β-Homoalanyl-PNA: A Special Case of β-Peptides with β-Sheet-Like Backbone Conformation; Organization in Higher Ordered Structures

Article abstract of DOI:10.1002/(SICI)1099-0690(199805)1998:5<827::AID-EJOC827>3.0.CO;2-B

Although β-homoalanyl peptide nucleic acid oligomers possess a β-peptide backbone, they contain nucleobases in their side chains and are thus an interesting special case of β-peptides. These nucleobases are mainly responsible for the formation of secondary structures through base pairing and stacking. We have investigated the pairing properties of β-homoalanyl-PNA oligomers using adenine and 7-carbaadenine. Simple model studies indicate the existence of Watson-Crick and Hoogsteen pairing planes as an intrinsic structural feature of purinyl β-PNA. As a consequence, the adeninyl-β-PNA hexamer and pentamer form higher ordered structures and the 7-carbaadenine-β-PNA hexamer, which lacks a Hoogsteen pairing site, pairs as a double strand. The synthesis of adeninyl and carbaadeninyl β-nucleoamino acids and their oligomerization is described. Pairing mode and strand orientation were investigated through experiment and simple model studies. In the context of β-peptides the β-PNA oligomers for the most part tend to exist as double strands and prefer the extended β-sheet-like backbone conformation which provides an interesting structure motif in addition to the 3₁-helix observed for β-peptides.

Full text of DOI:10.1002/(SICI)1099-0690(199805)1998:5<827::AID-EJOC827>3.0.CO;2-B

FULL PAPER
β-Homoalanyl-PNA: A Special Case of β-Peptides with β-Sheet-Like Backbone
Conformation; Organization in Higher Ordered Structures
Ulf Diederichsen* and Harald W. Schmitt
Institut für Organische Chemie und Biochemie der Technischen Universität München,
Lichtenbergstraße 4, D-85747 Garching, Germany
Fax: (internat.) ϩ 49(0)89/2891-3210
E-mail: ud@linda.org.chemie.tu-muenchen.de
Received December 29, 1997
Keywords: Adenine–adenine base pairs / β-Amino acids / Mitsunobu reaction / Peptide nucleic acid (PNA) /
Self organization
Although β-homoalanyl peptide nucleic acid oligomers
structures and the 7-carbaadenine-β-PNA hexamer, which
possess a β-peptide backbone, they contain nucleobases in lacks a Hoogsteen pairing site, pairs as a double strand. The
their side chains and are thus an interesting special case of
β-peptides. These nucleobases are mainly responsible for the
formation of secondary structures through base pairing and
stacking. We have investigated the pairing properties of β-
homoalanyl-PNA oligomers using adenine and 7-
carbaadenine. Simple model studies indicate the existence of
synthesis of adeninyl and carbaadeninyl β-nucleoamino
acids and their oligomerization is described. Pairing mode
and strand orientation were investigated through experiment
and simple model studies. In the context of β-peptides the β-
PNA oligomers for the most part tend to exist as double
strands and prefer the extended β-sheet-like backbone
Watson-Crick and Hoogsteen pairing planes as an intrinsic conformation which provides an interesting structure motif
structural feature of purinyl β-PNA. As a consequence, the
adeninyl-β-PNA hexamer and pentamer form higher ordered
in addition to the 31-helix observed for β-peptides.
The β-peptides are an especially interesting group of Hoogsteen pairing mode. Model studies indicate that base
polyamide oligomers[1]. They consist of β-amino acid pairing is responsible for the organization of β-PNA in β-
monomers oligomerized in the usual peptidic manner. β- strand-like structures. This provides the β-sheet-like back-
Peptide hexamers are already self-organizing in a stable, bone conformation, which occurs in addition to the 31-helix
helical secondary structure[2][3]. In contrast, α-peptides as a second structural motif for β-peptides.
show no helix formation in aqueous solution up to lengths
Scheme 1. β-Homoalanyl nucleo amino acids are the building units
of about 20 amino acids[4]. The 31-helix formed by β-pep-
of β-homoalanyl-PNA oligomers which should stabilize
in linear double strands by base pairing
tides requires exactly three amino acids for one turn, orient-
ing every fourth side chain in the same direction. The helix
formation is determined by the stereogenic centers provid-
ing, e.g. a 2.51-helix or even a pleated-sheet arrangement as
structural motifs along with the 31-helical structure[2]. The
possibility of designing β-peptides with a defined secondary
structure is as promising as the high resistance against enzy-
matic degradation.
