A G-quartet formed in the absence of a templating metal cation: A new 8- (N,N-dimethylaniline)guanosine derivative

Article abstract of DOI:10.1002/(SICI)1521-3773(20000403)39:7<1300::AID-ANIE1300>3.0.CO;2-I

Contrary to what was assumed previously, self-assembled G-quartets may be obtained even in the absence of a templating alkali metal cation. This observation leads to the suggestion that this structural motif, which has been implicated in a variety of biol

Full text of DOI:10.1002/(SICI)1521-3773(20000403)39:7<1300::AID-ANIE1300>3.0.CO;2-I

COMMUNICATIONS
[4] S. Mecking, L. K. Johnson, L. Wang, M. Brookhart, J. Am. Chem. Soc.
1998, 120, 888.
[5] M. P. Coles, C. I. Dalby, V. C. Gibson, W. Clegg, M. R. J. Elsegood, J.
Chem. Soc. Chem. Commun. 1995, 1709.
[6] V. C. Gibson, E. L. Marshall, C. Redshaw, W. Clegg, M. R. J. Else-
good, J. Chem. Soc. Dalton Trans. 1996, 21, 4197.
[7] J. D. Scollard, D. H. McConville, J. Am. Chem. Soc. 1996, 118, 10008.
[8] J. D. Scollard, D. H. McConville, J. J. Vittal, Organometallics 1995,
5478.
A G-Quartet Formed in the Absence of a
Templating Metal Cation: A New 8-(N,N-
dimethylaniline)guanosine Derivative**
Jonathan L. Sessler,* Muhunthan Sathiosatham,
Katherine Doerr, Vincent Lynch, and
Khalil A. Abboud
Â
[9] D. W. Stephan, F. Guerin, R. E. v. Spence, L. Koch, X. Gao, S. J.
Dedicated to Professor Jean-Marie Lehn
on the occasion of his 60th birthday
Brown, J. W. Swabey, Q. Wang, W. Xu, P. Zoricak, D. G. Harrison,
Organometallics 1999, 17, 2046.
Â
[10] D. W. Stephan, J. C. Stewart, F. Guerin, R. E. v. Spence, W. Xu, D. G.
Self-assembly is one of the more intriguing subfields within
the general area of molecular recognition. Indeed, quite a
number of natural and unnatural molecules form ordered
supramolecular ensembles by self-assembly.[1] Notable among
these is guanosine, a compound known to self-assemble into
dimeric,[2] tetrameric,[3] ribbonlike,[4] and helical structures[5]
as the result of Watson ± Crick and Hoogsteen base-pairing
interactions. The tetrameric forms of guanosine, known also
as G-quartets, have drawn considerable attention over the
past three decades. They have been implicated in a variety of
biological functions[6] and have seen application in a range of
nonbiological areas such as ionophores[7] and phase-transfer
catalysts.[8] X-ray analyses of crystals obtained from aqueous
solutions of 5'- and 3'-guanosine monophosphate revealed
that the constituent bases adopt a tetrameric planar arrange-
ment, which then stacks in a helical fashion with other
tetramers to form columnar aggregates.[9]
Harrison, Organometallics 1999, 17, 1116.
[11] X. Yang, C. L. Stern, T. J. Marks, J. Am. Chem. Soc. 1994, 116, 10015.
[12] X. Yang, C. L. Stern, T. J. Marks, J. Am. Chem. Soc. 1991, 113, 3623.
[13] Y.-X. Chen, C. L. Stern, T. J. Marks, J. Am. Chem. Soc. 1997, 119, 2582.
[14] P. A. Deck, T. J. Marks, J. Am. Chem. Soc. 1995, 117, 6128.
[15] X. Y. Chen, C. L. Stern, S. Yang, T. J. Marks, J. Am. Chem. Soc. 1996,
118, 12451.
[16] Z. Wu, R. F. Jordan, J. L. Petersen, J. Am. Chem. Soc. 1995, 117, 5867.
[17] R. F. Jordan, W. E. Dasher, S. F. Echols, J. Am. Chem. Soc. 1986, 108,
1718.
