Phe and Asn side chains in DNA double strands

Article abstract of DOI:10.1016/S0960-894X(00)00231-6

The contribution of amino acid side chains to the recognition of DNA by peptides or proteins is evaluated by substituting single nucleobases of a DNA double strand by amino acid side chains. C-nucleosides with the side chains of phenylalanine and asparagine were synthesized and incorporated in DNA. This modification should allow to keep the double strand conformation. Hydrogen bonds, π-π-interactions and solvation have an influence on the double strand stability. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00231-6

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1417±1420
Phe and Asn Side Chains in DNA Double Strands
Ulf Diederichsen a,* and Claudia M. Biro b
aInstitut fur Organische Chemie, Universitat Wurzburg, Am Hubland, 97074 Wurzburg, Germany
È
È
È
È
bInst. fur Organische Chemie und Biochemie, Technische Universitat Munchen, Lichtenbergstr. 4, 85747 Garching, Germany
È
È
È
Received 14 February 2000; accepted 17 April 2000
AbstractÐThe contribution of amino acid side chains to the recognition of DNA by peptides or proteins is evaluated by sub-
stituting single nucleobases of a DNA double strand by amino acid side chains. C-nucleosides with the side chains of phenylalanine
and asparagine were synthesized and incorporated in DNA. This modi®cation should allow to keep the double strand conforma-
tion. Hydrogen bonds, p±p-interactions and solvation have an in¯uence on the double strand stability. # 2000 Elsevier Science Ltd.
All rights reserved.
The recognition of DNA double strands by proteins or
peptides involves amino acid side chain interactions
with the nucleobases. Binding to DNA can be based on
hydrophobic or electrostatic contributions for non-speci®c
interactions, whereas hydrogen bonds and aromatic
interactions are responsible for sequence dependent
recognition.1 For a better understanding of the nature
and speci®city of these amino acid side chain interac-
tions2 experimental evidence is needed.
Within this system the substitution of guaninylalanine by a
phenylalanine or tryptophan has a stabilizing e€ect,
despite the missing hydrogen bonds with the com-
plementary cytosine.4 This unexpected stabilization is
likely to be the result of the reduction of unfavorable
aromatic-water interactions, which also depend on the
double strand topology. In linear alanyl-PNA the
DNA binding in the major or minor groove or by
intercalation is usually accompanied by conformational
reorientation of the double helix. Therefore, it is dicult
to isolate the individual amino acid side chain con-
tributions to binding. Positioning amino acid side
chains in the base stack of the double helix, in place of a
native DNA nucleobase, yields an opportunity to
observe interactions between peptide side chains and
neighboring nucleobases without disturbing the helix
topology. Two examples for incorporated amino acid
side chains in the base stack of DNA are described.
Instead of a regular nucleotide, C-nucleotides with the
desired side chain at the anomeric center were incorpo-
rated in DNA (Fig. 1).
These experiments were further encouraged by results
obtained by the linear, rigid, and well-de®ned model
system of alanyl peptide nucleic acids.3 The alanyl-PNA
double strands are based on a regular peptide backbone
with alternating con®guration of alanyl nucleo amino acids.
*Corresponding author. Tel.: +49-931-888-5326; fax: +49-931-884606;
e-mail: diederichsen@chemie.uni-wuerzburg.de
Figure 1. DNA oligomers with C-nucleosides having a benzyl d(b-
Phe) or aminocarbonylmethyl residue d(b-Asn) at the anomeric center.
