Photobiological activity of Ru(ll) dyads based on (Pyren-1 -yl)ethynyl derivatives of 1, 10-Phenanthroline

Article abstract of DOI:10.1021/ic902427r

Several mononuclear Ru(ll) dyads possessing 1, 10-phenanthrollne-appended pyrenylethynylene llgands were synthesized, characterized, and evaluated for their potential in photobiologlcal applications such as photodynamlc therapy (PDT). These complexes interact with DNA via intercalation and photocleave DNA in vitro at submlcromolar concentrations when irradiated with visible light (λ_irr ≥ 400 nm). Such properties are remarkably sensitive to the position of the ethynylpyrenyl substituent on the 1, 10-phenanthroline ring, with 3-substitutlon showing the strongest binding under all conditions and causing the most deleterious DNA damage. Both dyads photocleave DNA under hypoxic conditions, and this photoactlvity translates well to cytotoxicity and photocytotoxlcity models using human leukemia cells, where the 5- and 3-substituted dyads show photocytotoxiclty at 5-10μM and 10-20μM, respectively, with minimal, or essentially no, dark toxicity at these concentrations. This lack of dark cytotoxicity at concentrations where significant photoactlvity Is observed emphasizes that agents with strong intercalating units, previously thought to be too toxic for phototherapeutic applications, should not be excluded from the arsenal of potential photochemotherapeutic agents under investigation.

Full text of DOI:10.1021/ic902427r

Inorg. Chem. 2010, 49, 2889–2900 2889
DOI: 10.1021/ic902427r
Photobiological Activity of Ru(II) Dyads Based on (Pyren-1-yl)ethynyl Derivatives
of 1,10-Phenanthroline
†
†
‡
†
‡
,‡
Susan Monro, John Scott, Abdellatif Chouai, Richard Lincoln, Ruifa Zong, Randolph P. Thummel,*
,
†
and Sherri A. McFarland*
†
‡
Department of Chemistry, Acadia University, Wolfville, NS B4P 2R6, Canada and Department of Chemistry,
University of Houston, Houston, Texas 77204-5003
Received December 7, 2009
Several mononuclear Ru(II) dyads possessing 1,10-phenanthroline-appended pyrenylethynylene ligands were
synthesized, characterized, and evaluated for their potential in photobiological applications such as photodynamic
therapy (PDT). These complexes interact with DNA via intercalation and photocleave DNA in vitro at submicromolar
concentrations when irradiated with visible light (λ g 400 nm). Such properties are remarkably sensitive to the
irr
position of the ethynylpyrenyl substituent on the 1,10-phenanthroline ring, with 3-substitution showing the strongest
binding under all conditions and causing the most deleterious DNA damage. Both dyads photocleave DNA under
hypoxic conditions, and this photoactivity translates well to cytotoxicity and photocytotoxicity models using human
leukemia cells, where the 5- and 3-substituted dyads show photocytotoxicity at 5-10 μM and 10-20 μM, respectively,
with minimal, or essentially no, dark toxicity at these concentrations. This lack of dark cytotoxicity at concentrations
where significant photoactivity is observed emphasizes that agents with strong intercalating units, previously thought to
be too toxic for phototherapeutic applications, should not be excluded from the arsenal of potential photochemother-
apeutic agents under investigation.
1
. Introduction
The planar appendage provides a handle for intercalative
2-6
binding to DNA,
positioning the complex for enhanced
Traditionally, dyad-type molecules have piqued curiosity
reactivity toward DNA. Increased photoreactivity toward
DNA has been observed for Ru(II) complexes containing the
owing to their potential in photonic-based applications.
Specifically, several dyads assembled from Ru(II) poly-
pyridyl complexes have been shown to display some of the
longest lifetimes (>150 μs) observed for ruthenium-based
coordination complexes due to excited-state equilibrium
that is established between the two components of the
dyad, namely, the Ru(II) diimine complex and a pendant
0
0
intercalating dppz (dipyrido[3,2-a:2 ,3 -c]phenazine) ligand
3
and its derivatives and more recently for dirhodium(II)
complexes that incorporate the dppz ligand, which photo-
cleave DNA and exhibit photocytotoxicity toward certain
7
cancer cell lines. Moreover, the dyad motif positions the
1
lowest-energy π*-acceptor ligand at a relatively long distance
polycyclic aromatic hydrocarbon. The synergistic commu-
from the Ru(II) core, which profoundly increases the charge-
nication between these constituent components, while inter-
esting from a theoretical perspective, may prove to have far-
reaching consequences not only in applications such as
supramolecular chemistry but also in the area of photobio-
logy, particularly in the design of phototherapeutic agents.
3
separation that can be achieved in conventional MLCT
excited states. This increased charge-separation may lead to
longer excited-state lifetimes and reactive excited states that
are capable of oxidizing DNA without the need for an
8
external electron acceptor. Such states are of interest in the
development of photosensitizers that absorb low-energy light
and function in the absence of oxygen. The ability to damage
DNA under hypoxic conditions with photoactivation by
visible light would be a first-step toward the development
of ruthenium-based drugs for use in photobiological applica-
tions such as photodynamic therapy (PDT).
*
To whom correspondence should be addressed. E-mail: sherri.mcfarland@
acadiau.ca (S.A.M.), thummel@uh.edu (R.P.T.).
1) McClenaghan, N. D.; Leydet, Y.; Maubert, B.; Indelli, M. T.;
Campagna, S. Coord. Chem. Rev. 2005, 249, 1336–1350.
2) Zeglis, B. M.; Pierre, V. C.; Barton, J. K. Chem. Commun. 2007, 4565–
579.
3) Delaney, S.; Pascaly, M.; Bhattacharya, P. K.; Han, K.; Barton, J. K.
Inorg. Chem. 2002, 41, 1966–1974.
4) Erkkila, K. E.; Odom, D. T.; Barton, J. K. Chem. Rev. 1999, 99, 2777–
796.
(
(
4
(
(
2
(7) Angeles-Boza, A. M.; Bradley, P. M.; Fu, P.; Wicke, S. E.; Bacsa, J.;
Dunbar, K. R.; Turro, C. Inorg. Chem. 2004, 43, 8510–8519.
(8) Stemp, E.; Arkin, M. R.; Barton, J. K. J. Am. Chem. Soc. 1997, 119,
2921–2925.
(
(
5) Holmlin, R. E.; Stemp, E. D.; Barton, J. K. Inorg. Chem. 1998, 37, 29–34.
6) Jenkins, Y.; Friedman, A. E.; Turro, N. J.; Barton, J. K. Biochemistry
1
992, 31, 10809–10816.
r 2010 American Chemical Society
Published on Web 02/10/2010
pubs.acs.org/IC

