Gold(I) chloride adducts of 1,3-bis(di-2-pyridylphosphino)propane: Synthesis, structural studies and antitumour activity

Article abstract of DOI:10.1039/b705008a

The novel water soluble bidentate phosphine ligand 1,3-bis(di-2- pyridylphosphino)propane (d2pypp) has been synthesized by a convenient route involving treatment of 2-pyridyllithium with Cl₂P(CH ₂)₃PCl₂ and isolation in crystalline form as the hydrochloride salt. The synthesis of the precursor Cl₂P(CH ₂)₃PCl₂ has been optimized by the use of triphosgene as the chlorinating agent. The 2: 1 and 1: 2 AuCl: d2pypp adducts have been synthesized and characterized by NMR spectroscopy and single crystal X-ray studies, and shown to be of the form (AuCl)₂(-d2pypp-P, P′) and [Au(d2pypp-P,P′)₂]Cl(·3.75H₂O), respectively. The latter is more lipophilic than analogous 1: 2 adducts of gold(i) chloride with the diphosphine ligands 1,2-bis(di-n-pyridylphosphino) ethane (dnpype) for n = 2, 3 and 4, based on measurement of the n-octanol-water partition coefficient (log P = -0.46). A single crystal structure determination of the 1: 2 Au(i) complex of the 3-pyridyl ethane ligand shows it to be of the form [Au(d3pype-P,P′)₂]Cl·5H₂O. The in vitro cytotoxic activity of [Au(d2pypp)₂]Cl was assessed in human normal and cancer breast cells and selective toxicity to the cancer cells found. The significance of these results to the antitumour properties of chelated 1: 2 Au(i) diphosphine complexes is discussed. This journal is The Royal Society of Chemistry.

Full text of DOI:10.1039/b705008a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Gold(I) chloride adducts of 1,3-bis(di-2-pyridylphosphino)propane: synthesis,
structural studies and antitumour activity†‡§
a
a,b
a
a
Anthony S. Humphreys, Aleksandra Filipovska, Susan J. Berners-Price,* George A. Koutsantonis,
a
a
Brian W. Skelton and Allan H. White
Received 2nd April 2007, Accepted 22nd May 2007
First published as an Advance Article on the web 14th September 2007
DOI: 10.1039/b705008a
The novel water soluble bidentate phosphine ligand 1,3-bis(di-2-pyridylphosphino)propane (d2pypp)
has been synthesized by a convenient route involving treatment of 2-pyridyllithium with
Cl
2
P(CH
2
)
3
PCl
2
and isolation in crystalline form as the hydrochloride salt. The synthesis of the
PCl has been optimized by the use of triphosgene as the chlorinating agent. The
precursor Cl
2
P(CH
2
)
3
2
2
: 1 and 1 : 2 AuCl : d2pypp adducts have been synthesized and characterized by NMR spectroscopy
ꢀ
and single crystal X-ray studies, and shown to be of the form (AuCl)
2
(l-d2pypp-P, P ) and
O), respectively. The latter is more lipophilic than analogous 1 : 2 adducts
ꢀ
[
Au(d2pypp-P, P )
2
]Cl(·3.75H
2
of gold(I) chloride with the diphosphine ligands 1,2-bis(di-n-pyridylphosphino)ethane (dnpype) for n =
2
, 3 and 4, based on measurement of the n-octanol–water partition coefficient (log P = −0.46). A single
crystal structure determination of the 1 : 2 Au(I) complex of the 3-pyridyl ethane ligand shows it to be of
the form [Au(d3pype-P, P )
human normal and cancer breast cells and selective toxicity to the cancer cells found. The significance
of these results to the antitumour properties of chelated 1 : 2 Au(I) diphosphine complexes is discussed.
ꢀ
2
]Cl·5H
2
O. The in vitro cytotoxic activity of [Au(d2pypp) ]Cl was assessed in
2
Introduction
to selectively target tumour cells with lipophilic cations with
appropriate characteristics and two recent predictive models
have addressed key factors that determine selective accumulation
of delocalised lipophilic cations in tumour cell mitochondria,
considering lipophilic–hydrophilic balance as well as the acid–base
There is much recent evidence showing that mitochondria play
a critical role in the regulation of apoptosis (programmed cell
death), making them an attractive target for the design of new
anticancer drugs. A variety of different therapeutic strategies
targeting mitochondria for cancer therapy have been described
in recent literature (see e.g. ref. 2–6). One approach exploits
the long known phenomenon that carcinoma cells have an
elevated mitochondrial membrane potential compared to normal
human epithelial cells, leading to a higher level of uptake and
retention of lipophilic cations (e.g. Rhodamine-123), which are
concentrated within mitochondria in response to the mitochon-
drial membrane potential (negative inside). Several structurally
diverse, delocalised, lipophilic cations have demonstrated strong
activity in tumour models (reviewed in ref. 11). There is potential
1,2
12,13
dissociation constant and the electric charge.
Several different types of gold-based antitumour compounds
(
in both Au(I) and Au(III) oxidation states) have now been
shown to act by mechanisms involving mitochondrial cell death
14
15
pathways. Research stemming from the early 1980s points to
two distinct classes of Au(I) phosphine antitumour agents in
which the mode of action may depend on targeting mitochon-
7
8,9
16
drial function. These are neutral, linear, two-coordinate Au(I)
10
complexes, such as the tetraacetylthioglucose Au(I) phosphine
complex, ‘auranofin’, and bis-chelated 1 : 2 Au(I) diphosphine
+
complexes related to [Au(dppe)
2
] , which exhibit antitumour
17
activity against a range of tumour models in mice. Recent
evidence has shown that auranofin induces apoptosis by a
a
School of Biomedical, Biomolecular and Chemical Sciences, The University
of Western Australia, 35 Stirling Highway, Crawley, WA, 6009, Australia.
E-mail: sue.berners-price@uwa.edu.au; Fax: + 61 (0)8 6488 1005; Tel: +
mechanism involving selective inhibition of the selenoenzyme
18,19
mitochondrial thioredoxin reductase.
As bis-chelated Au(I)
6
1 (0)8 6488 3258
+
complexes, such as [Au(dppe) ] , do not undergo ligand exchange
2
b
Laboratory for Cancer Medicine, Western Australian Institute for Medical
Research and Centre for Medical Research, The University of Western
Australia, Perth, Western Australia, 6000, Australia
reactions as readily as two-coordinate linear Au(I) complexes,
a different mode of action is likely which may stem from the
lipophilic, cationic properties, as with other delocalised lipophilic
cations that accumulate in mitochondria. In previous work we
demonstrated that the selectivity for cancer cells over normal cells
can be ‘tuned’ by adjusting the hydrophilic/lipophilic balance in a
series of related Au(I) and Ag(I) complexes with bidentate pyridyl-
†
Based on the presentation given at Dalton Discussion No. 10, 3rd–5th
September 2007, University of Durham, Durham, UK.
