CuII cross-linked antiparallel dipeptide duplexes using heterofunctional ligand-substituted aminoethylglycine

Article abstract of DOI:10.1021/ic100252g

Two artificial dipeptides containing both a pendant monodentate (pyridine (py)) and tridentate (terpyridine (tpy) or phenyl terpyridine (tpy)) ligand on an aminoethylglycine (aeg) backbone have been synthesized. These oligopeptides are fully characterized by one and two-dimensional NMR spectroscopy, mass spectrometry, and elemental analysis. The ligands were chosen because they coordinate Cu²⁺ to form [Cu(py)(tpy)]²⁺ complexes; when bound to the dipeptide scaffold, Cu²⁺ chelation cross-links the strands to form double-stranded duplex structures with an antiparallel arrangement. Using spectrophotometric titrations, we observe coordination of one Cu²⁺ metal per dipeptide strand. Mass spectrometry, NMR spectroscopy, vapor pressure osmometry, and HPLC confirm that the resulting structures are the dipeptide duplex cross-linked by two metal centers.

Full text of DOI:10.1021/ic100252g

5126 Inorg. Chem. 2010, 49, 5126–5133
DOI: 10.1021/ic100252g
CuII Cross-Linked Antiparallel Dipeptide Duplexes Using Heterofunctional
Ligand-Substituted Aminoethylglycine
Matthew B. Coppock, Matthew T. Kapelewski, Hye Won Youm, Lauren A. Levine, James R. Miller, Carl P. Myers,
and Mary Elizabeth Williams*
Department of Chemistry, The Pennsylvania State University, 104 Chemistry Building, University Park,
Pennsylvania 16802
Received February 7, 2010
Two artificial dipeptides containing both a pendant monodentate (pyridine (py)) and tridentate (terpyridine (tpy) or
phenyl terpyridine (φ-tpy)) ligand on an aminoethylglycine (aeg) backbone have been synthesized. These oligo-
peptides are fully characterized by one and two-dimensional NMR spectroscopy, mass spectrometry, and elemental
analysis. The ligands were chosen because they coordinate Cu2þ to form [Cu(py)(tpy)]2þ complexes; when bound to
the dipeptide scaffold, Cu2þ chelation cross-links the strands to form double-stranded duplex structures with an anti-
parallel arrangement. Using spectrophotometric titrations, we observe coordination of one Cu2þ metal per dipeptide
strand. Mass spectrometry, NMR spectroscopy, vapor pressure osmometry, and HPLC confirm that the resulting
structures are the dipeptide duplex cross-linked by two metal centers.
Introduction
assembly of two- and three-dimensional DNA lattices1,2 and
nanoparticle crystals.3,4
An important aspect of designing potentially useful mole-
cular materials for sensors, catalysis, artificial photosynthe-
sis, and molecular wires that rival the utility and complexity
of biological systems is the ability to control the placement of
predetermined building blocks in specific geometries. Mole-
cular recognition and self-assembly are commonly used to
create complex supramolecular structures. For example,
using modified DNA to direct the arrangement of functional
species is of interest because sequence-specific base pairing
produces well-defined double-stranded geometries that can
be used as structural scaffolds. A key consideration in the
molecular design of modified DNAs is to elicit recognition at
precise locations, to induce folding into hairpin loops, or to
place tags in specific positions with respect to each other.
Modified DNA, for example, has been used to direct the
The insertion of electron donors and acceptors into oligo-
nucleotides at well-defined locations has enabled the study of
electron transfer in duplex DNA.5 The substitution of natural
base pairs with redox active species (e.g., inorganic complexes)
is potentially useful for creation of molecular circuitry. Incor-
poration of multiple metal centers into the core of duplex DNA
by replacement of nucleobases with bidentate ligands has been
previously reported.6 Heterometallic DNA structures based on
differential coordination affinity to bidentate and monodentate
ligands on the sugar phosphate backbone have recently been
described.7 Studies with the synthetic analogues peptide nucleic
acid (PNA)8 and glycol nucleic acid (GNA)9 scaffolds similarly
utilize pendant ligands to form metal coordinative cross-links.
In each of these cases, metal insertion impacted the stability of
the duplex; however molecular recognition was achieved in
part by hydrogen bonding of complementary nucleobases.
(6) (a) Meggers, E.; Holland, P. L.; Tolman, W. B.; Romesberg, F. E.;
Schultz, P. G. J. Am. Chem. Soc. 2000, 122, 10714–10715. (b) Atwell, S.;
Meggers, E.; Spraggon, G.; Schultz, P. G. J. Am. Chem. Soc. 2001, 123, 12364–
12367. (c) Tanaka, K.; Tengeiji, A.; Kato, T.; Toyama, N.; Shionoya, M. Science
2003, 299, 1212–1213.
(7) Tanaka, K.; Clever, G. H.; Takezawa, Y.; Yamada, Y.; Kaul, C.;
Shionoya, M.; Carell, T. Nat. Nanotechnol. 2006, 1, 190–194.
(8) (a) Popescu, D.-L.; Parolin, T. J.; Achim, C. J. Am. Chem. Soc. 2003,
125, 6354–6355. (b) Watson, R. M.; Skorik, Y. A.; Patra, G. K.; Achim, C. J. Am.
Chem. Soc. 2005, 127, 14628–14639. (c) Franzini, R. M.; Watson, R. M.; Patra,
G. K.; Breece, R. M.; Tierney, D. L.; Hendrich, M. P.; Achim, C. Inorg. Chem.
2006, 45, 9798–9811.
*To whom correspondence should be addressed. E-mail: mbw@
chem.psu.edu.
(1) Winfree, E.; Liu, F.; Wenzler, L. A.; Seeman, N. C. Nature 1998, 394,
539–544.
(2) Yan, H.; Park, S. H.; Finkelstein, G.; Reif, J. H.; LaBean, T. H.
Science 2003, 301, 1882–1884.