2
890 Inorganic Chemistry, Vol. 49, No. 6, 2010
Monro et al.
Chart 1. Diimine Ligands Linked to Pyrene
Diimine ligands linked directly to pyrene have been prepared,
and the photophysics of their Ru(II) complexes have been
Chart 2. Ruthenium(II) Complexes of Ethynylpyrene-Linked Diimine
Ligands
9,10
11
12
studied carefully by Schmehl,
Castellano, Schl u€ ter and
0
co-workers. When the pyrene is directly linked to 2,2 -bipyridine
(bpy) or 1,10-phenanthroline (phen), conformational inter-
action of hydrogens near the linkage cause non-planarity of
the system that interferes substantially with electronic commu-
nication between the two halves of the dyad (1; Chart 1). This
problem can be alleviated by using an ethynyl bridge in place of
13-18
the direct covalent C-C linkage.
This ethynyl bridge
maintains the rigidity and linearity of a C-C bond while
moving the dyad units apart such that conformational inter-
actions assume a less important role. Additionally, the cylin-
drical symmetry of the triple bond provides good conjugative
interaction between the ligand and the pyrene residue regardless
of the relative dihedral orientation of their respective planar
π-networks. This situation is illustrated for 1,10-phenanthroline
11,19
substituted at the 5-position with either a 1-pyrenyl (1)
pyren-1-yl)ethynyl (2) group. We have recently prepared dyad
ligands 2 and 3 and their corresponding heteroleptic complexes
or a
trishydroxymethylaminomethane (Tris base), Tris/HCl, try-
pan blue (0.4% w/v, sterile-filtered), and sodium chloride
were purchased from Sigma-Aldrich and used as received
without further purification. Distilled, deionized water
(
2
þ 2þ
Ru(2)(bpy-d ) ] (7) and [Ru(3)(bpy-d8)2] (8). In this paper
8 2
[
(
2
ddH O) was obtained from a Millipore system. Calf-thymus
we discuss the synthesis and characterization of these systems,
including their photophysical and electrochemical properties, as
well as their interactions with DNA. We also explore the
cytotoxicity and photocytotoxicity of 7 and 8 toward human
leukemia cells (Chart 2).
DNA (activated type XV) was purchased from Sigma and
reconstituted with 40 mM MOPS (pH 7.5) overnight at room
temperature following sonication. The concentration of the
resulting DNA solution was calculated from its absorbance
-
1
-1
(A260) using ɛ = 12824 M cm (base pairs), and purity was
estimated by relative absorption at 260 and 280 nm (A260/A280
∼1.8). Plasmid pUC19 DNA was purchased from New England
BioLabs and transformed using NovaBlue Singles Competent
Cells purchased from Novagen. Transformed pUC19 was
purified using the QIAprep Spin Miniprep Kit purchased from
Qiagen (yield ∼62 μg of plasmid DNA per 20 mL culture).
Supercoiled pDesR3 plasmid was a derivative of pET28a
2
. Experimental Procedures
.1. Materials. Acetic acid, acridine orange, agarose, Optima-
grade acetonitrile, ethylenediaminetetraacetic acid (EDTA), ethi-
2
-
1
dium bromide, ethanol, gentamicin (50 mg mL , in sterile-filtered
water), Hoechst 33258, 4-morpholinepropanesulfonic acid (MOPS),
2
0
containing the desR gene and was used in the present study
2
1
(
9) Del Guerzo, A.; Leroy, S.; Fages, F.; Schmehl, R. H. Inorg. Chem.
002, 41, 359–366.
10) Simon, J. A.; Curry, S. L.; Schmehl, R. H.; Schatz, T. R.; Piotrowiak,
P.; Jin, X.; Thummel, R. P. J. Am. Chem. Soc. 1997, 119, 11012–11022.
11) Tyson, D. S.; Bialecki, J.; Castellano, F. N. Chem. Commun. 2000,
355–2356.
12) Beinhoff, M.; Weigel, W.; Jurczok, M.; Rettig, W.; Modrakowski,
C.; Br u€ dgam, I.; Hartl, H.; Schl u€ ter, A. D. Eur. J. Org. Chem. 2001, 3819–
because of its increased GC content. Characterized fetal
bovine serum (FBS) and Iscove’s Modified Dulbecco’s Med-
ium (IMDM) supplemented with 4 mM L-glutamine were
purchased from Fisher Scientific, and human promyelocytic
leukemia cells (HL60) were supplied by the American Type
Culture Collection (ATCC CCL-240). Prior to use FBS was
divided into 40 mL aliquots that were heat inactivated (30 min,
2
(
(
2
(
3
829.
5
5 °C) and subsequently stored at -10 °C. Acetonitrile used for
(
13) Harriman, A.; Hissler, M.; Khatyr, A.; Ziessel, R. Chem. Commun.
cyclic voltammetry experiments was dried by reflux over CaH and
2
distilled under argon prior to use. 3-Bromo-1,10-phenanthroline,
1
999, 735–736.
22,23
(
14) Kozlov, D. V.; Tyson, D. S.; Goze, C.; Ziessel, R.; Castellano, F. N.
Inorg. Chem. 2004, 43, 6083–6092.
15) Goze, C.; Kozlov, D. V.; Tyson, D. S.; Ziessel, R.; Castellano, F. N.
New J. Chem. 2003, 27, 1679–1683.
16) Goze, C.; Kozlov, D. V.; Castellano, F. N.; Suffert, J.; R. Ziessel, R.
Tetrahedron Lett. 2003, 44, 8713–8716.
(
(20) Xue, Y.; Zhao, L.; Liu, H.; Sherman, D. H. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 12111–12116.
(
(21) S ꢀa ez Dı
´
az, R. I.; Regourd, J.; Santacroce, P. V.; Davis, J. T.;
(
17) Khatyr, A.; R. Ziessel, R. Tetrahedron Lett. 2002, 43, 7431–7434.
18) Hissler, M.; Harriman, A.; Khatyr, A.; Ziessel, R. Chem. Eur. J.
Jakeman, D. L.; Thompson, A. Chem. Commun. 2007, 2701–2703.
(22) Tzalis, D.; Tor, Y.; Failia, S.; Siegel, J. S. Tetrahedron Lett. 1995, 36,
3489–3490.
(23) Connors, P. J.; Tzalis, D.; Dunnick, A. L.; Tor, Y. Inorg. Chem. 1998,
37, 1121–1123.
(
1
999, 5, 3366–3381.
19) Tyson, D. S.; Henbest, K. B.; Bialecki, J.; Castellano, F. N. J. Phys.
Chem. A 2001, 105, 8154–8161.
(

Article
-bromo-1,10-phenanthroline,
Inorganic Chemistry, Vol. 49, No. 6, 2010 2891
24,25
27
2
O
13
4.5 Hz); C NMR could not be obtained because of poor
5
and [Ru(bpy-d
8
)
Cl ]-2H
2 2
were prepared according to known procedures. 1-Ethynylpyrene
was purchased from Aldrich and used without further purification.
solubility; MS: m/z 405.4 (100%, (M þ 1); Anal. Calcd for
30 16 2 2
C H N 0.5 H O: C, 86.08; H, 3.89; N, 6.65. Found: C, 86.29;
3
2
.2. Instrumentation. NMR spectra were recorded at 300
H, 3.54; N, 6.71.
[Ru(2)(bpy-d ) ](PF ) (7). A mixture of 2 (40.5 mg, 0.100
1
MHz for H and referenced to tetramethylsilane (TMS) in
CDCl and to the residual solvent peak in CD CN. Electronic
8
2
6 2
3
3
mmol) and cis-[Ru(bpy-d
absolute EtOH (40 mL) was refluxed under Ar for 24 h. After
cooling, excess NH PF (aq) was added, and the mixture was
stirred for 20 min. The solvent was evaporated, and the crude
2
product was chromatographed on SiO , eluting with CH Cl /
8 2 2 2
) Cl ]-2H O (59.0 mg, 0.110 mmol) in
spectra were measured on Jasco V-530 and Perkin-Elmer
Lambda 3B spectrophotometers. The Jasco V-530 spectro-
photometer was equipped with a ETC-505T Peltier temperature
controller for DNA melting experiments. Fluorescence spectra
were obtained on PTI Quantamaster and Perkin-Elmer LS-50
luminescence spectrometers. The PTI Quantamaster was
equipped with a standard R928 PMT for measuring conven-
tional emission (< 900 nm) and a Hamamatsu R5509-42 NIR
PMT for measuring near infrared (NIR) emission (< 1400 nm).
Cyclic voltammetry was carried out in a conventional three-
electrode cell with a BAS-27 potentiostat and a Houston
Instruments model 100 X-Y recorder according to a previously
4
6
2
2
MeOH (99:1) to afford 7 as an orange solid (74 mg, 66%), mp >
1
280 °C: H NMR (CD CN): δ 9.11 (d, H4, J = 8.4 Hz), 8.73 (d,
3
0
0
H2 or H10 , J = 9 Hz), 8.60 (s, H6), 8.59 (d, H7, J = 7.5 Hz),
8.4-8.03 (overlapping m, 10 H), 7.88 (dd, H3 or H8, J = 8.4, 5.1
Hz), 7.77 (dd, H3 or H8, J = 8.1, 5.1 Hz); MS: m/z 416.6 (100%,
(M-2PF )/2).
6
[Ru(3)(bpy-d ) ](PF ) (8). A mixture of 3 (40.5 mg, 0.100
8
2
6 2
mmol) and cis-[Ru(bpy-d ) Cl ]-2H O (59.0 mg, 0.110 mmol) in
2
8
2
2
27
described procedure. Mass spectra were obtained on a Hewlett-
Packard 5989B mass spectrometer (59987A electrospray) using
atmospheric pressure ionization at 160 °C. Elemental analyses
were performed by National Chemical Consulting, P.O. Box 99,
Tenafly, NJ 07670. Melting points were measured on a Hoover
capillary melting point apparatus and are not corrected. DNA
photocleavage experiments were carried out in a Luzchem LZC-
absolute EtOH (35 mL) was refluxed under Ar for 20 h. After
cooling, excess NH PF (aq) was added, and the mixture was
4
6
stirred for 20 min. The solvent was evaporated, and the crude
product was chromatographed on SiO , eluting with CH Cl /
MeOH (99:1) to afford 8 as a reddish solid (75 mg, 67%), mp
2
2
2
1
0
>280 °C: H NMR (CD CN): δ 8.91 (s, H4), 8.64 (d, H2 or
3
0
H10 , J = 8.1 Hz), 8.52 (d, H7, J = 9 Hz), 8.41-8.10 (over-
lapping m, 12 H), 7.76 (dd, H8, J 8.1, 5.1 Hz); MS: m/z 417
4
X photoreactor equipped with both side- and top-illumination
with LZC-420 bulbs (exposure standards available from
Luzchem). Ethidium bromide-stained agarose gels were imaged
using the Gel Doc-It imaging system from UVP or with a Kodak
EasyShare C330 digital camera with the GNU Image Mani-
pulation Program (GIMP). Viscosity measurements were made
using the model E120 50 semimicro viscometer purchased from
Cannon-Manning. Cellular counting and imaging was carried
out using a Nikon Eclipse TE2000-U inverted light microscope
in phase-contrast mode or epi-fluorescence mode (G-2A epi-
fluorescence filter block). Manual cell counting was done using a
(100%, (M- 2PF )/2).
6
2
.4. Methods. 2.4.1. DNA Binding by UV-vis. Optical
titrations were carried out on 0.5-2 mL solutions of the dyads
with increasing amounts of calf thymus or herring sperm DNA
to give [DNA bases]/[Ru] between 0.1 and 10. DNA was added
in 1-5 μL increments to solutions of compound (10 μM) in
1
0 mM MOPS, 10 mM MOPS with 50 mM NaCl, 5 mM Tris-
HCl, or 5 mM Tris-HCl with 50 mM NaCl at pH 7.5. The
dilution of metal complex concentration at the end of each
titration, although negligible, was accounted for in the bind-
ing constant analyses. The DNA binding constant (K ) was
b
obtained from fits of the titration data to eq 1,
Neubauer hemocytometer (Hausser Scientific).
0
28,29
2.3. Synthesis and Characterization. 5-(Pyren-1 -yl)ethynyl-1,
where b =
1
1
(
0-phenanthroline (2). To a pressure tube were added 5-bromo-
,10-phenanthroline (5, 250 mg, 0.97 mmol), 1-ethynylpyrene
4, 328 mg, 1.45 mmol), [Pd(PPh ](167 mg, 0.144 mmol), and
n-propylamine (40 mL). Ar was bubbled through the solution for
0 min, and the vessel then sealed with a Teflon stopcock and heated
at 80 °C for 2 days. After cooling, the precipitate was filtered, washed
with H O (10 mL), MeOH (10 mL) and CH Cl (10 mL), dried and
chromatographed on SiO , eluting with CH Cl /MeOH (99:1) to
give 2 (285 mg, 73%) as a yellow solid, mp 133-135 °C; H NMR
CDCl ):δ9.31 (dd, H9, J= 4.2, 1.5, Hz), 9.25 (dd, H2, J=4.2, 1.5
Hz), 9.09 (dd, H4, J = 8.1, 1.5 Hz), 8.77 (d, H2 or H10 , J = 9 Hz),
.37-8.05 (overlapping m, 10 H), 7.84 (dd, H3 or H8, J = 8.1, 4.2
1 þ K C þ K [DNA] /2s, C and [DNA] represent the total
b
t
b
t
t
t
dyad and DNA concentrations, respectively, s is the binding site
size, and ɛ , ɛ , and ɛ represent the molar extinction coefficients
of the apparent, free, and bound metal complexes, respectively.
3 4
)
a
f
b
1
ɛf was calculated at 414 nm for 7 and 412 nm for 8 before the
addition of DNA, and ɛ was determined at these wavelengths
a
2
2
2
after each addition of DNA. The value of ɛ was determined
b
2
2
2
from the plateau of the DNA titration, where addition of DNA
did not result in any further decrease in the absorption signal.
Detailed fits of the titration data were obtained using both
Kaleidagraph and Gnuplot.
1
(
3
0
0
8
13
Hz), 7.71, (dd, H3 or H8, J = 8.1, 4.5 Hz); C NMR could not be
obtained because of poor solubility; MS: m/z 405.3 (100%, (M þ 1);
2
2
b
1=2
Ea - Ef
Eb - Ef
b -ðb -2K Ct½DNAꢀ =sÞ
t
¼
ð1Þ
2KbCt
16 2 2
Anal. Calcd for C30H N 0.20H O: C, 87.17; H, 4.12; N, 6.80.
3
Found: C, 87.48; H, 3.37; N, 6.59.
0
DNA melting curves were constructed by measuring the
absorbance (A260) of a 2 mL, 25 μM DNA solution (40 mM
MOPS, pH 7.5) as a function of temperature (20-100 °C) in the
absence and presence of compound (5 μM). Solutions of DNA
and compound for melting experiments were allowed to equili-
brate for 30 min at 25 °C prior to measurement. The ETC-505T
temperature controller was cooled with ice water (4 °C) using a
fish-aquarium pump, and a stream of argon gas was supplied via
the gas inlet valve to the sample compartment to prevent
condensation on the cuvette windows during variable-tempera-
ture experiments.
3-(Pyren-1 -yl)ethynyl-1,10-phenanthroline (3). Following the
procedure described for 2, a mixture of 3-bromo-1,10-phenan-
2
2,23
throline
0
(6, 100 mg, 0.39 mmol), 1-ethynylpyrene (4, 131 mg,
](67 mg, 0.058 mmol), and n-propyl-
.58 mmol), [Pd(PPh )
3 4
amine (15 mL) was heated at 80 °C for 2 days to give 3 (125 mg,
1
7
9
9%) as a brownish solid, mp 168-172 °C: H NMR (CDCl
.45 (d, H2, J = 1.8 Hz), 9.24 (dd, H9, J = 4.2, 1.8 Hz), 8.71 (d,
3
): δ
0
0
H2 or H10 , J = 9 Hz), 8.54 (d, H4, J = 1.8 Hz), 8.30-8.03
overlapping m, 9 H), 7.82 (s, H5 and H6), 7.66 (dd, H8, J = 8.1,
(
(
24) Hissler, M.; Connick, W. B.; Geiger, D. K.; McGarrah, J. E.; Lipa,
D.; Lachicotte, R. J.; Eisenberg, R. Inorg. Chem. 2000, 39, 447–457.
25) Mlochowski, J. Roczniki Chem., Ann. Soc. Chim. Polonorum 1974,
8, 2145.
(
(28) Carter, M. T.; Rodriguez, M.; Bard, A. J. J. Am. Chem. Soc. 1989,
4
111, 8901–8911.
(
26) Ziessel, R.; Stroh, C. Tetrahedron Lett. 2004, 45, 4051–4055.
27) Goulle, V.; Thummel, R. P. Inorg. Chem. 1990, 29, 1767–1772.
(29) Kalsbeck, W. A.; Thorp, H. H. J. Am. Chem. Soc. 1993, 115, 7146–
(
7151.

