the free ligand.
For the mercury complex, the values are presumably time
averages for the chelated and nonchelated forms;26 this in turn
implies that the coupling to the CH2 group passes via the phenyl
ring.
Primary in Vitro Antitumor Screen. The ability of the
compounds to inhibit tumor-cell growth was assessed against a
human tumor cell-line panel. This showed that [AuCl2(damp)]
has a profile of toxicity similar to that of cisplatin, in that both
show greater toxicity (lower IC50 values) for the breast, bladder,
and ovarian tumor cell lines than for the other lines (Table 7).
Such differential cytotoxicity has been used as an indicator of
(29) (a) Affolter, S.; Pregosin, P. S. J. Organomet. Chem. 1990, 398, 197.
(b) Blixt, J.; Gyori, B.; Glaser, J. J. Am. Chem. Soc. 1989, 111, 7784.
(c) Hroto, M.; Koike, Y.; Tshizuka, H.; Yamasaki, A.; Fujiwara, S.
Chem. Lett. 1973, 853.
(30) Sebald, A.; Wrackmeyer, B. Spectrochim. Acta 1982, 38, 163. Mann,
B. E.; Taylor, B. F. 13C NMR Data of Organometallic Compounds;
Academic Press: London, 1981.
(31) Elsome, A. M.; Brumfitt, W.; Hamilton-Miller, J. M. T.; King, R. O.;
Fricker, S. P. Antimicrobial Activity and Potential Therapeutic Use
of New Gold Coordination Complexes. Abst. 31st Int. Conf. Antimi-
crobial Agents Chemother. 1991, A387.
(32) Fricker, S. P. Toxicol. in Vitro 1994, 8, 879.

Dichloro(2-((dimethylamino)methyl)phenyl)gold(III)
Inorganic Chemistry, Vol. 35, No. 6, 1996 1665
Table 7. IC50 Values (µg mL-1) for Various Tumor-Cell Lines
SW620
(colon)
SW1116
(colon)
SW403
(colon)
ZR-75-1
(breast)
HT29/219
(rectum)
HT1376
(bladder)
SK-OV-3
(ovary)
compound
[AuCl2(damp)]
[AuCl2(ppy)]
[PtCl2(NH3)2]
50
26
120
48
35
135
56
27
200
11
20
15
22
21
30
12
9
15
18
18
14
Table 8. Evaluation of [AuCl2(damp)] and Cisplatin in the ZR-75-1
Human Breast Tumor Xenograft Model
a
dose
time group
group
mg kg-1 days
size
RTVa ( SE T/Cb
[AuCl2(damp)]
25
7
14
21
28
7
14
21
28
7
14
21
28
7
14
21
28
6/6
6/6
4/6
2/6
6/6
6/6
6/6
6/6
6/6
190 ( 90
530 ( 220
560 ( 240
0.9
0.8
0.4
12.5
160 ( 50
300 ( 100
730 ( 280
1100 ( 370
230 ( 70
0.7
0.5
0.5
0.4
1.1
1.1
0.9
0.6
6.25
6/6
6/6
6/6
10/10
9/10
9/10 1370 ( 340
9/10 2930 ( 880
670 ( 180
1260 ( 310
1840 ( 490
220 ( 40
controls
630 ( 130
[PtCl2(NH3)2]
3
8
14
21
29
8
14
21
29
8
6/6
6/6
5/6
4/6
6/6
6/6
6/6
5/6
190 ( 50
360 ( 80
550 ( 180
870 ( 330
150 ( 40
240 ( 40
420 ( 80
650 ( 200
280 ( 60
490 ( 130
720 ( 180
0.7
0.7
0.8
0.7
0.5
0.5
0.6
0.6
b
1.5
controls
9/10
9/10
9/10
14
21
29
9/10 1180 ( 310
a RTV ) relative tumor volume ) 100(mean volume of tumors at
assessment time)/(mean volume of tumors at start of experiment). b T/C
) (RTV of treated group)/(RTV of control group).
antitumor activity,21 and the results obtained suggested that
[AuCl2(damp)] might have potential as an antitumor agent.
However, [AuCl2(ppy)] has similar toxicity toward all the cell
lines with the exception of HT1376 and was not investigated
further.
Xenograft Study. The in Vitro results showed that the ZR-
75-1 tumor cell line was the most sensitive to [AuCl2(damp)].
This compound was therefore evaluated against the solid ZR-
75-1 tumor growing as a xenograft in nude mice. The gold
compound was administered to tumor-bearing mice at its
previously determined maximum tolerated dose, 25 mg kg-1
and at two 2-fold dilutions. As a comparison, a similar
experiment was conducted with cisplatin. The tumor volumes
were measured at regular intervals and are expressed as a
percentage of the initial volume (RTV ) relative tumor volume)
in Table 8 and Figure 6.