(3) Nykypanchuk, D.; Maye, M. M.; van der Lelie, D.; Gang, O. Nature
2008, 451, 549–552.
(4) Park, S. Y.; Lytton-Jean, A. K. R.; Lee, B.; Weigand, S.; Schatz, G. C.;
Mirkin, C. A. Nature 2008, 451, 553–556.
(5) (a) Lewis, F. D.; Wu, Y.; Zhang, L.; Zuo, X.; Hayes, R. T.; Wasielewski,
M. R. J. Am. Chem. Soc. 2004, 126, 8206–8215. (b) Siegmund, K.; Daublain, P.;
Wang, Q.; Trifonov, A.;Fiebig, T.;Lewis, F. D. J. Phys. Chem. B 2009, 113, 16276–
16284. (c) O'Neill, M. A.; Dohno, C.; Barton, J. K. J. Am. Chem. Soc. 2004, 126,
1316–1317. (d) Genereux, J. C.; Barton, J. K. Chem. Rev. 2010, 110, 1642-1662.
(9) (a) Schlegel, M. K.; Zhang, L.; Pagano, N.; Meggers, E. Org. Biomol.
Chem. 2009, 7, 476–482. (b) Schlegel, M. K.; Essen, L.-O.; Meggers, E. J. Am.
Chem. Soc. 2008, 130, 8158–8159.
r
pubs.acs.org/IC
Published on Web 05/12/2010
2010 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 11, 2010 5127
demonstrate a method for creating supramolecular inorganic
complexes that imitate the molecular recognition of self-
complementary DNA sequences, which improves control over
the geometry of the duplex structures and points the way
toward formation of larger and more complex architectures.
Experimental Section
Instrumentation and Analysis. Positive-ion electrospray mass
spectrometry (ESIþ) was performed at the Penn State Mass
Spectrometry Facility using a Mariner mass spectrometer (Per-
septive Biosystems). All NMR spectra were collected on 300,
360, or 400 MHz spectrophotometers (Bruker). UV-vis absorp-
tion spectra were obtained using a double-beam spectrophot-
ometer (Varian, Cary 500). Elemental analysis was conducted
by Galbraith Laboratories. Solution molecular weights were
estimated using a Wespro Vapro model 5520 vapor pressure
osmometer using 10-25 mmol/kg solutions in acetonitrile
(ACN). A standard calibration curve was obtained using solu-
tions of known concentration of tetrabutylammonium per-
chlorate in ACN.
Analytical-scale HPLC was performed with a Varian system
equipped with two quaternary pumps (Model 210), an auto-
sampler (Model 410), a UV-vis detector (Model 320), a fraction
collector (Model 701), and a Thermo Scientific Betasil Silica-100
column (150 mm ꢀ 4.6 mm with 5 μm particle size). Dipeptides 3
and 4, and their Cu(II)-containing products, were introduced in
5.0 μL injection size and analyzed using a flow rate of 0.4 mL/
min. The eluent contained 0.1% trifluoroacetic acid (TFA) in a
mixture of acetonitrile (ACN), water, and saturated aqueous
potassium nitrate in a 6:3:1 volume ratio, respectively. Elution
of the compounds was monitored at 264 nm.
Figure 1. (A) Schematic of metal-induced duplex formation of a mono-
functional, self-complementary dipeptide that can form parallel/antipar-
allel isomers and registers. (B) Schematic of metal coordination induced
antiparallel duplex assembly of a heterofunctional, self-complementary
dipeptide.
Previous work in our group has described the synthesis of
artificial oligopeptides consisting of an aminoethylglycine
(aeg) backbone with pendant ligands for the coordination to
and cross-linking by transition metal ions.10 The use of a
neutral and flexible aeg backbone prevents the inherent
electrostatic repulsion between anionic strands of DNA,
precludes electrostatic attraction of metal cations to the
backbone, has a broader range of solvent compatibility,
and eliminates the irreversible oxidative electrochemistry of
natural nucleic acids. Molecular recognition and duplex
formation, which were first shown by Gilmartin et al.,10a
are dependent solely on the robust metal to ligand bonds
without reliance on nucleic acid hydrogen bonding. In these
examples, the oligopeptides were substituted with repeating
units of the same ligand (bpy or φ-tpy) on the aeg backbone.
Although metal-coordination based self-assembly of double
stranded duplexes was demonstrated, we recognized that
both the parallel and antiparallel isomers were likely to form
in solution. In addition, as the oligopeptide chains are
lengthened to make larger and more complex architectures,
there is increased possibility of chain misalignment and alter-
native register formation, as illustrated in Figure 1A. To
combat these issues we have designed heterofunctional oligo-
peptides that mimic self-complementary sequences of DNA:
the aeg dipeptide contains both pendant monodentate
(pyridine) and tridentate (terpyridine or phenyl terpyridine)
ligands. In the presence of a four-coordinate transition metal
ion (e.g., Cu(II)) the ligands would be expected to coordinate
two metals to gain coordinative saturation, forming the
antiparallel dipeptide duplex depicted in Figure 1B. These
dipeptide strands are the first self-complementary sequences
containing two different ligands in our library.
Chemicals. All chemicals were reagent grade and used as
received unless otherwise noted. N-hydroxybenzotriazole (HO-
Bt) and 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydro-
chloride (EDC) were purchased from Advanced ChemTech.
O-Benzotriazole-N,N,N0,N0-tetramethyl-uronium-hexafluoro-
phosphate (HBTU) was purchased from NovaBiochem. All
other chemicals were purchased from Aldrich. 40-Methyl-2,20-
bipyridine-4-acetic acid,11 (4-[2,20;60,200]terpyridin-40-yl-phenyl)-
acetic acid,10d fmoc-aeg-OtBu HCl,12 fmoc-aeg(py)-OH HCl,10a
3
0 3
fmoc-aeg(bpy)-OtBu,13 fmoc-aeg(φ-tpy)-OtBu,10d 4 -methyl-
2,20:60,200-terpyridine,14 and pyridacyl pyridinium iodide15 were
synthesized as previously reported. Water was purified with a
Nanopure water system (Barnstead, 18.2MΩ).
