[Ru(Me
4
phen)
2
dppz]
2+
, a Light Switch for DNA Mismatches
Adam N. Boynton
,
Lionel Marcélis
, and
Jacqueline K. Barton
*
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena,
California 91125, United States
Abstract
[Ru(Me
4
phen)
2
dppz]
2+
serves as a luminescent “light switch” for single base mismatches in DNA.
The preferential luminescence enhancement observed with mismatches results from two factors:
(i) the complex possesses a 26-fold higher binding affinity towards the mismatch compared to
well-matched base pairs, and (ii) the excited state emission lifetime of the ruthenium bound to the
DNA mismatch is 160 ns versus 35 ns when bound to a matched site. Results indicate that the
complex binds to the mismatch through a metalloinsertion binding mode. Cu(phen)
2
2+
quenching
experiments show that the complex binds to the mismatch from the minor groove, characteristic of
metalloinsertion. Additionally, the luminescence intensity of the complex with DNA containing
single base mismatches correlates with the thermodynamic destabilization of the mismatch, also
consistent with binding through metalloinsertion. This complex represents a potentially new early
cancer diagnostic for detecting deficiencies in mismatch repair.
Graphical abstract
DNA mismatches arise as a result of errors during replication, and deficiencies in mismatch
repair (MMR) machinery are implicated in several forms of cancer.
1
–
3
As such, the design
of small molecules that target DNA mismatches holds promise for chemotherapeutic and
diagnostic applications. A class of octahedral rhodium complexes, bearing sterically
expansive planar ligands, bind DNA mismatches with high selectivity and exhibit
preferential cytotoxicity towards MMR-deficient cancer cells.
4
–
6
These compounds bind to
Corresponding Author
. jkbarton@caltech.edu.
ASSOCIATED CONTENT
Supporting Information
Experimental methods and supporting figures S1–S5. This material is available free of charge
via
the Internet at
http://pubs.acs.org
The authors declare no competing financial interests.
HHS Public Access
Author manuscript
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.
Published in final edited form as:
J Am Chem Soc
. 2016 April 20; 138(15): 5020–5023. doi:10.1021/jacs.6b02022.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
DNA through metalloinsertion, in which the bulky ligand inserts into the duplex at the
thermodynamically destabilized mismatch site, displacing the mismatched bases into the
DNA groove.
7
,
8
Luminescent reporters of mismatches may represent early diagnostics of
carcinogenesis. Several groups have documented the use of organic small molecules,
including Thioflavin T, cationic perylenediimides, and bisanthracene macrocycles, for
fluorometric mismatch detection.
9
–
12
In an effort to develop new selective, signal-on probes for DNA mismatches, we have
focused our attention on derivatives of [Ru(bpy)
2
dppz)]
2+
(dppz = dipyridophenazine),
which serve as molecular “light switches” for duplex DNA.
13
While these complexes do
luminesce in aprotic solvents, in aqueous solution, their luminescence is quenched due to
hydrogen bonding interactions between solvent water molecules and the phenazine nitrogen
atoms of the dppz ligand. However, upon intercalation into well-matched duplex DNA, these
compounds luminesce brightly owing to protection of the dppz ligand from the aqueous
environment.
13
–
16
[Ru(bpy)
2
dppz]
2+
derivatives have seen utility as structural probes,
cellular imaging agents, and in the development of new cytotoxic and photoactive small
molecules.
17
–
25
Interestingly, [Ru(bpy)
2
dppz]
2+
exhibits a some-what brighter emission in the presence of a
DNA mismatch relative to completely well-matched DNA.
26
A crystal structure of the
complex bound to an oligonucleotide duplex containing a mismatch revealed that,
analogously to rhodium metalloinsertors, the ruthenium complex binds at the mismatch site
in the minor groove through metalloinsertion.
27
[Ru(bpy)
2
dppz]
2+
is not mismatch-specific,
however, since it readily binds to well-matched sites in the DNA duplex through
intercalation. Ruthenium complexes bearing expansive inserting ligands, such as 5,6-
chrysenequinone diimmine (chrysi), have been investigated, and while it was found that
these compounds show mismatch specificity in binding, they are not luminescent at ambient
temperature.