2
892 Inorganic Chemistry, Vol. 49, No. 6, 2010
.4.2. DNA Binding by Viscosity. Relative viscosity measure-
Monro et al.
2
aliquots) to 96-well tissue-culture plates (Corning Costar,
Acton, MA) containing culture medium either with or without
varying concentrations of complex to give final volumes in the
culture wells of 100 μL and 10,000-20,000 cells. Solutions of
complex in complete media were prepared in 1 mL portions,
where acetonitrile from the initial complex stock solution was
<0.5% v/v, and sterile-filtered in 3 mL syringes equipped with
0.22 μm Nalgene filters. Plates were incubated at 37 °C under
ments were made by charging the Cannon-Manning semimicro
viscometer with varying concentrations of compound (7.5-
45 μM) dissolved in 10 mM MOPS (pH 7.5) containing 150 μM
(
base pairs) calf thymus DNA, giving [compound]/[DNA bp]
ratios of 0.05-0.3. An identical solution without the sample
compound was used to reference each measurement. Trials were
performed in triplicate, and the viscometer was thoroughly
rinsed with ddH
2
O and 10 mM MOPS between samples fol-
2
5% CO for 30 min prior to exposure to 420 nm light in a
lowed by removal of all trace droplets under reduced pressure.
Samples were loaded, following equilibration for 30 min at 25 °C
prior to measurement, with the viscometer in an inverted posi-
tion by immersing tube N in the sample while applying suction
to tube L with a 3-way safety bulb. When the sample approached
mark G, the viscometer was returned to an upright position and
secured in its commercial holder at 25 °C. The efflux time of the
sample was measured as the time it took for the meniscus to
move between the timing notches, mark E to mark F. Relative
Luzchem photoreactor for 30 min; dark controls were incubated
under identical conditions in the dark. Dark controls, or cyto-
toxicity (CT) assays, refer to assays that include metal complex
but were not exposed to light, and light controls refer to light-
exposed assays that did not contain metal complex. Photo-
cytotoxicity (PCT) assays contained metal complex and were
exposed to light. Cell counts and viability staining were carried
out immediately and at ∼24 h following light exposure. Manual
counts were performed on 25 μL 1:1 mixtures of assay culture
and trypan blue solution in a Neubauer hemocytometer viewed
under an inverted light microscope in phase-contrast mode (4X
objective, 60X total magnification). Under these conditions,
viable cells appeared bright white, and non-viable cells were
blue. All experiments were carried out in triplicate, and the
graphed data is the average of three trials.
1
/3
viscosity for a given sample was calculated as (η/ηo) , where η is
the efflux time of the sample solution with DNA and η is the
efflux time of a solution containing only DNA. Ethidium
o
bromide and Hoechst 33258 were used as standards.
2.4.3. Photocleavage Titrations. DNA photocleavage experi-
ments were performed according to a general plasmid DNA
3
0,31
assay
microfuge tubes containing transformed pUC19 plasmid
200 ng, >95% Form I) in 10 mM MOPS buffer and 100 mM
with 20 μL total sample volumes in 0.5 or 1.5 mL
2.4.6. Viability Staining. Viability was established according
whereby a 100X stock solution of
3
2,33
to a published protocol,
ethidium bromide/acridine orange (EB/AO) was prepared by
dissolving ethidium bromide (50 mg) and acridine orange
(
NaCl, pH 7.4. DNA (1-5 μL) was delivered to the assay tubes as
a solution in 10 mM Tris-Cl (pH 8.5) and diluted with MOPS
2
(15 mg) in 95% ethanol (1 mL) and diluting 1/50 with ddH O.
(pH 7.5, final concentration 10 mM) and NaCl (final concentra-
tion 100 mM). Solutions of complexes 7 or 8 were added to give
the desired concentration, and the reaction mixtures were
diluted to a final volume of 20 μL, when necessary, with ddH O.
2
Complexes were dissolved initially in acetonitrile (2 μM stock
The 100X solution was divided into 1 mL aliquots and stored at
-20 °C. A 1X working solution was made by thawing a 1 mL
aliquot of the 100X stock solution and diluting 1/100 with
phosphate-buffered saline. The working solution was stored in
an amber bottle at 4 °C for up to 1 month. For cellular viability
6
solutions), and all subsequent dilutions were made with ddH O
staining, an aliquot of cell suspension was adjusted to 1-5 ꢁ 10
2
-
1
where final assay tubes contained <1% acetonitrile. For con-
centration-based assays, samples (no pre-incubation period)
were irradiated in air for 30 min with 420 nm light inside a
photoreactor (Luzchem LZC-4X). Where irradiation of deox-
ygenated samples was required, argon was bubbled through the
solutions for 15 min prior to irradiation under a positive
pressure of argon. All samples were quenched by the addition
of gel loading buffer (4 μL), loaded onto 1% agarose gels
cells mL in phosphate-buffered IMDM. A 25 μL aliquot of
this cell suspension was mixed with 1X EB/AO staining solution
(25 μL) in a microfuge tube; a 25 μL aliquot of this cell-stain
mixture was transferred to a hemocytometer and viewed under a
Nikon Eclipse TE2000-U inverted light microscope operating in
epi-fluorescence mode (10X or 40X objective, 150X or 600X
total magnification). Under these conditions, viable cells took
up AO and excluded EB, resulting in only green fluorescence
with UV-excitation. Non-viable cells (apoptotic or necrotic)
assimilated EB and fluoresced red with green excitation, over-
whelming any green fluorescence from AO. Apoptotic cells were
discerned from the formation of smaller, apoptotic bodies that
fluoresced red.
-
1
containing ethidium bromide (0.75 μg mL ), and electro-
phoresed for 30 min at 8-12 V cm- in 1X TAE (40 mM Tris-
acetate, 1 mM EDTA, pH 8.2). The bands were visualized with
UV-transillumination (UVP transilluminator) and quantified
using the Gel Doc-It Imaging system (UVP) or GNU Image
Manipulation Program (GIMP).
1
2.4.4. Cell Culture. HL-60 human promyelocytic leukemia
3. Results and Discussion
cells (ATCC CCL-240) were cultured at 37 °C under 5% CO in
2
-
1
3.1. Synthesis and Characterization. Following a modi-
Hyclone’s IMDM, supplemented with 20% FBS and 25 μg mL
24,25
fied procedure described by Eisenberg and co-workers,
-bromo-1,10-phenanthroline (5) was synthesized in 72%
yield by treating 1,10-phenanthroline with bromine and
gentamycin sulfate, and were passaged 3-4 times per week
according to standard aseptic procedures. Cultures were started
5
-
1
at 200,000 cells mL in 25 cm tissue culture flasks and were
2
-
1
subcultured before growth reached 750,000 cells mL to avoid
senescence associated with prolonged high cell density. Com-
plete media was prepared in 200 mL portions as needed by
combining IMDM (160 mL), FBS (40 mL, pre-aliquoted and
H SO in a sealed tube at 135 °C for 23 h. The 5-(pyren-
2
4
0
1 -yl)ethynyl-1,10-phenanthroline (2) was prepared in
73% yield by heating a mixture of 5 and a slight excess
of 1-ethynylpyrene (4) with 15 mol % [Pd(PPh ) ] and
-
1
3 4
heat inactivated), and gentamicin sulfate (100 μL of 50 mg mL
n-propylamine as both the base and the solvent at 80 °C
for 2 days.
Following a procedure recently described by Tor and co-
stock solution) in a 250 mL Millipore vacuum stericup (0.22 μm)
and filtering.
2.4.5. Cytotoxicity and Photocytotoxicity Assays. HL-60
22,23
workers, the treatment of 1,10-phenanthroline mono-
hydrochloride monohydrate with bromine in bromobenzene
cells growing in log phase were transferred (typically 25 μL
(
30) Croke, D. T.; Perrouault, L.; Sari, M. A.; Battioni, J. P.; Mansuy, D.;
Helene, C.; Le Doan, J. J. Photochem. Photobiol. 1993, 18, 41–50.
31) Praseuth, D.; Gaudemer, A.; Verlhac, J. B.; Kraljic, I.; Sissoeff, I.;
Guille, E. Photochem. Photobiol. 1986, 44, 717–724.
(32) Viability Staining Using Ethidium Bromide and Acridine Orange;
Technical Report, BD Biosciences: Mississauga, Ontario, Canada, 2008.
(33) Parks, D. R.; Bryan, V. M.; Oi, V. M.; Oi, V. T.; Herzenberg, L. A.
Proc. Natl. Acad. Sci. U.S.A. 1979, 76, 1962.
(

