Bifunctional bioorganometallic Iridium(III)-Platinum(II) complexes incorporating both intercalative and covalent DNA binding capabilities

Article abstract of DOI:10.1002/ejic.200500223

The DNA binding of bifunctional Ir^III-Pt^II complexes containing [(η⁵-Cp*)Ir(dppz)]²⁺ and [Pt(terpy)]²⁺ or trans-[PtL₂L′]²⁺ fragments (L = NH₃, L′ = DMF; L = H₂O, L′ = NH₃) bridged by flexible κS:κN(amino)-coordinated methionine-containing peptides has been studied by UV/Vis and CD spectroscopy. Stable intercalative binding of the Ir^III fragment of the complexes [{(η⁵-Cp*)Ir(dppz)}(μ-peptide-κS:κN) {Pt(terpy)}]-(CF₃SO₃)₄ (5) (peptide = H-Gly-Gly-Phe-Met-OH; H-Gly-OH = glycine, H-Phe-OH = L-phenylalanine, H-Met-OH = L-methionine) and 6 [peptide = H-(Ala)₄-Met-OH; H-Ala-OH = L-alanine] is indicated by their steady decrease in absorbance at maxima between 350 and 390 nm on titration with CT DNA and by the bathochromic shifts of these maxima. Binding constants K_b of 1.4(4) × 10⁶ M ^-1 for both 5 and 6 and site sizes s of 2.8(1) and 2.2(1), respectively, are in accordance with this monofunctional mode. Both peptide chain length and the site(s) of the labile Pt^II substituents DMF or H₂O play an important role in determining the degree of intercalation of the Ir^III fragment and the extent of helix distortion caused by simultaneous covalent binding of Pt^II centres for the DNA interaction of the complexes [{(η⁵-Cp*)Ir(dppz)}(μ-peptide- κS:κN){trans-(PtL₂L′)}]⁴⁺ 7-10. Whereas both 9 and 10 (L = H₂O, L′ = NH₃; peptide = H-Gly-Met-OH or H-Gly-Gly-Met-OH) exhibit a high degree of bifunctional binding [9, K_b = 1.1(5) × 10⁶ M^-1, s = 1.3(1); 10, K_b = 1.4(7) × 10⁶ M^-1, s = 2.1(1)], CD spectroscopy indicates that a more pronounced reorganisation of the DNA helix is required for the former complex with its shorter peptide. Effective intercalation is also observed for 8 (L = NH₃, L′ = DMF; peptide = H-Gly-Gly-Met-OH) but not for 7 with the analogous Pt^II fragment but a shorter bridging peptide (H-Gly-Met-OH). Wiley-VCH Verlag GmbH & Co. KGaA, 2005.

Full text of DOI:10.1002/ejic.200500223

FULL PAPER
Bifunctional Bioorganometallic Iridium(III)–Platinum(II) Complexes
Incorporating Both Intercalative and Covalent DNA Binding Capabilities
Susan Gençaslan[a] and William S. Sheldrick*[a]
Keywords: Iridium / Platinum / Peptides / DNA binding / Bioinorganic chemistry
The DNA binding of bifunctional IrIII–PtII complexes contain-
ing [(η5-Cp*)Ir(dppz)]2+ and [Pt(terpy)]2+ or trans-[PtL2LЈ]2+
fragments (L = NH3, LЈ = DMF; L = H2O, LЈ = NH3) bridged
by flexible κS:κN(amino)-coordinated methionine-containing
peptides has been studied by UV/Vis and CD spectroscopy.
Stable intercalative binding of the IrIII fragment of the com-
an important role in determining the degree of intercalation
of the IrIII fragment and the extent of helix distortion caused
by simultaneous covalent binding of PtII centres for the DNA
interaction of the complexes [{(η5-Cp*)Ir(dppz)}(µ-peptide-
κS:κN){trans-(PtL2LЈ)}]4+ 7–10. Whereas both 9 and 10 (L =
H2O, LЈ = NH3; peptide = H-Gly-Met-OH or H-Gly-Gly-Met-
plexes
(CF3SO3)4 (5) (peptide = H-Gly-Gly-Phe-Met-OH; H-Gly-OH
= glycine, H-Phe-OH = -phenylalanine, H-Met-OH = -me-
thionine) and 6 [peptide = H-(Ala)4-Met-OH; H-Ala-OH =
[{(η5-Cp*)Ir(dppz)}(µ-peptide-κS:κN){Pt(terpy)}]-
OH) exhibit a high degree of bifunctional binding [9, Kb
=
1.1(5)×106 –1, s = 1.3(1); 10, Kb = 1.4(7)×106 –1, s = 2.1(1)],
M
M
L
L
CD spectroscopy indicates that a more pronounced reorgan-
isation of the DNA helix is required for the former complex
with its shorter peptide. Effective intercalation is also ob-
served for 8 (L = NH3, LЈ = DMF; peptide = H-Gly-Gly-Met-
OH) but not for 7 with the analogous PtII fragment but a
shorter bridging peptide (H-Gly-Met-OH).
L-
alanine] is indicated by their steady decrease in absorbance
at maxima between 350 and 390 nm on titration with CT
DNA and by the bathochromic shifts of these maxima. Bind-
–1
ing constants Kb of 1.4(4)×106
M
for both 5 and 6 and site
sizes s of 2.8(1) and 2.2(1), respectively, are in accordance
with this monofunctional mode. Both peptide chain length
and the site(s) of the labile PtII substituents DMF or H2O play
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
dtdeg)PtCl]Cl3 {terpy = 2,2Ј:6Ј,2ЈЈ-terpyridine, dtdeg =
bis[4Ј-(2,2Ј:6Ј,2ЈЈ-terpyridyl)]diethyleneglycol
Introduction
ether}.[5]
Multinuclear platinum complexes containing two or
more square-planar platinum() centres capable of covalent
binding to DNA nucleobases represent a relatively new
class of anticancer agents.[1–3] They have been shown to ex-
hibit activity in cancer cell lines with intrinsic resistance to
cisplatin and can also overcome acquired resistance to this
successful antitumor drug, presumably as a result of their
participation in long-range interstrand DNA adducts.
Length, conformational flexibility, total charge and hydro-
gen-bonding donor capacity of the linking ligands (e.g. 1,
n-diaminoalkanes) have been identified by Farrell[1,2] as
paramount factors in the design of multinuclear platinum
drugs.
However, the light-sensitive and highly reactive former
complex proved to be unsuitable as a DNA binding probe
and the latter complex exhibited a lower cytotoxicity
towards cisplatin-sensitive cell lines in comparison to
[PtCl(terpy)]·2H2O, presumably due to the fact that its ru-
thenium moiety can only participate in electrostatic DNA
interactions. Attempts to improve on the effectiveness of
cisplatin have also involved the tethering of intercalating
acridine[6,7] and phenazine[8,9] groups to platinum()
centres. The presence of an intercalative chromophore leads
both to a marked increase in the rate of DNA platination
and to an altered sequence specificity.[7,9] Metallointercala-
tors such as [Pt(terpy)]2+[10,11] or [Ru(dppz)(phen)2]2+ (dppz
= dipyrido[3,2-a:2Ј,3Ј-c]phenazine)[12] have also been linked
to identical fragments to afford bifunctional complexes with
high DNA binding affinities. The use of dppz-containing
compounds to promote intercalative DNA binding is well
documented.[13]
A modified approach to the polynuclear concept involves
the design of bifunctional complexes with two distinct
DNA binding capabilities connected by a rigid or flexible
bridging moiety. For instance octahedral ruthenium() and
square-planar platinum() centres have been linked in the
heterodinuclear
complexes
[{cis-RuCl2(Me2SO)3}[µ-
We have recently demonstrated that bioorganometallic
metallointercalators of the type [(η5-Cp*)Ir(dppz)(pep-
tide)]n+ (n = 1–3)[14] and [(η6-arene)Ru(dppz)(peptide)]n+
(n = 1–3; arene = C6H6, Me3C6H3, C6Me6)[15] exhibit strong
side-on intercalative binding into DNA with equilibration
constants Kb of up to 5.5×106 –1. Following initial
H2N(CH2)4NH2]{cis-PtCl2(NH3)}][4] and [(terpy)Ru(µ-
[a] Lehrstuhl für Analytische Chemie, Ruhr-Universität Bochum,
44780 Bochum, Germany
Fax: +49-234-32-14742
E-Mail: William.Sheldrick@ruhr-uni-bochum.de
3840
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200500223
Eur. J. Inorg. Chem. 2005, 3840–3849

Bifunctional Bioorganometallic Iridium()–Platinum() Complexes
FULL PAPER
chemospecific κS coordination of the organometallic [(η5- tems and contains a κO-coordinated trifluoroacetate ligand
Cp*)Ir(dppz)]2+ or [(η6-arene)Ru(dppz)]2+ fragment in with an Ir–O2 distance of 2.156(7) Å. Although individual
complexes of methionine-containing peptides, a second sim- [(η5-Cp*)Ir(CF3COO)(dppz)]+ cations dimerise through π–
ilar or different metal fragment can be introduced by em- π interactions of their parallel phenanzine ligands in the
ploying either the terminal amino function or the donor crystal structure of 2 (Figure 2), more extensive base stack-
atom of a suitable side chain as the additional binding site. ing is prevented by the steric requirements of the Cp* and
The feasibility of this concept has been demonstrated for CF3COO– ligands. An interplanar distance of 3.48 Å is ob-
[{(η5-Cp*)Ir(dppz)}2(µ-H-Met-Met-OH-κS:κSЈ)](CF3SO3)4
and
[{(η5-Cp*)Ir(dppz)}(µ-H-Gly-Met-OH-κS:κNG) Dimerisation of complexes containing the [(η5-Cp*)-
served for the dppz ring systems within such cation pairs.