CCDC reference numbers 640094–640096. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/b705008a
Electronic supplementary information (ESI) available: Table S1 Torsion
‡
§
angles in 2. Table S2 Torsion angles in 3. Fig. S1 Potentiometric titration of
an aqueous solution of the d2pypp hydrochloride salt. Fig. S2 Projection of
cation 1 of 2 down the bisector of the largest and smallest P–Au–P angles,
which lie trans to each other. Fig. S3 Unit cell contents of 2, projected
down a, showing the layering of the structure. Fig. S4 Unit cell contents
of 3, projected down a.
20–23
phosphine ligands.
For 1 : 2 adducts of Au(I) with 1,2-bis(di-n-
pyridylphosphino)ethane (dnpype) for n = 2, 3 and 4, the positions
of the nitrogen atoms in the pyridyl rings can finely modulate the
22
lipophilic–hydrophilic balance, with consequent differences in
This journal is © The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4943–4950 | 4943

View Article Online
lipophilicity influencing the cellular uptake, tumour selectivity and
23
host toxicity. In order to explore further the potential for selective
targeting of tumour cells by complexes with optimal lipophilicity,
a wider range of similar compounds are needed for evaluation.
This work involves investigations of Au(I) adducts of the
novel water soluble bidentate phosphine 1,3-bis(di-2-pyridyl-
phosphino)propane, d2pypp, the 3-carbon bridge analog of the
counterpart ligand, d2pype. To rationalize trends in activity, a
detailed knowledge of the structural and solution chemistry of the
complexes is required. In addition to single crystal X-ray studies on
the 2 : 1 and 1 : 2 adducts of gold(I) chloride with d2pypp (1 and
2
), we have also determined the crystal structure of the ethane-
bridged 3-pyridyl complex [Au(d3pype) ]Cl (3), for comparison
with the structures of the 1 : 2 adducts of gold(I) chloride with
2
22
dnpype (where n = 2 and 4) reported in the earlier study.
Results and discussion
Synthesis of compounds
Scheme 1
We investigated two synthetic routes to prepare the target bidentate
pyridyl phosphine ligand 1,3-bis(di-2-pyridylphosphino)propane
The 2 : 1 bridged di-gold(I) complex (AuCl)
2
(d2pypp) (1) and
(
d2pypp), based on the procedures described previously for
the 1 : 2 bis-chelated complex [Au(d2pypp)
by direct reaction of stoichiometric quantities of the d2pypp
ligand and the Au(I) salt, [NBu ][AuCl ]. The neutral complex 1
is soluble in chloroform and acetone, sparingly soluble in ethanol
2
]Cl (2) were prepared
preparation of the analogous ethane-bridged ligand 1,2-
24
bis(di-2-pyridylphosphino)ethane (d2pype). As the precursor
chlorophosphine Cl P(CH PCl is not commercially available,
)
3
2
4
2
2
2
initially we chose a synthetic route involving treatment of 1,3-
dibromopropane with two equivalents of lithium di-2-pyridyl-
phosphide, generated from tris(2-pyridyl)phosphine and lithium
1
31
and insoluble in methanol. The H and P NMR chemical shifts
of the complex are very similar to those of the analogous ethane
2
2
3
1
bridged dinuclear complex (AuCl) (d2pype) (Table 1). Complex
2
metal. While P NMR showed that the synthesis proceeded almost
quantitatively, the d2pypp ligand could only be obtained as a crude
oil and attempts to purify this by recrystallization, trituration
and chromatography were hampered by the extreme sensitivity
of the ligand to oxidation of the P atoms. This synthetic route is
limited also by the need for prior synthesis of the precursor tris(2-
pyridyl)phosphine. A more convenient method, which produces
d2pypp in high yield, is the reaction of 2-pyridyllithium (prepared
from 2-bromopyridine under standard butyllithium metal halogen
2
1
13
is soluble in methanol and dmso and the H and C NMR
shifts are similar to those of the mononuclear bis-chelated complex
+
[
Au(d2pype)
2
] (Table 1).
The complicated multiplicity of the signals observed for the
1
3
propyl linker in the C NMR spectrum is consistent with
3
1
additional coupling to P in a second d2pypp ligand in the bis-
3
1
+
chelated complex. The P NMR shifts of the two [Au(P-P)
2
]
complexes differ by 24 ppm and the relative shielding of the
3
1
+
P NMR shift for [Au(d2pypp)
2
]
is consistent with trends
exchange conditions) with Cl
2
P(CH
2
)
3
PCl (Scheme 1).
2
observed previously for coordination chemical shifts (ccs) of
An early report suggested that the 1,3-bis(dichloro-
phosphino)propane could be accessed via the reaction of 1,3-
+
four-coordinate [Au(P-P)
phosphine ligands.
2
]
complexes with chelated bidentate
32,33
2
5
For complexes with five-membered chelate
bis(diphenylphosphino)propane and PCl at high temperatures;
3
26
rings the ccs are larger than for four-coordinate Au(I) complexes
with monodentate phosphines, attributed to the large positive ring
contribution, DR, to the chemical shift. In contrast, a negative
ring contribution has been noted for Au(I) complexes with six-
we found this reaction problematic as have others. An alternative
route involved the reduction of the tetraethyl propylene diphos-
phonate (prepared from Michaelis–Arbuzov condensation of 1,3-
27
dibromopropane and triethylphosphite) with LiAlH
4
, followed
32,33
28
membered chelate rings.
A further notable difference between
by chlorination of the resulting 1,3-bis(phosphino)propane. This
29,30
the bis-chelated Au(I) complexes of d2pype and d2pypp is that the
NMR spectra showed no evidence of dimerisation in solution for
could be achieved using phosgene,
but the more efficacious
) has been
and in this case provides
PCl in good yield. The
and considerably less toxic triphosgene (OC(OCCl
)
3 2
3
1
1
29,31
[Au(d2pypp) ]Cl. Previous P and H NMR studies revealed that
used previously as an alternative
a clean option, generating Cl P(CH
2
the 1 : 2 Au : d2pype adduct exists as an equilibrium mixture of
2
2
)
3
2
monomeric and dimeric species in methanol and only the dimeric
d2pypp ligand can be isolated in solid crystalline form as a
hydrochloride salt, which is relatively stable to oxidation in
solution. Potentiometric titration of an aqueous solution of the
d2pypp hydrochloride salt demonstrated protonation of four of
the six possible N and P sites (see ESI, Fig. S1§). The free d2pypp
ligand is readily generated in situ as required, by basification of an
22
species [Au(d2pype)
2
]
2
2
Cl has been obtained in crystalline form.