Article
Inorganic Chemistry, Vol. 49, No. 6, 2010 2893
Chart 3. Synthesis of Diimine Ligands 2 and 3
Table 1. Electronic Absorption and Emission Data for Ligands 2 and 3, Their Ru(II) Complexes, and Model Compounds (10 μM MeCN)
compound
λmax absorption (log ɛ), nm
λem, nm
2
3
1
235 (5.02), 282 (4.79), 315 (4.52), 376 (4.84), 397 (4.82)
236 (4.53), 285 (4.41), 312 (4.14), 376 (4.41), 398 (4.41)
230 (4.71), 264 (4.48), 280 (4.08)
449 (s)
458 (s)
,10-Phenanthroline
Pyrene
7
8
262 (4.31), 272 (4.61), 306 (3.98), 319 (4.38), 335 (4.59)
235 (4.91), 283 (5.03), 385 (4.59), 417 (4.65), 451 (4.48)
237 (4.88), 282 (5.01), 385 (4.60), 417 (4.50), 474 (4.14)
263 (4.87), 285 (4.91), 448 (4.31)
427 (s)
609 (vw)
625 (w)
615 (s)
2+
[
8 2
Ru(bpy-d ) (phen)]
for 36 h at 135 °C gave a 50% yield of a 1:1 mixture of
-bromo-1,10-phenanthroline (6) and 3,8-dibromo-1,10-
phenanthroline. Other brominated 1,10-phenanthrolines
were formed in small amounts depending upon the tem-
perature, concentration, and rate of addition. The desired
resonances shift upfield because of shielding by one of
the coordinated bipyridines. This shielding effect is less
dramatic when H3 is replaced by ethynylpyrene. The H3
and H8 resonances at higher field are once again quite
characteristic. For the complex of 2 there are two signals;
one of these signals disappears in the complex of 3.
The electronic absorption spectra of 2 and 3 were
measured, and the data are recorded in Table 1. Com-
pared to 1,10-phenanthroline (λmax=280 nm) and pyrene
3
6
could be isolated in pure form by careful chromato-
graphy on alumina or selective acidification using HCl in
ether. The subsequent reaction of 6 with 4 under the
previously employed Sonogashira coupling conditions
0
led to the formation of 3-(pyren-1 -yl)ethynyl-1,10-phen-
anthroline (3) in 79% yield (Chart 3).
(λmax = 335 nm), both measured in CH CN, the two
3
ethynylpyrenyl derivatives show a substantial red-shift of
their long wavelength ππ* band. The substitution of a
single ethynylpyrenyl group onto the 1,10-phenanthro-
line nucleus leads to a long wavelength absorption maxi-
mum at 397 nm (2) or 398 nm (3). It appears that 1,10-
phenanthroline and pyrene subunits are effectively
coupled electronically through the alkynyl linker, which
explains the lower energy absorptions for the dyad sys-
tems as compared to the component chromophores.
The ligands 2 and 3 are unsymmetrical with respect to
their diimine binding sites, and thus, the formation of
2
+
their heteroleptic [Ru(L)(bpy)2] complexes produces
non-equivalence in all protons, leading to highly complex
H NMR spectra. To simplify the spectra, we prepared
1
3
4
the corresponding perdeuterio bipyridine complexes,
having NMR-silent bipyridine protons. This isotopic
labeling was accomplished by treating 2 or 3 with 1 equiv
of [Ru(bpy-d ) Cl ] in refluxing ethanol to afford the
3
5
8
2
2
The electronic absorption spectra of the [Ru(bpy-d ) -
8 2
2
+
complexes in yields of 66-67%.
(L)] complexes where L = 2 or 3 were measured in
acetonitrile, and the data are also recorded in Table 1.
A long wavelength metal-to-ligand charge transfer
(MLCT) absorption is observed in the region of 451-
474 nm and can be compared with the corresponding
absorption at 448 nm for the analogous complex where
L = phen. The system having the ethynylpyrenyl sub-
stituent at the 5-position (λmax = 451 nm) is the most
similar to the parent 1,10-phenanthroline complex, indi-
cating that charge delocalization through the 5-position
is less effective than through the 3-position, where the
MLCT absorption is found at 474 nm. Although the
ligands show strong pyrene-based emission at 449-458
nm, the corresponding Ru(II) complexes are nearly non-
emissive at room temperature. Excitation spectra indicate
There are several notable features in the NMR spectra
that allow for straightforward characterization. When
H3 on 1,10-phenanthroline is replaced by an ethynyl-
pyren-1-yl group, this proton resonance disappears as
does the 3-bond coupling with H2 and H4, leaving these
two protons as signals 9.45 and 8.54 ppm having a 1.8 Hz
4
-bond coupling. With the disappearance of H3 at higher
field, only H8 appears in this region (7.66 ppm) as a very
characteristic doublet of doublets. Interestingly, H5 and
H6 show magnetic equivalence and give rise to a singlet at
7
.82 ppm. For the 5-substituted isomer, the pyridine ring
protons H2, H3, H4 and H7, H8, H9 are non-equivalent,
with the biggest difference being between H4 and H7 since
the former is strongly deshielded (9.09 ppm) by the
proximal ethynylpyrenyl group. H6 now appears as a
singlet, also deshielded and shifted downfield. Upon
2
+
that trace [Ru(bpy)3 ] is not responsible for this weak
emission.
2
+
complexation with [Ru(bpy-d ) ] , the H2 and H9
2
8
The half-wave redox potentials of the complexes were
measured by cyclic voltammetry, and the results are
summarized in Table 2. The first oxidation wave of Ru
(II) polypyridyl complexes is generally associated with the
removal of an electron from a metal-based d orbital and,
(
34) In our experience perdeuteration of the ancilliary ligands of the
Ru(II) complexes does not significantly alter their DNA photocleaving
properties or photophysics.
(
35) Chirayil, S.; Thummel, R. P. Inorg. Chem. 1989, 28, 812–813.