28
[Ru(bpy)
2
dppz]
2+
derivatives in which the inserting dppz ligand was directly
functionalized have also been examined, but an improved luminescence differential between
mismatched and well-matched DNA compared to [Ru(bpy)
2
dppz]
2+
was not achieved;
28
the
functionalization of ancillary ligands has not been investigated in this context.
Here we sought to attain mismatch specificity through ancillary ligand modification of the
[Ru(phen)
2
dppz]
2+
scaffold using 3,4,7,8-tetramethyl-1,10-phenanthroline (Me
4
phen, Figure
1). We rationalized that incorporation of methyl groups on the ancillary ligands would
disfavor binding to well-matched sites as a result of steric clashing between the ancillary
ligands and the DNA backbone. Moreover, bulkier ancillary ligands would both disfavor
deep intercalation of dppz at a matched site and favor shielding of the phenazine nitrogen
atoms with insertion at a mismatched site.
29
[Ru(Me
4
phen)
2
dppz]
2+
was synthesized in two steps (see Supporting Information), and a
racemic mixture of the chloride salt was used for all DNA experiments. As expected, the
complex is not luminescent in aqueous solution upon excitation at 440 nm (MLCT
transition).
Boynton et al.
Page 2
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
We studied the steady-state luminescence response of the complex towards a well-matched
27-mer DNA duplex and the analogous DNA duplex containing a single CC mismatch
(Figure 1). Indeed, the ruthenium complex acts as a DNA light switch. Excitation in the
presence of either duplex yields emission spectra centered at 650–660 nm (Figure 1).
Importantly, we observe appreciable luminescence with the 27-mer containing the single CC
mismatch in comparison to the same 27-mer sequence lacking a mismatch. From DNA
titrations of [Ru(Me
4
phen)
2
dppz]
2+
(Figure S1), we calculate relative binding affinities of
6.8 × 10
4
M
−1
and 1.8 × 10
6
M
−1
for well-matched and mismatched sites, respectively
(Table 1). Given the 26-fold difference in binding affinities, we can conclude that the
complex is quite selective for binding to the single base mismatch.
To determine whether the differential luminescence observed in the steady-state experiments
is due not only to a higher binding affinity towards the mismatch but also to an increase in
relative emissivity, we measured excited state emission lifetimes of [Ru(Me
4
phen)
2
dppz]
2+
with the well-matched and mismatched 27-mer (Table 1). In the presence of the well-
matched sequence, a short emission lifetime equal to 35 ns is detected. However, with the
mismatched duplex, the luminescence decays as a bi-exponential function with components
equal to 33 ns and 160 ns. We attribute this additional longer lifetime component, 81% of
the overall decay, to the population of excited complex bound to the mismatch. This longer-
lived component is similar in luminescence lifetime of the complex in dry acetonitrile (Table
1). This similarity in excited state lifetime illustrates how effectively the inserted complex is
protected from quenching within its mismatched binding site. Given the similarity in
emission lifetimes between the short components of the well-matched and mismatched
sequences and their % contributions, we assign the short component to ruthenium bound to
well-matched sites, rather than enantiomeric differences.
We also investigated whether [Ru(Me
4
phen)
2
dppz]
2+
is capable of probing other types of
DNA base mismatches using hairpin oligonucleotides (Figure 2) containing a variable base
pair (XY). The emission intensity of the complex with the well-matched GC and AT
sequences is compared to GG, AA, CT, TT, CA, and CC mismatches, as well as an abasic
site (CR). The greatest emission enhancement occurs in the presence of the most
thermodynamically destabilized mismatch, CC, followed by CA. We detect negligible
enhancement with the GG mismatch as expected given its stability. Only a small
enhancement is observed with the AA mismatch, which is generally more stable than CC,
CA, and CT mismatches.
30
,
31
Figure 2 shows similar emission intensities for CT and TT
mismatches, although we might anticipate a greater emission intensity for CT based on
relative stabilities; we have previously noted that for [Ru(bpy)
2
dppz]
2+
, hydrogen bonding
interactions between thymine and the dppz ligand at the mismatch may lead to partial
quenching.
26
Note that some luminescence is evident with the fully well - matched hairpins.
We attribute this luminescence to binding at the bulged hairpin site; metalloinsertion at
bulged DNA sites has been observed.