At the two highest dose levels, the RTVs of the [AuCl2-
(damp)]-dosed animals were lower than those of the untreated
controls at all assessment times. However, there was evidence
of toxicity at the highest dose, with only two animals surviving
until day 28, both having lost weight over the time of the
experiment. Some adhesion of the abdominal organs was also
seen, which is probably a local effect associated with the
intraperitoneal route of administration. This was particularly
noticeable at the top dose. At a dose of 12.5 mg kg-1, all of
the animals survived until day 28. For this dose group the RTV
( SE ranges were outside those of the controls at 14, 21, and
Figure 6. Growth of the ZR-75-1 breast tumor xenograft under
treatment with (a) [AuCl2(damp)] and (b) cisplatin. RTV ) tumor
volume as a percentage of the initial volume.
28 days, and at 28 days the mean RTV was 40% of that of the
control group (T/C ) 0.4). However, at no single assessment
time did the difference between the RTVs of the dosed and
control groups reach the 5% level of significance. Thus,
although the results indicate some antitumor activity for [AuCl2-
(damp)], it must be considered as modest.
Cisplatin also exhibited only limited activity in this tumor
model. At both dose levels the RTVs of the treated groups
were below those of the controls at all times, and at 1.5 mg
kg-1 the RTV ( SE ranges were outside those of the controls.
However, at no time did the difference reach the 5% level of
significance.
Discussion
Following the marked success of cisplatin and its analogues
as antitumor agents, much effort has been expended in monitor-

1666 Inorganic Chemistry, Vol. 35, No. 6, 1996
Parish et al.
ing other metal complexes. We considered that gold(III)
complexes were worthy of attention, despite their possible
greater reactivity.
dtc ligands is bidentate, resulting in the displacement of the
damp amine group. In solution, the two dtc ligands become
equivalent by rapid mono- to bidentate switching, and there is
no evidence for recoordination of the amine group.
Inorganic chemotherapeutic agents often act Via substitution
reactions. Thus cisplatin is hydrolyzed in ViVo to cis-[Pt(H2O)2-
(NH3)2]2+ and the aqua ligands can then be substituted by
guanine or adenine bases of DNA.33 The results reported above
and earlier6h show that [AuCl2(damp)] has a rich and interesting
substitution chemistry. The chloride ligands are directly
replaced by softer ligands, such as the heavier halides, cyanide
ion, and N- and S-donor ligands. It is therefore not surprising
that this complex interacts strongly with biological systems.
Most of the substitution reactions are straightforward and
involve simple replacement of chloride by another donor atom.
In two cases, however, displacement of the amine donor group
of the damp ligand is observed. This occurs with the two softest
ligands used, cyanide ion and dithiocarbamate. For cyanide ion,
the dechelation of damp is observed only when more than 2
molar equiv is used. This indicates only that the affinity of
gold(III) is considerably greater for cyanide than for a simple
amine, and that the difference is sufficient to outweigh the
chelate effect.
The behavior with dithiocarbamates is the most interesting.
A single dithiocarbamate ligand is able to displace both
chlorides, and the di-chelated cation [Au(dtc)(damp)]+ can be
isolated with chloride as its counterion. Although chelation by
dithiocarbamates is not rare, it is noteworthy that it involves a
four-membered ring. The affinity of the gold(III) atom for the
soft sulfur donor is sufficient to encourage formation of this
tight ring. In [Au(dtc)2(damp)], the expected structure with two
monodentate dtc ligands is not observed. Instead, one of the
When evaluated against selected microorganisms, the gold
complexes demonstrated only modest activity compared to
control drugs. Also, [AuCl2(damp)] displayed only marginal
selectivity for the microorganisms over CHO cells. This
suggests that it is not suitable as a potential antimicrobial
chemotherapeutic agent. However, the broad-spectrum activity
is compatible with antiseptic or biocide use.
The results from the xenograft model suggest that [AuCl2-
(damp)] and cisplatin have comparable activity against the ZR-
75-1 tumor, although for both compounds the activity was not
as marked as the in Vitro results might have predicted. The
low aqueous solubility of [AuCl2(damp)] might well have
hindered its transport from the injection site to the tumor. This
would have adversely affected its activity and might also have
contributed to the abdominal adhesions observed.
It is also useful to consider the xenograft model. The ZR-
75-1 tumor might be inherently insensitive to compounds of
this type, bearing in mind the low activity found for cisplatin.
Therefore another xenograft model could be more appropriate
for preliminary testing.
Acknowledgment. Financial support from the SERC (B.P.H.,
J.P.W.) and ERASMUS (J.M.) is gratefully acknowledged.
Supporting Information Available: Listings of analytical data,
crystallographic data, atomic coordinates, B(eq) and U values, and
intramolecular distances and angles (34 pages). Ordering information
is given on any current masthead page.
(33) Lempers, E. L. M.; Reedijk, J. AdV. Inorg. Chem. 1991, 37, 175.
IC950343B