2
894 Inorganic Chemistry, Vol. 49, No. 6, 2010
Monro et al.
Table 2. Cyclic Voltammetric Data for 1,10-Phenanthroline and Ru(II) Com-
a
plexes 7 and 8
red
red
E1/2 (ΔE)
compound
Ru(phen)-
E1/2 (ΔE)
[
+1.27 (74)
-1.36 (77) -1.55 (84)
-1.79 (86)
2 6 2
(bpy) ](PF )
ir
ir
7
8
+1.32 (128) -1.11
+1.31 (107) -1.21 (79) -1.45 (107) -1.81
-1.48 (98)
-1.81
ir
a
4 6
Spectra were recorded in MeCN containing 0.1 M NBu PF at a
scan rate of 100 mV/s. E1/2 is reported as V vs SCE, and ΔE as mV. The
notation ir represents an irreversible process, and the value was esti-
mated by differential peaks.
Figure 2. Binding isotherms for 7 (414 nm) and 8 (412 nm) with herring
sperm DNA (10 mM MOPS, pH 7.5).
the solution (0 vs 50 mM NaCl). Generally, the ligand-
centered ππ* and MLCT absorption transitions asso-
ciated with the dyads shifted to longer wavelengths
(bathochromism) and were reduced in intensity (hypo-
chromism) with increasing concentration of DNA
regardless of the conditions employed. The largest red-
shifts (16-18 nm) for both 7 and 8 were observed with calf
thymus DNA as titrant in 10 mM MOPS (pH 7.5) and
were accompanied by 15% and 20% absorbance attenua-
tions, respectively. As expected, these hypochromicities
were slightly reduced when the ionic strength of the
solution was increased by addition of 50 mM NaCl and
were enhanced when measured at the blue-edge of the
visible MLCT transitions or when the source of DNA was
herring sperm. Figure 1 illustrates optical titrations of 7
and 8 with herring sperm DNA (10 mM MOPS, pH 7.5)
Figure 1. Titrations of 10 μM solutions of 7 (a) and 8 (b) with herring
sperm DNA (10 mM MOPS, pH 7.5).
in this regard, is less influenced by subtle changes in the
identity of the ligands. The observed potentials of +1.31
and +1.32 V are close to the potential of +1.27 measured
2
+
over a narrow concentration range ([DNA]/[Ru] =
for the parent [Ru(bpy) (phen)] complex. The reduc-
2
36
0
.1-2), where clear isosbestic points are evident.
tions are ligand-based, and the first reduction potentials
of -1.11 and -1.21 V observed for complexes 7 and 8 are
slightly more positive than the parent complex (-1.36 V),
which is consistent with the increased delocalization
expected for these more conjugated systems. The second
and third reductions probably take place on the auxiliary
bpy ligands and, thus, these potentials fall within 0.10 V
for all three systems.
DNA binding constants (K ) for 7 and 8 were calculated
b
6
-1
7
-1
to be 8.5 ꢁ 10 M and 3.0 ꢁ 10 M , respectively, from
fits of the absorption changes as a function of DNA
concentration to eq 1 (Figure 2). Both the spectral
responses and the magnitudes of K for the dyads are
b
similar to what have been observed for known Rh(III)
and Ru(II) metallointercalators (K > 10 M ) contain-
6
-1
b
4
,5,7,37
ing the dppz ligand.
While base pair binding bite
3.2. DNA Binding. To discern possible interactions
with DNA, 10 μM solutions of 7 and 8 were titrated with
DNA in a variety of buffered solutions at pH 7.5. The
spectral responses of both 7 and 8 to added DNA were
highly dependent on the source of DNA used as titrant
(calf thymus vs herring sperm DNA), the buffering
system (MOPS vs Tris-HCl), and the ionic strength of
(
36) In many of the buffering systems employed, no clear isosbestic points
could be discerned. However, provided that ratios of [DNA] to [Ru] were
chosen to minimize aggregation of the dyads under low-salt conditions,
similar binding constants were obtained from the absorbance hypochromi-
cities.
(37) Friedman, A. E.; Chambron, J. C.; Sauvage, J. P.; Turro, N. J.;
Barton, J. K. J. Am. Chem. Soc. 1990, 112, 4960–4962.

Article
Inorganic Chemistry, Vol. 49, No. 6, 2010 2895
sizes for metallointercalators are generally greater
than 1,
2
9,38-40
s values for the dyads are 0.52 and 0.56,
respectively. Base pair binding site sizes less than unity, as
calculated for 7 and 8, have been interpreted by others to
mean that intercalator molecules are stacking with each
4
1,42
other on the DNA surface.
Given that polycyclic aromatic hydrocarbons inter-
43-50
calate into DNA,
albeit more weakly than cati-
onic intercalators such as ethidium bromide (K = 1.7 ꢁ
b
5 -1 51,52
0 M ),
1
it is reasonable to conclude that the ethynyl-
pyrenyl units in dyads 7 and 8 provide a handle for inter-
calation between DNA base pairs and that this interaction is
enhanced by the cationic metal center. In this scenario DNA
binding appears to be somewhat stronger with 3-substitu-
tion of the ethynylpyrenyl group on the 1,10-phenanthroline
ring. However, caution must be used in assigning intercala-
tion as the preferred DNA binding mode from only optical
titration data since DNA intercalation, groove binding, and
external association can all induce qualitatively similar
Figure 3. Thermal denaturation of calf-thymus DNA (24 μM base
pairs, 10 mM MOPS, pH 7.4) alone and in the presence of ligands (2, 3)
or dyads (7, 8).
53
effects via spectroscopic methods. Spectral hypochromi-
city and bathochromicity, which generally arise from π-
stacking interactions that are favored in polar media,
54-57
have been observed for some minor groove binders such as
0
58
,6-diamidino-2-phenylindole (DAPI). Additionally, it is
4
well-documented that polyanions like DNA facilitate the
aggregation of hydrophobic molecules at their surfaces in
(
38) Welch, T. W.; Corbett, A. H.; Thorp, H. H. J. Phys. Chem. 1995, 99,
1757–11763.
39) Johnston, D. H.; Thorp, H. H. J. Phys. Chem. 1996, 100, 13837–
3843.
40) Neyhart, G. A.; Grover, N.; Smith, S. R.; Kalsbeck, W. A.; Fairley,
T. A.; Cory, M.; Thorp, H. H. J. Am. Chem. Soc. 1993, 115, 4423–4428.
41) Nair, R. B.; Teng, E. S.; Kirkland, S. L.; Murphy, C. J. Inorg. Chem.
998, 37, 139–141.
42) Haq, I.; Lincoln, P.; Suh, D.; Norden, B.; Chowdhry, B. Z.; Chaires,
J. B. J. Am. Chem. Soc. 1995, 117, 4788–4796.
43) Wolfe, A.; Shimer, G. H., Jr.; Meehan, T. Biochemistry 1987, 26,
392–6396.
44) Berman, H. M.; Young, P. R. Annu. Rev. Biophys. Bioeng. 1981, 10,
7–114.
1
1
(
(
Figure 4. Relative viscosity changes of calf thymus DNA solutions with
increasing concentrations of 7 (.) and 8 (ꢁ). Ethidium bromide (9) and
Hoechst 33258 (2) are included for comparison.
(
1
(
59
aqueous solutions. Therefore, DNA thermal denaturation
measurements
60-63
64
and relative viscosity changes of DNA
(
6
8
solutions served as additional probes for DNA intercalation
as an important binding mode for the dyads.
The melting temperature (T ) of calf thymus DNA in
(
(
45) Dougherty, G.; Pilbrow, J. R. Int. J. Biochem. 1984, 16, 1179–1192.
46) Geacintov, N. E.; Prusik, T.; Khosrofian, J. M. J. Am. Chem. Soc.
m
(
10 mM MOPS (pH 7.4) was determined to be 60 ( 2 °C, and
1
976, 98, 6444–6452.
47) Geacintov, N. E.; Gagliano, A.; Ivanovic, V.; Weinstein, I. B.
Biochemistry 1978, 17, 5256–5262.
48) Ibanez, V.; Geacintov, N. E.; Gagliano, A. G.; Brandimarte, S.;
Harvey, R. G. J. Am. Chem. Soc. 1980, 102, 5661–5666.
49) Meehan, T.; Gamper, H.; Becker, J. F. J. Biol. Chem. 1982, 257,
0479–10485.
50) Nagata, C.; Kodama, M.; Tagashira, Y.; Imamura, A. Biopolymers
966, 4, 409–427.
changes in T induced by the dyads are shown in Figure 3.
The presence of ligand 2 or 3 (5 μM) as the free base, lacking
the favorable electrostatic component for DNA interaction,
m
(
(
6
5
did not significantly alter the melting temperature of DNA,
underscoring the importance of the cationic metal center for
(
1
(
DNA binding. Addition of 7 or 8 (5 μM) increased T by 10
m
1
and 20 °C, respectively. The large increase in T for 8,ΔT >
m
m
(
(
51) Tang, T. C.; Huang, H. J. Electroanalysis 1999, 11, 1185–1190.
2
0 °C, is consistent with ΔT values reported for known
m
52) Paoletti, C.; Le Pecq, J. B.; Lehman, I. R. J. Mol. Biol. 1971, 55,
3+
intercalators such as [Rh(phi) (phen)] (phi = 9,10-
7
5–100.
2
6
6
(
(
(
53) Li, H. J.; Crothers, D. M. J. Mol. Biol. 1969, 39, 461–477.
phenanthrenequinone diimine). The attenuated ΔTm
measured for 7, while still suggestive of intercalation or
partial intercalation, illustrates that positioning of the
ethynylpyrenyl group relative to the 1,10-phenanthroline
ring can be used to influence the magnitude of this DNA
interaction.
54) Felthouse, T. R. Prog. Inorg. Chem. 1982, 29, 73–166.
55) Bilakhiya, A. K.; Tyagi, B.; Paul, P.; Natarajan, P. Inorg. Chem.
2
002, 41, 3830–3842.
(
56) Jones, G.; Vullev, V. I. J. Phys. Chem. A 2001, 105, 6402–6406.
57) Jockusch, S.; Turro, N. J.; Tomalia, D. A. Macromolecules 1995, 28,
(
7
4
9
416–7418.
58) Kubista, M.; Akerman, B.; Nord ꢀe n, B. Biochemistry 1987, 26, 4545–
553.
59) Wang, M.; Silva, G. L.; Armitage, B. A. J. Am. Chem. Soc. 2000, 122,
977–9986.
60) Sinan, M.; Panda, M.; Ghosh, A.; Dhara, K.; Fanwick, P. E.;
Chattopadhyay, D. J.; Goswami, S. J. Am. Chem. Soc. 2008, 130, 5185–5193.
61) Kumar, C. V.; Asuncion, E. H. J. Am. Chem. Soc. 1993, 115, 8547–
553.
˚
(
Hydrodynamic methods provided further support for
DNA intercalation by both 7 and 8. Figure 4 shows an
(
(
(64) Suh, D.; Oh, Y.; Chaires, J. B. Process Biochem. 2001, 37, 521–525.
(65) The absorbance signal at 260 nm contains contributions from both
DNA and the potential DNA binder. Therefore, the maximum signal
obtainable is attenuated for ligands 2 and 3 owing to their reduced extinction
coefficients at this wavelength relative to the dyads.
(
8
3
(
62) Patel, D. J. Acc. Chem. Res. 1979, 12, 118–125.
63) Patel, D. J.; Canuel, L. L. Proc. Natl. Acad. Sci. U.S.A. 1976, 73,
(
343–3347.
(66) Fu, P.; Bradley, P. M.; Turro, C. Inorg. Chem. 2003, 42, 878–884.