[18] R. F. Jordan, C. S. Bajgur, R. Willett, B. Scott, J. Am. Chem. Soc. 1986,
108, 7410.
[19] M. Bochmann, S. J. Lancaster, M. B. Hursthouse, K. M. A. Malik,
Organometallics 1994, 13, 2235.
[20] Catalyst screening experiments were performed as detailed in
ref. [10]. Polymers obtained by using this catalyst have been described
in ref. [9].
[21] Crystal structure analysis of 3: monoclinic, space group C2/c, a ˆ
23.417(9), b ˆ 14.776(6), c ˆ 22.678(8) Š, b ˆ 109.98(3)o, Vˆ
7374.2(49) Š3, Z ˆ 4, 1calcd ˆ 1.535, m ˆ0.421, 5661 reflections, 486
parameters, R ˆ 0.0823, Rw ˆ 0.1827, GOF ˆ 1.002. Diffraction experi-
ments were performed on a Siemens SMART System CCD diffrac-
tometer, and the structure was solved with the SHELX-TL software
package. Crystallographic data (excluding structure factors) for the
structure reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary publica-
tion no. CCDC-134941. Copies of the data can be obtained free of
charge on application to CCDC, 12 Union Road, Cambridge
CB21EZ, UK (fax: (‡44)1223-336-033; e-mail: deposit@ccdc.cam.
ac.uk).
[22] W. Lasser, U. Thewalt, J. Organomet. Chem. 1986, 331, 69.
[23] R. F. Jordan, S. F. Echols, Inorg. Chem. 1987, 26, 383.
[24] P. N. Billinger, P. P. K. Claire, H. Collins, G. R. Willey, Inorg. Chim.
Acta 1988, 149, 63.
[25] T. K. Hollis, N. P. Robinson, B. Bosnich, J. Am. Chem. Soc. 1992, 114,
5464.
Currently, it is believed that templating alkali metal cations
‡
‡
such as Na and K are needed to stabilize the formation of
G-quartets.[10] The need for such cations to stabilize self-
assembled and higher order aggregates formed from lip-
ophilic deoxyguanosine derivatives in chlorinated organic
solvents (that is, G-quartet model systems) has also been
explicitly noted.[4, 7] Indeed, in the absence of alkali metal
cations these compounds were found to adopt a ribbonlike
structure (but not G-quar-
tets) in both chloroform
solution[4a] and in the solid
state.[4b] Here, we report
the synthesis of the new
(N,N-dimethylaniline)gua-
nosine derivative 1 that
forms a G-quartet in the
[26] B. Bosch, G. Erker, R. Frohlich, O. Meyer, Organometallics 1997, 16,
5449.
[27] M. L. H. Green, J. Sassmannshausen, Chem. Commun. 1999, 115.
absence of alkali metal
[*] Prof. J. L. Sessler, M. Sathiosatham, K. Doerr, Dr. V. Lynch
Department of Chemistry and Biochemistry
University of Texas at Austin
Austin, TX 78712 (USA)
Fax : (‡ 1)512-471-7550
E-mail: sessler@mail.utexas.edu
K. A. Abboud
Department of Chemistry
University of Florida
Gainesville, FL 32611 (USA)
[**] This work was supported by the Robert A. Welch foundation. We
thank Dr. Ben Shoulders and the staff at the NMR facility at the
University of Texas for their assistance.
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/angewandte/ or from the author.
1300
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3907-1300 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 7

COMMUNICATIONS
respect to the relevant N9 C1' glycosidic
bonds. Such an all-syn arrangement presumably
reflects either a) unfavorable steric interactions
between the isobutyryl protecting group and
the dimethylaniline moiety that would exist in
the anti conformation,[13] or b) the possibility of
hydrogen bond formation between the C5'
isobutyryl carbonyl group and the (N2)H*, or
c) some combination of these two effects.
