Pyranosyl-RNA supramolecules containing non-hydrogen bonding base pairs

Article abstract of DOI:10.1055/s-1999-3112

Synthesis and properties of a new pyranosyl-RNA nucleoside using tryptamine as nucleo-base is reported. Incorporation of this unit into oligomers using standard phosphoamidite chemistry yielded self-complementary and non-selfcomplementary oligonucleotide pairs. Thermal melting experiments of these examples showed the sequence dependent stabilising characteristics of the incorporated base in the symmetric pairing constitution with a standard T(m) near that of a similar A-T-pair as well as pairing selectivity with respect to non-symmetric pairing tolerating thymine but destabilizing if confronted to an adenine as complementary base in the antiparallel strand.

Full text of DOI:10.1055/s-1999-3112

940
LETTER
Pyranosyl-RNA Supramolecules Containing Non-Hydrogen Bonding Base-
Pairs
Christian Hamon,a Tilmann Brandstetter1, Norbert Windhaba*
a Department for Bio-Organic Systems, Aventis Research & Technologies, D-65926 Frankfurt am Main, Germany
Fax +49(69)305 81535; E-mail: windhab@aventis.com
Received 9 February 1999
Abstract: Synthesis and properties of a new pyranosyl-RNA nucle-
oside using tryptamine as nucleo-base is reported. Incorporation of
this unit into oligomers using standard phosphoamidite chemistry
yielded self-complementary and non-selfcomplementary oligonu-
cleotide pairs. Thermal melting experiments of these examples
showed the sequence dependent stabilising characteristics of the in-
corporated base in the symmetric pairing constitution with a stan-
dard Tm near that of a similar A-T-pair as well as pairing selectivity
with respect to non-symmetric pairing tolerating thymine but desta-
bilizing if confronted to an adenine as complementary base in the
antiparallel strand.
Key words: N-glycosides, synthetic nucleotides, supramolecular
chemistry, base pairing, nanotechnology
With the invention of pyranosyl-RNA Eschenmoser et al.
presented 1993 a new pairing system, a synthetic class of
molecules designed for the antithetic approach to the
question why furanoses and not pyranoses (in the nucleic
acids’ backbone)2a using synthetic molecules as hypothet-
ic challenges to our understanding of basic principles. p-
RNA surprisingly well matched the prerequisites for a ge-
notypic function and has even quantifiable advantages in
pairing selectivity and stability with respect to its isomer
natural RNA and to DNA. The ongoing work on the phe-
notypic properties with respect to conformation and reac-
tivity makes p-RNA to a distinguished representative of
constitutionally prebiotically possible systems, studied in
his laboratories.2b,c
Figure 1 ESI-mass-spectrum and constitution of a p-RNA oligonu-
cleotide containing the unnatural base Tr.
p-RNA strands consisting of nucleosides of the same
chirality pair exclusively in the Watson-Crick-modus in
antiparallel strand orientation. No triplex formation has
been reported and the idealized diamantoide backbone
conformation (Figure 1) illustrates the pairing orthogonal-
ity of the planar base-pair repetition throughout a p-RNA
duplex to natural systems, the latter leading to repetitive
angular increments as twist and pitch of the helix. The ex-
clusivity and orthogonality of its pairing properties make
p-RNA to a model for pairing principles and to a synthet-
ically manageable supramolecular entity being both a
theoretical and a technical model system in the field of
Life Science and Nanotechnology.3a,b,c
Co-operative stability afforded a good stacking residue,
our nanotechnological approach requested a versatile
linker-functionalisation to link the pairing strand with
any synthetic or biological molecule or solid support
at any position in its sequence. Another aspect was to in-
corporate a fluorescent probe without additional labelling.
Excellent planar stacking properties can be expected from
cristallographic data reported for tryptamine with differ-
ent biogenic heterocycles such as adenine-9-acetic acid,4a
and 7-methylguanosine.4b
To increase the repertoire for controlled supramolecular
synthesis for different technical applications and to fit
changing conditions, the supramolecular chemist seeks
for modular subunits in much higher variance than the
natural gene encoding nucleosides do provide for self as-
sembling systems.