2
896 Inorganic Chemistry, Vol. 49, No. 6, 2010
Monro et al.
increase in the relative viscosities of DNA solutions
containing these complexes. Intercalators unwind the
double helix by a defined angle when they insert between
base pairs, producing DNA that is somewhat elongated
relative to canonical B-form DNA. Lengthening of the
6
4
helical axis gives rise to measurable increases in viscosity
that are not observed for other common types of DNA
binding, namely, groove-binding or electrostatic associa-
tion, or DNA-templated aggregation of hydrophobic
chromophores. When 150 μM DNA (base pairs) was
treated withincreasingconcentrations of7 or 8, the relative
viscosity increased substantially. In fact, 8 produced a
larger change in viscosity than the known intercalator
ethidium bromide. Consistent with thermal denaturation
experiments, 7 produced a smaller increase in solution
viscosity but still notably greater than the minor-groove
binder Hoechst 33258, which is included for comparison.
Together the optical titrations, thermal denaturation
studies, and relative viscosity measurements point toward
intercalation as one important mode of binding to DNA
by this class of molecules. The ethynylpyrenyl handle is
likely the point of attachment to DNA as planar, aro-
matic systems are known to intercalate through efficient
π-stacking between the DNA base pairs. It is interesting
to note that ligands 2 and 3 (as their free bases) did not
interact substantially with DNA. Tethering these organic
chromophores in a Ru(II) diimine dyad such as 8 appears
to be a prerequisite for strong DNA binding and rein-
forces the association with DNA over known cationic
intercalators such as ethidium bromide. The attenuated
binding of 7 relative to 8 reveals that the magnitude of the
DNA-binding interaction is sensitive to the substitution
pattern on the 1,10-phenanthroline functionality and
may, in turn, influence the extent of photoreactivity
toward DNA (vide supra).
Figure 5. Gel electrophoretic analysis of DNA photocleavage titrations
-1
in air (1% agarose gel containing 0.75 μg mL ethidium bromide, 1X
TAE, 8 V cm , 30 min.). (a) and (b) contain pUC19 plasmid; (c) contains
pDesR3 plasmid. (a) Lane1, plasmid only; lane 2, 50 nM7; lane 3, 100 nM
-1
7; lane 4, 250 nM 7; lane 5, 500 nM 7; lane 6, 750 nM 7; lane 7, 1.00 μM 7;
lane8, 1.25μM 7; lane9, 1.50 μM 7; lane 10, 1.75μM 7; lane11, 2.00μM 7;
lane12, 2.25μM7; lane 13, 2.25μM7 (-hv). (b) Lane1, plasmid only;lane
2, 25 nM 8; lane 3, 50 nM 8; lane 4, 75 nM 8; lane 5, 100 nM 8; lane 6,
300 nM 8; lane 7, 500 nM 8; lane 8, 750 nM 8; lane 9, 1.00 μM 8; lane 10,
1.25 μM 8; lane 11, 1.50 μM 8; lane 12, 2.00 μM 8; lane 13, 2.00 μM 8
(-hv). (c) Lane 1, plasmid only; lane 2, 25 nM 8; lane 3, 50 nM 8; lane 4,
100 nM 8; lane 5, 500 nM 8; lane 6, 750 nM 8; lane 7, 1.00 μM 8; lane 8,
1.50 μM 8; lane 9, 2.00 μM 8; lane 10, 3.00 μM 8; lane 11, 5.00 μM 8;
lane 12, 7.00 μM 8; lane 13, 10.0 μM 8; lane 14, 10.0 μM 8 (-hv).
with 420 nm light (typical light-assay conditions) resulted
in no discernible DNA damage. However, irradiation of
either 7 or 8 in the presence of pUC19 or pBR322 plasmid
produced single-strand breaks in a concentration-depen-
dent manner, with EC50 (effective concentration) values
in the nanomolar regime and complete single-strand
cleavage (100% Form II) occurring at concentrations of
complex near 2 μM. Notably, when the GC content of the
plasmid was increased (e.g., pDesR3), 8 elicited double-
strand breaks at concentrations above 3 μM (Figure 5c).
The appearance of Form III DNA on the gel was only
apparent at concentrations above that which caused
complete single-strand scission, suggesting that double-
strand cleavage may arise from the accumulation of
single-strand breaks that occur within a critical distance
3.3. DNA Photocleavage. As shown in Figure 5, both 7
and 8 cause extensive damage to DNA when irradiated
with visible light. Plasmid DNA served as a convenient
probe for DNA photocleavage, whereby gel electro-
phoretic analysis with ethidium-bromide staining was
used to distinguish three distinct topological forms of
DNA: Form I (supercoiled), Form II (relaxed), and Form
III (linear). The topological state of plasmid DNA, in
turn, reflects the nature of DNA damage mediated by a
photoactivatable agent: single- or double-strand breaks
in the backbone as well as unwinding of the double helix.
Form I DNA (Figure 5a, lane 1), often referred to as
closed, circular DNA, represents native plasmid DNA
that is negatively supercoiled in the absence of DNA
damaging agents. When single-strand breaks occur,
superhelical tension is removed and the DNA relaxes to
produce Form II (Figure 5a, lane 12), which migrates
more slowly through the agarose gel as nicked, circular
DNA. If two single-strand lesions occur within approxi-
6
8,69
on the helix.
Unfortunately, the gradual disappear-
ance of both Form II and Form III bands at higher
concentrations of 8 precluded a detailed analysis of this
behavior since the ethidium bromide intercalating dye
was not able to penetrate the helix at elevated concentra-
tions of 8. Surprisingly, 7 did not linearize plasmid DNA
under similar conditions despite having a larger extinction
coefficient at the irradiating wavelength, underscoring the
importance of the 3-substitution pattern for the more
deleterious double-strand DNA cleavage. Enhanced differ-
ences in DNA-binding affinity (vide infra) or photoreacti-
vity toward DNA with higher %GC may reflect some
sequence selectivity for the dyads, especially 8.
6
7
mately 15 base pairs or if frank double-strand cleavage
takes place, the result is linear, Form III DNA (Figure 5c,
lane 13), which travels between Forms I and II on the gel.
DNA photocleavage assays with pUC19 or pBR322
plasmid and complex 7 or 8 showed no DNA damage in
the dark (Figure 5a, lane 13 and Figure 5b, lane 13).
Similarly, irradiation of the plasmids alone for 30 min
(
68) OhUigin, C. Ph.D. thesis, University of Dublin, 1988.
˚
(69) Akerman, B.; Tuite, E. Nucleic Acids Res. 1996, 24, 1080–1090.
(
67) Freifelder, D.; Trumbo, B. Biopolymers 1969, 7, 681–693.