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00231-6

1418
U. Diederichsen, C. M. Biro / Bioorg. Med. Chem. Lett. 10 (2000) 1417±1420
nucleobases or amino acid side chains are less water
accessible than in helical DNA. Further evidence for the
in¯uence of solvent accessibility is based on the PNA
(phenyl) amino acid (Fig. 2) with a phenyl side chain
incorporated in a helical PNA±DNA duplex which has
also a destabilizing e€ect.5
Many C-nucleosidic DNA-modi®cations are known
already.6 Most of them contain an arene directly bound
to the anomeric center in analogy to the binding of the
nucleobases (Fig. 2).7 Usually, the arene mismatch
destabilizes the DNA double strand, except when posi-
tioned at the end of the sequence.8 Larger aromatic
systems like d(pyrene) only show a small mismatch e€ect
because of a higher stacking contribution.10 Further-
more, a homoglutamine side chain incorporated cova-
lently to the anomeric center also induces a signi®cant
destabilization of the duplex.9
Figure 3. Synthesis of the Phe-Nucleoside ꢁ-1; d(a-Phe) nucleotide
ꢀ-1 was obtained in analogous manner from ꢀ-4.
Deprotection of the p-chlorobenzylesters was performed
with an excess of the Grignard reagent. The a/b-epimers
were obtained in a 2:1 ratio and separation of the epi-
mers was provided by MPL chromatography of the tri-
tyl derivatives ꢀ-/ꢁ-4 on silica gel with hexane/
ethylacetate. The assignment of the epimers was per-
formed with 2D-NMR spectroscopy.12 From the
nucleosides ꢀ-4 and ꢁ-4 the phosphoramidites ꢀ-1 and
ꢁ-1 were easily obtained with 2-cyanoethyl N,N-diiso-
propylchloro-phosphoramidite.13
To begin our examinations, C-nucleosides with a benzyl
side chain (d(Phe)) and an aminocarbonylmethyl residue
(d(Asn)) linked to the anomeric center were synthesized.
Both were prepared as the a- and b-anomers because
the di€erent orientation of the side chain was expected
to have a signi®cant in¯uence within the double strand
topology. They were incorporated into DNA oligomers
as phosphoramidites ꢀ/ꢁ-1 (Fig. 3) and ꢀ/ꢁ-2 (Fig. 4)
by solid phase synthesis on a CPG-resin using standard
DNA chemistry. The esters ꢀ-2 and ꢁ-2 were used for
the synthesis of DNA oligomers, but it was shown that
they were converted to the amides d(Asn) under the
deprotection conditions of the oligomer (NH3, H2O,
55 ꢀC, 16 h).
For the synthesis of the phosphoramidites ꢀ-2 and ꢁ-2,
leading to the d(Asn)-nucleotides in the oligomer, a
procedure for the allylation of the anomeric center of
acetyl-protected ribosylchloride 5 was used (Fig. 4)14
followed by separation of the 3:1 ratio of epimers ꢀ-6
and ꢁ-6 by MPL chromatography. The con®gurations
were assigned using 2D-NMR-spectroscopy.15 The
terminal double bond was oxidized with KMnO4 under
phase transfer conditions followed by methylation to
the esters ꢀ-7 and ꢁ-7. Selective deprotection of the
hydroxyl groups was performed with ammonia in
methanol at room temperature. At this point, forma-
tion of the amide was avoided because phosphitylation
of the amide would take place as a side reaction in the
®nal transformation to the phosphoramidites ꢀ-2 and
ꢁ-2. Therefore, ꢀ-2 and ꢁ-2 were obtained by DMT-
protection of the primary hydroxyl group followed by
phosphitylation.16
Synthesis of the Phe and Asn Nucleotides
The synthesis of d(Phe)-epimers ꢀ-1 and ꢁ-1 required a
benzyl side chain at the anomeric center. It was coupled
using a Grignard reaction with ribosylchloride 311 (Fig. 3).
DNA Double Strands with Phe and Asn Nucleotides
Hybridization of the DNA oligomers was performed in
0.1 M NaCl, 0.01 M Na2HPO4/H3PO4, pH 7. Their
stability was evaluated by thermal denaturation experi-
ments using UV-spectroscopy. The double strand sta-
bilities were indicated by the sigmoidal increase of
absorption caused by the cooperative destacking of
nucleobases. Pairing and depairing of all the double
strands described was reversible. The self-pairing DNA
octamer d(GCG-CGCGC) forms a double strand
(Tm=60 ꢀC) with eight G±C Watson±Crick base pairs
and was used as the reference. In the mismatch stu-
dies, either the central cytosinyl or the respective gua-
ninyl nucleotide were substituted by the d(a/b-Phe) or
Figure 2. Examples for nucleoside analogues incorporated already in
DNA±DNA or DNA±PNA double strands.