Successful incorporation of different isosteric compounds
have been reported for natural DNA and nonselective
pairing for the case of pairing with natural bases.5 Two
major differences to the natural system must be consid-
ered: p-RNA shows a strong inter-strand stacking which
Synlett 1999, S1, 940–944 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Pyranosyl-RNA Supramolecules Containing Non-Hydrogen Bonding Base-Pairs
941
is 2’→ 4’ directional and a lateral broadening of the du- N-Phthaloyl tryptamine 1 was synthesized from phthalic
plex with respect to pairing residues is theoretically possi- anhydride and tryptamine7. Reduction with borane-THF6
ble which would unwind any helical system. This is why gave the indoline derivative 2. The 3-substituted indoline
we neglected the isosteric considerations and used 2 was reacted with ribose in ethanol to yield the nucleo-
tryptamine even as pseudo-symmetric base pair, pairing side triol. The hydroxyl groups were protected with acetic
with itself.
anhydride, and the indoline was oxidized with DDQ to
give nucleoside 3. Treatment with sodium methoxide
The synthesis of the supramolecules starts with the syn-
thesis of the amidite-monomer for synthesizer chemistry
(Scheme 1). The phthalimide protecting group was chosen
in order to get a fully deprotected oligo after nucleophilic
cleavage from CPG-support. Direct catalysed N-glycosi-
dation proved to be very unsatisfactory.
removed the acetates. The 2’-position of
4 was
benzoylated selectively at –78 °C. Reaction with
dimethoxytrityl chloride took place predominantly at po-
sition 4’ of the hydroxyl group of 5. After rearrangement
of the benzoyl group to position 3’ yielding compound 6,
Synlett 1999, S1, 940–944 ISSN 0936-5214 © Thieme Stuttgart · New York

942
C. Hamon et al.
LETTER
the phosphoramidite 7 was obtained under standard con-
ditions and could be used for automatic oligonucleotide
synthesis without changing the protocol of p-RNA syn-
thesis.2a, 13-18
Melting studies were carried out at 272 nm on a computer-
interfaced Perkin-Elmer Lamba 2 UV-visible spectropho-
tometer in stoppered 1 cm path length quartz cells for the
1, 3 and 5 µM concentration of oligomer and with 0,1 cm
path length for the 10, 25, 50, 100 µM concentration.
Samples of 1 - 100 µM concentration of oligomer were
prepared in a buffer solution composed of 10 mM Tris/
HCl (pH 7.0) and 150 mM NaCl. Degasing and heating up
for one time above 80 °C yielded fully reversible hypo-
chromicity profiles. The melting curves (Figure 2) were
measured using a temperature gradient from 5 °C to 90 °C
for heating and cooling curves at a ramp rate of 1 °C/40s
with a tailor-made micro-thermoelement in the sample
cell read by a programmed Keithley Instruments DAS-
801-AT-Bus measure card (Quick-Basic 4.0 driver). Un-
certainty in the Tm data is estimated at ± 0.3 °C based on
the repetitions of experiments. Thermodynamic parame-
ters were estimated by the determination of Tm fitting on
the hypochromicity profile to a two state transition model
as described8,10 and a van’t Hoff interpretation of the con-
centration-dependent equilibria involving only duplex
formation and transitions that proceed in a two-state man-
ner (Figure 3, Table 1).
Figure 4 Calculated coalescence of the three different equilibria
around 25 °C.