Synthesis. General Procedures for Peptide Deprotections.
Briefly, a sample of fmoc-protected oligopeptide was stirred
for 30 min in a 20% piperidine solution in acetonitrile. The
reaction was extracted with hexanes (3 ꢀ 20 mL), and the
acetonitrile layer was flash evaporated. The residue was dis-
solved in dichloromethane (30 mL) and flash evaporated twice,
yielding the pure amine. Conversely, terminal acid formation
was accomplished by deprotection of the t-butyl terminus of the
oligopeptide using a 1:1 CF3CO2H/DCM mixture. The solution
was stirred for 1 h at 25 °C, and the solvent flash evaporated. The
residue was washed with Et2O, resulting in the acid-terminated
product.
In this initial study, we explore the use of this binding motif
to form metal-linked dipeptide duplexes and describe the
synthesis and characterization of two new heterofunctional
artificial dipeptides. The purities of the Cu-containing dipep-
tides were confirmed by high performance liquid chromato-
graphy (HPLC), and the products identified using spectro-
photometric titrations, vapor pressure osmometry, and mass
spectrometry. These results are consistent with dipeptide
duplexes each linked by two Cu(II) ions. These structures
(11) Ciana, L. D.; Hamachi, I.; Meyer, T. J. J. Org. Chem. 1989, 54, 1731–
1735.
(12) Thomson, S. A.; Josey, J. A.; Cadilla, R.; Gaul, M. D.; Hassman,
C. F.; Luzzio, M. J.; Pipe, A. J.; Reed, K. L.; Ricca, D. J.; Wiethe, R. W.;
Noble, S. A. Tetrahedron 1995, 51, 6179–6194.
(13) (a) Myers, C. P.; Gilmartin, B. P.; Williams, M. E. Inorg. Chem. 2008,
47, 6738–6747. (b) Myers, C. P.; Miller, J.; Williams, M. E. J. Am. Chem. Soc.
2009, 131, 15291–15300.
(14) Wolpher, H.; Sinha, S.; Pan, J.; Johansson, A.; Lundqvist, M. J.;
Persson, P.; Lomoth, R.; Bergquist, J.; Sun, L.; Sundstrom, V.; Akermark,
B.; Polivka, T. Inorg. Chem. 2007, 46, 638–651.
(10) (a) Gilmartin, B. P.; Ohr, K.; McLaughlin, R. L.; Koerner, R.;
Williams, M. E. J. Am. Chem. Soc. 2005, 127, 9546–9555. (b) Ohr, K.;
Gilmartin, B. P.; Williams, M. E. Inorg. Chem. 2005, 44, 7876–7885. (c)
Gilmartin, B. P.; McLaughlin, R. L.; Williams, M. E. Chem. Mater. 2005, 17,
5446–5454. (d) Ohr, K.; McLaughlin, R. L.; Williams, M. E. Inorg. Chem. 2007,
46, 965–974.
(15) Priimov, G. U.; Moore, P.; Maritim, P. K.; Butalanyi, P. K.; Alcock,
N. W. J. Chem. Soc., Dalton Trans. 2000, 445–449.

5128 Inorganic Chemistry, Vol. 49, No. 11, 2010
Coppock et al.
(40-Methyl-2,20:60,200-terpyridinyl)-acetic Acid (1). Using a
modification of a previously reported synthesis,16 an oven-dried
flask was purged and refilled 3ꢀ with N2, after which tetra-
hydrofuran (THF, 25 mL) was added. While the solvent was
stirred at -15 °C, 2,2,6,6-tetramethyl piperidine (0.77 mL,
4.4 mmol) and methyl lithium (2.5 mL, 3.9 mmol) were added
and the solution stirred for 15 min. A deaerated solution of
40-methyl-2,20:60200-terpyridine (0.87 g, 3.5 mmol) dissolved in a
minimal amount of THF was added and stirred for 30 min. CO2
(g) was then bubbled into the reaction for 1 h. The resulting solid
was collected and dissolved in 1 M HCl (20 mL). Ethanol
(EtOH, 30 mL) was added to the solution, and the solvent
volume was reduced via flash evaporation until a brown solid
began to precipitate. After cooling in the freezer overnight, a
0.801 mmol) was suspended in DCM (35 mL) and allowed to stir
for 15 min at 0 °C. To this, DIPEA (0.38 mL, 2.20 mmol) was
added, and the reaction was stirred for 1 h at 0 °C. This was
then transferred to the solution of aeg-(φ-tpy)-OtBu (0.30 g,
0.66 mmol) in DCM (15 mL), and the mixture was stirred for
48 h at 25 °C. The solution was extracted with water (3 ꢀ 20 mL),
the aqueous fractions combined and back-extracted with DCM
(20 mL). The organic layer was dried over Na2SO4, separated
from the drying agent, and flash evaporated to give a yellow oil.
The oil was chromatographed on a silica gel column, eluting
with a solvent gradient (100% DCM to 10% MeOH in DCM).
Like fractions were combined, and the solvent was removed
under vacuum to yield a yellow solid. Yield = 213 mg (38.3%).
1H NMR (360 MHz, chloroform-d, Supporting Information,
Figure S7): δ 8.73-8.67 (m, 4H); 8.65 (d, J = 7 Hz, 2H); 8.44 (d,
J = 5 Hz, 2H); 7.91-7.80 (m, 4H); 7.71 (d, J = 8 Hz, 2H); 7.55
(d, J = 8 Hz, 2H); 7.40 (d, J = 7 Hz, 2H); 7.36-7.33 (m, 4H);
7.29-7.20 (m, 2H); 7.11 (d, J = 5 Hz, 2H); 4.40-4.25 (m, 2H);
4.22-4.10 (m, 1H); 4.07-3.88 (m, 4H); 3.88-3.71 (m, 4H);
3.71-3.54 (m, 4H); 3.54-3.30 (m, 4H); 1.47 (d, 9H). HR MS
(ESI) [M þ H]þ Calcd 964.4339, Found 964.4350. Elemental
1
brown solid was collected. Yield = 400 mg (40%). H NMR
(400 MHz, DMSO-d6): δ 8.97 (s, 4H); 8.68 (s, 2H); 8.54 (t, J = 6
Hz, 2H); 7.95 (t, J = 6 Hz, 2H); 3.98 (s, 2H).