The various guanosine and deoxyguanosine
derivatives soluble in organic solvents whose
self-assembly properties were studied previous-
ly all lack the aryl substituent present in 1. They
thus adopt an anti conformation both in sol-
Scheme 1. Synthesis of guanosine derivative 1. a) BuLi, THF,
788C; b) nBu3SnCl,
788C !room temperature, 90%; c) [Pd(PPh3)4], toluene, reflux, 95%; d) NH3, MeOH,
CH2Cl2; e) (Me2CHCO)2O, DMAP, TEA, MeCN, 50%. DMAP ˆ 4-dimethylaminopyridine,
TEA ˆ triethylamine.
cations in both chlorinated organic solvents and in the solid
state.
Scheme 1 outlines the synthesis of 1. Briefly, N,N-dimethyl-
4-(tri-n-butylstannyl)aniline (3) was synthesized from com-
mercially available 4-bromo-N,N-dimethylaniline (2) by treat-
ment with BuLi in THF at
788C, followed by tri-n-
butylstannyl chloride. Coupling of this tin derivative with
the 8-bromoguanosine derivative 4[11] under Stille cross-
coupling conditions afforded N-protected 5 which was con-
verted to 1 by using standard deprotection and protection
procedures.[11]
Crystals of 1[12] were obtained by slow evaporation of a
solution of 1 in acetonitrile/isopropyl alcohol. An X-ray
crystallographic analysis of 1 revealed that it forms H-bonded
head-to-tail[10a] tetramers in the absence of an alkali metal
cation template (Figure 1). Each individual tetramer (see
Figure 2 for a schematic representation) possesses C4 sym-
metry along the c axis and has all four guanine bases within an
essentially planar array. The protected ribose subunits occupy
the space above this plane and adopt syn conformations with
Figure 2. Schematic representations of anti (a) and syn conformers of
guanosines (b) that can arise as the result of rotation about the purine ±
ribose bond. c) Truncated view of the ribbonlike structures observed for
lipophilic guanosine derivatives that are capable of adopting an anti
conformation. d) Idealized view of G-quartet-like tetramer formed from 1
both in solution in dichloromethane and in the solid state (the p-NMe2C6H4
substituent at C8' has been omitted for clarity). This structure displays a syn
arrangement of the ribose (Rib) and purine subunits. HG and H* indicate
the amino protons that are, and are not, participating in G-quartet
formation, respectively.
ution and in the solid state. They also self-assemble in the
form of ribbonlike structures rather than G-quartets in the
‡
absence of K or other templating cations.[4] By contrast 1,
presumably as a consequence of adopting a syn conformation,
self-assembles to form a cyclic tetramer. In the syn conforma-
tion, one of the faces on the guanosine subunit is blocked
(Figure 2). Apparently, this prevents the formation of ribbon-
like supramolecular structures as is seen in the case of
compounds that are capable of adopting an anti conformation.
The putative formation of a G-quartet-like structure from 1
was studied in organic solutions by 1H NMR spectroscopy. In
Figure 1. Ball-and-stick representation of the tetrameric arrangement of
molecules of 1 in the solid state. The geometry of the hydrogen bonding
interaction is: N11 H11b ´´´ N3A (related by y, 1 x, z), N ´´´ N 2.927(2) Š,
N ´´´ H 2.107(2) Š, N H ´´´ N 159.1(1)8; N6 H6 ´´´ O10A (related by y,
1
x, z), N ´´´ O 2.778(2) Š, H ´´´ O 1.92(3) Š, N H ´´´ O 172(2) Š.
Angew. Chem. Int. Ed. 2000, 39, No. 7 ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3907-1301 $ 17.50+.50/0
1301

COMMUNICATIONS
[D6]DMSO, where intermolecular H-bonding interactions are
minimized,[14] no significant self-assembly was observed at
room temperature (as indicated by the presence of NH2 and
(N1)H signals at d ˆ 6.40 and 10.80, respectively; Figure 3a).