{Pt(terpy)}](CF3SO3)4.[16] We now extend the design strat- Ir(dppz)]2+ fragment will, of course, be expected in solution
egy to iridium()–platinum() complexes exhibiting an in- and also possibly as an external binding mode on the DNA
tercalating [(η5-Cp*)Ir(dppz)]2+ moiety tethered to a trans- surface for the bifunctional complexes studied in the pres-
[PtL2LЈ] (L = NH3, H2O;LЈ = DMF, NH3) fragment. To ent work. This prompted us to determine the dimerisation
the best of our knowledge, such compounds of the type constant KD for [(η5-Cp*)Ir(dppz)(H-Gly-Gly-Met-OH-
[{(η5-Cp*)Ir(dppz)}(µ-peptide-κS:κN){trans-(PtL2LЈ)}]4+ κS)](CF3SO3)2 as a typical monofunctional member of this
represent the first examples of bifunctional complexes con- class of bioorganometallic compounds.
taining both a metallointercalator and a covalent DNA
binding capability.
Results and Discussion
Monofunctional Starting Compounds [(η5-Cp*)Ir(dppz)-
(peptide-κS)](CF3SO3)2 3 and 4
The bioorganometallic starting compounds of the type
[(η5-Cp*)Ir(dppz)(peptide-κS)](CF3SO3)2 [peptide = H-
Gly-Met-OH,[16] H-Gly-Gly-Met-OH,[14] H-Gly-Gly-Phe-
Met-OH (3), H-(Ala)4-Met-OH (4)] were prepared as de-
scribed previously for H-Gly-Met-OH[16] and H-Gly-Gly-
Met-OH[14] by initially refluxing [(η5-Cp*)Ir(acetone)3]-
(CF3SO3)2 (1)[17] with dppz in CH3OH/CH2Cl2 for 2 h. Ad-
dition of the appropriate peptide in CF3COOH to the re-
sulting [(η5-Cp*)Ir(acetone)(dppz)](CF3SO3)2 solution
leads to formation of the desired complexes [(η5-Cp*)-
Figure 1. Molecular structure of the cation of [(η5-Cp*)-
Ir(CF3COO)(dppz)](CF3SO3) (2): Selected bond lengths [Å] and
angles [°]: Ir–N1 2.109(8), Ir–N10 2.106(8), Ir–O2 2.156(7), Ir–C
Ir(dppz)(peptide-κS)](CF3SO3)2 (Scheme 1). Strongly acidic
conditions (CF3COOH) are, however, necessary to achieve
chemospecific κS coordination of the methionine side
2.158(4)–2.198(4), C100–O1 1.20(2), C100–O2 1.26(1), N1–Ir–N10
chains in the presence of non-protected terminal amino or
carboxylate groups. After stirring at 50 °C for 18 h and vol-
ume reduction, the products can be precipitated by addition
of diethyl ether. The remaining solution contains a residual
quantity of [(η5-Cp*)Ir(CF3COO)(dppz)](CF3SO3) (2),
which could be subsequently crystallised in low yield.
77.4(3), N1–Ir–O2 84.6(3), N10–Ir–O2 82.9(3).
The [(η5-Cp*)Ir(dppz)(H-Gly-Gly-Met-OH)]2+ cation
displays two characteristic absorption maxima at 364 and
382 nm for the π–π* transitions of its dppz ligand. At very
low concentrations (10 µ or lower) in a 10 m phosphate
buffer at pH = 7.2, the molar absorption coefficients ε364
and ε382 are independent of the cation concentration, i.e. 2
is present as a monomer. The observed hypochromic shifts
in the ε values for the concentration range 10–180 µ (Fig-
ure 3) are indicative of increasing intermolecular dipole–di-
pole interactions between dppz ligands and thereby of di-
merisation. Using the equation of Schwarz et al.,[18] a di-
merisation constant KD of 6.4×103 –1 was determined on
the basis of a regression analysis for (εM – εobsd.) (λ =
382 nm) in the concentration range 10–250 µ. Respective
values of 13395 L·mol–1·cm–1 and 8894 L·mol–1·cm–1 were
obtained for the molar absorption constants εM and εD of
the monomer and dimer. These represent a relative decrease
in absorption ∆A/A at 382 nm of 25.4%. The strength of
Scheme 1.
As depicted in Figure 1, complex 2 exhibits an in- the base stacking for the large planar dppz ligands of [(η5-
terplanar angle of 56.1° between its Cp* and dppz ring sys- Cp*)Ir(dppz)(H-Gly-Gly-Met-OH)]2+ is underlined by
Eur. J. Inorg. Chem. 2005, 3840–3849
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3841

S. Gençaslan, W. S. Sheldrick
FULL PAPER
Figure 2. Dimerisation of the [(η5-Cp*)Ir(CF3COO)(dppz)]+ cations of 2 in the crystal lattice.
comparing its KD value with the lower values of 4×103 and tions in the range δ = 1.79–1.81 ppm. As previously noted
5×102 –1 for [PtCl(terpy)]Cl[19] and proflavine.[18]
for both [(η5-Cp*)Ir(dppz)]2+ and [(η6-arene)Ru(dppz)]2+
derivatives[14,15] this behaviour represents a striking reversal
of the typical positive shift displayed by the signals of the
δ-CH3 protons in analogous half-sandwich organometallic
complexes without a chelating aromatic ligand such as
dppz.[20,21] The shielding cone of the neighbouring dppz li-
gand may well be responsible for this anomalous upfield
shift in 3–6.
Figure 3. Dimerisation of the [(η5-Cp*)Ir(dppz)(H-Gly-Gly-Met-
OH)]2+ cations in a phosphate buffer at pH = 7.2 as demonstrated
by the hypochromic shift of the molar absorption coefficients ε
(L·mol–1·cm–1) in the wavelength range 300–420 nm.
Bifunctional Iridium(III)–Platinum(II) Metallointercalators
5 and 6
The bifunctional iridium()–platinum() metallointerca-
lators
κS:κNG){Pt(terpy)}](CF3SO3)4 (5) and [{(η5-Cp*)Ir(dppz)}-
(µ-H-(Ala)4-Met-OH-κS:κNG){Pt(terpy)}](CF3SO3)4 (6)
[{(η5-Cp*)Ir(dppz)}(µ-H-Gly-Gly-Phe-Met-OH-
(Figure 4) are prepared by addition of 3 or 4, respectively,
to an aqueous [Pt(OH2)(terpy)](CF3SO3)2 solution. All four
complexes were characterised by elemental analysis, FAB
mass spectrometry, 1H and 13C NMR, IR and UV/Vis spec-
1
troscopy. Downfield H NMR shifts of 0.10 ppm are ob-
served for the signals of the Cp* methyl protons of the κS-
coordinated [(η5-Cp*)Ir(dppz)]2+ fragments in 3–6 in com-
parison to the δ value of 1.74 for this singlet in a CD3OD
solution
of
the
chloro
complex
[(η5-Cp*)-
Ir(Cl)(dppz)](CF3SO3).[14] In contrast, the signals of the
methionine δ-CH3 protons adjacent to the coordinated thio-
ether sulfur atoms in the met-containing complexes exhibit
Figure 4. Tetracations of the bifunctional complexes 5 and 6.
