Supramolecular helical porphyrin arrays using DNA as a scaffold

Article abstract of DOI:10.1039/b813584c

A diphenyl porphyrin substituted nucleotide was incorporated site specifically into DNA, leading to helical stacked porphyrin arrays in the major groove of the duplexes. The porphyrins show an electronic interaction which is significantly enhanced compared to the analogous tetraphenyl porphyrin (TPP) as shown in the large exciton coupling of the porphyrin B-band absorbance. Analogous to the TPP-DNA, an induced helical secondary structure is observed in the single strand porphyrin-DNA. The modified DNA can be hybridised to an immobilised complementary strand leading to fluorescent beads. The 2008 Royal Society of Chemistry.

Full text of DOI:10.1039/b813584c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Supramolecular helical porphyrin arrays using DNA as a scaffold†
Imenne Bouamaied, ThaoNguyen Nguyen, Thomas Ru¨hl and Eugen Stulz*
Received 5th August 2008, Accepted 12th September 2008
First published as an Advance Article on the web 22nd September 2008
DOI: 10.1039/b813584c
A diphenyl porphyrin substituted nucleotide was incorpo-
rated site specifically into DNA, leading to helical stacked
porphyrin arrays in the major groove of the duplexes. The
porphyrins show an electronic interaction which is signif-
icantly enhanced compared to the analogous tetraphenyl
porphyrin (TPP) as shown in the large exciton coupling of
the porphyrin B-band absorbance. Analogous to the TPP–
DNA, an induced helical secondary structure is observed in
the single strand porphyrin–DNA. The modified DNA can be
hybridised to an immobilised complementary strand leading
to fluorescent beads.
characterisation of oligo-deoxy nucleotides (ODNs) containing
various numbers of diphenyl porphyrins (DPP) and in different
sequences. The thus obtained DPP–DNA arrays show a strong
electronic coupling between the porphyrins due to the stacking
along the DNA backbone, which is much more pronounced than
in the analogous tetraphenyl porphyrin (TPP) counterparts.
The building block dUZnDPP 1 is readily accessible via Sono-
gashira coupling10 between the acetylene porphyrin and 5¢-DMT
protected 5-iodo deoxy-uridine and subsequent phosphitylation.†
Site specific incorporation of 1 into ODN strands 2 to 6 was
achieved using standard phosphoramidite solid phase synthesis
in a DNA synthesizer with an increased coupling time for 1 to
ten minutes; purification of the modified ODNs using fluorous
tag affinity chromatography11 yielded the pure strands. Thermal
denaturing with the corresponding unmodified complementary
strands (Table 1) were measured using both UV–vis and CD
spectroscopy. The Tm values obtained from the first derivatives
of the UV-melting curve (260 nm) and of the CD-melting curve
(at 252 nm) agree well. A large destabilisation of the duplex with
DTm = -21.1 ◦C was measured for one porphyrin modification in
the duplex 2·7. In the two-porphyrin duplex 3·7, the destabilisation
per porphyrin is much smaller with DTm,P = -13.3 ◦C. When more
than two porphyrins are adjacent to each other as in 4·7, the
DNA has recently become very attractive as a supramolecular
scaffold for creating functional molecules.1 The covalent attach-
ment of electronically active molecules such as metal complexes
or organic chromophores is most conveniently done at either
the nucleobase or the ribose moiety, and the substituents are
normally located in the major groove of the double helix.2
Replacement of the nucleobase, on the other hand, with aromatic
entities or metal complexes leads to arrays where the substituents
are placed in the interior of the DNA.3 We are in the course
of exploring the use of DNA as a supramolecular scaffold to
create helical multiporphyrin arrays, where the porphyrins are
attached covalently to the nucleobase deoxy-uridine via a rigid
acetylene spacer (Fig. 1).4 We have recently shown that up to
eleven tetraphenyl porphyrins can be attached to the DNA, which
is the highest number of modifications with a large substituent
to date.5 Other work in the field was concerned with the post-
synthetic modification of DNA with porphyrins to create bundles,6
direct replacement of the nucleobase with a porphyrin,7 use of
porphyrins as chirality markers8 or creation of di-porphyrin arrays
through hybridisation.9 Here, we report on the synthesis and
◦
destabilisation is only -7.1 C per porphyrin. It thus seems that
the hydrophobic interactions between the porphyrins compensate
significantly for the unfavourable steric effects. In fact, 4·7 has
a higher melting temperature than 3·7. A similar levelling of
the destabilisation was observed with the TPP analogues, but
there a significantly smaller destabilisation (DTm,P = -3.1 ◦C)
was measured.5 To test whether a levelling of the destabilisation
effect can be observed, strand 5 was synthesised where four
porphyrins are attached in a different sequence and separated by
an unmodified nucleotide. The drop in the melting temperature per
modification in 5·8 is indeed only DTm,P = -6.5 ◦C. In the case of
the five-porphyrin array 6·9, no actual melting temperature could
be observed, and probably no duplex is formed in this case.