Fmoc-aeg(tpy)-OtBu (2). A solution of compound 1 (0.300 g,
1.0 mmol), EDC (0.20 g, 1.0 mmol), HOBt (0.14 g, 1.0 mmol),
and DIPEA (0.52 mL, 3.2 mmol) in DCM (50 mL) was stirred at
0 °C for 15 min. The mixture was added to fmoc-aeg-OtBu
(0.34 g, 0.79 mmol) dissolved in DCM (25 mL) and stirred at
25 °C for 48 h. The solution was extracted with water (3 ꢀ
30 mL), and the aqueous fractions were combined and back-
extracted with DCM (25 mL). The organic solutions were
combined and dried over solid Na2SO4. Solvent was removed
via flash evaporation, resulting in a yellow oil which was purified
on a silica gel column eluting with a solvent gradient (100%
DCM to 5% MeOH in DCM). The first yellow band was
collected, like fractions were combined, and the solvent removed
under vacuum yielding a yellow solid. Yield =394 mg (75%). 1H
NMR (400 MHz, chloroform-d): δ 8.64 (d, J = 9 Hz, 2H); 8.56
(d, J = 9 Hz, 2H); 8.33 (s, 1H); 8.31 (s, 1H); 7.90-7.78 (m, 2H);
7.67 (d, J = 7 Hz, 2H); 7.58 (m, 2H); 7.40 (t, J = 7 Hz, 2H);
7.34-7.27 (m, 4H); 4.44 (d, J = 7 Hz, 1H); 4.33 (d, J = 7 Hz,
1H); 4.25 (t, J = 7 Hz, 1H); 4.07-3.83 (m, 4H); 3.68-3.62
(m, 2H); 3.53 (m, 1H); 3.28 (m, 1H); 1.48 (s, 9H). MS (ESIþ)
[M þ H]þ Calcd 670.7, Found 670.4.
Anal. fmoc-aeg(py)-aeg(O-tpy)-OtBu H2O Calc: 69.64 C; 5.95
3
H; 11.40 N. Found: 69.51 C; 5.95 H; 11.41 N.
General Procedure for Reaction and Isolation of Cu(II) Com-
plexes. A solution of the peptide was prepared in MeOH and
combined with a solution containing a slight molar excess of
Cu(NO3)2 2.5H2O, stirred, and heated at 40 °C for 24 h. The
3
solvent was removed under reduced pressure, and the solid was
dissolved in a 1:4 MeOH/H2O solution. A saturated solution of
NH4PF6 was added, producing a blue to green precipitate. The
solid was collected by filtration and rinsed with H2O and Et2O.
[Cu2(3)2](PF6)4 MS (ESI) (Figure 3, Supporting Information,
Figure S14) [[Cu2(3)2](PF6)2]2þ calcd 1097.3, found 1097.3;
[[Cu2(3)2](PF6)]3þcalcd 683.2, found 683.2. 1H NMR (400
MHz, ACN-d3, Supporting Information, Figure S12): δ 7.90-
7.25 (br m, 16H); 4.40-2.10 (br m, 38H); 1.35 (br s, 18H).
[Cu2(4)2](PF6)4 MS (ESI) (Figure 3, Supporting Information,
Figure S15) [[Cu2(4)2](PF6)2]2þ Calcd 1173.3, found 1173.4;
[[Cu2(4)2](PF6)]3þ Calcd 733.9, found 734.0. 1H NMR (400
MHz, ACN-d3, Supporting Information, Figure S13): 8.00-
7.25 (br m, 24H); 4.60-2.55 (br m, 38H); 1.45 (br s, 18H).
Spectrophotometric Titrations. Titrations were performed
using methanolic solutions of known concentrations of mono-
or dipeptide and Cu(NO3)2. For measurements of visible wave-
length absorption, ∼10 mM peptide solutions were used; in the
UV region, solutions of ∼10 μM peptide were used. Solutions
were background subtracted using the double beam of the
spectrometer. Titrations either incrementally added metal ion
to peptide solution or peptide to metal ion solution, as indicated
below.
Fmoc-aeg(py)-aeg(tpy)-OtBu (3). The fmoc was cleaved from
compound 2 using the above protocol, giving aeg-(tpy)-OtBu. A
mixture of fmoc-aeg(py)-OH (0.30 g, 0.61 mmol), EDC (0.12 g,
0.61 mmol), HOBt (0.082 g, 0.61 mmol), and DIPEA (0.31 mL,
1.9 mmol) was suspended in DCM (60 mL) and allowed to stir
for 15 min at 0 °C. The solution was added to aeg-(tpy)-OtBu
(0.21 g, 0.47 mmol) dissolved in DCM (15 mL) and stirred at
25 °C for 48 h. The reaction was extracted with water (3 ꢀ
30 mL), the aqueous fractions combined and back-extracted
with DCM (20 mL). The combined organic fractions were dried
over Na2SO4, and the solvent was flash evaporated to give a
light orange solid. The solid was purified by chromatography on
a silica gel column eluting a solvent gradient (100% DCM to
5% MeOH in DCM). Like fractions were combined and flash
evaporated to yield a light orange solid. Yield = 170 mg (41%).