U. Diederichsen, C. M. Biro / Bioorg. Med. Chem. Lett. 10 (2000) 1417±1420
1419
are quite similar. Therefore, it seems unlikely that the C-
nucleotides disturb the overall helix conformation.
As indicated by the lower destabilization of the oligo-
mers 14±17 the d(Asn) nucleotides ®t better into the B-
DNA helix topology. On the other hand, stabilization
by hydrogen bonds seems unlikely since double strand
stabilities are comparable to the sequences with abasic
sites. Only the slightly more stable oligomer 16 might be
able to form hydrogen bonds with the Asn-side chain.19
DNA modi®cation with the C-nucleotides d(Phe) and
d(Asn) turned out to destabilize the double strands
similar to, or even more than, an abasic site. Other
nucleotide derivatives are currently under investigation
in order to obtain a DNA double strand with preference
for Z-DNA conformation. Guanosine isosters imitating
the syn conformation should be promising. As a ®nal
comment, we note that knowledge about the in¯uence
of anomeric side chains on the DNA double strand may
have implications in replication studies and the design
of a universal nucleobase.
Figure 4. Synthesis of the nucleoside ꢁ-2 leading to d(b-Asn): The
ester b-2 will be converted to the amide under deprotection conditions
of the oligomer. The precursor for d(a-Asn)nucleotide ꢀ-2 was
obtained in analogous manner from ꢀ-7.
Acknowledgements
This work was supported by the Deutsche For-
schungsgemeinschaft and the Fonds der Chemischen
Industrie. We are especially grateful to Professor Horst
Kessler, Garching, for generous support.
d(a/b-Asn) C-nucleotides.17 Further oligomers containing
an abasic deoxy-ribose (d(Gly) represents a glycine side
chain) as a mismatch were used for comparison. The
stabilities for antiparallel self pairing double strands are
summarized in Table 1. All mismatch sequences contain
two neighboring defects.
References and Notes
The two abasic sites in the d(Gly) oligomers lower the
stability by 14 ꢀC in case of the purine rich duplex 9, and
by 20 ꢀC when guanine is eliminated (8). In contrast to
the result in alanyl-PNA, a benzylgroup destabilized a
DNA double strand even more than the abasic site. The
d(Phe) modi®ed oligomers with b-con®guration 12 and
13 are 5 ꢀC more stable than the sequences containing
the a-epimers 10 and 11. The destabilizing e€ect is more
relevant with the benzylgroup between the larger gua-
nines. Through simple model studies we suggest a sterical
con¯ict is responsible for this destabilization, since the
only possibility to place the phenylring within the
nucleobase stack of B-DNA requires ecliptic bonds at
the anomeric center.18 The CD-spectra of all oligomers
1. Saenger, W. Principles of Nucleic Acid Structure; Springer:
New York, 1984; pp 385±431.
2. Seeman, N. C.; Rosenberg, J. M.; Rich, A. Proc. Nat. Acad.
Sci. USA 1976, 73, 804.
3. (a) Diederichsen, U. Angew. Chem., Int. Ed. Engl. 1996, 35,
445. (b) Diederichsen, U. Angew. Chem., Int. Ed. Engl. 1997,
36, 1886. (c) Diederichsen, U. In Bioorganic Chemistry, High-
lights and New Aspects; Diederichsen, U., Lindhorst, T.,
Wessjohann, L., Westermann, B., Eds.; Wiley-VCH: Wein-
heim, 1999; pp 255±260.