The UV–vis spectra of the porphyrin–DNA strands are dis-
played in Fig. 2. It should be noted that, at ambient temperature,
School of Chemistry, University of Southampton, Highfield, Southampton,
UK, SO17 1BJ. E-mail: est@soton.ac.uk; Fax: +44-(0)23 80 59 68 05;
Tel: +44-(0)23 80 59 93 69
† Electronic supplementary information (ESI) available: Synthesis of 1
to 6, full spectroscopic characterisation and molecular modelling of the
DPP–DNA strands. See DOI: 10.1039/b813584c
Fig. 1 Structure of the porphyrin–dU building block and ODN sequences.
3888 | Org. Biomol. Chem., 2008, 6, 3888–3891 This journal is The Royal Society of Chemistry 2008
©

View Article Online
Table 1 Analytical data of 1 and of the porphyrin–DNA: melting points
of the DNA duplexes and spectroscopic data of the porphyrins in the
double strands
H-aggregate to give rise to the blue-shifted Soret band,
respectively.12 The broadening of the signals also suggests that
there is a certain freedom in orientation of the porphyrins, and
they are not in a strictly fixed position. The comparable energy
splitting in 3 and 5 indicate similar exciton interactions, thus the
separation of the porphyrins by an unmodified nucleotide reduces
the electronic coupling in the porphyrin array as compared to
the neighbouring arrangement in the arrays 4 and 6, where the
exciton interaction increases strongly with increasing amount of
porphyrins. The strength of the exciton coupling does not change
upon duplex formation with the complementary strand. This
is of importance when designing porphyrin arrays that could
potentially act as electronic wires.
Tm/◦Ca
DTm,P/◦Cb
lmax (loge)/nma,c
l
em/nma
1
—
406 (4.95)
631, 694
2a·7
2·7
3·7
4·7
5a·8
5·8
6a·9
6·9
63.5
42.4
36.9
42.2
65.0
38.9
42.2
n. d.
-21.1
-13.3
-7.1
412 (4.60)
386 (4.00), 412 (4.45)
376 (3.80), 416 (4.21)
635, 657, 694
635, 655, 697
636, 655, 697
-6.5
383 (4.22), 413 (4.45)
367 (3.70), 418 (4.01)
625, 658, 697
655, 714
n. d.
a Determined in 100 mM NaCl, 50 mM KH2PO4, pH 7.0, c(ODN) =
10-6 M. Strands 2a, 5a and 6a are unmodified control strands. b Decrease
in Tm per porphyrin modification. c Absorbance maxima and molar
extinction coefficients determined by Gauss deconvolution.
With the point-dipole approximation and using the energy
splitting of the Soret bands, the centre-to-centre distance in the
˚
˚
porphyrins can be estimated to be 8.66 A in 3·7, 7.38 A in 4·7
˚
and 6.69 A in 6·9.‡ The distance of the porphyrins in 5·8 would
˚
be 8.18 A, which is comparable to 3·7. However, according to the
standard DNA model, the interchromophore distance in 3·7 and
˚
5·8 would be 13.6 A (separated porphyrins), and in 4·7 and 6·9
˚
it would be 9.74 A (adjacent porphyrins). Either the point-dipole
approximation is not applicable in this case, or the chromophores
are actually much closer than would be expected by the standard
model. If the latter is the case, then this would indicate a large
distortion in the DNA structure, which would be in agreement
with the different CD spectra of the modified dsDNA compared
to natural DNA (see below).
The exciton coupling between the porphyrins also has a strong
influence on the steady state emission spectra of the arrays.
The porphyrins show a very different fluorescence spectrum as
compared to the building block 1. In 1, two emission maxima at
631 and at 694 nm with a ratio of 0.9 : 1 of the relative intensities
are observed. In the porphyrin–DNA conjugates, three maxima
at around 630, 660 and 700 nm with relative intensities of about
0.8 : 1 : 0.8 can be discerned. This is quite different to what we had
observed in the TPP–DNA arrays. However, a similar observation
between the DPP–DNA and TPP–DNA is that the fluorescence is
strongly quenched in the multiporphyrin arrays, and in 6 it is only
very weak. The difference in the shape of the spectra could arise
from the intramolecular stacking of the porphyrins on the DNA
scaffold. No intermolecular stacking is observed and the shape of
the spectra do not change in the range of 0.1 to 2 mM. At this
point it is, however, not quite clear why the emission in 2 is also
very different to 1.