Additional κN coordination of the [Pt(terpy)]2+ fragment
a more pronounced opposite upfield shift from typical val- by the terminal amino group of the peptide ligands in 5 and
ues of ca. 2.10 ppm in the free peptide to resonance posi- 6 leads to significant upfield shifts in the δ values of the H6
3842
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 3840–3849

Bifunctional Bioorganometallic Iridium()–Platinum() Complexes
FULL PAPER
and H5 protons from their resonance positions of δ = 9.00
and 7.86 ppm in [PtCl(terpy)]Cl (CD3OD solution). Re-
spective shifts of δ = 8.83/8.88 and 7.82 ppm are observed
for the bifunctional complexes. Further clear evidence for
κN coordination is also provided by the characteristic up-
field shifts recorded for the signals of the adjacent α pro-
ton(s) of the terminal amino acid residue in 5 and 6.
Whereas the αGly1 singlet exhibits only a modest shift from
δ = 3.65 ppm in the free peptide to 3.61 ppm upon coordi-
nation in 5, a striking move from δ = 3.94 to 3.72 ppm is
observed for signal of the αAla1 proton in 6.
DNA Binding Studies for 3–6
A pronounced decrease in absorbance at 364 and 382 nm
and the bathochromic shifts of these absorption maxima
(Figure 5) on UV/Vis titration of 20 µ solution of 3–6 with
calf thymus DNA (CT DNA) are clearly indicative of dppz
intercalation into DNA. In contrast to the hypochromic
shifts –∆A/A of 32.1–38.7% for 3–6 at 382 nm with their
associated red shifts of 5 nm, a much smaller decrease in
absorbance of only 12.2 or 11.0%, respectively, is observed
for the terpyridine ligand at its 343 nm maximum in 5 and
6, which remains unshifted. As no correction for the in-
crease in DNA absorbance at 343 nm (due to its 15-fold
increase in concentration during the UV/Vis titration) was
made, it is reasonable to assume that terpyridine intercal-
ation will not be of great significance for the binding of 5
and 6. Absorption data collected at 382 nm for 3–6 were
fitted by least squares using the non-cooperative non-spe-
cific binding model by Bard and Thorp.[22,23] Best fits were
obtained for the intrinsic binding constants Kb and the
average binding site sizes s listed in Table 1. Both these pa-
rameters and the melting temperature shifts ∆Tm for 3 and
4 are closely similar to the values of 1.2(1)×106 –1 (Kb),
1.6 (s) and 6.9 °C (∆Tm) previously reported for the analo-
gous compound [(η5-Cp*)Ir(dppz)(H-Gly-Gly-Met-OH-
κS)](CF3SO3)2.[14] We chose the longer oligopeptides H-
Gly-Gly-Phe-Met-OH and H-(Ala)4-Met-OH in the hope
that these might enable long-range bis-intercalation of their
bifunctional complexes 5 and 6 into the DNA helix.
Figure 5. (a) UV/Vis spectra for the titration of [{(η5-Cp*)Ir(dppz)}-
[µ-(Ala)4-Met-OH-κS:κN]{Pt(terpy)}](CF3SO3)4 6 (20 µ) in a
10 m phosphate buffer (pH = 7.2) with CT DNA [0–300 µ (nu-
cleotide)]; (b) the best least-squares fit for 6/DNA to the non-coop-
erative non-specific binding model by Bard and Thorp.[22,23]
Table 1. Binding constants Kb, site sizes s and melting temperature
shifts ∆Tm for the interaction of CT DNA with complexes 3–6 in
a 10 m phosphate buffer at pH = 7.2.
Complex
Kb [–1]
s
∆Tm [°C]
3
2.3(3)×106
1.0(3)×106
1.4(4)×106
1.4(4)×106
1.2(1)×106
2.6(1)
1.8(1)
2.8(1)
2.2(1)
1.6(1)
5.2
5.2
7.2
8.2
6.9
4
5
6
Ref.[14][a]
[a] For [(η5-Cp*)Ir(dppz)(H-Gly-Gly-Met-OH-κS)](CF3SO3)2.
However, as simultaneous intercalation of both the aro- cessful fitting of the UV/Vis data for all complexes in a wide
matic ligands of the [(η5-Cp*)Ir(dppz)]2+ and [Pt(terpy)]2+ complex/DNA molar ratio {r = 0.067–1.0 with [DNA] =
fragments would be predicted to afford significantly higher M(nucleotide)} indicate that the DNA binding mode of
binding constants and site sizes [i.e. the average number such metallointercalators must be effectively independent of
of nucleobase pairs between the specific or middle binding the number of amino acid residues. Whereas the extended
position of neighbouring mono- or bis-metallointercalators, aromatic system of dppz preferentially intercalates into
respectively], the calculated values of 1.4×106 –1 and DNA, the stabilising role of the smaller terpyridine ligands
2.8(1)/2.2(1) for Kb and s in 5 and 6 are clearly in accord- will presumably be restricted to intermolecular base stack-
ance with effectively exclusive [(η5-Cp*)Ir(dppz)]2+ mono- ing on the DNA surface. The relatively limited decrease in
intercalation. We have previously reported similar binding absorbance at the terpy maximum (λ = 343 nm) observed
constants [1.5(1)×106 –1 and 3(3)×106 –1] and site sizes during UV/Vis titrations is in accordance with this binding
(2.6, 2.9) which also indicate only mono-intercalation of the description (Figure 5). Interestingly, the highest Kb and s
IrIII fragment for two likewise intercalators [{(η5-Cp*)- values were determined for the H-Gly-Met-OH complex,
Ir(dppz)}2(µ-H-Met-Met-OH-κS:κSЈ)](CF3SO3)4
and thereby suggesting that such surface π–π interactions may,
[{(η5-Cp*)Ir(dppz)}(µ-H-Gly-Met-OH-κS:κNG){Pt(terpy)}]- indeed, be more favourable for the [Pt(terpy)]2+ moiety
(CF3SO3)4.[16] These values, taken together with the suc- when bridged to the IrIII fragment by this shorter peptide.
Eur. J. Inorg. Chem. 2005, 3840–3849
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3843

S. Gençaslan, W. S. Sheldrick
FULL PAPER
Bifunctional Iridium(III)–Platinum(II) Complexes 7–10 with
Both Intercalative and Covalent DNA Binding Capabilities
tion following initial precipitation and removal of AgCl on
treatment of the PtII complex with 3 equiv. of Ag(CF3SO3).
Although the molecular ions, themselves, could not be de-
tected by FAB mass spectrometry, the bridging µ-
1κS:2κNGly1 coordination mode is clearly confirmed for the
Treatment of the appropriate monofunctional starting
compound [(η5-Cp*)Ir(dppz)(peptide-κS)](CF3SO3)2 with
trans-[Pt(DMF)2(NH3)2](NO3)2 in DMF affords the bifunc-
peptides H-Gly-Met-OH (9) and H-Gly-Gly-Met-OH (10)
tional
complexes
[{(η5-Cp*)Ir(dppz)}(µ-peptide-
1
by the pronounced H NMR upfield shifts of their δMet
-
κS:κNG1)(trans-{Pt(NH3)2DMF})](CF3SO3)2(NO3)2 (pep-
tide = H-Gly-Met-OH 7, H-Gly-Gly-Met-OH 8) shown in
Figure 6. The required trans-[Pt(DMF)2(NH3)2](NO3)2 is
prepared in situ by addition of 2 equiv. of AgNO3 to trans-
[PtCl2(NH3)2] in DMF solution followed by filtration of the
precipitated AgCl. Retention of the κS coordination mode
by the [(η5-Cp*)Ir(dppz)]2+ fragment is confirmed by the
CH3 and αGly1–CH2 singlets to δ = 1.79/3.64 and 1.81/3.64
ppm, respectively. The much stronger trans effect of NH3 in
comparison to the aqua ligands (Et values 200:1)[26] leads
to exclusive κN coordination of the peptide amino group in
trans position to the ammine ligand in 9 and 10. Sharp sing-
lets for Me αGly1–CH2 protons are also in accordance with
the presence of only the trans isomer.