4. Diederichsen, U., Weicherding, D. Synlett 1999, 917.
5. Challa, H.; Woski, S. A. Tetrahedron Lett. 1999, 40, 8333.
6. (a) Polt, R.; Wijayaratne, T. Tetrahedron Lett. 1991, 32,
4831. (b) Chaudhuri, N. C.; Kool, E. T. Tetrahedron Lett.
1995, 36, 1795. (c) Ren, R. X.-F.; Chaudhuri, N. C.; Paris, P.
L.; Rumney, S. IV; Kool, E. T. J. Am. Chem. Soc. 1996, 118,
7671. (d) Wichai, U.; Woski, S. A. Bioorg. Med. Chem. Lett.
1998, 8, 3465; (e) Wichai, U.; Woski, S. A. Org. Lett. 1999, 1,
1173. (f) Adlington, R. M.; Baldwin, J. E.; Pritchard, G. J.;
Spencer, K. C. Tetraherdon Lett. 2000, 41, 575.
Table 1. Stabilities of DNA double strands with benzyl d(Phe) and
aminocarbonylmethyl d(Asn) mismatch compared to regular DNA
sequences and abasic d(Gly) mismatch double strands
7. (a) Schweitzer, B. A.; Kool, E. T. J. Org. Chem. 1994, 59,
7238. (b) Schweitzer, B. A.; Kool, E. T. J. Am. Chem. Soc.
1995, 117, 1863. (c) Chaudhuri, N. C.; Ren, R. X.-F.; Kool, E.
T. Synlett 1997, 341. (d) Diederichsen, U. Angew. Chem., Int.
Ed. Engl. 1998, 37, 1655.
8. Guckian, K. M.; Schweitzer, B. A.; Ren, R. X.-F.; Sheils, C.
J.; Paris, P. L.; Tahmassebi, D. C.; Kool, E. T. J. Am. Chem.
Soc. 1996, 118, 8182.
Mismatch
Sequences
d(GCGCGCGC) Tm=60 ꢀC (8% Ha, 7 mM)
d(GCGCXCGC)
d(GCGXGCGC)
d(Gly)
Tm=40 ꢀC (8, 9% H, 9 mM)
Tm=46 ꢀC (9, 9% H, 9 mM)
d(a-Phe)
d(b-Phe)
d(a-Asn)
d(b-Asn)
Tm=35 ꢀC (10, 3% H, 6 mM) Tm=30 ꢀC (11, 9% H, 9 mM)
Tm=40 ꢀC (12, 5% H, 7 mM) Tm=35 ꢀC (13, 6% H, 8 mM)
Tm=40 ꢀC (14, 9% H, 7 mM) Tm=46 ꢀC (15, 9% H, 7 mM)
Tm=46 ꢀC (16, 9% H, 7 mM) Tm=40 ꢀC (17, 8% H, 8 mM)
9. Francois, P.; Perilleux, D.; Kempener, Y.; Sonveaux, E.
Tetrahedron Lett. 1990, 31, 6347.
aH=hyper-chromicity, conditions: 0.1 M NaCl, 0.01 M Na2HPO4/H3PO4,
pH 7.

1420
U. Diederichsen, C. M. Biro / Bioorg. Med. Chem. Lett. 10 (2000) 1417±1420
10. Matray, T. J.; Kool, E. T. J. Am. Chem. Soc. 1998, 120, 6191.
11. The glycoside donor 3 was prepared following the proce-
dure of Kotick, M. P., Szantay, C., Bardos, T. J. J. Org.
Chem. 1969, 34, 3806.
C2H3), 5.83 (m, 1H, CH2-C2H3). Assignment with P. E.