The CD spectra of 2·7 to 5·8 show the characteristic B-type
DNA signature in the UV part of the spectra with a bisignate
signal having maxima at (-252)/(+277) nm (Fig. 3). Interestingly,
the signal intensity increases with increasing amount of porphyrin
modification, and the duplex 4·7 resembles the natural DNA
closest in terms of relative intensities. The porphyrin modification,
therefore, has an influence on the helical structure of the DNA
leading to local perturbance of the B-DNA, which could explain
the large drop in thermal stability of the duplexes and the
strong exciton coupling between the chromophores. Generally,
the CD signal of the porphyrin is dominated by a broad and
weak negative signal, similar to what was observed in the TPP
arrays but the signals are much less intense here. A negative
Soret band signal in non-covalent porphyrin–DNA complexes
normally is indicative of intercalation. This feature was already
Fig. 2 UV–vis and fluorescence spectra of the porphyrin–DNA single
strands. The inset shows the Gauss deconvolution of 6 as an example.
100 mM NaCl, 50 mM KH2PO4, pH 7.0, c(ODN) = 10-6 M.
a decrease of about 20% in the porphyrin absorbance upon duplex
formation is detectable. In 2, the porphyrin B-band absorbance
at 411 nm is broadened compared to the building block itself.
In the multi-porphyrin arrays, the porphyrin absorbance becomes
more and more broadened and can be deconvoluted into two
overlapping absorbances. In the array 6, the absorbance is actually
split into two distinguishable peaks at 367 and 418 nm. This is
characteristic of an exciton coupling between the chromophores
and of stacking of the porphyrins along the DNA-scaffold.
The Soret splitting is dependent on the sequence and the amount
of porphyrins in the array, and is 1422 cm-1 in 3, 2302 cm-1 in
4 and 3085 cm-1 in 6 as determined by Gauss deconvolution of
the absorbances (see Fig. 2, inset, for 6 as an example). The
array 5, which has four porphyrins but a comparable sequence
to 3 in that the porphyrins are separated by one unmodified
nucleotide, shows a Soret splitting of 1688 cm-1, similar to the
energy splitting in 3. The absorbances show a relatively large blue-
shift and a small red-shift in the split signals compared to the
single porphyrin absorbance itself. According to the point-dipole
approximation of the exciton coupling model, these split signals
arise from the interactions of the lower energy Bx and higher energy
By dipoles. The shifts indicate that By are coupled as a J-aggregate
to give rise to the red-shifted Soret band, and Bx are coupled as a
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3888–3891 | 3889
©

View Article Online
is greatly diminished. The spectra are consistent with a stacked
array in the single strands at room temperature, which is disrupted
at elevated temperatures. A transition can indeed be observed
when the thermal denaturing of the single strands is measured
at 415 nm where the porphyrins have their absorbance maximum
(Fig. 4c). All single strands show a transition which is consistent
with an unstacking of the porphyrin array. The transitions occur
at increasing temperatures from 36 to 58 ◦C with increasing
amounts of stacked porphyrins; 3 and 5 show a similar transition
temperature. The spectra obtained at 80 ◦C for the CD show that
the unstacking is not complete at this temperature. The covalent
linkage through the DNA thus retains a local arrangement where
the porphyrins can still interact electronically even at higher
temperatures.
We have also investigated the possibility of immobilising the
porphyrin–DNA strands 2 to 4 on a solid phase. Synthesis of the
complementary strand on an oligo affinity support was done using
standard DNA synthesis. The porphyrin DNA indeed hybridises
reversibly to the solid phase, leading to fluorescent beads (Fig. 5).
The melting curves show a large hysteresis between the melting and
the annealing, which reveals different kinetics in both processes.13
The averaged melting temperatures are quite similar for strands 2
and 3 at 48–50 ◦C, and much lower for 4 at ~35 ◦C (see the ESI‡).