1
characteristic H NMR upfield shifts of the methionine δ-
Before turning in detail to DNA interaction studies of
the bifunctional complexes 7–10 with their complementary
covalent and intercalative binding capabilities, it is neces-
sary to consider possible competitive substitution reactions
for the nucleobase nitrogen atoms, i.e. whether IrIII or PtII
will be preferred as coordination partners. The kinetics of
CH3 singlets to δ = 1.79 ppm in 7 and 8. Replacement of
the IrIII fragment by trans-[Pt(DMF)(NH3)2] would, in con-
trast, lead to a downfield shift for the signals of these pro-
tons from δ Ϸ 2.10 ppm in the free peptides to a value in
the range δ = 2.3–2.5 ppm.[24,25] Coordination of the PtII
moiety by the N-terminal glycine (Gly1) amino function in
7 and 8 is indicated by the pronounced upfield shifts of the
αGly1 singlets from δ = 3.72/3.75 ppm in the free peptides
to δ = 3.64/3.62 ppm in the bifunctional complexes. The
resonances of the trans-sited DMF ligand (CD3OD solu-
tion) are observed with appropriate integral values at δ =
2.84, 2.98 (CH3) and 7.96 (CH) in 7 and 2.85, 2.97 (CH3)
and 7.96 (CH) ppm in 8.
1
nucleobase binding was, therefore, studied by H NMR in
an exemplary manner for the reaction of 7 with the model
purine base 9-ethylguanine at a 1:2 molar ratio in a 10 m
phosphate buffer (pH = 7.2) at T = 298 K (Figure 7). As
indicated by the rapid growth in the free DMF CH singlet
at δ = 7.94 ppm in comparison to the declining resonance
at δ = 7.90 ppm (coordinated DMF), substitution by the
hydrogen- or dihydrogenphosphate buffer anions is rela-
tively rapid. The extent of the subsequent slow replacement
of the new anionic ligand by κN7-coordinated 9-ethylgua-
nine can be gauged by the integral ratio of the H8 reso-
nances at δ = 8.32 (coordinated) and 7.80 (free) ppm. After
5 d, ca. 90% of 7 is converted into [{(η5-Cp*)Ir(dppz)}(µ-
H-Gly-Met-OH-κS:κNG1)[trans-{Pt(9-ethylguanine)-
(NH3)2}]]4+ (11) depicted in Figure 8. Exclusive PtII coordi-
nation is confirmed by the unchanged resonance positions
and integral values for the dppz and δMet protons after this
period of time.
Figure 6. Tetracations of the bifunctional complexes 7–10.
Bifunctional organometallic compounds of the type
[{(η5-Cp*)Ir(dppz)}(µ-peptide-κS:κNG1)(trans-{Pt(NH3)-
(OH2)2})][Et4N](CF3SO3)5 9 and 10 are prepared by reac-
tion of the starting compounds [(η5-Cp*)Ir(dppz)(peptide-
1
Figure 7. Aromatic region of the H NMR spectrum of a 1:2 reac-
tion mixture of 7 (c = 5 m) and 9-ethylguanine in a 10 m phos-
κS)](CF3SO3)2 with (Et4N)[PtCl3(NH3)] in aqueous solu- phate buffer at pH = 7.2 and T = 298 K.
3844 © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjic.org
Eur. J. Inorg. Chem. 2005, 3840–3849

Bifunctional Bioorganometallic Iridium()–Platinum() Complexes
FULL PAPER
cline in ∆A/A from a maximum value of ca. 45.1%[27] after
2 min.
Table 2. –∆A/A (for λ = 382 nm), binding constants Kb, site sizes s,
melting temperature shifts ∆Tm and induced circular dichroism Θ
(for λ = 300 nm) for the interaction of CT DNA with complexes
7–10 in a 10 m phosphate buffer at pH = 7.2.
Complex –∆A/A
Kb
s
∆Tm
[Θ]
[%]
[–1]
[°C] [deg·cm2·dmol–1]
7
8
9
27.5
35.2
43.9
38.5
1.1(2)×106 0.9(1)
8.2
–1.5×103
–2.9×103
–3.8×103
–3.8×103
4(3)×106
2.4(2) 11.2
1.1(5)×106 1.3(1) 17.2
1.4(7)×106 2.1(1) 17.2
10
Figure 8. Product 11 of the reaction between 7 and 9-ethylguanine.
UV/Vis DNA Binding Studies for 7–10
The interaction of 7–10 with CT DNA was studied by
UV/Vis titrations, DNA melting temperature determi-
nations and CD spectroscopy. As for 3–6, hypo- and ba-
thochromic shifts were observed for the dppz absorption
maxima at 363 and 382 nm on increasing the DNA/com-
plex molar ratio from 1:1 to 15:1. Although ca. 90–95% of
the total decrease in absorbance ∆A/A at 382 nm is achieved
within 2 min, it takes between 8 (complex 7) and 40 h (com-
plex 9) for the DNA/complex reaction mixtures to reach an
equilibrium. This behaviour is in striking contrast to 3–6,
for which –∆A/A reaches its maximum value within 2 min,
and indicates that a slow structural reorganisation and
binding optimisation must take place for 7–10 following ra-
pid initial intercalation and/or surface base stacking. All
UV/Vis spectra employed for the determination of binding
constants and site sizes were measured after equilibration,
i.e. after no further change in the monitored absorbance at
382 nm was recorded.
DNA binding parameters for 7–10 are listed in Table 2.
As satisfactory least-squares fits (Figure 9) could be ob-
tained in all cases for the absorption data using the model
by Bard and Thorp[22,23] non-cooperative non-specific bind-
ing can be assumed for the interaction of the complexes
with CT DNA. The low –∆A/A value of 27.5% and the
physically unrealistic site size (s = 0.9) of less than one base
pair between binding sites indicate that only a limited de-
gree of intercalation is achieved by 7 with its dipeptide
Figure 9. Best least-squares fits to the model by Bard and
Thorp[22,23] for complexes 7 and 10.
Circular Dichroism DNA Binding Studies for 7–10
In contrast to 7, the much higher –∆A/A and s values of
bridging ligand H-Gly-Met-OH. A comparison of these val- 35.2% and 2.4(2), respectively, for 8 indicate that this com-
ues with those of the monofunctional complex [(η5-Cp*)- plex with its longer bridging peptide, H-Gly-Gly-Met-OH,
[27]
Ir(dppz)(H-Gly-Met-OH-κS)](CF3SO3)2
[–∆A/A
=
is apparently capable of combining both κN coordination
45.1%, Kb = 1.9(7)×106 –1, s = 2.2(1)], for which equili- of the DNA nucleobases by the trans-[Pt(NH3)2(peptide-
bration is reached within 2 min, suggests that coordination κN)]2+ fragment and effective intercalation of the [(η5-Cp*)-
of the PtII centres by DNA nucleobases must be rapid for Ir(dppz)(peptide-κS)]2+ moiety. Characteristic changes in
7. Exclusive initial intercalation of the [(η5-Cp*)Ir- the observed circular dichroism (CD) spectrum of DNA
(dppz)]2+ fragment (as observed for the monofunctional provide a means of monitoring both conformational
complex), followed by a slow increase in κN coordination changes for the biopolymer and possible intercalation or
of the group 10 metal and a concomitant reduction in the groove binding of an interacting metal complex.[28,29] As
degree of intercalation, would otherwise lead to a slow de- depicted for CT DNA in Figure 10, a negative band at
Eur. J. Inorg. Chem. 2005, 3840–3849
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3845

S. Gençaslan, W. S. Sheldrick
FULL PAPER
246 nm and a positive band at 275 nm due to base stacking stabilisation of base stacking in the helix due to the combi-
are observed for a B-DNA helix. Addition of the metall- nation of both dppz intercalation (see Figure 10) and coval-
ointercalator
[(η5-Cp*)Ir(dppz)(H-Gly-Gly-Met-OH- ent binding of the PtII centres. In contrast, the decrease in
κS)](CF3SO3)2 at a complex/DNA molar ratio of r = 0.1 [Θ] for 7 at this wavelength is comparable with that pre-
leads to a small reduction in the negative Cotton effect at viously reported for [PtCl(dien)]Cl[30] and can, therefore, be
246 nm and a small hypsochromic shift of 3 nm for the pos- regarded as being indicative for monodentate PtII coordina-
itive band to 272 nm, thereby indicating only minor distor- tion with a significantly lower degree of intercalation. The
tion of the B helix. The additional negative-induced CD large decrease in the negative [Θ] values at 246 nm in both 7
band at 300 nm must presumably be caused by the intercal- and 8 in comparison to [(η5-Cp*)Ir(dppz)(H-Gly-Gly-Met-
ation of the dppz ligands into the DNA helix.[29]
OH-κS)](CF3SO3)2 (Figure 10) may be due to the fourfold
positive charge of the bifunctional complexes, which could
lead to a change in the helical conformation. A schematic
illustration of a possible long-range simultaneous covalent
and intercalative binding mode for complex 8 with s = 2.0
(calcd. 2.4) is given in Figure 12a.