As part of our ongoing studies of α-alanyl and α-homo-
alanyl peptide nucleic acids (PNAs)[5] we synthesized β-
homoalanyl-PNAs which are β-peptides consisting of nu-
Synthesis of Nucleo-β-amino Acids and Oligomerization
cleo-β-amino acids (Scheme 1). The nucleobases, which are
The β-PNA oligomers used for the structural studies were
connected at the γ-position to β-homoalanine, are a func- based on the nucleo amino acids Boc-adeninyl-β-homoala-
tional part of the side chain. The donor/acceptor pattern of nine (1) and Boc-7-carbaadeninyl-β-homoalanine (2)
purines and pyrimidines provides an additional possibility (Scheme 2). Many methods are known for the synthesis of
for recognition and for determining the secondary struc- β-amino acids[7] which are not transferable to the prep-
ture. We describe the intrinsic tendency of β-homoalanyl- aration of nucleo-β-amino acids because of the competition
PNA to organize in strands with extended conformation of N-9/N-7 purine attack or N-1/N-3 pyrimidine attack, in-
which aggregate in higher ordered structures. Furthermore, tramolecular azridine formation and the poor solubility of
the high propensity of the adenine β-PNA hexamer and the nucleobases in most organic solvents. Furthermore, the
pentamer to aggregate[6] was examined by a chemical modi- Arndt-Eistert homologisation of aromatic α-amino acids is
fication; the 7-carbaadenine β-PNA oligomers exclude the expected to proceed with racemisation[7e]. Therefore, the
Eur. J. Org. Chem. 1998, 827Ϫ835
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ193X/98/0505Ϫ0827 $ 17.50ϩ.50/0
827

U. Diederichsen, H. W. Schmitt
FULL PAPER
key step for the synthesis of nucleo-β-amino acids is the amino acid 9 were obtained, so the Mitsunobu reaction
substitution of a purine or pyrimidine at the γ-position of leading to the 7-carba analogon 10 was performed at
the β-homoalanine using the Mitsunobu reaction[8]. The re- Ϫ78°C adding DEAD to a mixture of β-homoserine (4), 6-
quired β-homoserine derivative 4 was available by reduction chloro-7-carbapurine (6), and triphenylphosphane in THF.
of the C-terminal carbonyl group of Boc--Asp(Bzl)-OH (3) The conversion of N-Boc-amino acids into N-Boc-aziri-
in 95% yield[9]. 6-Chloropurine (5) and 6-chloro-7-carbapu- dines under Mitsunobu conditions appears to predominate
rine (6) were used because adenine failed in the Mitsunobu as the competing reaction[11]. Therefore, the Mitsunobu re-
reaction. 6-Chloro-7-carbapurine (6) was synthesized by the agents and 6-chloro-7-carbapurine (6) were dissolved in
Davoll procedure starting with bromacetaldehyde diethyl THF at Ϫ78°C and added to a solution of β-homoserine
acetal (7) and ethyl cyanoacetate (8) in four steps (Scheme derivative ent-4 in THF at Ϫ20°C in a latter experiment.
3)[10]
.
The yield was improved to 60%, probably due to suppres-
Scheme 2. Synthesis of the nucleo amino acids Boc-adeninyl-β-homoalanine (1) and Boc-7-carbaadeninyl-β-homoalanine (2) using the
Mitsunobu reaction as the key step for connecting the nucleobases with the side chain of the β-amino acid
The Mitsunobu reaction is known to be useful, e.g. for sion of the side reaction leading to aziridine. Conversion of
connecting nucleobases to propanoic acid derivatives[8a], to the chloropurinyl derivatives 9 or 10 to the adeninyl nucleo
the anomeric position of furanoses[8g] and to carbacy- amino acids 11 or 12 was achieved by the substitution of
cles[8c][8d]. The reaction conditions vary with regard to tem- the azide for chloride[12]. Reduction with H2, PdO·H2O or
perature (reflux or Ϫ78°C) and to the sequence for adding H2, Pd/C yielded the desired nucleo amino acids 1[13] and
the components. By analogy to these procedures, the β- 2. The azide of the 6-chloropurine derivative 9 was obtained
homoserine derivative 4 was added to 6-chloropurine (5) using sodium azide in 2-O-methylglycol. For the conversion
together with DEAD and triphenylphosphane in THF at of 7-carbapurine derivative 10, DMF was used as the sol-
room temperature. Only modest yields (26Ϫ40%) of nucleo vent because the lower reactivity of the purine moiety of 10
828
Eur. J. Org. Chem. 1998, 827Ϫ835

β-Homoalanyl-PNA: A Special Case of β-Peptides with β-Sheet-Like Backbone Conformation
FULL PAPER
Scheme 3. Synthesis of 6-chloro-7-carbapurine (6) following a pro-
Scheme 4. β-Homoalanyl-PNA in the extended backbone confor-
mation provides uniform alignment of all nucleobases;
all WatsonϪCrick sites, as well as all Hoogsteen sites
simultaneously spread pairing planes suitable for recog-
nition which could lead to higher ordered structures
cedure of Davoll[10]
led to transesterification with 2-O-methylglycol. As known
from the literature the azidopurines 11 and 12 are in a sol-
vent-dependent equilibrium with the tetrazole derivatives 13
and 14, respectively[14]. In chloroform, the tetrazole isomer
13 is favored by 2:1 compared to the azidopurine 11,
whereas the equilibrium of tetrazole 14 and 6-azido-7-car-
bapurine 12 is 19:1, as determined by 1H-NMR spec-
troscopy. In methanol or DMSO the tetrazole isomer 13
predominates.