Article
Inorganic Chemistry, Vol. 49, No. 6, 2010 2897
hypoxic tumors remains a significant challenge. It is often
these oxygen-deprived tumor cells that are the most
8
0,81
82,83
resistant to radiotherapy
and particularly susceptible to metastasis.
and chemotherapy
8
4,85
The observation that both 7 and 8 photocleave DNA in
the absence of oxygen without the use of a sacrificial
electron acceptor implies that reactive excited states,
Figure 6. Comparison of DNA photocleavage by 7 and 8 for air-
saturated and deoxygenated solutions by gel electrophoretic analysis
3
other than the commonly employed MLCT state for
O sensitization, may be involved. In fact, dirhodium
1
-1
(
8
8
1% agarose gel containing 0.75 μg mL ethidium bromide, 1X TAE,
V cm , 30 min.). Lanes 1-4 and 7 are control lanes, and lanes 5-6 and
-9 are reaction lanes: lane 1, pUC19 plasmid only (-hv); lane 2, pUC19
2
-1
complexes of dppz, which exhibit some oxygen-indepen-
dence, are thought to react with DNA through a low-
lying triplet state that may be associated more with the
π-extended ligand, creating a long-lived, charge-separated
state that results in oxidative DNA photocleavage by the
plasmid in air-saturated solution; lane 3, pUC19 plasmid in argon-purged
solution; lane 4, pUC19 plasmid incubated for 30 min at 37 °C in air-
saturated solution containing 2 μM 7 (-hv); lane 5, pUC19 plasmid in air-
saturated solution containing 2 μM 7; lane 6, pUC19 in argon-purged
solution containing 2 μM 7; lane 7, pUC19 plasmid incubated for 30 min
at 37 °C in air-saturated solution containing 2 μM 8 (-hv); lane 8, pUC19
plasmid in air-saturated solution containing 2 μM 8; lane 9, pUC19 in
argon-purged solution containing 2 μM 8.
7
dirhodium core. Unusually long excited-state lifetimes
(τ ∼42 μs) have been measured for pyrene-Ru(II) poly-
pyridine dyads derived from the bipyridyl counterpart of
ligand 3 and are thought to arise from intramolecular
triplet energy transfer involving the lowest-lying MLCT
1
Despite their abilities to photosensitize O2 in the
3
7
absence of DNA, 7 and 8 showed little attenuation in
0
3
18
state and a ligand-centered ππ* state. A similar equi-
librium may be at play in the pyrene-Ru(II) 1,10-phenan-
DNA photocleavage in the absence of oxygen, in the
presence of reactive oxygen scavengers, or in D O buffer.
3
+
2
throline dyads, allowing the quasi-oxidized Ru center
to react directly with DNA through guanine oxidation.
The enhanced photocleavage observed for DNA
with higher %GC supports this proposed mechanism,
and a detailed investigation to elucidate the photophysics
of the ethynylpyrenyl-1,10-phenanthroline Ru(II) dyads
and the nature of the reactive excited states that photo-
cleave DNA is underway.
This oxygen-independence is quite notable in that mono-
nuclear polypyridyl Ru(II) complexes typically rely on
singlet-oxygen generation for DNA photocleavage,
7
1-75
and the ensuing DNA damage is much less efficient.
Figure 6 (Lanes 2-3) indicates that irradiation of plasmid
DNA with visible light for 30 min under ambient or
deoxygenated conditions results in no detectable strand
breaks in the DNA backbone. Similarly, incubation of
3.4. Cytotoxicity and Photocytotoxicity. The cytotoxi-
2
μM 7 (Lane 4) or 8 (Lane 7) with plasmid DNA in the
city and photocytotoxicity of 7 and 8 toward human
leukemia cells (HL-60) in culture was determined by
measuring percent viability as a function of increasing
concentrations of the dyads with and without irradiation
with visible light. These measurements were made relative
to growing cultures that were not subjected to potentially
cytotoxic agents or light. Interpolation of the collected
data was used to highlight the concentration of dyad
necessary to achieve a 50% loss in viability, which is
designated LC50 (LC = lethal concentration). As shown
in Figure 7a, the LC50 for 7 in the dark is approximately
dark (10 mM MOPS, pH 7.5, 0.1 mM NaCl) produces no
discernible DNA damage. Lanes 5 and 6 clearly show that
there is no difference in the extent of single-strand photo-
cleavage by 7 when oxygen is removed, and only a slight
inhibition of strand breakage occurs for deoxygenated
samples of 8 (Lanes 8-9). While some Rh(III) complexes
are known to photocleave DNA via both oxygen-depen-
7
,76
dent and oxygen-independent pathways,
few Ru(II)
complexes have been shown to operate under hypoxic
conditions, and these utilize at least two metals working
in concert and, in some cases, require the presence of
15 μM. In the absence of 7, there is only a marginal loss in
7
7-79
glutathione.
Thus, the discovery of a relatively sim-
viability (<10%) caused by irradiation of the cells, and, in
most instances, there was no detectable loss in viability
under the irradiation conditions employed. When grow-
ing cultures containing 7 were irradiated for 30 min with
visible light, the LC50 decreased to well below 5 μM.
Notably, at concentrations between 5 and 10 μM 7, there
is a 50% loss in the viability of the irradiated cells over the
control cells that were incubated with 7 in the dark. This
enhanced photocytotoxicity relative to the dark controls
is an essential property for a photochemotherapeutic
agent and is exemplified in the case of 8. Figure 7b
demonstrates that there is <5% loss in the viability of
ple mononuclear Ru(II) motif that leads to efficient DNA
photocleavage under hypoxic conditions is particularly
intriguing for phototherapeutic applications such as
PDT, whereby the search for new drugs that target
(
70) Emission from singlet oxygen at 1268 nm was observed for solutions
of 7 and 8 upon photoirradiation in acetonitrile or 4:1 EtOH/MeOH.
Quantum yields for singlet oxygen generation were not measured.
(
71) Liu, Y.; Hammitt, R.; Lutterman, D. A.; Joyce, L. E.; Thummel,
R. P. Inorg. Chem. 2009, 48, 375–385.
72) Chouai, A.; Wicke, S. E.; Turro, C.; Bacsa, J.; Dunbar, K. R.; Wang,
(
D.; Thummel, R. P. Inorg. Chem. 2005, 44, 5996–6003.
(
(
73) Clarke, M. J. Coord. Chem. Rev. 2003, 236, 209–233.
74) Hergueta-Bravo, A.; Jimenez-Hernandez, M. E.; Montero, F.;
Oliveros, E.; Orellana, G. J. Phys. Chem. B 2002, 106, 4010–4017.
75) Fu, P. K.-L.; Bradley, P. M.; van Loyen, D.; D u€ rr, H.; Bossman,
S. H.; Turro, C. Inorg. Chem. 2002, 41, 3808–3810.
76) Bradley, P. M.; Angeles-Boza, A. M.; Dunbar, K. R.; Turro, C.
(80) Gatenby, R. A.; Kessler, H. B.; Rosenblum, J. S.; Coia, L. R.;
Moldofsky, P. J.; Hartz, W. H.; Broder, G. J. Int. J. Radiat. Oncol. Biol.
Phys. 1988, 14, 831–838.
(81) Hockel, M.; Knoop, C.; Schlenger, K.; Vorndran, B.; Baubmann, E.;
Mitze, M.; Knapstein, P. G.; Vaupel, P. Radiother. Oncol. 1993, 26, 45–50.
(82) Sartorelli, A. C. Cancer Res. 1988, 48, 775–778.
(
(
Inorg. Chem. 2004, 43, 2450–2452.
(
(
77) Swavey, S.; Brewer, K. J. Inorg. Chem. 2002, 41, 6196–6198.
78) Holder, A. A.; Swavey, S.; Brewer, K. J. Inorg. Chem. 2004, 43,
(83) Tannock, I.; Guttman, P. Br. J. Cancer 1981, 42, 245–248.
(84) Sundfor, K.; Lyng, H.; Rofstad, E. K. Br. J. Cancer 1998, 78,
822–827.
3
03–308.
79) Janaratne, T. K.; Yadav, A.; Ongeri, F.; MacDonnell, F. M. Inorg.
Chem. 2007, 46, 3420–3422.
(
(85) Rofstad, E. K.; Danielson, T. Br. J. Cancer 1999, 80, 1697–1707.