1H NMR (400 MHz, chloroform-d, Supporting Information,
Figure S2): δ 8.67-8.61 (m, 2H); 8.58 (d, J = 7 Hz, 2H); 8.39
(d, J = 6 Hz, 2H); 8.33 (s, 2H); 7.89-7.81 (m, 2H); 7.73 (d, J =
8 Hz, 2H); 7.66-7.50 (m, 2H); 7.37-7.27 (m, 6H); 7.06 (d, J =
6 Hz, 2H); 4.41-4.30 (m, 2H); 4.24-4.11 (m, 1H); 4.07-3.90 (m,
4H); 3.90-3.65 (m, 4H); 3.65-3.45 (m, 4H); 3.45-3.22 (m, 4H);
1.47 (s, 9H). HR MS (ESIþ) [M þ H]þ Calcd 889.3969, Found
Results and Discussion
Synthesis and Characterization of Artificial Dipeptides.
Incorporation of a tridentate ligand, in this case terpyr-
idine or phenyl terpyridine, in series with the monoden-
tate ligand pyridine onto the aminoethylglycine (aeg)
scaffold creates a bifunctional peptide with variable
affinity for metal ions. To coordinatively saturate Cu(II),
the tridentate tpy and monodentate py ligands are
expected to form [Cu(tpy)(py)]2þ complexes in a distorted
square planar geometry.17 Thus, the motif for molecular
recognition to form an antiparallel dipeptide duplex uses
ligands with coordinatively “complementary” denticities
with respect to the tetracoordinate metal ion: the dipep-
tides have been designed to self-assemble into duplexes
889.4037. Elemental Anal. fmoc-aeg(py)-aeg(tpy)-OtBu H2O
3
Calc: 67.53 C; 6.00 H; 12.35 N; Found: 67.3 C; 6.00 H; 12.2 N.
Fmoc-aeg(py)-aeg(O-tpy)-OtBu (4). The fmoc was cleaved
from fmoc-aeg-(φ-tpy)-OtBu using the above procedure, giving
aeg-(φ-tpy)-OtBu. A mixture of fmoc-aeg(py)-OH(0.27 g,
0.801 mmol), EDC (0.15 g, 0.801 mmol), and HOBt (0.10 g,
(16) Potts, K. T.; Usifer, D. A.; Guadalupe, A.; Abruna, H. D. J. Am.
Chem. Soc. 1987, 109, 3961–3967.
(17) Su, C.-C.; Li, C.-B. Polyhedron 1994, 13, 825–834.

Article
Inorganic Chemistry, Vol. 49, No. 11, 2010 5129
Scheme 1. Terpyridine Monomer Synthetic Strategy; (i) HBTU, HOBt, DIPEA; (ii) 20% Piperidine in ACN
Scheme 2. Dipeptide Synthetic Strategy; (i) EDC, HOBt, DIPEA
upon chelation with tetracoordinate metals such as
Cu(II). The two tridentate ligands selected for these
studies have differing length attachments to the back-
bone, which would impact the relative locations of the Cu
cross-links (vide infra).
In both cases, the dipeptide product was purified by
column chromatography, and purity assessed by HPLC:
Figure 2 contains the chromatograms of dipeptides 3
and 4 (red lines), which have a single, sharp peak
indicative of a single, pure species. Additional confir-
mation of purity by elemental analysis, and identity by
NMR spectroscopy and positive ion electrospray mass
spectrometry (vide infra), are used to fully characterize
the dipeptides. The overall reactions produced 41%
yield for fmoc-aeg(py)-aeg(tpy)-OtBu (3) and 38% yield
for fmoc-aeg(py)-aeg(φ-tpy)-OtBu (4) dipeptides, both
of which are higher yields and in larger quantities than
those obtained for the solid-phase peptide synthesis of
φ-tpy substituted dipeptides (i.e., 23%).10d
The termini of the aminoethylglycine backbone have
been protected with fmoc and t-butyl ester as in previous
studies.10 Since the fmoc group is base labile and the
t-butyl ester group is acid labile, selective deprotection
followed by extension of the peptide chain was achieved
from both termini. Solution phase synthesis of these
dipeptides was used to enable bidirectional chain lengthen-
ing in future studies as well as to increase yields com-
pared to solid phase synthesis. The purification techniques
used in solution based synthesis allow for isolation of
greater amounts of product than solid phase methods.10a
For the synthesis of a dipeptide with pendant pyridine
and either terpyridine or phenyl terpyridine ligands,
fmoc-aeg(py)-OtBu and fmoc-aeg(φ-tpy)-OtBu mono-
mers were prepared according to previous work.10b,d
Analogously, terpyridine acetic acid was prepared using
modified syntheses and coupled to the aeg backbone with
our standard conditions (Scheme 1).10d The fmoc protect-
ing group was selectively removed from the tridentate
monomers under mild basic conditions, resulting in pri-
mary amine termini. This method was chosen to reduce
the aromatic bulkiness of the monomers and maximize
the coupling yield. Shown in Scheme 2, the tridentate
monomers (aeg(tpy)-OtBu or aeg(φ-tpy)-OtBu) were sepa-
rately combined with acid-terminated pyridine monomer
(fmoc-aeg(py)-OH) and EDC/HOBt coupling reagents.
1
The H NMR spectra of the dipeptides yielded the
anticipated relative proton integrations (i.e., 9 t-butyl
protons, 18 aliphatic protons, and 22 aromatic protons
for dipeptide 3, and 9 t-butyl protons, 18 aliphatic pro-
tons, and 26 aromatic protons for dipeptide 4). The
protons assigned to the fmoc protecting group, as well
as the pyridine ligand, appear at the same positions in
both NMR spectra (see Supporting Information, Figures
S2 and S7). Protons on the heterocyclic terpyridine ring in
each of the tridentate ligands have similar resonances, but
the phenyl ring shifts the terpyridine protons slightly
downfield, whereas proximity to the backbone shifts
these slightly more upfield in dipeptide 4.