The nucleo amino acids 1 and 2 were oligomerized by
solid-phase peptide synthesis (SPPS) on a 4-methylbenzhy-
drylamine-(MBHA-)polystyrene support loaded with (R)-
lysine(Z)ϪOH (Z ϭ benzyloxycarbonyl)[15]. A fourfold ex-
cess of nucleo amino acid was coupled in DMF using O-
(7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexa-
fluorophosphate (HATU) and N,N-diisopropylethylamine
for activation. Each coupling step was performed with at
least 97% yield. Purification of the oligomers by HPLC
(RP-C18) was followed by characterization using electro-
spray-ionisation mass spectrometry (ESI-MS) and 1H-
NMR spectroscopy (see Experimental Section).
mer 15 (Tm > 90°C, > 70% hyperchromicity, 3 µ) and
pentamer 16 (Tm ϭ 72°C, > 60% hyperchromicity, 4 µ)
were extraordinarily high (Schemes 5A and 5B). Even tetra-
mer 17 formed a reasonably stable double strand (Tm
ϭ
26°C, 16% hyperchromicity, 5 µ) (Scheme 5C). The sta-
bilities (Tm) and hyperchromicities were determined by UV
melting curves. As destacking of the nucleobases results in
an increase of UV absorption, cooperative dissociation of a
double strand can be recognized as a sigmoidal-shaped
curve when measuring absorption with increasing tempera-
ture. Therefore, the melting curves mainly result from nu-
cleobase interactions. UV and CD measurements were per-
formed in an aqueous solution of 0.1  NaCl at pH ϭ 7
(0.01  Na2HPO4/H3PO4). The dissociation of all β-PNA
oligomer complexes was shown to be reversible with hyster-
esis. To prove the influence of NaCl on the pairing stability
we also measured the UV and CD spectra for hexamer 15
at pH ϭ 7 without NaCl; this did not significantly change
the profile of the melting curve or CD spectra (not shown).
In contrast to α-alanyl-PNA[18], no salt dependence of the
Adenine؊Adenine Pairing of β-PNA Oligomers
Simple model studies of a β-homoalanyl-PNA single pairing complex stability for β-PNA was observed.
strand show particular potential for the formation of higher The unusually high stability and hyperchromicity of
ordered structures (Scheme 4). In the extended backbone hexamer 15 and pentamer 16 are not only due to double-
conformation, a homochiral β-homoalanyl-PNA strand strand formation but result from higher aggregation. As
provides a uniform orientation of all side chains. Therefore, pointed out in Scheme 4, a β-PNA strand in the extended
all adenine bases are uniformly aligned and all Watson- conformation has two pairing planes. By simultaneous pair-
Crick and Hoogsteen pairing sites each have the same order ing of the WatsonϪCrick and Hoogsteen pairing planes,
of donor/acceptor functionalities and the same orientation. aggregation of β-PNA strands is possible, forming band
As a consequence, purinyl-β-homoalanyl-PNA in the ex- structures based on nucleobase recognition. There are two
tended conformation provides two pairing planes based on possibilities for the formation of higher ordered PNA ag-
all WatsonϪCrick sites and on all Hoogsteen sites, respec- gregates: (A) either all adenines pair in the Hoogsteen mode
tively.
or (B) the reversed WatsonϪCrick and reversed Hoogsteen
The adenineϪadenine self-pairing is not observed in B- mode alternate (Scheme 6). In case (A) only one kind of
DNA[16] because of the helix topology but is known for pairing mode is involved and all oligomers have antiparallel
oligomers with nearly linear backbones[17]. We examined strand orientation. In contrast, in case (B) two different
AϪA pairing in the β-PNA series by synthesis of the three alternating pairing modes are required which would be
sequences H-(β-HalA)6-Lys-NH2 (15), H-(β-HalA)5-Lys- recognized as two steps in the melting curve. Furthermore,
NH2 (16), and H-(β-HalA)4-Lys-NH2 (17) which differ only all strands should be oriented parallel with all positively
in length [β-HalA ϭ γ-(adenine-9-yl)-β-(S)-homoalanine, charged (at pH ϭ 7) lysine amides directed to the same side.