The fluorescence spectra show similar features as obtained from
Fig. 3 CD spectra of the porphyrin–DNA duplexes. The inset shows the
magnified porphyrin Soret band region.
observed in the TPP-system and discussed in detail.5 However, to
demonstrate that the porphyrins are located in the major groove
rather than being intercalated, we performed additional melting
experiments with C–T mismatches at the porphyrin site in the
array 5·8. With four mismatches, an additional drop in Tm by
-10 ◦C was observed, which shows that the porphyrin bearing
thymidines are still involved in Watson–Crick hydrogen bonding
to the complementary nucleobase, thus the porphyrins are stacked
in the major groove of the duplex.
Analogous to the TPP–DNA, in the DPP–DNA, a secondary
helical structure seems to be induced in the single strands. This is
concluded from the spectroscopic data obtained from the single
stranded DPP–DNA at variable temperatures. In Fig. 4a and b, the
spectral data for 5 as a single strand are shown as a representative
example. The UV–vis, luminescence and CD spectra are similar to
the spectra obtained for double stranded DPP–DNA at ambient
temperature. The formation of the duplex induces, however, some
quenching of the fluorescence intensity. The CD signals of the
porphyrins are bisignate having maxima at (+370)/(-405) nm,
but are less intense for the single stranded DNA. This indicates a
stable helical conformation of the single stranded DPP–DNA as
compared to the random coil conformation of natural DNA. The
absorbances drop by about 20% upon duplex formation, thus a
more efficient stacking seems to occur after hybridisation with the
complementary strand. After heating to 80 ◦C, the UV–vis spectra
show a sharpening and increase in absorbance for the porphyrin
Soret band absorption, and in the CD spectra, the Soret signal
Fig. 5 Reversible binding of the porphyrin–DNA to a solid phase,
performed as rapidly stirred suspensions. (a) Melting profile with sigmoidal
curve fitting (2), (b) fluorescence spectra of the immobilised strands 2–4,
(c) fluorescence microscopy of the porphyrin containing beads (2).
Fig. 4 (a) UV–vis absorbance and emission spectra of 5 and 5·8 at various temperatures; (b) CD spectra of 5 at 20 and 80 ◦C, and of single stranded
DNA; (c) normalised melting profiles of the porphyrin–DNA single strands recorded at 415 nm, the temperatures indicate the reflection point of the
curves.
3890 | Org. Biomol. Chem., 2008, 6, 3888–3891
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
the strands in solution but with blue-shifted maxima at 565, 642
and 698 nm for 2. For 3 and 4, one of the emission peaks shows
additional splitting with maxima at 631 and 658 nm. The electronic
coupling between the chromophores, therefore, is retained in the
immobilised porphyrin arrays.
MacGregor, Jr. and D. A. Erie, Nano Lett., 2002, 2, 269–274; (c) S. H.
Park, C. Pistol, S. J. Ahn, J. H. Reif, A. R. Lebeck, C. Dwyer and T. H.
LaBean, Angew. Chem., Int. Ed., 2006, 45, 735–739; (d) G. A. Burley, J.
Gierlich, M. R. Mofid, H. Nir, S. Tal, Y. Eichen and T. Carell, J. Am.
Chem. Soc., 2006, 128, 1398–1399; (e) J. Wengel, Org. Biomol. Chem.,
2004, 2, 277–280.
2 (a) D. Lindegaard, A. S. Madsen, I. V. Astakhova, A. D. Malakhov,
B. R. Babu, V. A. Korshun and J. Wengel, Bioorg. Med. Chem., 2008,
16, 94–99; (b) M. Nakamura, Y. Murakami, K. Sasa, H. Hayashi and
K. Yamana, J. Am. Chem. Soc., 2008, 130, 6904–6905; (c) D. J. Hurley
and Y. Tor, J. Am. Chem. Soc., 2002, 124, 3749–3762; (d) Q. Xiao, R.
T. Ranasinghe, A. M. P. Tang and T. Brown, Tetrahedron, 2007, 63,
3483–3490; (e) M. R. Reddy, N. Shibata, Y. Kondo, S. Nakamura and
T. Toru, Angew. Chem., Int. Ed., 2006, 45, 8163–8166.
3 (a) A. Zahn and C. J. Leumann, Bioorg. Med. Chem., 2006, 14, 6174–
6188; (b) C. Brotschi, G. Mathis and C. J. Leumann, Chem.–Eur. J.,
2005, 11, 1911–1923; (c) K. Tanaka, G. H. Clever, Y. Takezawa, Y.
Yamada, C. Kaul, M. Shionoya and T. Carell, Nat. Nanotechnol., 2006,
1, 190–194.