Structural characterization of the dipeptides was there-
fore further confirmed using 1H-1H COSY, 13C-1H
HMQC, and 13C-1H HMBC. The COSY spectra clearly
show the couplings between aromatic protons and between

5130 Inorganic Chemistry, Vol. 49, No. 11, 2010
Coppock et al.
Figure 2. High performance liquid chromatograms using a silica col-
umn (150 mm ꢀ 4.6 mm with 5 μm particle size) and 6:3:1 mixture of
ACN/H2O/KNO3 with 0.1% TFA and a flow rate of 0.4 mL/min.
Injection volume is 5 μL. (A) [Cu2(3)2] (black line) versus 3 (red line);
(B) [Cu2(4)2] (black line) versus 4 (red line).
neighboring protons on the aminoethyl glycine backbone.
For example, in the COSY spectrum of compound 3, the
presence of the pyridine ligand is confirmed in both dipep-
tides by the presence of a peak corresponding to two
protons at 7.1 ppm (Supporting Information, Figure S2,
peak k) correlated to a second peak corresponding to the
two protons at 8.4 ppm (Supporting Information, Figure
S2, peak l). The terpyridine protons closest to the nitrogens
∼8.8 ppm (Supporting Information, Figure S2, peak q)
couple with their neighboring protons ∼7.3 ppm (Sup-
porting Information, Figure S2, peak p) and enabled the
assignments for the other terpyridine protons in the COSY
spectrum. The final two protons on the terpyridine rings
located at 8.3 ppm (Supporting Information, Figure S2,
peak m) in the 1D 1H NMR spectrum exhibited no
coupling in the COSY spectrum since these are electro-
nically isolated. In contrast, the COSY spectrum of mole-
cule 4 contains a peak corresponding to two protons at
7.9 ppm (Supporting Information, Figure S7, peak m) that
couples with a second peak corresponding to two protons
around 7.3 ppm (Supporting Information, Figure S7, peak
n); these are assigned to the phenyl ring on the ligand. In
both COSY spectra, there is correlation between the pro-
tons of the fmoc protecting group (4.1-4.4 ppm, Support-
ing Information, Figures S2 and S7 peaks e, f) and the
bridging ethyl group (3.2-3.7 ppm, Supporting Informa-
tion, Figures S2 and S7 peaks i, j, h, g). Taken together, the
HMBC and HMQC spectra conclusively confirm the
identity of the dipeptides.
Figure 3. Positive ion electrospray mass spectra of (A) [Cu2(4)2]4þ and
(B) [Cu2(3)2]4þ, which are observed together with associated PF6- anions
(as indicated). The theoretical isotope patterns for the þ2 and þ3 charged
-
ions (containing two and one PF6 anion, respectively) are shown for
comparison.
calculated 889.3969, found 889.4037) and dipeptide 4
((M þ H)þ: calculated 964.4339, found 964.4350) and
further confirm their identities. Elemental analysis
affirmed the purity of the two dipeptides, which were
observed to be hygroscopic.
Synthesis and Characterization of Duplexes. Addition
of Cu(II) to each dipeptide is expected to cause forma-
tion of [Cu(py)(tpy)]2þ complexes, resulting in coordi-
native cross-links between the strands. To form the
duplexes, the peptide strands must coordinate two metal
ions and align in an antiparallel fashion, as in Figure 1B.
Copper has initially been used for duplex self-assembly
because it is sufficiently labile to allow the formation of
Mass spectrometry was used for additional characteri-
zation of the dipeptides. High resolution molecular ion
peaks were observed for dipeptide 3 ((M þ H)þ m/z:

Article
Inorganic Chemistry, Vol. 49, No. 11, 2010 5131
Table 1. UV-vis Absorbance Data for the Monomers and Dipeptides
molecule
λabs a, nm (ε, 103 M-1 cm-1
)
max
py monomer
tpy monomer (2)
264 (34.5), 278 (19.7)
276 (36.1), 300 (23.3)
Φ-tpy monomer
3
4
255 (60.1), 267 (65.2), 275 (66.1)
264 (34.3), 276 (33.7), 300 (20.3)
255 (53.7), 267 (56.8), 278 (57.7)
a Wavelength of the maximum absorbance of the peak.
the most thermodynamically stable structure to be obtai-
ned under mild heating. Addition of Cu(II) to a metha-
nolic solution of the dipeptides while refluxing resulted in
methanol-soluble product, which was then isolated by
diluting the solution with water and precipitating the
complex with ammonium hexafluorophosphate. To as-
sess the purity of the resulting Cu-containing products,
HPLC was used and the chromatograms compared to the
unmetalated dipeptides; Figure 2 shows these chromato-
grams obtained under identical conditions. The indi-
vidual chromatograms of dipeptides 3 and 4, and those
of the Cu(II)-containing products, contain a single, sharp
peak. We note the presence of a small side peak in the
chromatogram of both 4 and its Cu complex, but in such a
low relative quantity that it represents a very minor
impurity. Because the eluent (0.1% TFA in 6:3:1 ACN/
H2O/aqueous saturated KNO3) is highly polar, in com-
parison with the pure peptides, the metalated products
elute at an earlier time because of their greater charge.18
In the chromatograms of the metalated dipeptides, we do
not observe a peak at a time expected for pure dipeptide,
and the peak widths are approximately the same for
metalated versus unmetalated molecules. These HPLC
data therefore confirm the presence of a single Cu-con-
taining product following the reaction of Cu(II) with each
of the dipeptides. Separate vapor pressure osmometry
experiments of the Cu complexes in acetonitrile solutions
provide molecular weights that are consistent with the
dipeptide duplexes linked by two metal ions (2200-
2900 g/mol), although imprecision in these values results
in part from varying amounts of residual solvent present
in the Cu complexes.