Lys-NH2 ϭ (R)-lysine amide[15] [6]
]
. The stabilities of hexa- Therefore, we favor case (A) as the most likely structure of
Eur. J. Org. Chem. 1998, 827Ϫ835
829

U. Diederichsen, H. W. Schmitt
FULL PAPER
Scheme 5. Melting curves of H-(β-HalA)6-Lys-NH2 (15), H-(β-HalA)5-Lys-NH2 (16), H-(β-HalA)4-Lys-NH2 (17), H-(β-Hal7CA)6-Lys-
NH2 (18), H-(β-Hal7CA)5-Lys-NH2 (19) and H-(β-Hal7CA)4-Lys-NH2 (20) (0.01  Na2HPO4/H3PO4, pH ϭ 7.0, 0.1  NaCl,
270 nm)
7-Carbaadenine/7-Carbaadenine Pairing of β-PNA
the β-PNA aggregates. Tetramer 17 has a significantly lower
stability and hyperchromicity indicating double-strand for-
mation. AϪA-pairing double strands in the α-PNA series
typically result in a hyperchromicity that is lower than
20%[5a], whereas aggregation of GϪC-pairing α-PNA leads
to an enormous increase in hyperchromicity[5b]. It is quite
remarkable that an adenine tetramer still forms a stable
pairing complex, therefore providing a strong argument for
complexation which cannot result from a 31-helix, as is
Oligomers
To further prove the aggregation model and to obtain
information about the AϪA-pairing mode we synthesized
β-PNAs with 7-carbaadenine instead of adenine as the rec-
ognition unit. This modification does not change the geo-
metrical requirements or potential backbone stabilization
but excludes pairing of the Hoogsteen side. Therefore, if
pairing is found for the 7-carbaadenine oligomers H-(β-
Hal7CA)6-Lys-NH2 (18) and H-(β-Hal7CA)5-Lys-NH2 (19)
[β-Hal7CA ϭ γ-(7-carbaadenine-9-yl)-β-(S)-homoalanine],
it would only be possible as a double strand according to
the reversed WatsonϪCrick pairing mode (Scheme 7). The
UV melting curves shown in Scheme 5 (D and E) indicate
a stability of Tm ϭ 35°C (24% hyperchromicity, 2 µ) for
hexamer 18 and a melting point Tm ϭ 31°C (14% hyper-
chromicity, 2 µ) for pentamer 19. We also measured the
tetramer H-(β-Hal7CA)4-Lys-NH2 (20) (Scheme 5F) which
showed Tm ϭ 28°C (15% hyperchromicity, 2 µ)[19]. All
three 7-carbaadenine oligomers showed melting curves typi-
cal for dissociation of a double strand. A difference of one
base pair in length results in an increased stability of only
known for β-peptides[2][3]
.
Scheme 6. Two possibilities for aggregation of adeninyl-β-homoala-
nyl-PNA oligomers: (A) all nucleobases pair in the
Hoogsteen mode providing alternating backbone orien-
tation; (B) each nucleobase aggregates with the reverse
WatsonϪCrick and the reverse Hoogsteen mode; all
strands are oriented alike
3Ϫ4°C. The stability of the adeninyl tetramer 17 (Tm
ϭ
26°C, 16% hyperchromicity, 5 µ) is in the same order as
the 7-carbaadeninyl β-PNA double strand. Further, the cir-
cular dichroism (CD) spectra of the adeninyl tetramer 17
and 7-carbaadeninyl pentamer 19 as well as hexamer 18
(not shown) are similar, especially when compared to the
higher organized adeninyl pentamer 16 (Scheme 8). With
CD spectroscopy it is possible to distinguish between the
pairing complex in which the chromophore (nucleobase) is
conformationally fixed in a chiral environment (backbone)
and the statistical distribution of a single strand at higher
temperatures. Within the same class of backbone a similar
830
Eur. J. Org. Chem. 1998, 827Ϫ835

β-Homoalanyl-PNA: A Special Case of β-Peptides with β-Sheet-Like Backbone Conformation
FULL PAPER
shape observed for CD spectra measured at different tem- show the highest Cotton effect that we have observed so far
peratures indicates a comparable pairing mode. Therefore, for alanyl- or homoalanyl-PNA. The small Cotton effects
by analogy to the 7-carbaadeninyl oligomers we suggest the might result from a C2-symmetrical pairing complex for the
pairing mode for the double strand of adeninyl tetramer 17 reverse WatsonϪCrick pairing mode and parallel strand
to be reverse WatsonϪCrick. It is also noteworthy that the orientation. In contrast, symmetry is lost for the Hoog-
CD effects we observed for the double strands are very steen-pairing higher ordered structure with all strands being
small whereas the higher ordered structures of 15 and 16 antiparallel (Scheme 6A).