Figure 10. CD spectra of CT DNA and a mixture of [(η5-Cp*)-
Ir(dppz)(H-Gly-Gly-Met-OH-κS)](CF3SO3)2 and DNA [molar ra-
tio (1:10)] in a 10 m phosphate buffer (pH = 7.2) after 48 h incu-
bation at 295 K.
Figure 11 compares the CD spectra of the bifunctional
complexes 7 and 8 with CT DNA at a molar ratio of 0.1.
The significantly higher degree of intercalation for 8 leads
to a molar elipticity ([Θ] = –2.9×103 deg·cm2·dmol–1 for λ
= 300 nm) that is almost twice as large as that recorded
for the less effective metallointercalator 7 ([Θ] = –1.5×103
deg·cm2·dmol–1 for λ = 300 nm). In the case of the former
complex, the small but significant increase in [Θ] for the
positive CD band at 276 nm may be attributed to additional
Figure 12. Schematic depictions of possible simultaneous covalent
and intercalative binding modes; (a) for 8 with s = 2.0 (calcd. 2.4);
(b) for 9 with s = 1.0 (calcd. 1.3).
The cis positions of the labile aqua ligands relative to the
coordinating amino groups of the peptide ligands in 9 and
10 might be expected to favour a high degree of simulta-
neous covalent and intercalating DNA binding even for the
shorter dipeptide of 9. Experimental support for this hy-
pothesis is provided by their induced molar elipticity [Θ] at
λ = 300 nm (–3.8×103 deg·cm2·dmol–1) and by the high val-
ues of –∆A/A (43.9, 38.5%) and Kb (1.1×106, 1.4×106 –1)
observed for their UV/Vis titrations with CT DNA at
382 nm. The former parameters are particularly interesting
for 9, as they indicate highly effective dppz intercalation for
a low average site size of only 1.3 base pairs. In contrast to
10, for which only the negative band at 247 nm is signifi-
Figure 11. CD spectra of (a) complexes 7 and 8 and (b) complexes
9 and 10 with CT DNA at a complex/[DNA] molar ratio of 1:10
in a 10 m phosphate buffer (pH = 7.2) at 295 K after 48 h incu-
bation. [Θ] values are given in deg·cm2·dmol–1·103.
3846
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 3840–3849

Bifunctional Bioorganometallic Iridium()–Platinum() Complexes
FULL PAPER
and [(η5-Cp*)Ir(dppz)(H-Gly-Gly-Met-OH-
κS)](CF3SO3)2 were prepared from 1 and the relevant peptide in a
manner previously described for the latter complex.[14]
(Et4N)[PtCl3(NH3)] was obtained according to a literature pro-
cedure on treating cis-[PtCl2(NH3)2] with (Et4N)Cl·H2O.[36]
[16]
κS)](CF3SO3)2
cantly influenced by metal binding, marked decreases in the
intensity are observed for both the positive and negative
bands in the CD spectrum recorded for 9/DNA at a 1:10
molar ratio. The magnitude of these decreases is similar to
those observed for trans-[PtCl2(NH3)2],[30] which can dis-
turb the DNA helix by participation in trans-bidentate [(η5-Cp*)Ir(dppz)(H-Gly-Gly-Phe-Met-OH-κS)](CF3SO3)2 (3): The
ligand dppz (56.5 mg, 0.2 mmol) was added to 1 (159.8 mg,
0.2 mmol) in CH3OH/CH2Cl2 (10 mL) and the resulting solution
stirred at reflux for 2 h. After addition of H-Gly-Gly-Phe-Met-OH
(82.1 mg, 0.2 mmol) in CF3COOH (1.5 mL) and stirring at 50 °C
for 18 h, the volume was reduced to 3 mL. The product was precipi-
tated by addition of diethyl ether, washed and dried in vacuo. Yield
181.9 mg (69%). C48H51F6IrN8O11S3 (1318.3): calcd. C 43.7, H 3.9,
N 8.5, S 7.3; found C 43.3, H 3.8, N 8.4, S 7.3. FAB MS: m/z (%)
= 1319 (17) [M]+, 1168 (10) [M – OTf]+, 1019 (44) [M – 2 OTf]+.
1H NMR (CD3OD): δ = 1.79 (s, 3 H, δ Met), 1.84 (s, 15 H, CH3
binding. However, the characteristic bathochromic shifts of
the CD band at maxima for this coordination mode are
missing for both 9 and 10. It can be concluded that 9 and
10 must both exhibit a high degree of simultaneous covalent
and intercalative DNA binding and that a more pro-
nounced reorganisation of the DNA conformation is re-
quired to achieve this goal with complex 9 with its shorter
linking peptide. The high melting temperature shifts ∆Tm of
17.2 °C for the interaction mixtures 9/DNA and 10/DNA
result presumably from increased electrostatic stabilisation Cp*), 1.9–2.1 (mm, 4 H, β and γ Met), 2.93, 3.14 (m, 2 H, β Phe),
3.65, 3.69 (2s, 4 H, α Gly1, α Gly2), 4.29 (m, 1 H, α Phe), 4.36 (m,
1 H, α Met), 7.15 (m, 4 H, Phe), 7.24 (s, 1 H, Phe), 8.11 (m, 2 H),
8.43 (m, 2 H), 8.49 (dd, 2 H), 9.37 (dd, 2 H), 10.04 (dd, 2 H, dppz)
ppm. 13C NMR (CD3OD): δ = 8.6 (CH3 Cp*), 17.0 (δ Met), 30.6
(β Met), 33.7 (γ Met), 38.2 (β Phe), 41.6, 43.4 (α Gly), 51.0 (α Met),
56.2 (α Phe), 97.2 (Cp*), 127.8, 129.6, 130.3 (Phe), 130.4, 130.9
(dppz), 131.0 (Phe), 133.8, 134.3, 139.2, 140.6, 144.3, 151.3, 155.2
of the double helix due to the presence of an additional
[Et4N]+ cation. The schematic depiction of possible simulta-
neous covalent and intercalative binding for 9 violates the
general rule of the neighbour-exclusion principle,[31] that in-
tercalative binding can only occur at every other base pair
(s Ն 2). Neighbour-exclusion violating structures are more
rigid[32] and support for this type of structure is provided
by the pronounced helix distortions evident from the CD
spectra of the interaction mixture 9/DNA (Figure 11b).
The present work, therefore, confirms that bifunctional
complexes of the type [{(η5-Cp*)Ir(dppz)}(µ-peptide-
κS:κN){trans-(PtL2LЈ)}]4+ with flexible bridging peptide li-
gands are capable of simultaneous covalent and intercal-
ative DNA binding. Both the chain length of the peptide
backbone and the position of the labile ligands in the
(dppz), 168.1, 171.1, 172.7, 173.6 (COO, NHCO) ppm. IR: ν =
˜
3436 vs. (NH), 1737, 1675 s (ν CO), 1545, 1499 (δ NH) cm–1.
[(η5-Cp*)Ir(dppz)[H-(Ala)4-Met-OH-κS]](CF3SO3)2 (4): Prepara-
tion as for 3 with H-(Ala)4-Met-OH (86.7 mg, 0.2 mmol). Yield
139.5 mg (52%). C47H56F6IrN9O12S3 (1341.4): calcd. C 42.1, H 4.2,
N 9.4, S 7.2 (%); found C 42.3, H 3.8, N 9.1, S 6.7 (%). FAB MS:
m/z (%) = 1341 (12) [M]+, 1192 (11) [M – OTf]+, 1042 (10) [M – 2
1
OTf]+. H NMR (CD3OD): δ = 1.33 (mm, 9 H, β Ala), 1.49 (m, 3
H, β Ala), 1.79 (s, 3 H, δ Met), 1.84 (s, 15 H, CH3 Cp*), 1.9–2.1
square-planar PtII coordination sphere are shown to exhibit (mm, 4 H, β and γ Met), 3.94, 4.10, 4.21 (3m, 3 H, α Ala), 4.35
(m, 2 H, α Ala und α Met), 8.17 (dd, 2 H), 8.44 (m, 2 H), 8.54 (dd,
2 H), 9.39 (m, 2 H), 10.05 (dd, 2 H, dppz) ppm. 13C NMR
(CD3OD): δ = 9.1 (CH3 Cp*), 17.3 (δ Met), 17.9, 18.1, 18.3, 18.4
(β Ala), 31.4 (β Met), 34.5 (γ Met), 50.4, 50.7, 50.9 (α Ala), 52.6
(α Met), 97.4 (Cp*), 130.8, 131.2, 133.0, 134.3, 138.5, 141.0, 144.6,
151.7, 155.4 (dppz), 167.1, 171.1, 174.9, 175.4, 182.1 (COO,
an important influence on the degree of intercalation and
the extent of helix distortion.