Structures
The results of the ‘low’-temperature single crystal X-ray structure
determinations are consistent with the stoichiometries, connectiv-
ities and stereochemistries implied by the above formulations. In
aqueous solution of d2pypp·4HCl with NaHCO
3
and extraction
of the neutral ligand with dichloromethane.
4
944 | Dalton Trans., 2007, 4943–4950
This journal is © The Royal Society of Chemistry 2007

View Article Online
1
31
13
a
Table 1 H, P and C NMR data for d2pypp and 2 : 1 and 1 : 2 Au(I) adducts in comparison to analogous Au(I) adducts of d2pype (data from ref. 21)
1
d H
3
1
b
Compound
Solvent
CD OD
D O
2
CDCl
3
H
3
H
4
H
5
H
6
CH
2
(a )
CH
2
(b )
d P
d2pypp·4HCl
3
8.17
8.09
7.77
7.55
7.43
7.78
7.51
8.44
8.44
7.90
7.59
7.69
7.94
7.69
7.97
7.99
7.37
7.25
7.29
7.38
7.43
8.91
8.82
8.75
8.47
8.56
8.74
8.60
2.81
2.74
3.03
2.85
2.44
3.15
3.08
1.68
1.64
2.04
2.00
−15.9
−16.6
28.8
[
[
(AuCl)
2
(d2pypp)] (1)
Au(d2pypp)
2
]Cl (2)
CD
CD
3
OD
OD
3.4d
c
d2pype
3
−6.0
c
e
[
[
(AuCl)
Au(d2pype)
2
(d2pype)]
CDCl
3
31.0
cf
2
]Cl
CD
3
OD
26.8
13
g
d
C (J(C, P)/Hz)
C
2
C
3
C
4
C
5
C
6
CH
2
(a)
2
CH (b)
d2pypp·4HCl
Au(d2pypp)
CD
CD
3
OD
OD
156.6 d (26)
158.5 m
162.3
133.2 d (17)
129.6 m
129.1 vt (11)
130.2
145.2
138.5
136.3
138.9
128.2
125.9
123.0
126.2
147.1
151.1
149.8
151.3
27.9 vt (13)
27.6 m
22.1
22.9 vt (18)
20.0
[
2
]Cl (2)
3
c
d2pype
Au(d2pype)
CDCl3
CD OD
c
[
2
]Cl
3
157.8
27.5
a
2
Spectra recorded at ambient temperature and referenced to either TMS or TSP (for solutions in D O); assignments made with the aid of correlations in
d e f
1
1
1
13
b
c
[
H, H] COSY and [ H, C] HSQC spectra. Referenced to external 85% H
3
PO
4
(d = 0). Data from ref. 21. In CDCl
3
. In dmf. Monomeric complex,
1
g
signals for dimeric complex are also observed in the H NMR spectrum. Singlet unless otherwise stated.
1, one formula unit, unsolvated and devoid of crystallographic
symmetry, comprises the asymmetric unit of the structure.
Torsion angles in the C–C bonds of the ligand string are
◦
1
70.1(5), 57.6(8) (Fig. 1(a)). The gold environments are linear,
with bonding parameters in keeping with those of a number of
other (bisphosphine)(AuCl) arrays; the most pertinent example
is the dppp derivative, (AuCl) (dppp), for which two modifica-
tions have been structurally characterized, one in space group
2
2
34
35
Pbcn and the other in P2
1
.
The intramolecular geometries
of these dppp structures are not significantly different (Au–P
2
2
.231(6)–2.244(3) (< > 2.237(5)); Au–Cl 2.296(7)–2.310(3) (< >
˚
.305(6) A), but they differ in their supramolecular structure.
The orthorhombic form exhibits aurophilic interactions which
catenate molecules related by a b-glide, Au(1) · · · Au(2) (1/2-
˚
x, 1/2+y, z) 3.316 A whereas in the monoclinic example the
˚
aurophilic interaction (3.2368(9) A) causes cyclisation of the
structure giving essentially a seven-membered ring. In the case
of complex 1, the intramolecular distances are similar to those
in the previous dppp analogues (see above), Au–P, Cl being
˚
2
.216(2), 2.285(2) (Au(1)); 2.205(2), 2.260(2) A (Au(2)) with P–
◦
Au–Cl 179.38(8), 175.77(7) respectively. The molecules pack
in the crystal lattice with the P–Au–Cl components obligate
parallel to their inversion images, giving aurophilic interactions
(
1
Au(1) · · · Au(1)(2−x, 1−y, 1−z) 4.109(3); Au(2) · · · Au(2) (1−x,
˚
−y, z¯ ) 3.622(2) A) (Fig. 1(b)), with the closest interactions to the
gold atoms being intramolecular hydrogen contacts.
In 2, two formula units, devoid of crystallographic symmetry,
comprise the asymmetric unit of the structure, the compound
being ionic, i.e. of the form [Au(d2pypp)
2
]Cl·3H
2
O. The chelate
rings are of the ‘chair’ form, with very approximate m symmetry,
m passing through the gold atom in each case; in combination,
they render the cation chiral, but the two components of
Fig. 1 (a) Projection of a single molecule of (AuCl) (d2pypp) (1).
2
(b) Unit cell contents of 1, projected down a, showing the inversion related
P–Au–Cl pairings.
This journal is © The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4943–4950 | 4945

View Article Online
the asymmetric unit are of opposed chirality (Fig. 2) so
that, despite the chiral space group, each individual crystal is
racemic. Au–P distances (Table 2) are consistent with many
ꢀ
established for P, P -chelate systems.
Fig. 3 Projection of the cation of [Au(d3pype)
-axis).
2
]Cl·5H
2
O (3) (down its
Fig. 2 The pseudo-symmetrically related pair of cations of the asymmet-
ric unit of [Au(d2pypp) ]Cl.3H O (2).