Figure 4. Spectophotometric titrations of the reaction of Cu2þ with
dipeptides: (A) 7.64 mM dipeptide 4 titrated in 50 μL increments into a
2.0 mL volume of 1.91 mM Cu(NO3)2; and (B) addition of 50 μL incre-
ments of 9.55 mM dipeptide 3 into 2.0 mL of 1.91 mM of Cu(NO3)2 . All
solutions are in methanol. Insets: change in the absorbance at 650 nm as a
function of the relative molar ratio of added peptide to Cu2þ
.
of these would likely be more difficult to ionize and there-
fore not appear in the mass spectra, it is important to note
that only a single peak is observed in each of the chromato-
grams. For a single, higher order structure to form in solu-
tion and then decompose in the gas phase into only the
[Cu2(peptide)2]4þ structure (with associated anions), and
not [Cu(peptide)]2þ or [Cu3(peptide)3]6þ, etc.), is unlikely.
Therefore, we use the sum of the data to conclude that even
if higher order polymeric structures form in the reaction
solution, these are present in extremely small quantities, are
most likely insoluble and therefore not detected by HPLC
or mass spectrometry, and the majority of the reaction
product is the dipeptide duplex linked by two metal ions.
Therefore we employed positive ion electrospray mass
spectrometry (ESIþ MS) to confirm the identities of the
products. Using conditions for the mass spectrometric
experiments from Gatlin et al.,19 mass spectra were obtai-
ned as shown in Figure 3 (full spectra are shown in Sup-
porting Information, Figures S14 and S15). Molecular
ion peaks corresponding to the þ2 and þ3 duplexes were
found for each of the metalated compounds with asso-
1
Finally, we acquired H NMR spectra of these com-
pounds because the presence of Cu(II) results in a para-
magnetic shift of the ligand protons to which it coordi-
nates, and broadens nearby protons. An absence of ligand
protons in the 1H NMR spectra of [Cu2(3)2]4þ and [Cu2-
(4)2]4þ (Supporting Information, Figures S12 and S13)
verified that the Cu(II) was coordinated to both py and
tpy ligands (in 3) or py and φ-tpy ligands (in 4). Although
the aminoethylglycine backbone is flexible, because of the
short dipeptide length, the two ligands are spaced closely
and are not able to coordinate to the same metal ion.10a,b
The NMR results confirm that the Cu ion is not solely
coordinated to the tridentate ligands, but also binds py close
in its coordination shell. To accomplish this, the Cu ions
must cross-link ligands on separate strands. Taken together,
the HPLC, NMR, vapor pressure osmometry, and mass
spectrometry data point toward the formation of the species
-
ciated PF6 anions: the þ2 peaks are identified as the
species [Cu2(3)2(PF6)2]2þ or [Cu2(4)2(PF6)2]2þ, whereas
the þ3 peaks are [Cu2(3)2(PF6)]3þ and [Cu2(4)2-
(PF6)]3þ 13b
These observed peaks confirm the duplex
.
identities. Similar to our prior reports,1 although the mass
spectra are complex and a significant number of frag-
ments are observed, we do not observe molecular ion
peaks for higher order structures (e.g., [Cun(peptide)n-
(PF6)x]2n-x) or monometallic species (e.g., [Cu(pep-
tide)(PF6)x]2-x). Although it is possible that the former
(18) Elliott, C. M.; Freitag, R. A.; Blaney, D. D. J. Am. Chem. Soc. 1985,
107, 4647–4655.
(19) Gatlin, C. L.; Turecek, F. J. Mass Spectrom. 2000, 35, 172–177.

5132 Inorganic Chemistry, Vol. 49, No. 11, 2010
Coppock et al.
Figure 5. Difference spectra acquired during the spectophotometric titration of the reaction of Cu2þ with (A) 2.0 mL of 10 μM dipeptide 4; and (B) 2.0 mL
of 10 μM dipeptide 3 in methanol. Cu2þ is added in 50 μL increments of 40 μM of Cu(NO3)2. (C) and (D) are plots of the change in the absorbance at the
indicated wavelength as a function of the relative molar ratio of added Cu2þ to 6 and 5, respectively. Blue symbols correspond to ΔA values on left axis; red
symbols on right axis.
containing two dipeptides cross-linked by two metal ions, as
depicted in Figure 1B.
absorbance at the peak maxima are plotted versus the relative
ratio of peptide to Cu, and show an increase in absorbance as
dipeptide is added. In the case of 4, the maximum absorbance
occurs at a mole ratio of ∼1:1 dipeptide/Cu and then slightly
decreases. In contrast, the titration curve of 3 in Figure 4A
has a maximum absorbance at ∼0.75 Cu/dipeptide and falls
with additional aliquots of dipeptide. These stoichiometric
points are consistent with either a single Cu coordinated to a
single dipeptide, or two Cu ions linking two dipeptide
strands. Because of the steric hindrance imparted by the close
proximity of the two ligands on each strand, and using our
understanding of the products of the dipeptide reactions with
Cu (vide supra), we conclude from these data that the
[Cu2(peptide)2]4þ structures are formed at the ∼1:1 stoichio-
metric point. In the Cu titrations of both 3 and 4, we
hypothesize that the decrease in absorbance after this stoi-
chiometric point to formation of [Cu(tpy)]2þ and [Cu(py)]2þ
complexes at the expense of metal-linked duplex. That is,
addition of extra ligand shifts the equilibrium from [Cu-
(tpy)(py)]2þ to (primarily) [Cu(tpy)] and [Cu(py)]; the later
occurrence in the titration curve of 4 suggests that this Cu-
linked duplex is more stable than for 3, and therefore larger
amounts of Cu(II) are required to dis-associate the structure.
Ongoing experiments aim to specifically measure association
constants and stabilities.
Metal Coordination to Dipeptides
The above experiments demonstrated the self-assembly of
the di-Cu linked dipeptide duplexes. To begin to understand
how these structures form in the presence of copper, a series
of spectrophotometric titrations were also performed. Dipep-
tides 3 and 4 were first studied using UV-visible spectro-
photometry to give quantitative measures of their extinction
coefficients (Table 1). The UV-vis spectra of the non-
metallated dipeptides are dominated by absorbances due to
the πtpy-π*tpy and the πφ-tpy-π*φ-tpy transitions at 260-
310 nm in 3 and 240-290 nm in 4, respectively, which are
consistent with other terpyridine and phenyl terpyridine
systems.20 The pyridine ligand also has a πpy-π*py transition
absorbance at 264 nm, but it is not easily identified because
this region of the spectrum contains many overlapping peaks.