Experimental Section
NHCO) ppm. IR: ν = 3435 s (NH), 1664 (ν CO), 1543, 1499 (δ
˜
General: All manipulations and reactions were performed under
argon in carefully dried solvents using standard Schlenk tech-
niques. FTIR: Perkin–Elmer 1760X as KBr discs. CD: Jasco J-715
in the range 220–400 nm for 1:10 complex/[DNA] mixtures [DNA
concentration in M(nucleotide)] in a 10 m phosphate buffer at pH
= 7.2. FAB MS: Fisons VG Autospec with 3-nitrobenzyl alcohol
NH) cm–1.
[{(η5-Cp*)Ir(dppz)}(µ-H-Gly-Gly-Phe-Met-OH-κS:κN){Pt-
(terpy)}](CF3SO3)4 (5): Ag(CF3SO3) (102.8 mg, 0.4 mmol) was
added to [PtCl(terpy)]Cl·2H2O (107.1 mg, 0.2 mmol) in H2O
(10 mL) and stirred in the dark for 24 h. After removal of precipi-
as the matrix. 1H and 13C NMR spectroscopy: Bruker DRX 400 tated AgCl and addition of 3 (263.7 mg, 0.2 mmol), the clear solu-
with chemical shifts reported as δ values relative to the signal of
the deuterated solvent. 13C NMR signals for the CF3SO3 anions
are observed in the range δ = 121.8–122.8 (q) ppm and are not
listed for individual complexes. Elemental analyses: Vario EL of
tion was stirred at 50 °C for 18 h. The solvent was removed and
the resulting solid redissolved in CH3OH (3 mL). Following pre-
cipitation with diethyl ether, the product was washed and dried in
vacuo. Yield 171.8 (42%). C65H62F12IrN11O17PtS5 (2044.9): calcd.
Elementar Analysensysteme GmbH. IrCl3·xH2O, K2PtCl4, cis- C 38.2, H 3.1, N 7.5, S 7.8; found C 38.8, H 3.4, N 7.0, S 7.8. FAB
[PtCl2(NH3)2] and trans-[PtCl2(NH3)2] were purchased from Chem-
pur, Ag(CF3SO3) and AgNO3 from Acros and peptides, with the
exception of H-(Ala)4-Met-OH, from Bachem. The mentioned
pentapeptide was prepared by solid-state synthesis, [{(η5-Cp*)-
IrCl}2(µ-Cl)2],[33] [PtCl(terpy)]Cl[34] and dipyrido[3,2-a:2Ј,3Ј-c]-
phenazine (dppz)[35] in accordance with literature procedures. Sub-
sequent treatment of an acetone solution of [{(η5-Cp*)IrCl}2(µ-
Cl)2] with 4 equiv. of Ag(CF3SO3) and filtration of the resulting
AgCl precipitate afforded the starting compound [(η5-Cp*)Ir(ace-
MS: m/z (%) = 1895 (5) [M – OTf]+, 1319 (13) [M – 2 OTf –
Pt(terpy)]+, 1019 (12) [M – 4 OTf – Pt(terpy)]+. 1H NMR
(CD3OD): δ = 1.81 (s, 3 H, δ Met), 1.84 (s, 15 H, CH3 Cp*), 1.9–
2.1 (mm, 4 H, β and γ Met), 3.61, 3.68 (2s, 4 H, α Gly1, α Gly2),
4.32 (m, 1 H, α Phe), 4.39 (m, 1 H, α Met), 7.15 (mm, 5 H, Phe),
7.82 (m, 2 H, terpy H5), 8.11 (m, 2 H, dppz), 8.3–8.5 (mm, 11 H,
terpy and dppz), 8.83 (m, 2 H, terpy H6), 9.37 (m, 2 H), 10.03 (dd,
2 H, dppz) ppm. 13C NMR (CD3OD): δ = 8.5 (CH3 Cp*), 17.0 (δ
Met), 30.5 (β Met), 33.8 (γ Met), 37.1 (β Phe), 41.6, 43.5 (α Gly),
52.5 (α Met), 56.4 (α Phe), 97.2 (Cp*), 125.4, 126.7 (terpy), 127.8,
tone)3](CF3SO3)2
(1).[17]
[(η5-Cp*)Ir(dppz)(H-Gly-Met-OH-
Eur. J. Inorg. Chem. 2005, 3840–3849
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3847

S. Gençaslan, W. S. Sheldrick
FULL PAPER
129.5, 130.2 (Phe), 130.4, 131.0, 133.8, 134.0, 139.2, 141.0 (dppz),
[{(η5-Cp*)Ir(dppz)}(µ-H-Gly-Met-OH-κS:κNG1)(trans-{Pt(NH3)-
144.0, 144.2 (terpy), 144.5 (dppz), 148.8 (terpy), 151.6, 155.6 (OH2)2})][Et4N](CF3SO3)5 (9): Preparation as for 7 with
(dppz), 168.3, 170.9 (COO, NHCO) ppm. IR: ν = 3448 s (NH),
Ag(CF3SO3) (154.2 mg, 0.6 mmol) and (Et4N)[PtCl3(NH3)]
(89.7 mg, 0.2 mmol) in H2O (10 mL) as starting compounds. Yield
151.3 mg (39%). C48H66F15IrN8O20PtS6 (1939.7): calcd. C 29.7, H
3.4, N 5.8, S 9.9; found C 29.5, H 2.9, N 5.5, S 9.0. FAB MS: m/z
˜
1664 (ν CO), 1543 (δ NH) cm–1.
[{(η5 -Cp*)Ir(dppz)}(µ-H-(Ala)4 -Met-OH-κS:κNA 1){Pt-
(terpy)}](CF3SO3)4 (6): Preparation as for 5 with 4 (268.3 mg,
0.2 mmol). Yield 140.6 mg (34 %). C64H67F12IrN12O18PtS5
(2067.9): calcd. C 37.2, H 3.3, N 7.8, S 8.1; found C 37.1, H 3.1,
N 8.1, S 8.0. FAB MS: m/z (%) = 1919 (1) [M – OTf]+, 1042 (17)
[M – 4 OTf – Pt(terpy)]+. 1H NMR (CD3OD): δ = 1.32 (mm, 9 H,
β Ala), 1.49 (d, 3 H, β Ala), 1.79 (s, 3 H, δ Met), 1.84 (s, 15 H,
CH3 Cp*), 1.9–2.1 (mm, 4 H, β and γ Met), 3.72, 4.10, 4.21 (3m,
3 H, α Ala), 4.35 (m, 2 H, α Ala und α Met), 7.82 (m, 2 H, terpy
H5), 8.16 (dd, 2 H, dppz), 8.3–8.6 (mm, 11 H, terpy and dppz),
8.88 (m, 2 H, terpy H6), 9.37 (m, 2 H), 10.05 (dd, 2 H, dppz) ppm.
13C NMR (CD3OD): δ = 8.5 (CH3 Cp*), 16.7 (δ Met), 17.6, 17.8,
18.1 (β Ala), 31.4 (β Met), 33.9 (γ Met), 50.2, 50.6, 50.7, 50.8 (α
Ala), 51.8 (α Met), 97.2 (Cp*), 123.4, 125.4 (terpy), 130.7, 131.0,
133.6, 134.0, 139.1, 140.5, 144.5 (dppz), 148.7 (terpy), 151.5 (dppz),
151.7 (terpy), 155.7 (dppz), 156.9, 159.1 (terpy), 170.9, 174.5, 175.1
1
(%) = 1115 (35) [M – 3 OTf – Et4N – Pt(NH3)(OH2)2]+. H NMR
(CD3OD): δ = 1.29 (tt, 12 H, CH3 Et4N), 1.79 (s, 3 H, δ Met), 1.84
(s, 15 H, CH3 Cp*), 1.98 (m, 2 H, β Met), 2.15 (m, 2 H, γ Met),
3.64 (s, 2 H, α Gly), 4.40 (m, 1 H, α Met), 8.18 (dd, 2 H), 8.46 (m,
2 H), 8.54 (dd, 2 H), 9.39 (dd, 2 H), 10.07 (dd, 2 H, dppz) ppm.
13C NMR (CD3OD): δ = 7.8 (CH3 Et4N), 8.7 (CH3 Cp*), 16.9 (δ
Met), 30.3 (β Met), 34.4 (γ Met), 41.8 (α Gly), 52.4 (α Met), 53.5
(CH2 Et4N), 97.5 (Cp*), 130.8, 131.2, 133.8, 134.3, 139.4, 141.0,
144.7, 151.8, 155.8 (dppz), 174.3 (COO) ppm. IR: ν = 3436 vs.