32
We also examined the luminescence response
towards an abasic site (CR), and we find that the enhancement is comparable to that with the
CC mismatch, consistent with relative stabilities. Generally, then, the relative
thermodynamic destabilization of the mismatch site correlates with the luminescence
intensities seen in Figure 2. This dependence on the instability of the mismatch is consistent
with metalloinsertion.
33
Boynton et al.
Page 3
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Metalloinsertion by octahedral metal complexes occurs from the minor groove side of DNA.
To probe whether [Ru(Me
4
phen)
2
dppz]
2+
does in fact bind at the mismatch from the minor
groove, we tested Cu(phen)
2
2+
as a minor groove quencher (Figure 3).
27
,
34
,
35
With the DNA
mismatch, as [Cu]/[Ru] increases, there is significant quenching of [Ru(Me
4
phen)
2
dppz]
2+
luminescence (Figure 3). Conversely, with well-matched DNA, there is little change in
luminescence with increasing Cu(phen)
2
2+
concentration. These observations indicate
ruthenium binding at the mismatch
via
the minor groove, consistent with metalloinsertion.
36
Moreover, the results suggest that binding to well-matched sites by the Me
4
phen derivative
occurs through the major groove.
37
To help explain the differential luminescence observed between the mismatched and well-
matched DNA samples, we explored models of the complex bound to well-matched and
mismatched sites. Using the DNA coordinates from the crystal structure of Δ-
[Rh(bpy)
2
(chrysi)]
3+
bound by metalloinsertion to an AC mismatch,
7
we oriented Δ-
[Ru(Me
4
phen)
2
dppz]
2+
into the mismatch site from the minor groove while minimizing
steric clashes with the Me
4
phen ancillary ligands and DNA (Figure 4). From this view, we
can see that the dppz ligand is capable of deeply inserting into the mismatch site, allowing
for significant protection from quenching by water. We also modeled major groove binding
using the coordinates for intercalation by another rhodium complex.
38
For intercalation at a
well-matched site, we consider two possible binding orientations:
15
,
29
(i) the dppz ligand
intercalates in a “head-on” fashion (Figure S2), leaving both phenazine nitrogen atoms
relatively well surrounded by the base stack; (ii) the dppz ligand binds “side-on”, achieving
overlap with the base but with one of the phenazine nitrogen atoms being highly exposed to
solvent quenching (Figure 4). Given the very short 35 ns lifetime observed for the complex
with the well-matched duplex, we hypothesize that this side-on intercalation is the dominant
binding mode when [Ru(Me
4
phen)
2
dppz]
2+
is bound to a well-matched site.
39
By incorporating methyl groups onto the ancillary ligands of the [Ru(phen)
2
dppz]
2+
scaffold, we have thus prepared a luminescent light switch that is highly selective in probing
a DNA mismatch. This selectivity is the result of both a higher binding affinity towards
mismatched DNA and a longer excited state emission lifetime when bound to a mismatch.
This work demonstrates that ancillary ligand modification offers a new approach in the
design of mismatch-specific transition metal complexes. Importantly, this complex
represents a potential diagnostic probe for detecting early mismatch repair-deficient cancers.
Supplementary Material
Refer to Web version on PubMed Central for supplementary material.
Acknowledgments
We are grateful to the NIH (GM033309) for funding this work. L.M. thanks the Belgian American Educational
Foundation for the Cabeaux-Jacobs Fellowship. We also thank Sarah Antilla for assisting in studies of the complex.
REFERENCES
1. Hoeijmakers JH. Nature. 2011; 411:366–374. [PubMed: 11357144]
Boynton et al.