2 2
2
essentially equivalent (Table 3) and similar to those reported for
Table 2 Selected geometries, 2
+
22
˚
[
Au(d4pype)
2
]
(2.36(1)–2.417(9), < > 2.383(15) A) and related
+
Atoms
Parameter
Atoms
Parameter
Cation 2
[Au(dppe)
2
extant for the latter);
2
] (‘dppe’ = Ph P(CH PPh ) species (e.g. 2.3908(6)–
2
2
)
2
2
˚
.412(2), < > 2.404(12) A in the two most precise determinations
Distances/A˚
Cation 1
38,39
there are similar interligand angular
discrepancies (two the same, by symmetry this time, (120.59(2) )
one much larger (129.45(2) ) and one rather smaller (117.37(2) );
the ‘bite’ angle of the ligand here (86.26(2) ) is much less than
◦
Au(1)–P(1)
Au(1)–P(2)
Au(1)–P(3)
Au(1)–P(4)
2.402(2)
2.355(2)
2.399(2)
2.373(2)
2.38(2)
Au(2)–P(6)
Au(2)–P(5)
Au(2)–P(8)
Au(2)–P(7)
< >
2.407(2)
2.380(2)
2.398(2)
2.400(2)
2.399(10)
◦
◦
◦
◦
those in 2 (one ca. 96, one ca. 93 ) where the chelate ring is
<
>
+
six- rather than five-membered. In the [Au(d4pype)
2
]
species,
although the bite angles are similar to the present (86.0(3),
◦
Angles/
◦
8
6.5(3) ), the interligand angles are very different: two small (and
◦
P(1)–Au(1)–P(2)
P(1)–Au(1)–P(3)
P(1)–Au(1)–P(4)
P(2)–Au(1)–P(3)
P(2)–Au(1)–P(4)
P(3)–Au(1)–P(4)
96.41(6)
106.03(6)
115.78(6)
116.42(6)
128.19(6)
93.39(6)
P(5)–Au(2)–P(6)
P(6)–Au(2)–P(8)
P(6)–Au(2)–P(7)
P(5)–Au(2)–P(8)
P(5)–Au(2)–P(7)
P(7)–Au(2)–P(8)
95.66(7)
106.39(6)
117.02(6)
117.49(7)
126.83(6)
93.29(6)
equivalent) (111.3(3), 112.8(3) ) and two much larger (129.7(3),
137.5(3) ). The ring conformation is essentially 2 here (through
◦
Au and the mid-point of the C–C bond), distorted somewhat,
presumably by the lack of conformity of the aromatic pendents, in
turn consequent on lattice forces (Table 3). Chloride counterions
are distributed over two sites, one fully occupied (Cl(1)) on a site
¯
of 4 symmetry, with four associated fully occupied O(03) water
Although there are no structurally defined examples as yet
˚
molecules (Cl(1) · · · O(03) 3.13(1) A). Further chloride contacts to
ꢀ
+
ꢀ
+
for [Au(dppp-P,P )
2
] , the [Au(Ph
PPh
chairs as in the present structure with Au–P 2.392(11)–2.426(7)
2
PCH
2
3
CH(R)CH
2
P )
2
] core has
four aromatic hydrogen atoms are found for Cl(1) (Cl(1) · · · H(126)
6
been defined for R = CH
2
2
(AuCl), the chelate rings being
˚
2.7A) with Cl(2) distributed over four symmetry related sites,
occupancies 0.25, with contacts to O(01), O(02) (x−1, y, z)
˚
(
1
< > 2.41(2) A), P–Au–P (chelate) 94.3(4), 94.4(3), interligand
07.6(3), 116.9(6), 117.0(4), 125.8(5) . Common to many of these
systems and their silver [Ag(dppp) ] counterparts is an asymme-
try in the interligand P–M–P angles, here, two being closely ranged
˚
◦
(2.945(6), 2.872(6) A) (ordered water molecules). The only N · · · O
˚
˚
contact < 3.7 A is O(01) · · · N(213) ( x¯ , 1/2−y, z) at 3.107(4) A.
+
37
2
A similar water/chloride/pyridyl hydrogen-bonding network is
postulated in the related d4pype complex.
22
◦
about their mean (116.6(8) , two cations), the other two divergent
◦
by (ca.) ± 10 in each cation (cf. above), suggesting the overall
cation conformation to be consequent on intracation constraints
and interactions, rather than interspecies effects, and independent
of substitution of pyridyl for phenyl substituents; in the present
situation, the two independent cations are pseudosymmetrically
related (the coordinates of Au(1) are (0.35, 0.30, 0.39) and of
Au(2) (0.36, 0.69, 0.61)) forming embracing pairs, disposed in
layers about (0,0,0), (0,0,1/2) (Fig. 2 and Fig. S3§).
Table 3 Selected geometries, 3
Atoms
Parameter
Atoms
Parameter
2.4050(6)
Distances/A˚
Au–P(1)
2.4009(6)
Au–P(2)
The structure of the 3-pyridyl complex, 3, exhibits similar
behaviour in its lattice arrangements, a projection down a being
shown in Fig. S4§ and a cation in Fig. 3.
◦
Angles/
i
P(1)–Au–P(2)
P(1)–Au–P(2 )
86.30(2)
120.57(2)
P(1)–Au–P(1 )
129.44(2)
117.35(2)
i
i
P(2)–Au–P(2 )
Here again, within the cation, which lies with the metal
atom disposed on a crystallographic 2-axis, Au–P distances are
Transformation of the asymmetric unit:
i 1−x, 1/2−y, z
4
946 | Dalton Trans., 2007, 4943–4950
This journal is © The Royal Society of Chemistry 2007

View Article Online
Biological studies
Early structure–activity relationships for complexes of the type
ꢀ
[
Au(R
2
P(CH
2
)
n
PR
2
2
) ]X showed that antitumour activity is re-
tained where bidentate phosphines have aromatic substituents and
1
5,40
two or three carbon atoms in the bridge.
Preclinical studies on
+
the parent [Au(dppe)
2
] were abandoned when severe hepatoxicity
was observed in dogs. A previous study demonstrated the
relationship between lipophilicity and antitumour activity for a
23
series of analogs of [Au(dppe)
pyridyl phosphine ligands (R
2
]Cl with bidentate ethane-bridged
ꢀ
ꢀ
2
ꢀ
P(CH
2
)
2
PR
2
, where R = R = 2-,
3
- or 4-pyridyl and R = Ph, R = 2-pyridyl). These complexes are
structurally similar to the parent compound, but the position of
the nitrogen atom on the pyridyl ring finely tunes interaction with
the solvent so that the hydrophilic character covers a very large
range (log P values between 1.14 and −1.77). Within this series
the d2pype complex (with intermediate lipophilicity) exhibits the
highest in vivo antitumour activity against the colon 38 tumour
in mice and this activity was shown to correlate with the highest
drug concentrations in plasma and tumour tissue. For the highly
+
lipophilic parent complex [Au(dppe)
2
] , non-specific binding to
proteins and other macromolecules leads to high host toxicity,
whereas that of the more hydrophilic compounds may be limited by
23
high rates of excretion as a consequence of low protein binding.
Fig. 4 Effect of [Au(d2pypp)
2
2
]Cl (2) (a) and [Au(dppe) ]Cl (b) on
An objective of the current study was the targeted synthesis of
another bis-chelated Au(I) diphosphine complex with lipophilicity
in the optimal range close to that of the 1 : 2 Au(I) : d2pype adduct.
growth of normal cells (white bars) and breast cancer cells (black bars).
Increasing concentrations of the Au(I) complexes were incubated with
either MDA-MB-468 cells or HMEC cells for 24 h, and cell growth was
measured using a CellTiter 96ꢀR cell proliferation assay. Data are expressed
as a percentage of cells grown in the absence of the Au(I) compounds and
the data are means ± S.D. of five independent experiments.