Under the assumption that enthalpy and entropy are not
temperature dependent in the experimental range, the
slope of the free enthalpy as shown in Figure 4 illustrates
the difference of the pairing behaviour although the free
enthalpy at 25 °C is the same in the range of experimental
error. The strong entropic disadvantage of two isolated
“non-polar“-tryptamine groups in a single strand (pr-
CCTrCGTrGG) can be explained due to the surface
shown to the rest of the molecule. The sequential
repetition (pr-CCCTrTrGGG) will have less “unnatural“
base interaction along the co-operative unit and in fact
shows a temperature dependence of the free enthalpy as in
the case of a single Tr-residue (pr-CCCTrGGG) incorpo-
rated. Repetition of Tr seems to overcompensate (Figure
4). However this difference could occur through a differ-
ent mechanism: the mode of transition. The data presented
here are derived using equilibria without taking any pop-
ulation of intermediate states into account. This could be
the case for a strong enthalpic stabilising entity which de-
creases the length of the co-operative unit.10 This non-co-
operative quality, as well as the strong unfavorable
entropic effect, could be an explanation for the absence of
any detectable pairing for the sequence with three Tr-res-
idues (pr-CTrCTrGTrG) up to a concentration of 500
µM (Table 1). The sequence with the same number of G-
C-pairs was not investigated as the appropriate p-RNA se-
quences with a number of bases superior to 8 (such as
pr-CCTrCTrGTrGG) would be likely to form a hairpin
structure.2c
Figure 3 Concentration dependence of Tm for three self-complemen-
tary dimerisations.
A second group of non self-complementary oligomers re-
veals first insights into the pairing selectivity with respect
to mismatch discrimination (Table 3). The clear unfavour-
able pairing to an adenine complement with full toleration
of a thymine and a tryptamine complement can be ex-
plained by the water cloud that the adenine presents into
the pairing pocket - one explanation of a ground state se-
lectivity of natural pairing, with hydrogen bonds that are
destabilising if not saturated (by the Watson-Crick-com-
plement) but not leading to a significant (> 1 kcal/mol)
stabilisation if water is present per se, as the interstrand
stack of adenine is still possible.11
Table 1 Comparison of the thermodynamic melting data. pr-CC-
CGGG* data reported.9
Synlett 1999, S1, 940–944 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Pyranosyl-RNA Supramolecules Containing Non-Hydrogen Bonding Base-Pairs
943
This interstrand stacking behaviour is in contrast to anoth-
er unnatural non-hydrogen base designed for interstrand
stack in DNA.12 Kool’s pyrene nucleoside destabilises the
DNA duplex for any canonical pairing partner. Interest-
ingly two pyrenes paired against one another show nearly
the same stability as a natural A-T-pair. This very same
comparison was obtained in our first mismatch study (Ta-
ble 3).
References and Notes
(1) New address: GPC-AG, Fraunhoferstr. 20, D-82151
Martinsried / München, Germany. E-mail:
tilmann.brandstetter@gpc-ag.com
(2) (a) Pitsch, S.; Wendeborn, S.; Jaun, B.; Eschenmoser A. Helv.
Chim. Acta 1993, 76, 2161. (b) Schlönvogt, I.; Pitsch, S.;
Lesueur, C.; Eschenmoser, A.; Jaun, B.; Wolf, R.M. Helv.
Chim. Acta 1996, 79, 2316. (c) Micura, R.; Bolli, M.;
Windhab, N.; Eschenmoser, A. Angew. Chem. 1997, 109, 899;
Angew. Chem., Int. Ed. Engl. 1997, 36, 870.
(3) (a) Mucic, C. R.; Storhoff, J.; Mirkin, C. A.; Letsinger, R. L.
J. Am. Chem. Soc., 1998, 120, 12674. (b) Seeman, N. C.
Annu. Rev. Biomol. Struct. 1998, 27, 225. (c) Holmlin, R. E.;
Dandliker, P. J.; Barton, J. K. Angew. Chem. 1997, 109, 2831;
Angew. Chem., Int. Ed. Engl. 1997, 36, 2714.
Table 2. Mismatch examples synthesized.a)
(4) (a) Ishida, T.; Tomita, K.-I.; Inoue, M. Arch. Biochem.