Page 4
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
2. Iyer RR, Pluciennik A, Burdett V, Modrich PL. Chem. Rev. 2006; 106:302–323. [PubMed:
16464007]
3. Jirincy J. Nat. Rev. Mol. Cell Biol. 2006; 7:335–346. [PubMed: 16612326]
4. Komor AC, Barton JK. J. Am. Chem. Soc. 2014; 136:14160–14172. [PubMed: 25254630]
5. Komor AC, Scheider CJ, Weidmann AG, Barton JK. J. Am. Chem. Soc. 2012; 134:19223–19233.
[PubMed: 23137296]
6. Ernst RJ, Komor AC, Barton JK. Biochemistry. 2011; 50:10919. [PubMed: 22103240]
7. Pierre VC, Kaiser JT, Barton JK. Proc. Natl. Acad. Sci. 2007; 104:429–434. [PubMed: 17194756]
8. Zeglis BM, Pierre VC, Kaiser JT, Barton JK. Biochemistry. 2009; 48:4247–4253. [PubMed:
19374348]
9. Liu L, Shao Y, Peng J, Liu H, Zhang L. Mol. BioSyst. 2013; 9:2512–2519. [PubMed: 23903702]
10. Granzhan A, Teulade-Fichou M-P. Chem. Eur. J. 2009; 15:1314. [PubMed: 19115291]
11. Takada T, Ashida A, Nakamura M, Yamana K. Bioorg. Med. Chem. 2013; 21:6011–6014.
[PubMed: 23953688]
12. Sato Y, Honjo A, Ishikawa D, Nishizawa S, Teramae N. Chem. Commun. 2011; 47:5885–5887.
13. Friedman AE, Chambron JC, Sauvage JP, Turro NJ, Barton JK. J. Am. Chem. Soc. 1990;
112:4960–4962.
14. Jenkins Y, Friedman AE, Turro NJ, Barton JK. Biochemistry. 1992; 31:10809–10816. [PubMed:
1420195]
15. Hartshorn RM, Barton JK. J. Am. Chem. Soc. 1992; 114:5919.
16. Olson EJC, Hu D, Horman A, Jonkman AM, Arkin MR, Stemp EDA, Barton JK, Barbara PF. J.
Am. Chem. Soc. 1997; 119:11458–11467.
17. Gill MR, Thomas JA. Chem. Soc. Rev. 2012; 41:3179–3192. [PubMed: 22314926]
18. Baggaley E, Weinstein JA, Williams JA. Coord. Chem. Rev. 2012; 256:1762–1785.
19. Shade CM, Kennedy RD, Rouge JL, Rosen MS, Wang MX, Seo SE, Clingerman DJ, Mirkin CA.
Chem. Eur. J. 2015; 21:10983-. [PubMed: 26119581]
20. Wachter E, Moya D, Parkin S, Glazer EC. Chem. Eur. J. 2016; 22:550–559. [PubMed: 26560887]
21. Wachter E, Howerton BS, Hall EC, Parkin S, Glazer EC. Chem. Commun. 2014; 50:311–313.
22. Pena B, David A, Pavani C, Baptista MS, Pellois JP, Turro C, Dunbar KR. Organometallics. 2014;
33:1100–1103.
23. Knoll JD, Albani BA, Turro C. Chem Commun. 2015; 51:8777.
24. Knoll JD, Turro C. Coord. Chem. Rev. 2015; 282–283:110–126.
25. Zhao R, Hammitt R, Thummel RP, Liu Y, Turro C, Snapka RM. Dalton Trans. 2009:10926–10931.
[PubMed: 20023923]
26. Lim MH, Song H, Olmon ED, Dervan EE, Barton JK. Inorg. Chem. 2009; 48:5392–5397.
[PubMed: 19453124]
27. Song H, Kaiser JT, Barton JK. Nat. Chem. 2012; 4:615–620. [PubMed: 22824892]
28. McConnell AJ, Lim MH, Olmon ED, Song H, Dervan EE, Barton JK. Inorg. Chem. 2012;
51:12511–12520. [PubMed: 23113594]
29. Holmlin RE, Yao JA, Barton JK. Inorg. Chem. 1999; 38:174–189.
30. Peyret N, Senevirante A, Allawi HT, SantaLucia J. Bio-chemistry. 1999; 38:3468–3477.
31. SantaLucia J, Hicks D. Annu. Rev. Biophys. Biomol. Struct. 2004; 33:415–440. [PubMed:
15139820]
32. Zeglis BM, Boland JA, Barton JK. Biochemistry. 2009; 45:839–849. [PubMed: 19146409]
33. Jackson BA, Barton JK. Biochemistry. 2000; 39:6176–6182. [PubMed: 10821692]
34. Sigman DS, Chen C-HB. Annu. Rev. Biochem. 1990; 59:207–236. [PubMed: 1695832]
35. Lim MH, Lau IH, Barton JK. Inorg. Chem. 2007; 46:9528–9530. [PubMed: 17918931]
36. As with other metalloinsertors, [Ru(Me
4
phen)
2
dppz]
2+
exhibits cell-selective cytotoxicity towards
mismatch repair deficient cells (See SI).