The octanol–water partition coefficient for [Au(d2pypp) ]Cl (2)
2
(
log P −0.46 ± 0.04) shows that the complex is more lipophilic than
23
the d2pype analog (log P = −0.92), consistent with the increased
length of the alkane bridge in the bidentate phosphine. Both
complexes are significantly more lipophilic than the water-soluble
Au(I) : d2pypp complex (1) exhibits aurophilic interactions in
the solid state. For the bis-chelated 1 : 2 Au(I) : d2pypp adduct
we found no evidence of dimerisation in solution or the solid
state, whereas the analogous adduct of the bidentate two-carbon
bridged 2-pyridyl phosphine (d2pype) crystallises as a dimer and
exists in solution as a solvent dependent equilibrium mixture of
complexes [Au(dnpype)
2
]Cl where n = 3 (3) and 4 (log P = −1.46
23
and −1.77, respectively), but significantly more hydrophilic than
+
the highly toxic parent complex [Au(dppe)
2
] (log P = 1.41).
The cytotoxicity of 2 was investigated in normal human
mammary epithelial cells (HMEC) and cancer breast cells (MDA-
MB-468) by measuring their cell growth over 24 h (Fig. 4a).
In the MDA-MB-468 breast cancer cells 2, concentrations
up to 0.2 lM, did not affect cell viability, while concentrations
above 0.8 lM caused substantial cell death (Fig. 4a). The same
concentrations did not affect cell viability of the HMEC normal
breast cells and 2 was not toxic even at 100 lM (Fig. 4a).
22
monomeric, dimeric and possibly tetrameric forms. Hydrogen-
bonding networks observed in the structure of the 3-pyridyl
complex [Au(d3pype) ]Cl (3) resemble those observed previously
2
for the 4-pyridyl analogue and are consistent with the high water
solubility of these complexes.
Previous studies on lipophilic cationic [Au(P-P) ] complexes
+
2
In contrast, the parental compound [Au(dppe)
2
]Cl was non-
have demonstrated that optimizing the lipophilicity is a critical
factor in achieving high antitumour activity by fine-tuning the
balance between cellular uptake and binding to plasma proteins.
By increasing the length of the alkane bridge in the bidentate
phosphine we have achieved the objective of synthesis of another
complex of this type with an octanol–water partition coefficient in
selectively toxic to both normal and cancer breast cells at
concentrations as low as 0.4 lM (Fig. 4b). These results are
consistent with a mechanism that involves selective accumulation
of lipophilic cationic complexes in tumour cell mitochondria
and studies probing the mechanism for the selective toxicity of
[
Au(d2pypp)
2
]Cl are in progress.
the optimal range. This complex [Au(d2pypp) ]Cl (2) is selectively
2
toxic to cancer breast cells but not normal breast cells. Further
studies are in progress to investigate the antitumour mechanism
and to confirm whether mitochondria are indeed the critical target
within cells.
Conclusion
We have optimized conditions for the synthesis of a novel water sol-
uble bidentate phosphine, 1,3-bis(di-2-pyridylphosphino)propane
(
d2pypp), used here in the synthesis of an antitumour Au(I)
Experimental
complex, but also with potential application for investigations of
water-soluble organometallic reagents for use in biphasic catalysis.
Single crystal structural studies show that the neutral 2 : 1
Nuclear magnetic resonance spectra were recorded using Bruker
ARX-300 (300.1 MHz for H, 75.5 MHz for C, 121.5 for P) or
1
13
31
This journal is © The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4943–4950 | 4947

View Article Online
1
13
Bruker AV-500 (500.1 MHz for H, 125.8 MHz for C, 202.5 MHz
Found: C, 45.9; H, 5.4; N, 9.1; Cl, 23.0. C23
requires C, 45.5; H, 5.1; N, 9.2; Cl, 23.3. ES MS (MeOH : H
H
22
N
4
P
2
·4HCl·2.5H
O :
)] , 51%; 479.1, [Unassigned] , 71%;
2
O
3
1
for P) spectrometers at ambient temperature unless otherwise
specified. H and C chemical shifts were referenced to solvent
resonances and P chemical shifts to external 85% H
Microanalyses were performed at the Microanalytical Laboratory,
ANU Research School of Chemistry, Canberra. Mass spectra were
acquired on a VG Autospec spectrometer employing the Fast-
Atom-Bombardment (FAB) or Electro-Spray (ESI) techniques.
2
1
13
+
+
AcOH) m/z: 904.2, [(HL·Cl)
2
41
31
+
+
3
PO
4
.
417.2, [LH] , 100%; 338.1, [L-(Py-(2))] , 29%. The presence of
4 mol equiv of chloride per mol of d2pypp ligand was confirmed by
potentiometric titration of a solution of the d2pypp hydrochloride
salt (67.7 mg in 25 mL 0.5 M HNO
3
) with a standard solution of
(see Fig. S1‡). H-NMR (CD OD): d 8.91 (d,
H6–5 5.5 Hz, H6), 8.44 (dd, JH4–5 7.9 Hz, JH3–4 7.9 Hz H4), 8.17
1
0.016 M AgNO
3
3
3
3
3
J
(
d, H3), 7.97 (dd, H5), 2.81 (vt, J 7.9 Hz, Ha), 1.68 (m, Hb).
Preparation of compounds
3
1
1
13
1
P{ H} NMR (CD
3
OD) d −15.87(s). C{ H}-NMR (CD
3
OD) d
1
Standard Schlenk techniques using flame-dried glassware un-
der an argon atmosphere were used in all procedures except
where indicated. Tetrahydrofuran was freshly distilled under
an argon atmosphere from potassium-benzophenone ketyl, di-
ethyl ether from sodium-potassium-benzophenone ketyl and
dichloromethane form calcium hydride. All other solvents were
deoxygenated prior to use by the freeze–pump method. 1,3-
156.60 (d, JP-C2 25.8 Hz, C2), 147.09 (s, C6), 145.23 (s, C4), 133.24
(d, JP-C3 17.0 Hz, C3), 128.23 (s, C5), 27.85 (t, JP-Ca 13.5 Hz, Ca),
22.94 (t, JP–Cb 17.6 Hz, Cb).
2
1
2
(AuCl) (d2pypp) (1)
2
To a solution of crude 1,3-bis(di-2-pyridylphosphino)propane (ca.
0.86 mmol) in acetone (50 mL) was added dropwise a solution
of [NBu
42
24
Bis(phosphino)propane, tris(2-pyridyl)phosphine, 1,2-bis(di-3-
24
33
4
][AuCl ] (871.9 mg, 1.71 mmol) in acetone (25 mL).