Backbone Orientation
Scheme 7. The only possibility for self-pairing of a 7-carbaadenine
is the reverse WatsonϪCrick mode
A 7-carbaadeninyl-β-PNA oligomer will form a self-pair-
ing complex only as a double strand in the reversed Wat-
sonϪCrick mode with parallel strand orientation. The equi-
molar mixture of the enantiomeric oligomers H-(β-
Hal7CA)6-Lys-NH2 (18) and H-(β-Hal7CA)6-Lys-NH2 (ent-
18) might also associate in duplexes by forming the reversed
WatsonϪCrick mode but this would require antiparallel
oriented backbones. As shown in Scheme 9A, the mixture
of the two enantiomers 18 and ent-18 has a UV melting
curve with Tm ϭ 38°C (25% hyperchromicity, 2 µ each).
Therefore, the antiparallel strand orientation does not have
significantly higher stability than the parallel self-pairing
orientation. Since the pairing geometry is the same in both
cases, the differentiation between parallel and antiparallel
oriented strands might result only from charge separation
of the protonated lysine amides in the antiparallel case. The
Scheme 8. CD spectra of H-(β-Hal7CA)5-Lys-NH2 (19), H-(β-
HalA)4-Lys-NH2 (17) and H-(β-HalA)5-Lys-NH2 (16)
(0.01  Na2HPO4/H3PO4, pH ϭ 7.0, 0.1  NaCl, 270
nm); the CD spectra of oligomers 17 and 19 are similar,
indicating double-strand formation in the reverse Wat-
sonϪCrick mode for both; the higher aggregation of β-
PNA 16 is also indicated by a drastic change of the
CD spectra
Scheme 9. Melting curves of the equimolar mixture of H-(β-
Hal7CA)6-Lys-NH2 (18) and H-(β-Hal7CA)6-Lys-NH2
(ent-18) (A) and H-(β-HalA-Hal7CA)3-Lys-NH2 (21)
(0.01  Na2HPO4/H3PO4, pH ϭ 7.0, 0. 1  NaCl, 270
nm)
Eur. J. Org. Chem. 1998, 827Ϫ835
831

U. Diederichsen, H. W. Schmitt
FULL PAPER
comparable stability of parallel and antiparallel oriented Therefore, the geometrical requirements of one base pair
strands is in agreement with base-pair planes being orthog- are also valid for the whole strand. AdenineϪadenine self-
onal to the backbone; this is the only angle which does not pairing is possible with three pairing modes (Scheme 10).
effect the strand orientation[17]. We can further conclude The reverse WatsonϪCrick and reverse Hoogsteen pairing
that the influence of the C-terminal lysine amides on the modes are both C2-symmetrical with both adenines being
strand orientation is small or negligible.
equal pairing partners. They are interconvertible only by a
twist of the strands without changing the strand orien-
tation. In linear self-pairing double strands, the C2-sym-
metrical base pairs, and the respective backbones have the
same axis of symmetry. Therefore, the strand orientation of
self-pairing β-homoalanyl-PNA strands in the reverse Wat-
sonϪCrick or reverse Hoogsteen mode is parallel (Scheme
10A).
Alternating Adeninyl/7-Carbaadeninyl β-Homoalanyl-PNA
The alternating adeninyl/7-carbaadeninyl oligomer H-(β-
HalA-Hal7CA)3-Lys-NH2 (21) was expected to pair as a
double strand because three adenine nucleobases should
not be sufficient to form higher ordered structures. Indeed,
for oligomer 21, double-strand formation with a stability
of Tm ϭ 30°C (11% hyperchromicity, 5 µ) was observed
The two adenine nucleobases in the AϪA Hoogsteen
(Scheme 9B). Therefore, it is possible to suppress the forma- mode are not equivalent; one nucleobase pairs on the Wat-
tion of higher ordered structures by incorporation of 7-car- son-Crick side, the other pairs with the Hoogsteen side. The
baadenine which avoids Hoogsteen pairing. Oligomer 21 latter requires a reversed orientation of donor/acceptor po-
has a stability 5°C lower compared to oligomer H-(β- sitions compared to the symmetrical AϪA pairing. There-
Hal7CA)6-Lys-NH2 (18) which indicates that in a β-homo- fore, AϪA self-pairing in the Hoogsteen mode is possible
alanyl-PNA 7-carbaadenine contributes more to hydrogen only with antiparallel strand orientation (Scheme 10C).
bonding or stacking than adenine.