By contrast the 1H NMR spectrum of a 58 mm solution of 1 in
CD2Cl2 at room temperature showed evidence for self-
assembly. Under these conditions a broad signal, exchange-
able with D2O, is observed between d ˆ 7.0 and 7.8 (Fig-
ure 3b). While such a signal, assigned on the basis of
integration to (N2)HG, is consistent with the formation of a
tetramer that is in fast equilibrium with other self-assembled
species (dimers, trimers, oligomers, etc.), it could also indicate
the incomplete formation of the proposed G-quartet-like
ensemble. We currently favor the first of these interpretations
since re-recording the spectrum at 308C produces well-
defined spectral patterns that are considered consistent with
the formation of a G-quartet (Figure 3c). In particular, new
exchangeable signals at d ˆ 9.81 and 5.15, ascribed to (N2)HG
and (N2)H*, respectively, are observed that are coincident, in
terms of chemical shift, with those of previously described
guanine-derived tetramers that are stabilized by alkali metal
cations.[7, 15] Furthermore, the fact that under these low-
temperature conditions only a single signal is seen for each
of the (N3)HG and (N3)H* protons leads us to suggest that, in
contrast to other G-quartets,[7, 15] only one of two possible syn/
anti ribose conformers, presumably syn, is present in solution.
To probe more fully the solution-phase characteristics of 1,
the number-averaged molecular weight of putative aggre-
gate(s) in dichloroethane was estimated at 408C by using
vapor-pressure osmometry (VPO).[16] These studies, carried
out at initial concentrations[17] of 28, 42, 57, and 71 mm (in 1),
revealed an averaged molecular weight of 2170 Æ 150 amu,
that is, close to the molecular weight expected for a tetramer
of 1 (2448 amu). We thus conclude that the tetrameric form,
although subject to exchange, constitutes the predominant
species in solution as well as in the solid state.
We have shown that G-quartets can be stabilized in the
absence of a templating alkali metal cation. This leads us to
suggest that this kind of structural motif may be more
fundamental in its origin and easier to form than previously
assumed.
Received: September 3, 1999
Revised: November 26, 1999 [Z13962]
[1] a) D. Philp, J. F. Stoddart, Angew. Chem. 1996, 108, 1242 ± 1286;
Angew. Chem. Int. Ed. Engl. 1996, 35, 1154 ± 1196; b) M. M. Conn, J.
Rebek, Jr., Chem. Rev. 1997, 97, 1647 ± 1668; c) J.-M. Lehn, M. J.
Krische, Struct. Bonding 2000, in press.
[2] J. L. Sessler, R. Wang, Angew. Chem. 1998, 110, 1818 ± 1821; Angew.
Chem. Int. Ed. 1998, 37, 1726 ± 1729.
[3] V. Andrisano, G. Gottarelli, S. Masiero, E. H. Heijne, S. Pieraccini,
G. P. Spada, Angew. Chem. 1999, 111, 2543 ± 2544; Angew. Chem. Int.
Ed. 1999, 38, 2386 ± 2388.
[4] a) G. Gottarelli, S. Masiero, E. Mezzina, G. P. Spada, P. Mariani, M.
Recanatini, Helv. Chim. Acta 1998, 81, 2078 ± 2092; b) K. Arakik, M.
Abe, A. Ishizaki, T. Ohya, Chem. Lett. 1995, 359 ± 360.
[5] J. A. Walmsley, J. F. Burnett, Biochemistry 1999, 38, 14063Ð14068.
[6] S. Borman, Chem. Eng. News 1999, 77(27), 36 ± 37.
[7] A. L. Marlow, E. Messina, G. P. Spada, S. Masiero,
J. T. Davis, G. Gottarelli, J. Org. Chem. 1999, 64,
5116 ± 5123.
[8] A. L. Marlow, J. T. Davis, Tetrahedron Lett. 1999, 40,
3539 ± 3542.
[9] S. B. Zimmerman, J. Mol. Biol. 1975, 106, 663 ± 672.
[10] a) J. A. Walmsley, R. T. G. Barr, E. Bouhoutsos-
Brown, T. J. Pinnavaia, J. Phys. Chem. 1984, 88,
2599 ± 2605; b) E. Bouhoutsos-Brown, C. L. Mar-
shall, J. T. Pinnavaia, J. Am. Chem. Soc. 1982, 104,
6576 ± 6584.
[11] J. L. Sessler, B. Wang, A. Harriman, J. Am. Chem.
Soc. 1995, 117, 704 ± 714.