2
pyridylphosphino)ethane (d3pype) and [Au(dppe)
synthesised according to published procedures. [NBu
was prepared by the reduction of [NBu ][AuCl ] with sodium
2
]Cl were
After stirring for 2 h, the solution was concentrated in vacuo
4
][AuCl ]
2
◦
to half its volume and stored under argon at −10 C for ca. 2
4
4
43
months. An insoluble brown solid (35 mg) was filtered off and
the solvent was removed by rotary evaporation to give an oil. The
compound precipitated on addition of ethanol and was recrystal-
acetylacetonate, as described elsewhere.
1
,3-Bis(di-2-pyridylphosphino)propane (d2pypp)
lized from hot ethanol and acetone to give a white crystalline solid
(
◦
162.6 mg, 21.7%), mp 232 C. Found: C, 31.8; H, 2.6; N, 6.3.
(
a) The ligand d2pypp was synthesised as a crude oil using tris(2-
C
8
23
H
22
N
4
P
2
Au
2
+
Cl
2
requires C, 31.4; H, 2.5; N, 6.4. FABS MS m/z:
45, [M − Cl] , 100%. The small crystals obtained were suitable
for X-ray structure determination using a synchrotron facility.
pyridylphosphine) and 1,3-dibromopropane by a modification
of the published procedure for the analogous 2-carbon bridged
24
ligand d2pype. The crude oil was dissolved in acetone and the
3
1
concentration of the stock solution estimated by P NMR by
the addition of a known concentration of triethyl phosphate and
comparison of integrals.
[
Au(d2pypp)
2
]Cl·3H
2
O (2)
To a solution of d2pypp·4HCl (1.0144 g, 1.804 mmol) in water
(
b) Alternatively, d2pypp was synthesised by a more con-
venient procedure involving treatment of 2-pyridyllithium with
Cl P(CH PCl (prepared by a modification of the published
procedures
(
100 mL) was added a saturated solution of sodium hydrogen
carbonate (40 mL). After stirring for 1 h, dichloromethane
100 mL) was added and the yellow organic layer was separated. To
this was added a solution of [NBu ][AuCl ] (459.6 mg, 0.901 mmol)
2
2
)
3
2
2
(
9, 31
) and isolation in crystalline form as the hydrochlo-
4
2
◦
ride salt. To a cooled solution (−10 C) of H
2
P(CH
2
)
3
PH
2
in 2 : 1 acetone : methanol (50 mL), dropwise over 5 min whilst
stirring. The resultant solution was stirred for 1 h and the solvent
removed in vacuo. The crude product was dissolved in hot ethanol
(
4.05 g, 37.5 mmol) in dichloromethane (150 mL), a solu-
tion of bis(trichloromethyl)carbonate (14.90 g, 50.2 mmol) in
dichloromethane (90 mL) was added dropwise. The mixture was
(
40 mL) and diethyl ether (40 mL) was added. The complex was
◦
warmed to 25 C and stirred for 4 h. The solvent was removed
◦
afforded as a white solid after cooling to −10 C for several weeks
in vacuo and the pale yellow crude product distilled under reduced
and was isolated by filtration (yield 0.245 g, 25%). Found: C, 49.4;
◦
pressure (102 C, 1.5 mmHg) to give Cl
2
P(CH
2
)
3
PCl
2
(7.13 g, 77%).
H, 4.4; N, 10.1; P, 11.1. C46
H
44
N
8
P
4
AuCl·3H
2
O requires C, 49.4;
1
31
13
29
H, P and C NMR were identical with literature values.
+
H, 4.5; N, 10.0; P, 11.1. FABS MS m/z: 1028, [M − H] , 100%;
2
-Bromopyridine (3.4 mL, 5.5 g, 35 mmol) in diethyl ether
+
6
13 [M − L] 75%.
(
50 mL) was added dropwise over 30 min to a solution of n-
Small crystals of [Au(d2pypp)
2
]Cl·3.75H
2
O suitable for X-ray
butyllithium (1.6 M solution in hexanes, 21.8 mL, 34.9 mmol) in
structure determination were obtained by recrystallization of a
sample of [Au(d2pypp) ]Cl from warm ethanol.
◦
diethyl ether (200 mL) at −40 C. The mixture was stirred for
2
2
h then Cl
2
P(CH
2
)
3
PCl (1.0 mL, 1.63 g, 6.6 mmol) in diethyl
2
ether (25 mL) was added dropwise. After stirring for a further
[
Au(d3pype)
2
]Cl·5H
2
O (3)
◦
2
h, the mixture was allowed to warm to −20 C and was
quenched with 2M H
2
SO
4
(50 mL). The aqueous fraction was
The complex [Au(d3pype) ]Cl was prepared according to the liter-
ature procedure by the reaction of 1.0 mol equiv of [NBu ][AuCl ]
with 2.05 mol equiv of the ligand d3pype in dichloromethane. The
solid product was dissolved in warm 1 : 1 methanol : ethanol and
insoluble material filtered off. Small colourless crystals suitable
for X-ray structure determination were formed after reducing the
2
22
collected and basified with 2M NaOH (110 mL), then extracted
with dichloromethane (2 × 100 mL). The solvent was removed
in vacuo and the crude yellow oil dissolved in concentrated HCl
4
2
(
40 mL). The solvent was removed and the crude yellow solid
recrystallised from hot methanol and acetone to give a pale yellow
◦
31
hygroscopic powder, d2pypp·4HCl (2.1 g, 57%), mp 206 C (dec.).
volume to ca. 1 mL and were washed with cold ethanol. P and
4
948 | Dalton Trans., 2007, 4943–4950
This journal is © The Royal Society of Chemistry 2007

View Article Online
1
H NMR chemical shifts (in CD
3
OD) were identical to published
Measurement of lipophilicity (P)
22
values.
The octanol–water partition coefficients (P) of [Au(d2pypp)
2
]Cl·
O (2) were determined using a shake-flask method. Water
50 mL, distilled after milli-Q purification) and n-octanol (50 mL,
45
3
(
H
2
Structure determinations
For 2, 3 full spheres of CCD area-detector diffractometer data
were measured (Bruker AXS instrument, x-scans; monochromatic
vacuum distilled) were shaken together using a laboratory shaker
(Mistral Multi-Mixer), for 72 h to allow saturation of both phases.
A stock solution of 2 (100 lM) was prepared in the aqueous phase
and a 1 mL aliquot was then added to an equal volume of the
n-octanol phase. The resultant biphasic solution was mixed for 1 h
and then centrifuged (3000 × g, 5 min) (Beckman Coulter Allegra
X-22R centrifuge) to separate the phases. The concentrations
of the complex in the organic and aqueous phases were then
determined using UV absorbance spectroscopy (260 nm). Log
P was defined as the logarithm of the ratio of the concentrations
of the complex in the organic and aqueous phases (Log P = Log
{[Au(org)]/[Au(aq)]}; the value reported is the mean of three separate
determinations.
˚
molybdenum Ka radiation, k = 0.71073 A; T ca. 153 K); for
˚
1
1
, synchrotron radiation (k = 0.5594 A) was employed (T ca.