Biophys. 1980, 200, 492. (b) Ishida, T.; Doi, M.; Ueda, H.;
Inoue, M.; Scheldrick, G.M. J. Am. Chem. Soc. 1988, 110,
2286.
(5) Schweitzer, B.; Kool E. T. J. Am. Chem. Soc. 1995, 117, 1863.
(6) Giannis, A.; Sandhoff, K. Angew. Chem. 1989, 101, 220;
Angew. Chem., Int. Ed. Engl. 1989, 28, 218.
(7) Kuehne, M. E.; Huebner, J. A.; Matsko, T. H. J. Org. Chem.
1979, 44, 2477.
Table 3. Non self-complementary pairs and their Tm at 5+5 µM strand
concentration. X corresponds to either a dye-label (Cy-3) or a biotine
or an amino-linker. Our previous studies have shown that only minor
differences for Tm (< ± 1 °C) are obtained when the 4’ end of the
pRNA is occupied with either the dye (Cy-3), biotine, aminolinker or
simply a free 4’-OH group. Strand orientation: Æ.
(8) Marky, L.A.; Breslauer, K.J. Biopolymers, 1987, 26, 1601.
(9) Krishnamurty, R.; Pitsch, S.; Minton, M.; Miculka, C.;
Windhab, N.; Eschenmoser, A. Angew. Chem. 1996, 108,
1619; Angew. Chem., Int. Ed. Engl. 1996, 35, 1537.
(10) Windhab, N.; Ohms, J.; Ackermann, Th. Biophys. Chem.
1993, 47, 225.
(11) Kool, E. Chem. Rev. 1997, 97, 1473.
(12) Matray, T.; Kool, E. J. Am. Chem. Soc. 1998, 120, 6191.
(13) 2: 1H NMR (CDCl3, 300 MHz): 1.85-2.00, 2.14-2.28 (2 m,
2 x 1 H, CH2CH2NPhth), 2.70 (bs, 1 H, NH), 3.24-3.38, 3.66-
3.86 (2 m, 5 H, CH2CH2NPhth, H-2a, H-2b, H-3), 6.62 (d,
J = 8.0 Hz, 1 H, H-7), 6.66-6.72 (m, 1 H, H-5), 6.99 (app t,
J = 7.5 Hz, 1 H, H-6), 7.14 (d, J = 8.0 Hz, 1 H, H-4), 7.64-
7.74, 7.78-7.86 (2 m, 2 x 2 H, Phth). 13C NMR(CDCl3, 75
MHz): 32.70, 36.10 (2 t, C-2, CH2CH2NPhth), 39.62 (d, C-3),
53.04 (t, CH2NPhth), 109.65 (d, C-7), 118.74 (d, C-5), 123.25
(d, Phth), 123.92, 127.72 (2 d, C-4, C-6), 131.81 (s, C-3a),
132.14 (s, Phth), 133.99 (d, Phth), 151.26 (s, C-7a), 168.38 (s,
C = O). Cal: C: 73.96, H: 5.52, N: 9.58; found: C: 73.89, H:
5.57, N: 9.55. MS (ES+): 293 (MH+, 100%)
(14) 3: 1H NMR (CDCl3, 300 MHz): 1.64, 1.98, 2.19 (3 s, 3 x 3 H,
Ac), 3.06 (t, J = 8.0 Hz, 2 H, CH2CH2NPhth), 3.81-4.00 (m, 4
H, H-5ax, H-5eq, CH2NPhth), 5.13 (ddd, J = 2.5, 6.0, 10.5 Hz,
1 H, H-4), 5.36 (dd, J = 3.5, 9.5 Hz, 1 H, H-2), 5.71 (d, J = 9.5
Hz, 1 H, H-1), 5.74 (app t, J = 3.0 Hz, 1 H, H-3), 7.02 (s, 1 H,
H-2), 7.04-7.10, 7.13-7.19 (2 m, 2 x 1 H, H-5, H-6), 7.33 (d,