[12] X-ray crystal data for 1: Thin colorless needles
were grown from a CH3CN/iPrOH mixture (1:1
ratio). C30H40N6O8 ´ 1ꢁ2 CH3CN, tetragonal, I4, Z ˆ 4,
a ˆ 28.3035(11), c ˆ 8.2683(5) Š, Vˆ 6623.6(5) Š3,
1calcd ˆ 1.27 gcm 3, F(000) ˆ 2696. A total of 23013
reflections were measured, 7585 unique (Rint ˆ 0.029),
on a Siemens SMART PLATFORM equipped with a
CCD area detector using graphite-monochromated
MoKa radiation (l ˆ 0.71073 Š) at
1008C. The
structure was refined on F 2 to an Rw ˆ 0.107, with a
conventional R ˆ 0.0455 (5821 reflections with Fo >
4[s(Fo)]), GOF ˆ 1.21 for 437 refined parameters.
The tetramer resides around the crystallographic
fourfold rotation axis. A molecule of acetonitrile
resides between layers of tetramers, spaced 8.2 Š
apart, and is disordered along the fourfold rotation
axis. Crystallographic data (excluding structure fac-
tors) for the structure reported in this paper have
been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication no.
CCDC-134040. Copies of the data can be obtained
free of charge on application to CCDC, 12 Union
Road, Cambridge CB21EZ, UK (fax: (‡44)1223-
336-033; e-mail: deposit@ccdc.cam.ac.uk).
Figure 3. Portion of the proton NMR spectrum of 1 showing changes consistent with the
formation of a G-quartet: a) 70 mm solution of 1 in [D6]DMSO at room temperature;
b) 58 mm solution of 1 in CD2Cl2 at room temperature; c) 58 mm solution of 1 in CD2Cl2 at
308C. The signals ascribed to (N1)H are observed at d ˆ 10.80, 12.47, and 12.70 in the case of
spectra a), b), and c), respectively. By contrast, the (N2)H2 signal observed at d ˆ 6.40 in
[D6]DMSO at room temperature appears in the form of two signals at d ˆ 5.15 and 9.81 when
the spectrum is recorded at 308C in CD2Cl2. See text for further discussion. For definitions
of HG and H* see caption to Figure 2.
1302
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3907-1302 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 7

COMMUNICATIONS
[13] Destabilization of the anti conformation due to steric interactions has
also been reported: a) S. N. Rao, P. A. Kollman, J. Am. Chem. Soc.
1986, 108, 3048 ± 3053; b) S. H. Krawczyk, N. Bischofberger, L. C.
Griffin, V. S. Law, R. G. Shea, S. Swaminathan, Nucleosides Nucleo-
tides 1995, 14, 1109 ± 1116; c) F. Seela, Y. Chen, C. Mittelbach, Helv.
Chim. Acta 1998, 81, 570 ± 583, and references therein.
[14] J. L. Sessler, R. Wang, J. Org. Chem. 1998, 63, 4079 ± 4091.
[15] In these previous systems both syn and anti ribose conformers were
observed as implied in the text. Proper comparison was thus made to
the signals assigned to the syn form.
endothelium established carbohydrate sulfotransferases as
potential targets for anti-inflammatory therapy.[2] Ongoing
genome sequencing projects have uncovered numerous
carbohydrate sulfotransferase genes in the past few years.[1, 3]
It is now apparent that carbohydrate sulfotransferases com-
prise a large family of enzymes with overlapping tissue
distribution and substrate specificities. To deconvolute the
precise role of each sulfotransferase gene product and
elucidate its contribution to normal and pathological pro-
cesses, cell-permeable and highly specific small-molecule
antagonists need to be identified. Surprisingly, no inhibitor
of a carbohydrate sulfotransferase has been reported to date.
At present, the limited structural and mechanistic information
about this class of enzymes impedes rational approaches to
inhibitor design. To identify lead inhibitors of carbohydrate
sulfotransferases, we therefore adopted a strategy that
involved the screening of small-molecule libraries.