10 K). Nt(otal) reflections were obtained, merging to N unique
(
R
int cited) after ‘empirical’/multiscan absorption correction (pro-
prietary software), N with I > 2r(I) considered ‘observed’.
Full matrix least squares refinement on F refined anisotropic
displacement parameters for the non-hydrogen atoms, (x,y,z,Uiso
o
2
)
H
being included following a riding model. Reflection weights were
2
2
2
−1
2
2
(
r (F
o
) + (aP) + bP) (P = (F
o
+ 2F )/3). Neutral atom
c
complex scattering factors were employed within the SHELXL
44
9
7 program. Results are presented below and in the tables
and figures, the latter showing crystallographic numbering with
0% probability amplitude displacement ellipsoids for the non-
5
˚
Cell culture and toxicity assay
hydrogen atoms, hydrogen atoms having arbitrary radii of 0.1 A.
Individual variations in procedure are cited as ‘variata’.
A human breast adenocarcinoma cell line MDA-MB-468 (from
◦
ATCC) was cultured at 37 C under humidified 95% air/5% CO
2
Crystal/refinement data
in Dulbecco’s modified Eagle’s medium (DMEM) without phenol
red, containing Earle’s balanced salt solution and supplemented
¯
1
.
C
23
H
22Au
2
Cl
2
N
4
P
2
, M = 881.2. Triclinic, space group P1
−
1
1
with 2 mM Glutamax, penicillin (100 U mL ), streptomycin
˚
(
9
2
‘
0
2
C
i
, No. 2), a = 9.189(6), b = 11.499(7), c = 12,259(8) A, a =
−
1
◦
3
(100 lg mL ) and 10% heat inactivated fetal calf serum (FCS). A
˚
5.62(4), b = 103.85(3), c = 92.06(3) , V = 1249 A . D (Z = 2) =
c
−
3
−1
human mammary epithelial cell line (HMEC from Cambrex) was
.342 g cm . l = 6.6 mm ; specimen: 0.07 × 0.05 × 0.04 mm;
◦
◦
cultured at 37 C under humidified 95% air/5% CO
2
in MCDB
T’ min/max = 0.85. 2hmax = 55 ; N
t
= 38 163, N = 9718 (Rint =
1
70 medium, supplemented with 5 mL MCDB 100X supplement,
.033), N
o
= 9123; R1 = 0.053; wR2 = 0.135 (a = 0.0336, b =
−
1
−1
penicillin (100 U mL ) and streptomycin (100 lg mL ). For
toxicity studies, cells were grown in 96-well tissue culture dishes to
90% confluence and incubated for 24 h with 100 lL of their growth
2
9.46); S = 1.20.
2
.
C
46
H
44AuClN
8
P
(D
4
·3.75H
2
O, M = 1132.8. Orthorhombic,
4
space group P2
1
2
1
2
1
2
, No. 19), a = 15.276(2), b = 18.394(1),
serum containing increasing concentrations of 2 or [Au(dppe) ]Cl.
3
−3
˚
˚
c = 34.794(3) A, V = 9777 A . D
c
(Z = 8) = 1.539 g cm . lMo
=
Stock solutions of 2 and [Au(dppe) ]Cl were prepared in water
2
−
1
3
.2 mm ; specimen: 0.14 × 0.14 × 0.08 mm; ‘T’min/max = 0.72.
and 1% DMSO in water, respectively. Cell death was determined
◦
2h
max = 65 ; N
t
= 155 499, N = 35 424 (Rint = 0.081), N
o
= 23 553;
using the cell titre assay following the manufacturer’s instructions
(Promega, G3580). The assay was performed by adding a 20 lL
of the CellTiter 96ꢀR AQueous One Solution Reagent directly to
R1 = 0.057; wR2 = 0.148 (a = 0.0653, b = 17.59); S = 1.05.
Variata. In all determinations, nitrogen atom assignment
was based on refinement/difference map behaviour. Difference
map residues were assigned as water molecule oxygen atoms,
associated hydrogens atoms not being located. Initially xabs was
not significantly different from 0.5; refinement in terms of a twin
model reduced it to 0.02(13).
◦
culture wells, incubating for 1 h at 37 C under humidified 95%
air/5% CO and then recording the absorbance at 490 nm with a
2
BMG Labtech Fluostar Optima spectrophotometric plate reader.
The quantity of formazan product as measured by the amount
of 490 nm absorbance is directly proportional to the number of
living cells in culture. The data is expressed as a percentage of cells
grown in the absence of gold compounds. Data are means ± S.D.
of five independent experiments.
3
.
C
44
H
40AuClN
8
P
4
·5H
2
O, M = 1127.3. Tetragonal, space
6
˚
group I4
647 A . D
1
/a (C4h , No. 88), a = 13.4748(9), c = 53.133(3) A, V =
3
−3
−1
˚
9
0
c
(Z = 8) = 1.552 g cm . lMo = 3.3 mm ; specimen:
◦
.18 × 0.13 × 0.11 mm; ‘T’min/max = 0.66. 2hmax = 75 ; N
t
= 98670,
N = 12 696 (Rint = 0.033), N
o
= 9829; R1 = 0.029; wR2 = 0.092
Acknowledgements
(
a = 0.0470, b = 15.43); S = 1.12.
This research was supported by the Australian Research Council
(Discovery grants for S. B. P. and A. H. W.), the Australian
National Health & Medical Research Council (Howard Florey
Fellowship for A. F.) and an Australian Postgraduate Research
Award (A. S. H.). We thank Dr Peter Barnard for helpful discus-
sions, Dr Lindsay Byrne and Dr Donald Thomas for assistance
with NMR experiments and Dr Tony Reeder for assistance with
mass spectrometry.
Variata. One of the anion components was modelled as 0.25
site occupancy (at (−0.15, 0.28, −0.03)), assignment assisted
by consideration of contact distances to other difference map
residues, the latter assigned as water molecule oxygen atoms (no
associated hydrogen atoms located) on that basis and that of
refinement behaviour, ‘O(3)’ being assigned a site occupancy of
0.5.‡
This journal is © The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4943–4950 | 4949

View Article Online
Use of the ChemMatCARS Sector 15 at the Advanced
Photon Source, was supported by the Australian Synchrotron
Research Program, which is funded by the Commonwealth of
Australia under the Major National Research Facilities Program.
ChemMatCARS Sector 15 is also supported by the National
Science Foundation/Department of Energy under grant numbers
CHE9522232 and CHE0087817 and by the Illinois Board of
Higher Education. The Advanced Photon Source is supported
by the U.S. Department of Energy, Basic Energy Sciences, Office
of Science, under Contract No. W-31-109-Eng-38.
19 M. P. Rigobello, G. Scutari, A. Folda and A. Bindoli, Biochem.
Pharmacol., 2004, 67, 689–696.