37. Quenching studies with NaI also support binding to well-matched sites from the major groove
(Figure S3).
Boynton et al.
Page 5
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
38. Kielkopf CL, Erkkila KE, Hudson BA, Barton JK, Rees DC. Nat. Stuct. Biol. 2000; 7:117–121.
39. We also prepared [Ru(5,5’-dimethylbpy)
2
dppz]
2+
(Figure S4) This light switch exhibits a 7-fold
luminescence differential between mismatched and well-matched sequences, an improvement over
[Ru(bpy)
2
dppz]
2+
but not as large as with the Me
4
phen complex.
Boynton et al.
Page 6
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Figure 1.
(Top) DNA sequences used in this study. (Bottom left) Schematic of [Ru(Me
4
phen)
2
dppz]
2+
.
(Bottom right) Steady-state luminescence spectra of
rac
-[Ru(Me
4
phen)
2
dppz]
2+
with the
well-matched (blue) duplex and with the duplex containing a single base pair CC mismatch
(red). Samples were in 5 mM tris, 200 mM NaCl, pH 7.5. [Ru] = 2 μM, [DNA duplex] = 2
μM,
λ
ex
= 440 nm.
Boynton et al.
Page 7
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Figure 2.
Plot of integrated emission intensity of [Ru(Me
4
phen)
2
dppz]
2+
(2 μM) with DNA hairpins (2
μM) containing a variable XY base pair. “R” denotes a tetrahydrofuranyl abasic site.
λ
ex
=
440 nm. Samples prepared in 5 mM tris, 50 mM NaCl, pH 7.5. Error bars indicate standard
deviations of three replicates.
Boynton et al.
Page 8
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Figure 3.
Steady-state Cu(phen)
2
2+
quenching of [Ru(Me
4
phen)
2
dppz]
2+
(2 μM) bound to well-
matched (top, blue) and mismatched (bottom, red) DNA (2 μM). Solid lines indicate no Cu
present, and dotted lines, increasing concentrations of Cu such that [Cu]/[Ru] = 7, 40, and
100, respectively.
λ
ex
= 440 nm. Samples prepared in 5 mM tris, 50 mM NaCl, pH 7.5. The
DNA sequences are as in Figure 1.
Boynton et al.
Page 9
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Figure 4.
Views down the helix axis of Δ-[Ru(Me
4
phen)
2
dppz]
2+
modeled into the crystal structures
of DNA duplexes. The ruthenium complex is shown in green with nitrogen atoms in blue.
Left: metalloinsertion at a mismatch site from the minor groove; the extruded mismatched
bases are shown in orange. Right: side-on intercalation at a well-matched site from the major
groove.
Boynton et al.
Page 10
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Boynton et al.
Page 11
Table 1
DNA Binding Affinities of [Ru(Me
4
phen)
2
dppz]
2+
and Luminescence Life-times
Well-
matched
DNA
Mis-
matched
DNA
CH
3
CN
K
a
(M
−1
)
a
6.8 × 10
4
1.8 × 10
6
-
Emis-
sion Life-
time (ns)
b
35
33 (19%)
160 (81%)
189
c
a
Titrations were performed with DNA sequences shown in Figure 1 in 5 mM tris, 200 mM NaCl, pH 7.5. [Ru] = 2 μM,
λ
ex
= 460 nm. The binding
affinity is expressed per binding site, see SI.
b
Samples containing 4 μM Ru and 4 μM DNA were prepared in 5 mM tris, 200 mM NaCl, pH 7.5 using DNA sequences shown in Figure 1.
λ
ex
=
440 nm,
λ
em
= 660 nm. Percentages reflect relative contributions of each lifetime to the overall decay.
c
Obtained in degassed, anhydrous acetonitrile.
J Am Chem Soc
. Author manuscript; available in PMC 2016 August 18.