COSY, TOCSY, HMQC and NOESY (120 ms, H10±H30 and
H10±H20 cross peaks). EI±MS: 242 (M+). ꢁ-6: 1H NMR
(CDCl3): 1.62 (m, 1H, H20), 1.84 (m, 1H, H200), 1.99 (s, 3H,
Ac), 2.09 (s, 3H, Ac), 2.16 (m, 1H, CH2±C2H3), 2.23 (m, 1H,
CH2±C2H3), 3.62 (m, 1H, H50, H500), 3.64 (m, 1H, H10), 3.73
(m, 1H, H50, H500), 4.82 (m, 1H, H40), 5.03 (m, 2H, CH2±
C2H3), 5.33 (m, 1H, H30), 5.77 (m, 1H, CH2-C2H3). Assign-
ment with P. E. COSY, TOCSY, HMQC and NOESY (120
ms). EI±MS: 242 (M+). For a comparable assignment: Mat-
suura, N., Yashiki, Y., Nakashima, S., Meada, M., Sasali, S.
Heterocycles 1999, 51, 975.
12. Analytical data for ꢀ-4: 1H NMR (C6D6): 2.17 (m, 2H,
H20, H200), 3.22 (m, 1H, CH2-Ph), 3.46 (m, 1H, CH2-Ph), 3.78
(m, 1H, H50, H500), 3.82 (2s, 6H, OMe), 3.88 (m, 1H, H50,
H500), 4.53 (m, 1H, H40), 4.58 (m, 1H, H30), 4.90 (m, 1H, H10),
7.26 (m, 4H, DMT), 7.53±7.75 (m, 8 H, DMT, Ph), 8.02 (m,
4H, DMT), 8.18 (m, 2H, DMT). Assignment with P. E.
COSY, TOCSY, HMQC and NOESY (120 ms, H10±H30 cross
1
peak). ESI-MS: 533.2 (M+Na)+. ꢁ-4: H NMR (C6D6): 2.13
(m, 1H, H200), 2.52 (m, 1H, H20), 3.28 (m, 1H, CH2-Ph), 3.51
(m, 1H, CH2-Ph), 3.78 (m, 1H, H50, H500), 3.80 (2s, 6H, OMe),
3.97 (m, 1H, H50, H500), 4.62±4.66 (m, 2H, H30, H40), 4.77 (m,
1H, H10), 7.22 (m, 4H, DMT), 7.55±7.73 (m, 8 H, DMT, Ph),
8.00 (m, 4H, DMT), 8.16 (m, 2H, DMT). Assignment with P.
E. COSY, TOCSY, HMQC and NOESY (120 ms, H10±H20
cross peak). ESI-MS: 533.2 (M+Na)+.
13. Analytical data for ꢀ-1: 1H NMR (CD2Cl2): 1.02±1.18 (m,
12H, iPr), 1.84 (m, 1H, H20, H200), 2.04 (m, 1H, H20, H200),
2.43 (t, 1H, J=6 Hz, CH2CN), 2.53 (t, 1H, J=6 Hz, CH2CN),
2.91 (m, 2H, CH2-Ph), 3.08 (m, 2H, H50, H500), 3.42±3.75 (m,
4H, iPr, CH2-CH2-CN), 3.78 (s, 3H, OMe), 3.79 (s, 3H, OMe),
4.02 (m, 1H, H40), 4.33±4.40 (m, 2H, H10, H30), 6.86 (m, 4H,
DMT), 7.15±7.38 (m, 12H, DMT, Ph), 7.42±7.50 (m, 2H,
DMT). 31P NMR (CD2Cl2): 148.2 (both diastereomers). ESI-
MS: 733.1 (M+Na)+. ꢁ-1: 1H NMR (C6D6): 0.96±1.20 (m,
12H, iPr), 1.78 (m, 1H, H20, H200), 2.22 (m, 1H, H20, H200),
2.28 (t, 1H, J=5 Hz, CH2CN), 2.46 (t, 1H, J=5 Hz, CH2CN),
2.80 (m, 1H, H50, H500), 2.97 (m, 2H, CH2-Ph), 3.09 (m, 1H,
H50, H500), 3.40±3.56 (m, 4H, iPr, CH2-CH2-CN), 3.70 (2s, 6H,
OMe), 4.15 (m, 1H, H40), 4.32 (m, 1H, H10, H30), 4.42 (m, 1H,
H10, H30), 6.67 (m, 4H, DMT), 7.05±7.20 (m, 12 H, DMT,
Ph), 7.34±7.38 (m, 2H, DMT). 31P NMR (C6D6): 145.0 (both
diastereomers). ESI-MS: 733.1 (M+Na)+.