˜
(NH), 1629 (ν CO), 1546 (δ NH) cm–1.
[{(η5-Cp*)Ir(dppz)}(µ-H-Gly-Gly-Met-OH-κS:κNG1)(trans-
{Pt(NH3)(OH2)2})][Et4N](CF3SO3)5 (10): Preparation as for 9 with
[(η5-Cp*)Ir(dppz)(H-Gly-Gly-Met-OH-κS)](CF3SO3)2 (234.2 mg,
0.2 mmol). Yield 171.7 mg (43%). C50H69F15IrN9O21PtS6 (1996.8):
calcd. C 30.1, H 3.5, N 6.3, S 9.6; found C 30.1, H 3.6, N 6.6, S
9.1. FAB MS: m/z (%) = 1172 (15) [M – 3 OTf – Et4N –
Pt(NH3)(OH2)2]+, 1022 (29) [M – 4 OTf – Et4N – Pt(NH3)-
(OH2)2]+. 1H NMR (CD3OD): δ = 1.29 (tt, 12 H, CH3 Et4N), 1.81
(s, 3 H, δ Met), 1.85 (s, 15 H, CH3 Cp*), 1.9–2.1 (mm, 4 H, β and
γ Met), 3.64 (s, 2 H, α Gly1), 3.90 (s, 2 H, α Gly2), 4.39 (br., 1 H,
α Met), 8.18 (dd, 2 H), 8.47 (m, 2 H), 8.53 (dd, 2 H), 9.38 (dd, 2
H), 10.06 (dd, 2 H, dppz) ppm. 13C NMR (CD3OD): δ = 7.9 (CH3
Et4N), 8.8 (CH3 Cp*), 17.2 (δ Met), 30.1 (β Met), 34.1 (γ Met),
41.9, 43.6 (α Gly), 52.1 (α Met), 53.6 (CH2 Et4N), 97.5 (Cp*),
130.9, 131.3, 133.9, 134.4, 139.5, 141.2, 144.8, 151.8, 155.8 (dppz),
(COO, NHCO) ppm. IR: ν 3434 vs. (NH), 1742, 1631 (ν CO), 1528
˜
(δ NH) cm–1.
[{(η5-Cp*)Ir(dppz)}(µ-H-Gly-Met-OH-κS:κNG1)(trans-{Pt-
(NH3)2(DMF)})](CF3SO3)2(NO3)2 (7): AgNO3 (68.0 mg, 0.4 mmol)
was added to trans-[PtCl2(NH3)2] (60.0 mg, 0.2 mmol) in DMF
(10 mL) and stirred in the dark for 24 h. After removal of precipi-
tated AgCl and addition of [(η5-Cp*)Ir(dppz)(H-Gly-Met-OH-
κS)](CF3SO3)2 (222.8 mg, 0.2 mmol), the clear solution was stirred
at room temperature for 18 h. The solvent was removed and the
resulting solid redissolved in CH3OH (3 mL). Following precipi-
tation with diethyl ether, the product was washed and dried in
vacuo. Yield 135.6 mg (44 %). C40H52F6IrN11O16PtS3 (1540.4):
calcd. C 31.2, H 3.4, N 10.0, S 6.2; found C 31.3, H 3.3, N 9.8, S
168.4, 171.7 (COO, NHCO) ppm. IR: ν = 3436 vs. (NH), 1629 (ν
˜
CO), 1499 (δ NH) cm–1.
DNA Binding Studies of 3–10: The thermal denaturation tempera-
ture Tm of 1:10 complex/DNA mixtures was determined in a 10 m
phosphate buffer at pH = 7.2. Melting curves were recorded at
260 nm with a Lambda 15 Perkin–Elmer spectrometer connected
with a temperature controller (Haake FS thermostat). A ramp rate
of 0.25 °C min–1 was used over the range 25–97 °C. Tm values were
calculated by determining the midpoints of the melting curves from
the first-order derivatives. Experimental ∆Tm values are estimated
to be accurate within 1. Concentrations of CT DNA were deter-
mined spectrophotometrically using the molar extinction coeffi-
cient ε260 = 6600 –1 cm–1.[37] All absorption titrations were per-
formed at room temperature. After sonication, buffered solutions
of CT DNA gave a ratio of UV absorbance A260/A280 of ca. 1.90,
indicating that DNA was sufficiently free of protein.[38] 20 µ solu-
tions of the individual metal complexes were treated with DNA
over a range of molarities from 20 to 400 µ (nucleotide). All UV/
Vis spectra were measured after equilibration, i.e. no further
change in monitored absorbance. Titration curves were constructed
from the fractional change in the absorbance as a function of DNA
concentration according to the model by Bard and Thorp[22,23] for
non-cooperative non-specific binding for one type of discrete DNA
binding site.
6.3. FAB MS: m/z (%) = 1043 (41) [M – 2 OTf – 2 NO3 – DMF]+
,
1025 (12) [M – 2 OTf – 2 NO3 – DMF – NH3]+. 1H NMR
(CD3OD): δ = 1.79 (s, 3 H, δ Met), 1.82 (s, 15 H, CH3 Cp*), 1.97
(m, 2 H, β Met), 2.16 (m, 2 H, γ Met), 2.84, 2.98 (2s, 6 H, CH3
DMF), 3.64 (s, 2 H, α Gly), 4.40 (m, 1 H, α Met), 7.96 (s, 1 H, CH
DMF), 8.16 (dd, 2 H), 8.45 (m, 2 H), 8.52 (dd, 2 H), 9.39 (m, 2
H), 10.06 (m, 2 H, dppz) ppm. 13C NMR (CD3OD): δ = 9.1 (CH3
Cp*), 17.1 (δ Met), 31.0 (β Met), 35.1 (γ Met), 37.4 (CH3 DMF),
41.4 (α Gly), 52.5 (α Met), 97.6 (Cp*), 130.9, 131.3, 133.4, 134.1,
138.7, 141.1, 144.7, 155.2 (dppz), 168.4, 171.5 (COO, NHCO) ppm.
IR: ν = 3446 vs. (NH), 1684 s (ν CO), 1636 (ν CO DMF), 1496 (δ
˜
NH) cm–1.
[{(η5-Cp*)Ir(dppz)}(µ-H-Gly-Gly-Met-OH-κS:κNG1)(trans-
{Pt(NH3)2(DMF)})](CF3SO3)2(NO3)2 (8): Preparation as for 7 with
[(η5-Cp*)Ir(dppz)(H-Gly-Gly-Met-OH-κS)](CF3SO3)2 (234.2 mg,
0.2 mmol). Yield 147.0 mg (46%). C42H55F6IrN12O17PtS3 (1597.4):
calcd. C 33.8, H 3.5, N 10.5, S 6.0; found C 33.5, H 3.7, N 10.0, S 5.8.
FAB MS: m/z (%) = 1172 (14) [M – 2 NO3 – DMF – Pt(NH3)2]+
,
1022 (6) [M – OTf – 2 NO3 – DMF – Pt(NH3)2]+. 1H NMR
(CD3OD): δ = 1.79 (s, 3 H, δ Met), 1.84 (s, 15 H, CH3 Cp*), 1.9–
2.1 (mm, 4 H, β and γ Met), 2.85, 2.97 (2s, 6 H, CH3 DMF), 3.62
(s, 2 H, α Gly1), 3.88 (s, 2 H, α Gly2), 4.40 (br., 1 H, α Met), 7.96
(s, 1 H, CH DMF), 8.18 (dd, 2 H), 8.45 (m, 2 H), 8.53 (dd, 2 H),
9.37 (dd, 2 H), 10.06 (dd, 2 H, dppz) ppm. 13C NMR (CD3OD): δ
(εa – εf)/(εb – εf) = (b – {b2 – 2Kb Ct[DNA]/s}1/2)/2KbCt
(1)
2
= 9.1 (CH3 Cp*), 17.6 (δ Met), 30.8 (β Met), 31.9 (CH3 DMF), Equation (1) was used to fit the absorption data by least-squares
35.6 (γ Met), 37.2 (CH3 DMF), 41.8, 43.6 (α Gly), 53.4 (α Met),
97.3 (Cp*), 130.7, 131.1, 133.8, 134.1, 139.5, 140.9, 144.6, 151.8,
155.0 (dppz), 165.1 (CH DMF), 171.7, 172.8 (COO, NHCO) ppm.
refinement of binding constants (Kb) and site sizes (s) with b = 1
+ KbCt + Kb[DNA]/2s, where εa is the extinction coefficient ob-
served at a given DNA concentration, εf the extinction coefficient
of the complex in the absence of DNA, εb the extinction coefficient
IR: ν = 3446 vs. (NH), 1636 (ν CO DMF), 1543 (δ NH) cm–1.