In α-peptide chemistry simultaneous altering of the
strand orientation (inversion of the sequence) and total in-
version of configuration does not change the overall side-
Discussion of Pairing Specificities
For double-strand formation complementary donor and chain geometry[20]. This retro-inverso concept is still valid
acceptor positions are required. In the extended confor- for β-peptides and β-PNA. Therefore, pairing of β-homo-
mation of β-homoalanyl-PNA, all purines and all donor/ alanyl-PNA in the reverse WatsonϪCrick mode should also
acceptor positions of a single strand are aligned alike. be possible with antiparallel backbone orientation and one
Scheme 10. A β-PNA double strand based on AϪA pairing is schematically shown in Scheme 10 with the backbone (arrows for the
orientation) and the nucleobases with the donor (D) and acceptor (A) positions indicated; the difference in base pairing
between the reverse WatsonϪCrick or reverse Hoogsteen and the Hoogsteen mode is one adenine unit which changes from
the WatsonϪCrick to the Hoogsteen site by flipping and rotation operations changing the backbone orientation
832
Eur. J. Org. Chem. 1998, 827Ϫ835

β-Homoalanyl-PNA: A Special Case of β-Peptides with β-Sheet-Like Backbone Conformation
FULL PAPER
chloroform/acetonitrile (9:1), and dichlormethane/acetonitrile
gradient (9:1Ϫ4:1) to give 1.30 g of 10 (30%) as a colorless solid,
m.p. 156Ϫ157°C. RF [chloroform/acetonitrile (9:1)] ϭ 0.46. Ϫ
strand having the inversed configuration (Scheme 10B).
Furthermore, a parallel oriented double strand should be
preferred for Hoogsteen pairing of an all-(S)- with an all-
(R)-configured backbone (Scheme 10D).
In contrast to β-peptides, the secondary structure of β-
homoalanyl peptide nucleic acids is determined by the hy-
drogen bonds and aromatic π interaction (stacking) of the
nucleo amino acid side chains. Therefore, an extended back-
bone conformation is stabilized by base pairing and pro-
20
[α]589 ϭ 11.7 (c ϭ 1.1, CH2Cl2) Ϫ ESI MS; m/z (%): 445.2 (100)
[M ϩ H]ϩ Ϫ IR (KBr): ν˜ ϭ 3365 cmϪ1, 2959, 1739, 1683, 1587,
1
1546, 1530, 1510, 1420, 1354, 1286, 1200, 1134. Ϫ H NMR (250
MHz, CDCl3): δ ϭ 1.38 (s, 9 H, Boc), 2.51Ϫ2.63 (m, 2 H, α-H),
4.26Ϫ4.45 (m, 3 H, β-H, γ-H), 5.11 (s, 2 H, PhCH2), 5.41 (d, J ϭ
8 Hz, 1 H, BocNH), 6.55 (d, J ϭ 4 Hz, 1 H, 7-H), 7.17 (d, J ϭ 4
Hz, 1 H, 8-H), 7.28Ϫ7.39 (m, 5 H, Ph), 8.53 (s, 1 H, 2-H). Ϫ 13C
vides an additional motif for β-peptide scaffolds. Compared NMR (62 MHz, CDCl3): δ ϭ 28.1, 36.1, 47.5, 48.1, 66.8, 79.8,
99.8, 104.0, 128.4, 128.5, 128.6, 129.9, 135.3, 150.6, 150.7, 151.4,
152.2, 154.9, 170.8. Ϫ C22H25ClN4O4 (444.9): calcd. C 59.39, H
5.66, N 12.59; found C 59.27, H 5.55, N 12.51.
to α-alanyl-PNA stabilizing in linear double strands in the
β-sheet backbone conformation, the β-homoalanyl-PNA
oligomers are remarkable because of their uniformly
aligned nucleobases. This allows higher aggregation to hy-
drogen-bonded polymeric structures but also provides
stacking of nucleobases in the same orientation and directly
on top of each other. Despite a larger base-pair distance,
pairing of β-homoalanyl PNA double strands is more stable
than that of comparable α-alanyl PNA duplexes which indi-
cates β-homoalanyl-PNA with all canonical nucleobases be-
ing a valuable tool for experiments determining nucleobase-
Ϫamino acid side-chain interactions, intercalation and tem-
plate-directed oligomerization.
(R)-N-tert-Butoxycarbonyl-γ-(6-chloro-7-carba-9-purinyl)-β-
homoalanine Benzyl Ester (ent-10): The synthesis followed the pro-
cedure for enantiomer 10 except for the order of adding the re-
agents. A mixture of the Mitsunobu reagents and 6-chloro-7-carba-
purine (6) in THF was stirred at Ϫ78°C and added to N-Boc-β-
homoserine benzyl ester (ent-4) within 5 min at Ϫ20°C. Amino
acid ent-10 was obtained in 60% yield. The analytical data of ent-10
20
and 10 are identical except for [α]589 ϭ Ϫ11.9 (c ϭ 1.0, CH2Cl2).
(S)-N-tert-Butoxycarbonyl-γ-(6-azido-7-carba-9-purinyl)-β-
homoalanine Benzyl Ester (12): A suspension of (S)-N-Boc-γ-(6-
chloro-7-carba-9-purinyl)-β-homoalanine benzyl ester (10) (1.03 g,
2.31 mmol) and sodium azide (2.25 g, 34.67 mmol) in DMF (30
ml) was heated for 12 h at 60°C. After evaporation of the solvent,
the residue was suspended in water and extracted with dichloro-
methane. Purification of the organic phase was done by flash chro-
matography using 170 g silica gel and hexane/acetone (7:3) gave
643 mg of 12 (62%) against 362 mg recovered starting material
which was converted in a second cycle to an additional 297 mg
of 12 (28%) next to 38 mg starting material, colorless solid, m.p.