J = 8.0 Hz, 1 H, H-7), 7.58-7.66, 7.72-7.80 (2 m, 5 H, Phth, H-
4). 13C NMR(CDCl3, 75 MHz):20.23, 20.65, 20.87 (3 q, Ac),
24.41, 38.28 (2 t, CH2CH2), 63.53 (t, C-5), 66.24, 68.00, 68.64
(3 d, C-2, C-3, C-4), 80.33 (d, C-1), 109.79 (d, C-7), 113.95
(s, C-3), 119.33, 120.39, 122.04, 122.47 (4 d, C-4, C-5, C-6,
C-7), 123.18 (d, Phth), 128.70, 132.17 (2 s, C-3a, Phth),
133.87 (d, Phth), 136.78 (s, C-7a), 168.24, 168.77, 169.44,
169.87 (4 s, C = O). Cal: C: 63.50, H: 5.15, N: .11; found.: C:
Tryptamine may not be a potentially prebiotic substance,
but it is a bio- molecule. Incorporation of this new p-RNA
nucleoside with no donor- or acceptor-sites for the exclu-
sive Watson-Crick pairing constitution shows that differ-
ent rules control this sequentially selective model pairing
system. The naive question about a natural use of similar
compounds as pseudo bases or about any possible geno-
typic function of non-hydrogen-bonding nucleobases
would lead to the basic problem of a quantifiable reach of
the incremental optimisation in evolution with respect to
a genetic code.
+
63.48, H: 5.16, N: 5.05. MS (ES+): 566 (M+NH4 , 82%), 549
(MH+, 74%), 114 (100%).
(15) 4: 1H NMR (MeOD, 300 MHz): 3.09 (app. t, J = 7.0 Hz, 2 H,
CH2CH2NPhth), 3.64-3.98 (m, 5 H, H-4, H-5ax, H-5eq,
CH2NPhth), 4.05 (dd, J = 3.5, 9.5 Hz, 1 H, H-2), 4.22 (app t,
J = 3.0 Hz, 1 H, H-3), 5.65 (d, J = 9.5 Hz, 1 H, H-1), 6.95-
7.05, 7.09-7.16 (2 m, 2 x 1 H, H-5, H-6), 7.25 (s, 1 H, H-2),
7.44 (d, J = 8.0 Hz, 1 H, H-7), 7.60 (d, J = 8.0 Hz, 1 H, H-4),
7.74-7.84 (m, 4 H, Phth). 13C NMR (d6-DMSO, 75 MHz):
Acknowledgement
The authors are grateful for the expert technical assistance of Petra
Ickstadt for the monomer synthesis, Daniela Kammann, Joachim
Schmidt and Manfred Wildt for the synthesizer synthesis of the oli-
gos.
Synlett 1999, S1, 940–944 ISSN 0936-5214 © Thieme Stuttgart · New York

944
C. Hamon et al.
LETTER
23.87, 37.79 (2 t, CH2CH2NPhth), 64.82 (t, C-5), 66.74 (d, C-
4), 68.41 (d, C-2), 71.42 (d, C-3), 81.37 (d, C-1), 110.42 (d, C-
7), 111.05 (s, C-3), 118.17, 119.21, 121.36, 122.92, 123.80 (5
d, C-2, C-4, C-5, C-6, NPhth), 127.86, 131.59 (2 s, C-3a,
Phth), 134.27 (d, Phth), 136.62 (s, C-7a), 167.72 (s, C = O).