Upon Cu(II) coordination to the dipeptides, the π-π*
absorption bands of the ligands shift to 310-350 nm, and a
band appears in the visible region at ∼650 nm.
Spectrophotometric titrations at higher concentrations
were performed so that the low energy transition (650 nm)
could be monitored. Figure 4 contains the difference spectra
obtained during the titration of compounds 3 or 4 into a
solution containing Cu(II). Since neither dipeptide absorbs
light in this region of the visible spectrum, this peak is
assigned to the d-d transition of the resulting Cu complex.10b
In the Figure 4 insets, plots of the incremental change in
Using lower concentrations to examine the UV region of
the spectrum provides insight into the changes observed
ligand-centered transitions, and approaches the complexa-
tion equilibria from the opposite direction. Separate titra-
tions of Cu(II) into solutions of the dipeptides reveal
isosbestic points at 270 and 310 nm for reaction with 3, and
at 240 and 290 nm for reaction with 4. The difference spectra
(20) Araki, K.; Arita, S.; Cheon, J.-D.; Mutai, T. J. Chem. Soc., Perkin
Trans. 2001, 2, 1045–1050.

Article
Inorganic Chemistry, Vol. 49, No. 11, 2010 5133
Figure 6. Titration of a 2.0 mL methanolic solution containing (A) 10 μM fmoc-aeg(py)-OH and 10 μM 2 and (B) a 2.0 mL solution containing 10 μM
fmoc-aeg(py)-OH and 10 μM fmoc-aeg(φ-tpy)-OtBu, with 25 μL increments of 50 μM of Cu(NO3)2. Plots of the change in absorbance at the indicated
wavelengths as a function of the relative molar ratio of added Cu2þ to peptides: (C) fmoc-aeg(py)-OH and 2; (D) fmoc-aeg(py)-OH and fmoc-aeg(φ-tpy)-
OtBu . Blue symbols correspond to ΔA values on left axis; red symbols on right axis.
in Figure 5 show that as Cu(II) is added to the solution of
dipeptide, the absorbance of the free ligand decreases and a
concomitant increase in the metal-coordinated ligand absor-
bance is observed. Plotting the intensity of the new absor-
bance peak versus the molar ratio of Cu(II) to dipeptide
results in the titration plots shown in Figures 5C and D which
increase and reach a constant absorbance at a ∼1:1 ratio of
Cu/dipeptide. This stoichiometric ratio is again associated
with two Cu ions linking two dipeptides. At select wavelengths,
the absorbance continues to vary beyond the duplex stoichio-
metric point as additional Cu2þ is added, consistent with the
hypothesis that beyond this point other complexes (e.g., [Cu-
(tpy)(H2O)]2þ and [Cu(py)(H2O)3]2þ) could be forming.
For comparison, spectrophotometric titrations of Cu(II)
into solutions containing unlinked ligand-modified aeg
monomers at the identical concentrations to the experiments
in Figure 5 were also performed, and these are shown in
Figure 6. The difference spectra acquired for these titrations
contain isosbestic points at the same wavelengths as the
dipeptide counterparts (Figure 5). Decreasing absorbance is
observed at wavelengths associated with unmetalated ligand
in both Figures 5C and D and 6C and D as the ligands
coordinate to Cu(II). Concomitant absorbance increases are
observed at new wavelengths that are attributed to metal
complex formation. Comparison of the extinction coeffi-
cients (ε) obtained for the unlinked ligands versus metalated
dipeptide is listed in Table 2. These data show that ε is
approximately twice as large in the Cu-containing dipeptide,
indicative of two Cu centers per dipeptide duplex.
Table 2. UV-vis Absorbance Data for Cu Complexes
complex
λabs a, nm (ε, 103 M-1 cm-1
)
max
[Cu(py)(tpy)]2þ
[Cu2(3)2]4þ
282 (30.8), 326 (15.0), 338 (14.1)
282 (43.7), 326 (28.5), 338 (24.7)
275 (61.5), 290 (53.6), 332 (24.6)
275 (74.5), 290 (84.8), 332 (47.5)
[Cu(py)(Φ-tpy)]2þ
[Cu2(4)2]4þ
a Wavelength of the maximum absorbance of the peak.
Figure 5, and the inflection point occurs beyond the expected
1:1 molar ratio of Cu to dipeptide. This result suggests a higher
binding affinity when the ligands are tethered on the dipeptide,
consistent with positive cooperativity. Because of the extensive
overlap and complexity of the absorbance spectra, continuing
studies are underway to quantify the apparent cooperative
effects and binding affinities during formation of the dipeptide
duplexes. In addition to quantitatively understanding the
association constants for the formation of the metal-linked
duplexes, our longer term objectives include examining the role
of artificial peptide length on stability, and testing the specificity
and selectivity of recognition between strands analogous to
sequence-specific DNA hybridization.
Acknowledgment. We gratefully acknowledge support
of this work by a grant from the National Science Found-
ation (CHE-0718373). We thank Megan Andes and
Christine Keating for assistance with the vapor pressure
osmometry measurement.
Supporting Information Available: Complete 13C NMR of the
tpy monomer, 1H NMR, COSY, HMBC, HMQC spectra,
and mass spectra of the dipeptides and Cu-linked duplexes.
This material is available free of charge via the Internet at
http://pubs.acs.org.
The titration plots in Figures 6 C and D have an inflection
point, suggesting a similar bonding motif of Cu(II) with the
ligands (e.g., [Cu(py)(tpy)]2þ). However, when the monomers
are unlinked, the rise of the titration curve is shallower than in