This work was supported by the Hermann-Schlosser-Stiftung (Ph.
D. fellowship for H. W. S.) and the Deutsche Forschungsgemein-
schaft (Di 542 2-1). We are grateful to Professor H. Kessler, Gar-
ching, for generous support.
Experimental Section
General: All reagents were of analytical grade and used as sup-
plied. Solvents were of the highest grade available. Ϫ NMR spectra
were recorded with a Bruker AC 250, Bruker AMX 500 or Bruker
DMX 500. Ϫ ESI mass spectra were measured with a TSQ 700
Finnigan spectrometer. Ϫ For optical rotation a Perkin-Elmer
Polarimeter 241 MC, for elemental analysis a Heraeus EA 415-0
and for melting-point determination a Büchi SMP-20 apparatus
was used. Ϫ HPLC was done with Beckman System Gold using
YMC-Pack ODS, RP-C18, 250 ϫ 20 mm, 5 µm, 120 A for prep-
aration and 250 ϫ 4.6 mm, 5 µm, 120 A for analytical samples.
Ϫ UV melting curves were measured with a Perkin-Elmer UV/Vis
Lambda 10 with Peltier Temperature Programmer PTP 1 and CD
spectra were performed with Jasco J-710 with Peltier-Type Tem-
perature Control System PTC-348W. Ϫ The optical purity was de-
termined by HPLC analysis of the amino acid dimers obtained with
Boc-(S)-Ala-OSu. Ϫ The oligomer concentration was calculated by
taking the extinction coefficient at 90°C as being the sum of the
20
139Ϫ140°C. RF [hexanes/acetone (7:3)] ϭ 0.21. Ϫ [α]589 ϭ 29
(c ϭ 1.0, CHCl3). Ϫ ESI MS; m/z (%): 452.3 (100) [M ϩ H]ϩ. Ϫ
IR (KBr): ν˜ ϭ 3357 cmϪ1, 3120, 2966, 1742, 1681, 1633, 1531,
1
1484, 1417, 1393, 1366. Ϫ H NMR (250 MHz, CDCl3): δ ϭ 1.28
(s, 9 H, Boc), 2.52Ϫ2.71 (m, 2 H, α-H), 4.35Ϫ4.48 (m, 1 H, β-H,
γ-H), 4.48Ϫ4.62 (m, 2 H, β-H, γ-H), 5.06Ϫ5.19 (m, 2 H, PhCH2),
5.36 (d, J ϭ 8 Hz, 1 H, BocNH), 6.41 (d, J ϭ 4 Hz, 0.05 H, 7-H
azide) 7.02 (d, J ϭ 4 Hz, 0.95 H, 7-H tetrazole), 7.09 (m, 0.05 H,
8-H azide), 7.26 (d, J ϭ 3 Hz, 0.95 H, 8-H tetrazole), 7.29Ϫ7.39
(m, 5 H, Ph), 8.50 (s, 0.05 H, 2-H azide), 9.12 (s, 0.95 H, 2-H
tetrazole). Ϫ 13C NMR (62 MHz, CDCl3): δ ϭ 28.1, 35.9, 48.0,
48.4, 66.8, 78.0, 101.9, 103.2, 128.3, 128.4, 128.5, 128.6, 131.2,
135.2, 141.6, 146.6, 154.9, 170.7. Ϫ C12H25N7O4 (451.5): calcd. C
58.53, H 5.58, N 21.72; found C 58.87, H 5.60, N 21.87.
(R)-N-tert-Butoxycarbonyl-γ-(6-azido-7-carba-9-purinyl)-β-
extinction coefficients of the nucleo amino acids [ε(1) ϭ 14000 at homoalanine Benzyl Ester (ent-12): The synthesis followed the pro-
263 nm]. Ϫ The experimental procedures and analytical data of
compounds 1, 9, 11/13 and oligomers 15, 16, 17 are described in
cedure for enantiomer 12. The analytical data of ent-12 and 12 are
identical except for [α]589 ϭ Ϫ29.7 (c ϭ 0.9, CHCl3).
20
the preceding communication[6]
.