MS (ES-):457 (M+OH-+H2O, 49%), 439 (M+OH-, 100%),
421 (M-H+, 28%)
CH2CH2NPhth), 3.67-3.74 (m, 1 H, H-5ax), 3.69, 3.70 (2 s,
2 x 3 H, OMe), 3.85 (t, J = 7.5 Hz, 2 H, CH2CH2NPhth), 3.94
(ddd, J = 3.0, 5.0, 10.5 Hz, 1 H, H-4), 4.03 (dd, J = 3.5, 9.0 Hz,
1 H, H-2), 5.51 (d, J = 9.0 Hz, 1 H, H-1), 5.86 (bs, 1 H, H-3),
6.68-7.66 (m, 25 H), 8.19-8.30 (m, 2 H). 13C NMR(CDCl3, 75
MHz): 24.16, 38.80 (2 t, CH2CH2NPhth), 55.25, 55.26 (2 q,
Ome), 65.58 (t, C-5), 68.29, 69.19, 73,83 (3 d, C-2, C-3, C-4),
83.03 (d, C-1), 87.31 (CAr3)110.03 (d, C-7), 113.37, 113.47 (2
d), 113.53 (s, C-3), 118.95, 120.20, 122.28, 122.31, 123.10,
127.07, 128.02, 128.08, 128.68 (9 d), 128.74 (s), 130.02,
130.19, 130.22 (3 d), 130.37, 131.95 (2 s), 133.40, 133.83 (2
d), 135.98, 136.14, 136.56, 145.12, 158.82, 166.76, 168.52 (7
s, C-7a, 2 COMe, 2 C = O).
(16) 5: 1H NMR (CDCl3, 300 MHz): 2.45, 2.70 (2 bs, 2 x 1 H, OH),
3.04 (t, J = 8.0 Hz, 2 H, CH2CH2NPhth), 3.80-4.20 (m, 5 H, H-
4, H-5ax, H-5eq, CH2NPhth), 4.63 (bs, 1 H, H-3), 5.46 (dd,
J = 3.5, 9.5 Hz, 1 H, H-2), 6.03 (d, J = 9.5 Hz, 1 H, H-1), 7.08-
7.31 (m, 5 H, H-2, H-5, H-6, Bz-m-H), 7.41-7.48 (m, 1 H, H-
Bz-p-H), 7.50 (d, J = 8.0 Hz, 1 H, H-7), 7.64-7.79 (m, 7 H,
Phth, H-4, Bz-o-H). 13C NMR (d6-DMSO, 75 MHz): 24.40,
38.22 (2 t, CH2CH2NPhth), 65.95 (t, C-5), 66.65 (d, C-4),
69.55 (d, C-3), 71.87 (d, C-2), 79.57 (d, C-1), 109.96 (d, C-7),
113.70 (s, C-3), 119.21, 120.21, 122.11, 122.41, 123.14 (5 d,
C-2, C-4, C-5, C-6, NPhth), 128.28 (d, Bz), 128.58, 128.59 (2
s, C-3a, Bz), 129.62 (d, Phth), 132.05 (s, Phth), 133.81 (Bz),
136.97 (s, C-7a), 165.12, 168.29 (2 s, C = O). MS (ES-): 525
(M-H+, 12%), 421 (M-PhCO+, 23%), 107 (100%).
(18) 7: 1H NMR (CDCl3, 300 MHz): characteristic peaks:2.42,
2.53 (2 dd, J = 5.0, 11.0 Hz, 2 H, 2 H-5eq), 3.76, 3.77, 3.78,
3.79 (4 s, 4 x 3 H, OMe), 5.70, 5.73 (2 d, J = 9.0 Hz, 2 H, 2 H-
1), 6.16, 6.29 (2 bs, 2 H, 2 H-3). 31P NMR (CDCl3): 150.6,
151.0
Article Identifier:
1437-2096,E;1999,0,S1,0940,0944,ftx,en;W07199ST.pdf
(17) 6: 1H NMR (CDCl3, 300 MHz): 2.64 (bs, 1 H, OH), 2.68 (dd,
J = 5.0, 11.5 Hz, 1 H, H-5eq), 2.94 (dd, J = 7.5, 16.0 Hz, 1 H,
CH2CH2NPhth), 3.03 (dd, J = 8.0, 16.0 Hz, 1 H,
Synlett 1999, S1, 940–944 ISSN 0936-5214 © Thieme Stuttgart · New York