2
0 S. J. Berners-Price, R. J. Bowen, P. J. Harvey, P. C. Healy and G. A.
Koutsantonis, J. Chem. Soc., Dalton Trans., 1998, 1743–1750.
21 S. J. Berners-Price, R. J. Bowen, P. Galettis, P. C. Healy and M. J.
McKeage, Coord. Chem. Rev., 1999, 185–186, 823–826.
2 S. J. Berners-Price, R. J. Bowen, T. W. Hambley and P. C. Healy, J. Chem.
Soc., Dalton Trans., 1999, 1337–1346.
2
2
3 M. J. McKeage, S. J. Berners-Price, P. Galettis, R. J. Bowen, W. Brouwer,
L. Ding, L. Zhuang and B. C. Baguley, Cancer Chemother. Pharmacol.,
2
000, 46, 343–350.
2
4 R. J. Bowen, C. Garner, S. J. Berners-Price and I. D. Jenkins,
J. Organomet. Chem., 1998, 554, 181–184.
2
2
5 K. Sommer, Z. Anorg. Allg. Chem., 1970, 376, 37–43.
6 K. Diemert, W. Kuchen and J. Kutter, Chem. Ber., 1982, 115, 1947–
1955.
References
2
7 J. A. Griffith, D. J. McCauley, J. R. E. Barans and A. W. Herlinger,
Synth. Commun., 1998, 28, 4317–4323.
1
2
D. R. Green and G. Kroemer, Science, 2004, 305, 626–629.
L. Galluzzi, N. Larochette, N. Zamzami and G. Kroemer, Oncogene,
28 N. Weferling, Z. Anorg. Allg. Chem., 1987, 548, 55–62.
29 L. Dahlenburg and A. Kaunert, Eur. J. Inorg. Chem., 1998, 885–887.
30 V. Plack, J. R. Goerlich, A. Fischer and R. Schmutzler, Z. Anorg. Allg.
Chem., 1999, 1979–1984.
31 E. Lindner, M. Schmid, J. Wald, J. A. Queisser, M. Gepr a¨ gs, P. Wegner
and C. Nachtigal, J. Organomet. Chem., 2000, 602, 173–187.
32 P. E. Garrou, Chem. Rev., 1981, 81, 229–266.
33 S. J. Berners-Price and P. J. Sadler, Inorg. Chem., 1986, 25, 3822–3827.
34 M. K. Cooper, L. E. Mitchell, K. Henrick, M. McPartlin and A. Scott,
Inorg. Chim. Acta, 1984, 84, L9–L10.
35 W. Kaim, A. Dogan, A. Klein and S. Z a´ li sˇ , Z. Anorg. Allg. Chem., 2005,
631, 1355–1358.
36 P. Sevillano, M. E. Garcia, A. Habtemariam, S. Parsons and P. J. Sadler,
Met.-Based Drugs, 1999, 6, 211–221.
37 D. Affandi, S. J. Berners-Price, Effendy, P. J. Harvey, P. C. Healy, B. E.
Ruch and A. H. White, J. Chem. Soc., Dalton Trans., 1997, 1411–1420.
38 M. Bardaji, P. G. Jones, A. Laguna, A. Moracho and A. Fischer,
J. Organomet. Chem., 2002, 648, 1–7.
39 P. G. Jones and M. C. Gimeno, (‘BIBXOA’) CCDC, 2004.
40 S. J. Berners-Price, G. R. Girard, D. T. Hill, B. M. Sutton, P. S. Jarrett,
L. F. Faucette, R. K. Johnson, C. K. Mirabelli and P. J. Sadler, J. Med.
Chem., 1990, 33, 1386–1392.
41 H. E. Gottlieb, V. Kotlyar and A. Nudelman, J. Org. Chem., 1997, 62,
7512–7515.
2
006, 25, 4812–4830.
3
4
5
6
7
8
N. Dias and C. Bailly, Biochem. Pharmacol., 2005, 70, 1–12.
V. R. Fantin and P. Leder, Oncogene, 2006, 25, 4787–4797.
J. S. Armstrong, Br. J. Pharmacol., 2006, 147, 239–248.
A. S. Don and P. J. Hogg, Trends Mol. Med., 2004, 10, 372–378.
L. B. Chen, Annu. Rev. Cell. Biol., 1988, 4, 155–181.
K. D. Nadakavukaren, J. J. Nadakavukaren and L. B. Chen, Cancer
Res., 1985, 45, 6093–6099.
I. C. Summerhayes, T. J. Lampidis, S. D. Bernal, J. J. Nadakavukaren,
K. D. Nadakavukaren, E. L. Shepherd and L. B. Chen, Proc. Natl.
Acad. Sci. USA, 1982, 79, 5292–5296.
9
1
0 M. F. Ross, G. F. Kelso, F. H. Blaikie, A. M. James, H. M. Cochem e´ , A.
Filipovska, T. Da, Ros, T. R. Hurd, R. A. J. Smith and M. P. Murphy,
Biochemistry (Moscow), 2005, 70, 222–230.
1
1
1 J. S. Modica-Napolitano and J. R. Aprille, Adv. Drug Delivery Rev.,
2
001, 49, 63–70.
2 I. K. Kandela, W. Lee and G. L. Indig, Biotech. Histochem., 2003, 78,
1
57–169.
1
1
3 S. Trapp and R. W. Horobin, Eur. Biophys. J., 2005, 34, 959–966.
4 P. J. Barnard and S. J. Berners-Price, Coord. Chem. Rev., 2007, 251,
1
889–1902.
5 S. J. Berners-Price and P. J. Sadler, Struct. Bonding, 1988, 70, 27–
02.
6 M. J. McKeage, L. Maharaj and S. J. Berners-Price, Coord. Chem. Rev.,
002, 232, 127–135.
1
1
1
1
42 L. Maier, Helv. Chim. Acta, 1966, 49, 842–851.
43 R. W. Buckley, P. C. Healy and W. A. Loughlin, Aust. J. Chem., 1997,
50, 775–778.
44 G. M. Sheldrick, SHELXL-97, A Program for Crystal Structure
Refinement, G o¨ ttingen, Germany.
2
7 S. J. Berners-Price, C. K. Mirabelli, R. K. Johnson, M. R. Mattern,
F. L. McCabe, L. F. Faucette, C.-M. Sung, S.-M. Mong, P. J. Sadler
and S. T. Crooke, Cancer Res., 1986, 46, 5486–5493.
1
8 M. P. Rigobello, G. Scutari, R. Boscolo and A. Bindoli, Br. J. Pharma-
col., 2002, 136, 1162–1168.
45 A. Albert, The Physiochemical Basis of Therapy, Chapman and Hall,
London, 1979.
4
950 | Dalton Trans., 2007, 4943–4950
This journal is © The Royal Society of Chemistry 2007