4 (a) I. Bouamaied, L. A. Fendt, M. Wiesner, D. Ha¨ussinger, S. Tho¨ni,
N. Amiot and E. Stulz, Nucleosides, Nucleotides Nucleic Acids, 2007,
26, 1533–1538; (b) I. Bouamaied, L. A. Fendt, M. Wiesner, D.
Ha¨ussinger, N. Amiot, S. Tho¨ni and E. Stulz, Pure Appl. Chem.,
2006, 78, 2003–2014; (c) I. Bouamaied and E. Stulz, Chimia, 2005,
59, 101–104; (d) I. Bouamaied and E. Stulz, Synlett, 2004, 1579–
1583.
5 L. A. Fendt, I. Bouamaied, S. Tho¨ni, N. Amiot and E. Stulz, J. Am.
Chem. Soc., 2007, 129, 15319–15329.
6 (a) M. Endo, T. Shiroyama, M. Fujitsuka and T. Majima, J. Org. Chem.,
2005, 70, 7468–7472; (b) M. Endo, N. C. Seeman and T. Majima, Angew.
Chem., Int. Ed., 2005, 44, 6074–6077.
In summary, we have shown that changing the structure of the
porphyrin in DNA–porphyrin arrays has a large influence on
both the thermodynamic stability of the DNA duplexes as well
as of the electronic properties of the porphyrins. The diphenyl
porphyrin, as compared to the TPP analogue, is expected to have
less steric hindrance towards neighbouring porphyrins in the array,
thus closer contacts and more efficient electronic interactions are
possible. This is clearly demonstrated in the large differences of
the absorbance and emission spectra. However, the structure of
the porphyrin–DNA arrays seems to be more disrupted than in the
TPP analogous system as seen by the CD spectra and the melting
measurements. These systems are in further evaluation for their
applicability in energy or electron transfer through the porphyrins.
DNA is indeed a very suitable scaffold to produce arrays of
electronically active molecules with tunable properties, thus the
synthesis of designer molecules will be achievable when combining
different chromophores within the same or in complementary
DNA strands.
7 (a) H. Morales-Rojas and E. T. Kool, Org. Lett., 2002, 4, 4377–4380;
(b) E. T. Kool, Acc. Chem. Res., 2002, 35, 936–943.
Acknowledgements
8 A. Mammana, T. Asakawa, K. Bitsch-Jensen, A. Wolfe, S.
Chaturantabut, Y. Otani, X. Li, Z. Li, K. Nakanishi, M. Balaz, G.
A. Ellestad and N. Berova, Bioorg. Med. Chem., 2008, 16, 6544–6551.
9 (a) M. Endo, M. Fujitsuka and T. Majima, Tetrahedron, 2008, 64, 1839–
1846; (b) M. Endo, M. Fujitsuka and T. Majima, J. Org. Chem., 2008,
73, 1106–1112.
The authors wish to thank the EPSRC for financial support,
ATDbio Southampton for help in the DNA synthesis and the
group of Dr Rachel O’Reilly (Cambridge, UK) for assistance in
the recording of the CD spectra.
10 R. Chinchilla and C. Najera, Chem. Rev., 2007, 107, 874–
922.
Notes and references
11 C. Beller and W. Bannwarth, Helv. Chim. Acta, 2005, 88, 171–
179.
12 I. W. Hwang, M. Park, T. K. Ahn, Z. S. Yoon, D. M. Ko, D. Kim, F.
Ito, Y. Ishibashi, S. R. Khan, Y. Nagasawa, H. Miyasaka, C. Keda, R.
Takahashi, K. Ogawa, A. Satake and Y. Kobuke, Chem.–Eur. J., 2005,
11, 3753–3761.
13 D. A. Rusling, V. J. Broughton-Head, A. Tuck, H. Khairallah, S. D.
Osborne, T. Brown and K. R. Fox, Org. Biomol. Chem., 2008, 6, 122–
129.
‡ The distances of the porphyrins were calculated according to the
2
following equation: DE = m /2pe0R3, where DE is the energy splitting
-1
˚
(cm ), R is the centre-to-centre distance (A), and m is the transition dipole
(D). m was determined from the absorption spectrum of 2 according to ref.
14 using water as solvent (n0 = 1.333); the value obtained is 9.6 D, which
agrees well with other values.
1 (a) K. M. Stewart, J. Rojo and L. W. McLaughlin, Angew. Chem., Int.
Ed., 2004, 43, 5808–5811; (b) M. A. Batalia, E. Protozanova, R. B.
14 H. L. Anderson, Inorg. Chem., 1994, 33, 972–981.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3888–3891 | 3891
©