16. Analytical data for ꢀ-2: 1H NMR (CD2Cl2): 1.08±1.21 (m,
12H, iPr), 1.86 (m, 1H, H20, H200), 2.43 (m, 1H, H20, H200),
2.44 (t, 1H, J=6 Hz, CH2CN), 2.60 (t, 1H, J=6 Hz, CH2CN),
2.63 (m, 1H, CH2-COOMe), 2,80 (m, 1H, CH2-COOMe),
3.02±3.20 (m, 2H, H50, H500), 3.44±3.75 (m, 4H, iPr, CH2±
CH2±CN), 3.69 (s, 3H, COOMe), 3.79 (2s, 6H, OMe), 4.18 (m,
1H, H40), 4.48 (m, 1H, H10, H30), 4.60 (m, 1H, H10, H30), 6.85
(m, 4H, DMT), 7.15±7.35 (m, 12 H, DMT, Ph), 7.42±7.48 (m,
2H, DMT). 31P NMR (CD2Cl2): 148.0 (both diastereomers).
ESI-MS: 715.1 (M+Na)+. ꢁ-2: 1H NMR (CD2Cl2): 1.15±1.30
(m, 12H, iPr), 1.80 (m, 1H, H20, H200), 2.17±2.46 (m, 2H, H20,
H200, CH2CN), 2.55±2.83 (m, 4H, H20, H200, CH2CN, CH2-
COOMe), 3.45±3.75 (m, 2H, H50, H500), 3.55 (m, 2H, iPr), 3.62
(s, 3H, COOMe), 3.79 (2s, 6H, OMe), 3.95±4.12 (m, 3H, H40,
H10, H30), 6.85 (m, 4H, DMT), 7.15±7.45 (m, 12 H, DMT,
Ph), 7.50±7.60 (m, 2H, DMT). 31P NMR (CD2Cl2): 147.6
(both diastereomers). ESI-MS: 715.0 (M+Na)+.
17. ESI-MS of the oligomers was performed in the negative
mode from a solvent mixture H2O:CH3CN:Et3N=68:30:2.
Analytical data for 10: ESI-MS: 1174.7 ((M-2)/2)2 ; 11: ESI-
MS: 1195.5 ((M-2)/2)2 ; 12: ESI-MS: 1174.9 ((M-2)/2)2 ; 13:
ESI±MS: 1194.5 ((M-2)/2)2 ; 14: ESI±MS: 1178.5 ((M-2)/2)2
;
15: ESI±MS: 1158.4 ((M-2)/2)2 ; 16: ESI-MS: 1178.3 ((M-2)/
2)2 ; 17: ESI±MS: 1158.4 ((M-2)/2)2
.
14. McDevitt, J. P.; Lansbury, P. T. Jr. J. Am. Chem. Soc.
1996, 118, 3818.
18. Biro, M.-C. Synthese und Eigenschaften isotopenmar-
kierter und durch C-Nucleotid-Fehlstellen modi®zierter Oli-
gonucleotide, Ph.D. thesis, TU-Munchen, 1999.
19. For a theoretical study of amino acid side chain nucleo-
base interactions see: Gresh, N.; Pullman, B. Biochim. Bio-
phys. Acta 1980, 608, 4753.
1
15. Analytical data for ꢀ-6: H NMR (CDCl3): 1.78 (m, 1H,
H20), 2.08 (2s, 6H, Ac), 2.29 (m, 1H, CH2-C2H3), 2.43 (m, 1H,
CH2-C2H3), 2.47 (m, 1H, H200), 4.05±4.20 (m, 3H, H40, H50,
H500), 4.17 (m, 1H, H10), 5.07 (m, 1H, H30), 5.12 (m, 2H, CH2-