˜
3848
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 3840–3849

Bifunctional Bioorganometallic Iridium()–Platinum() Complexes
FULL PAPER
[13] K. E. Erkkila, D. T. Odom, J. K. Barton, Chem. Rev. 1999, 99,
2777–2795.
of the complex when fully bound to DNA (i.e. no absorption
change on further addition of DNA), Kb the equilibrium constant
in –1, Ct the total metal complex concentration, [DNA] the DNA
concentration in M(nucleotide), and s the binding site size. Values
of εb were obtained by extrapolation from the y intercept of plots
of εa/εf vs. 1/[DNA]. The Kb and s values of Tables 1 and 2 are
those for the best least-squares fits to the individual UV/Vis ti-
tration curves using the program ORIGIN 6.0.
[14] D. Herebian, W. S. Sheldrick, J. Chem. Soc. Dalton Trans. 2002,
966–974.
[15] A. Frodl, D. Herebian, W. S. Sheldrick, J. Chem. Soc. Dalton
Trans. 2002, 3664–3673.
[16] R. Stodt, S. Gençaslan, A. Frodl, C. Schmidt, W. S. Sheldrick,
Inorg. Chim. Acta 2003, 355, 242–253.
[17] C. White, S. J. Thompson, P. M. Maitlis, J. Organomet. Chem.
1977, 127, 415–421.
X-ray Structural Analysis of 2: [(η5-Cp*)Ir(CF3COO)(dppz)]-
[18] G. Schwarz, S. Klose, W. Balthasar, Eur. J. Biochem. 1970, 12,
¯
(CF3SO3), C31H25F6IrN4O5S, M = 871.8, triclinic, space group P1
454–460.
(no. 2), a = 8.701(2), b = 11.535(2), c = 16.603(3) Å, α = 103.60(3),
[19] K. W. Jennette, J. T. Gill, J. A. Sadownick, S. J. Lippard, J. Am.
Chem. Soc. 1976, 98, 6159–6168.
β = 90.93(3), γ = 102.21(3)°, V = 1579.2(5) Å3, Z = 2, Dcalcd.
=
1.833 g·cm–3, µ(Mo-Kα) = 4.376 mm–1; Siemens P4 diffractometer;
graphite-monochromated Mo-Kα radiation (λ = 0.71073 Å). Crys-
tal size: 0.46×0.30×0.12 mm; ω-scan in the range 2θ Յ 50° (–10
Յ h Յ 10, 0 Յ k Յ 13, –19 Յ l Յ 19), 5683 reflections collected,
5406 independent reflections (Rint = 0.051); semiempirical absorp-
tion corrections with ψ-scans; max./min. transmission 0.716/0.373;
428 parameters refined; R1 = 0.057 for 4326 reflections with I Ͼ
[20] W. S. Sheldrick, S. Heeb, J. Organomet. Chem. 1989, 377, 357–
366.
[21] W. S. Sheldrick, A. Gleichmann, J. Organomet. Chem. 1994,
470, 183–187.
[22] M. T. Carter, M. Rodriguez, A. J. Bard, J. Am. Chem. Soc.
1989, 111, 8901–8911.
[23]
[24]
[25]
[26]
S. R. Smith, G. A. Neyhart, W. A. Karlsbeck, H. H. Thorp,
New J. Chem. 1994, 18, 397–406.
M. Hahn, M. Kleine, W. S. Sheldrick, J. Biol. Inorg. Chem.
2001, 6, 556–566.
M. Kleine, D. Wolters, W. S. Sheldrick, J. Inorg. Biochem. 2003,
97, 354–363.
2
2 2
2σ(I), wR2 = {[Σw(Fo – Fc ) ]/[Σw(Fo)2]}1/2 = 0.136 for all data;
min./max. residual electron density 1.82/–1.83 e·Å–3. Anisotropic
temperature factors for all non-hydrogen atoms with the exception
of the disordered F atoms of the CF3COO– ligand, which exhibit
s.o.f. values of 0.8 and 0.2.[39]
L. I. Elding, O. Gröning, Inorg. Chem. 1978, 7, 1872–1880.
[27] S. Gençaslan, Dissertation, Ruhr-Universität Bochum, 2004.
[28] V. Brabec, V. Kleinwächter, J.-L. Butour, N. P. Johnson, Bio-
phys. Chem. 1990, 35, 129–141.
[29] D. M. Gray, in: Circular Dichroism and the Conformational
Analysis of Biomolecules (Ed.: G. D. Fasman), Plenum Press,
New York, 1996, pp. 469–500.
[30] J.-P. Macquet, J.-L. Butour, Eur. J. Biochem. 1978, 83, 375–387.
[31] D. M. Crothers, Biopolymers 1968, 6, 575–584.
[32] S. N. Rao, P. A. Kollman, Proc. Nat. Acad. Sci. USA 1987, 84,
5735–5739.
[1] N. Farrell, Y. Qu, U. Bierbach, M. Valsecchi, E. Mertz, in:
Cisplatin: Chemistry and Biochemistry of a Leading Anticancer
Drug (Ed.: B. Lippert), Wiley-VCH, Weinheim, 1999, pp. 479–
496.
[2] N. Farrell, in: Platinum-Based Drugs in Cancer Therapy (Eds.:
L. R. Kelland, N. Farrell), Humana Press, Totowa, 2000, pp:
321–338.
[3] N. J. Wheate, J. G. Collins, Coord. Chem. Rev. 2003, 241, 133–
145.
[4] Y. Qu, N. Farrell, Inorg. Chem. 1995, 34, 3573–3576.
[5] K. van der Schilden, F. Garcia, H. Kooijman, A. L. Spek, J. G.
Haasnoot, J. Reedijk, Angew. Chem. Int. Ed. 2004, 43, 5786–
5788.
[6] B. E. Bowler, K. J. Ahmed, W. I. Sundquist, L. S. Hollis, E. E.
Whang, S. J. Lippard, J. Am. Chem. Soc. 1989, 111, 1299–1306.
[7] M. D. Temple, W. D. McFadyen, R. J. Holmes, W. A. Denny,
V. Murray, Biochemistry 2000, 39, 5593–5599.
[8] L. C. Perrin, C. Cullinane, W. D. McFadyen, D. R. Phillips,
Anti-Cancer Drug Des. 1999, 14, 243–252.
[9] L. C. Perrin, P. D. Prenzler, C. Cullinane, D. R. Phillips, W. A.
Denny, W. D. McFadyen, J. Inorg. Biochem. 2000, 81, 111–117.
[10] W. D. McFadyen, L. P. G. Wakelin, I. A. G. Roos, B. L. Hill-
coat, Biochem. J. 1986, 238, 757–763.
[33] a) J. W. Kang, K. Moseley, P. M. Maitlis, J. Am. Chem. Soc.
1969, 91, 5970–5977; b) B. L. Booth, R. N. Haszeldine, M. Hill,
J. Chem. Soc. A 1969, 1299–1303.
[34] G. Annibale, M. Brandolisio, B. Pitteri, Polyhedron 1995, 3,
451–453.
[35] M. Yamada, Y. Tanaka, Y. Yoshimoto, Bull. Chem. Soc. Jpn.
1992, 65, 1006–1011.
[36] M. J. Abrams, C. M. Giandomenico, J. F. Vollano, D. A.
Schwartz, Inorg. Chim. Acta 1987, 131, 3–4.
[37] H.-Q. Liu, T.-C. Cheung, S.-M. Peng, C.-M. Che, J. Chem. Soc.
Chem. Commun. 1995, 17, 1787–1788.
[38] J. Marmur, J. Mol. Biol. 1961, 3, 208–218.
[39] CCDC-266154 (2) contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[11] H. Kurosaki, N. Yamakawa, M. Sumimoto, K. Kimura, M.
Goto, Bioorg. Med. Chem. Lett. 2003, 13, 825–828.
[12] B. Önfelt, P. Lincoln, B. Norden, J. Am. Chem. Soc. 1999, 121,
10846–10847.
Received: March 16, 2005
Published Online: August 4, 2005
Eur. J. Inorg. Chem. 2005, 3840–3849
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3849