To narrow our search, we focused on the similarities
between the substrates utilized by sulfotransferases and
kinases, a widely studied family of enzymes for which diverse
and potent inhibitors are available.[4] Carbohydrate sulfo-
transferases catalyze the transfer of a sulfonyl group from the
universal sulfate donor 3'-phosphoadenosine 5'-phosphosul-
fate (PAPS) to a hydroxy (or amino) group of the acceptor
oligosaccharide (Scheme 1).
[16] a) D. E. Burge, Am. Lab. 1977, 9, 41 ± 51; b) D. E. Burge, J. Appl.
Polym. Sci. 1979, 24, 293 ± 299.
[17] In contrast to various previously reported lipophilic deoxyguanosine
derivatives that were all found to form gels in chlorinated organic
solvents in the absence of cations (that is, self-assembled ribbonlike
aggregates)[4a], compound 1 was not found to form a gel even at
concentrations as high as 0.2m. As discussed in the text, we rationalize
this in terms of 1, but not these other compounds, being present in a
syn conformation. The observation that the less sterically hindered
compound, (2',3',5'-tri-O-isobutyrylribofuranosidyl)-2-aminopurin-6-
one, which lacks an aryl substituent and which should be present
mostly in anti conformation, forms a gel at concentrations as low as
0.02m is considered consistent with this theory.
Discovery of Carbohydrate Sulfotransferase
Inhibitors from a Kinase-Directed Library**
O3SO
HO
carbohydrate
sulfotransferase
O
O
OH
OH
Joshua I. Armstrong, Adam R. Portley, Young-
Tae Chang, David M. Nierengarten, Brian N. Cook,
Kendra G. Bowman, Anthony Bishop,
Nathanael S. Gray, Kevan M. Shokat, Peter G. Schultz,
and Carolyn R. Bertozzi*
NH2
3',5'-ADP
N
N
O
S
O
N
N
P
O
O
O
O
O
O
2
O3PO
PAPS
OH
Carbohydrate sulfotransferases have recently emerged as
an important and relatively unexplored class of therapeutic
targets.[1] For example, the seminal discovery that sulfated
sialyl LewisX mediates the adhesion of leukocytes to inflamed
(3'-phosphoadenosine
5'-phosphosulfate)
2
R
OPO3
R
OH
kinase
5'-ADP
[*] Prof. C. R. Bertozzi, J. I. Armstrong, A. R. Portley,
D. M. Nierengarten, B. N. Cook, K. G. Bowman, N. S. Gray
Department of Chemistry
NH2
N
N
O
P
O
O
P
University of California, Berkeley
Berkeley, CA 94720 (USA)
N
O
O
P
O
O
N
O
O
O
O
Fax: (‡1)510-643-2628
E-mail: bertozzi@cchem.berkeley.edu
OH OH
ATP
Dr. Y.-T. Chang, Prof. P. G. Schultz
Department of Chemistry
(adenosine 5'-triphosphate)
The Scripps Research Institute, La Jolla (USA)
Scheme 1. Reactions catalyzed by carbohydrate sulfotransferases and
kinases.
A. Bishop, Prof. K. M. Shokat
Department of Cellular and Molecular Pharmacology
University of California, San Francisco (USA)
Kinases catalyze a similar anionic group transfer reaction
using adenosine 5'-triphosphate (ATP) as a phosphoryl donor.
Thus, both enzyme classes recognize adenosine-based sub-
strates, PAPS and ATP. Furthermore, the hydrophobic
adenine binding pockets of the recently crystallized estrogen
sulfotransferase,[5] and heparin N-sulfotransferase[6] are sim-
ilar to those of several kinases. On the basis of these parallels,
we chose to screen a panel of previously reported kinase-
[**] We thank Sharon Long and Dave Keating for providing both the
NodH sulfotransferase and APS Kinase during our preliminary
experiments and Jack Kirsch for numerous helpful conversations.
J.I.A. and K.G.B were supported by NIH Molecular Biophysics
Training Grant (No. T32GM0895). This research was funded by grants
to C.R.B. from the Pew Scholars Program, the W. M. Keck Foundation
and the American Cancer Society (Grant No. RPG9700501BE).
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/angewandte/ or from the author.
Angew. Chem. Int. Ed. 2000, 39, No. 7
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3907-1303 $ 17.50+.50/0
1303