2
898 Inorganic Chemistry, Vol. 49, No. 6, 2010
Monro et al.
and 9; increasing the magnification by 4-fold or 10-fold
(b and e, respectively) reveals the formation of apoptotic
bodies quite clearly. These apoptotic bodies (zoomed for
clarity in the right panel) are formed from surface blebbing
of the membrane that occurs subsequent to DNA conden-
sation and fragmentation in response to programmed cell
death. Photocytotoxicity resulting from the induction of
apoptosis in HL-60 cells that were exposed to the dyads and
irradiated is a remarkable finding. For photobiology appli-
cations such as PDT, apoptosis is the preferred mode of cell
death as necrosis is a rather traumatic form of cellular
destruction that is accompanied by intense inflammation
of surrounding tissue and the rupturing of extended blood
86,87
vessels in vivo.
Because the latter often hinders further
PDT treatment, much effort to improve PDT agents is
aimed at controlling the outcome of photosensitization by
careful minimization of tissue necrosis.
The lack of significant cytotoxic activity in the dark at
relatively low concentrations of 7 and 8 is somewhat
surprising given the ability of the dyads to intercalate
DNA. Often, the topological changes to DNA that
accompany intercalation disrupt crucial protein-nucleic
acid interactions that control cellular function such as
DNA replication and transcription. The result is a dark
toxicity that usually parallels the strength of DNA bind-
ing. Interestingly, 8 interacts with DNA more strongly
than 7 according to optical titrations, thermal denatura-
tion studies, and relative viscosity changes, yet 7 is
notably more cytotoxic than 8 toward HL-60 cells. Since
Figure 7. Plots of percent viability of human promyelocytic leukemia
cells (HL-60) as a function of concentration of (a) 7 and (b) 8 in the dark
8
showed stronger DNA interactions and caused more
(9) and irradiated with visible light for 30 min (~).
deleterious DNA damage (i.e., double-strand scission of
some plasmids) in non-cell-based assays, factors other
than DNA-binding strength and reactivity must contri-
bute to the differences in dark cytotoxicity documented
for these dyads. It is well-known that cellular uptake of
Ru(II) complexes is strongly influenced by the identity
of the coordinating ligands, which control the overall
the cells when incubated with 8 in the dark over the
concentration range investigated, precluding an estima-
tion of the LC . Remarkably, the viability of irradiated
cultures containing 10 μM 8 was reduced by 60%, and
20 μM 8 resulted in ∼90% increase in toxicity relative to
the dark controls.
5
0
8
8,89
complex charge, size, shape, and hydrophobicity.
Results obtained from nuclear morphology staining
with the AO-EB dye combination
Therefore, the nature of both the DNA-interacting and
the ancillary ligands may have implications in the relative
cytotoxicities of these systems. Regardless of the under-
lying mechanism responsible for the cytotoxicity and
photocytoxicity associated with 7 and 8, it is evident from
these studies that the position occupied by the ethynyl-
pyrenyl substituent on the 1,10-phenanthroline ring has a
considerable influence. Efforts are underway to deter-
mine the source of cytotoxicity and photocytotoxicity for
the dyads and the role that positional isomerism plays in
their corresponding mechanisms of action.
3
2,33
correlated well
with manual photo- and dark cytotoxicity determina-
tions. Figures 8 and 9 illustrate the enhanced cytotoxicity
observed when 7 and 8 are activated by visible light,
respectively. Briefly, small aliquots were removed from
the photo- and dark cytotoxicity assays (used for the
manual cell counts) and stained with a mixture of AO-EB
in PBS. In these experiments a microscope operating in
epi-fluorescence mode was used to illuminate the stained
cells and capture the resulting fluorescence from the dyes.
It is well-established that AO is able to penetrate the
membranes of viable cells while EB is excluded. There-
fore, green fluorescence from AO is a convenient marker
for viable cells, and red fluorescence from EB indicates a
loss of cell viability. At the concentrations shown, there is no
significant loss in cell viability for HL-60 cells exposed to 7
or 8 in the dark, and, thus, they fluoresce green as expected.
However, increasing concentrations of 7 or 8, respectively,
result in larger fractions of cells that fluoresce red. These
nonviable cells, having compromised membranes, may be
apoptotic or necrotic, which can be discerned by viewing
with increased magnification. Figure 10 (left panel) illus-
trates the photocytotoxicity produced by 7 and 8 when
viewed with the same magnification as shown in Figures 8
4. Concluding Remarks
Two new mononuclear Ru(II) complexes possessing the
phenanthroline-appended ethynylpyrenyl ligands 2 and 3
were synthesized, characterized, and evaluated for their
potential in photobiological applications such as PDT. Nu-
merous techniques support intercalation as an important
binding mode for dyads 7 and 8, and both complexes
(86) Li, C. Z.; Cheng, L. F.; Wang, Z. Q.; Gu, Y. Endoscopy 2003, 35,
1043–1048.
(
(
87) Ali, S. M.; Olivo, M. Int. J. Oncol. 2003, 22, 1181–1191.
88) Puckett, C. A.; Barton, J. K. Biochemistry 2008, 47, 11711–11716.
(89) Puckett, C. A.; Barton, J. K. J. Am. Chem. Soc. 2007, 129, 46–47.

Article
Inorganic Chemistry, Vol. 49, No. 6, 2010 2899
Figure 8. Cytotoxicity and photocytotoxicity of HL-60 cells with varying concentrations of 7 observed by nuclear morphology staining with AO-EB.
a) 1 μM 7, -hv; (b) 5 μM 7, -hv; (c) 10 μM 7, -hv; (d) 1 μM 7, +hv; (e) 5 μM 7, +hv; (f) 10 μM 7, +hv.
(
Figure 9. Cytotoxicity and photocytotoxicity of HL-60 cells with varying concentrations of 8 observed by nuclear morphology staining with AO-EB.
a) 5 μM 8, -hv; (b) 10 μM 8, -hv; (c) 20 μM 8, -hv; (d) 5 μM 8, +hv; (e) 10 μM 8, +hv; (f) 20 μM 8, +hv.
(
photocleave DNA in vitro when irradiated with visible light
λirr g 400 nm) at concentrationssimilar tothose documented
in DNA photocleavage assays for lutetium(III) texaphyrin
dyads photocleave DNA under hypoxic conditions, an im-
portant step toward the design of new phototherapeutic
agents since few systems demonstrate such activity in the
absence of oxygen. The photoactivity of 7 and 8 translates
well to cytotoxicity and photocytotoxicity models using
human leukemia cells, where 7 and 8 show minimal or
essentially no cytotoxicity in the dark at concentrations of
5-10 μM and 10-20 μM, respectively, and much higher
toxicity when irradiated. Because the dark cytotoxicity asso-
ciated with these dyads does not necessarily parallel DNA-
binding strength or the magnitude of the DNA damage
apparent in photocleavage assays, differences in cytotoxic
activity may be influenced by differences in membrane
penetration, subcellular localization, biological targets, or
mechanisms of action. Consequently, this work provides
evidence that agents incorporating strong-intercalating units
donot a priori producedarktoxicity, discounting the popular
tenet that DNA intercalation may not be a desirable property
(
90,91
(LuTx),
a photosensitizer currently used in PDT. DNA-
binding and ensuing photocleavage are sensitive to the
position of the ethynylpyrenyl substituent on the 1,10-phe-
nanthroline ring, with 8 showing the strongest binding under
all conditions and causing more deleterious damage to DNA.
It is not clear whether the increased photoreactivity of 8 is
correlated directly to a higher affinity for DNA, but a
thorough photophysical investigation of the interactions of
the dyads with DNA is currently underway and should help
clarify this issue. Interestingly, these mononuclear Ru(II)
(
90) Allison, R. R.; Downie, G. H.; Cuenca, R.; Hu, X.-H.; Childs,
C. J. H.; Sibata, C. H. Photodiagn. Photodyn. Ther. 2004, 1, 27–42.
91) Magda, D.; Wright, M.; Miller, R. A.; Sessler, J.; Sansom, P. I.
J. Am. Chem. Soc. 1995, 117, 3629–3630.
(

2
900 Inorganic Chemistry, Vol. 49, No. 6, 2010
Monro et al.
Figure 10. Formation of apoptotic bodies in HL-60 cells with 7 [(a)-(c)] and 8 [(d)-(e)] observed by nuclear morphology staining with AO-EB. (a) 10 μM
, +hv, 10X obj.; (b) 10 μM 7, +hv, 40X obj.; (c) 10 μM 7, +hv, 40X obj. (zoomed); (d) 20 μM 8, +hv, 10X obj.; (e) 20 μM 8, +hv, 100X obj.; (f) 20 μM 8,
hv, 100X obj. (zoomed).
7
+
of a successful photochemotherapy agent. Importantly, the
photocytotoxicity associated with the dyads results in apopto-
sis, which is a necessary consideration in the development of
new molecules for clinical purposes. Together, the findings
reported herein introduce supramolecular systems derived
from conjugated Ru(II)-diimine dyads as promising candi-
dates to be included in the arsenal of potential photochemo-
therapeutic agents under investigation.
Acknowledgment. S.A.M., S.M., J.S., and R.L. thank the
Natural Sciences and Engineering Research Council of
Canada, the Canadian Foundation for Innovation, and
the Nova Scotia Health Research Foundation for financial
support and Prof. Todd Smith for invaluable help with
cell culture protocols. A.C., R.Z., and R.P.T. thank the
Robert A. Welch Foundation (E-621) and the National
Science Foundation (CHE-0714751) for financial support.