(S)-N-tert-Butoxycarbonyl-γ-(7-carba-9-adeninyl)-β-homoala-
(S)-N-tert-Butoxycarbonyl-γ-(6-chloro-7-carba-9-purinyl)-β- nine (2): Palladium oxide hydrate (200 mg, 1.42 mmol) was added
homoalanine Benzyl Ester (10): (S)-N-Boc-β-homoserine benzyl es-
ter 4 (3.00 g, 9.69 mmol), 6-chloro-7-carbapurine (6) (1.49 g, 9.69
mmol), and triphenylphosphane (10.16 g, 38.76 mmol) were dis-
to a solution of (S)-N-Boc-γ-(6-azido-7-carba-9-purinyl)-β-homo-
alanine benzyl ester (12) (917 mg, 2.03 mmol) in methanol (80 ml)
and acetic acid (400 µl). After saturation with hydrogen, the reac-
solved in anhydrous THF (140 ml) and cooled to Ϫ78°C. Within tion mixture was stirred overnight. The precipitated nucleo amino
15 min DEAD (6.09 ml, 38.76 mmol) was added and the reaction acid was dissolved in acetonitrile/water (2:3) and separated from
mixture kept at this temperature for 90 min. After stirring for three the catalyst. Evaporation of the solvent gave a violet solid which
days at Ϫ20°C the reaction was quenched with water (2 ml) and
was crystallized from hot methanol to yield 409 mg (60%) 2, color-
the solvent evaporated. Purification was done by consecutive flash less solid, m.p. 232Ϫ234°C (96% e.e.). Ϫ RF [ethyl acetate/meth-
chromatography (170 g silica gel) with hexanes/ethylacetate (3:2), anol/water/acetic acid (10:1:1:0.5), saturated with NaCl] ϭ 0.54; Ϫ
Eur. J. Org. Chem. 1998, 827Ϫ835
833

U. Diederichsen, H. W. Schmitt
FULL PAPER
20
[α]589 ϭ 38.5 (c ϭ 0.8, DMSO). Ϫ ESI MS; m/z (%): 336.1 (100) H, 8-H), 7.19 (d, J ϭ 3 Hz, 1 H, 8-H), 7.23 (d, J ϭ 3 Hz, 1 H, 8-
[M ϩ H]ϩ. Ϫ IR (KBr): ν˜ ϭ 3420 cmϪ1, 2983, 1699, 1680, 1565, H), 7.35 (d, J ϭ 3 Hz, 1 H, 8-H), 8.13 (s, 1 H, 2-H), 8.14 (s, 1 H,
1512, 1394, 1368, 1281, 1247, 1166. Ϫ 1H NMR (500 MHz, 2-H), 8.15 (m, 2 H, 2-H), 8.17 (m, 2 H, 2-H).
[D6]DMSO): δ ϭ 1.05 (s, br., 1.5 H, Boc-rotamer), 1.25 (s, br., 7.5
H-[(R)-β-Hal7CA]6-(S)-Lys-NH2 (ent-18): The synthesis fol-
H, Boc), 2.28Ϫ2.45 (m, 2 H, α-H), 4.00Ϫ4.15 (m, 2 H, γ-H),
4.18Ϫ4.29 (m, 1 H, β-H), 6.32Ϫ6.40 (m, 0.16 H, 7-H rotamer),
6.50 (d, J ϭ 3 Hz, 0.84 H, 7-H), 6.81 (d, J ϭ 8 Hz, 0.84 H,
BocNH), 6.90 (s, br., 2 H, NH2), 6.93Ϫ6.98 (m, 0.16 H, 8-H rot-
lowed the general method for SPPS and the spectroscopic data are
in agreement with those for enantiomer 18.
H-[(S)-β-Hal7CA]5-(R)-Lys-NH2 (19): The synthesis followed
amer), 7.00 (d, J ϭ 3 Hz, 0.84 H, 8-H), 8.01 (s, 1 H, 2-H). Ϫ 13C
NMR (125 MHz, [D6]DMSO): δ ϭ 28.1, 36.9, 46.9, 48.1, 77.7,
98.4, 102.3, 124.4, 149.8, 151.4, 154.7, 157.3, 172.0. Ϫ C15H21N5O4
(335.4): calcd. C 53.72, H 6.31, N 20.88; found C 53.77, H 6.34,
N 20.44.
the general method for SPPS. Ϫ ESI MS; m/z (%): 1231.6 (100) [M
ϩ H]ϩ, 616.6 (95) [M ϩ 2 H]2ϩ. Ϫ HPLC: 25.2 min, gradient
1
5Ϫ20% acetonitrile in 30 min. Ϫ H NMR (500 MHz, D2O): δ ϭ
1.32Ϫ1.49 (m, 2 H, H-Lys), 1.60Ϫ1.72 (m, 3 H, H-Lys), 1.73Ϫ1.82
(m, 1 H, H-Lys), 2.13Ϫ2.45 (m, 7 H, α-H), 2.48Ϫ2.59 (m, 2 H, α-
H), 2.70Ϫ2.79 (m, 1 H, α-H), 2.95 (t, J ϭ 8 Hz, 2 H, ε-H-Lys),
3.20 (m, 2 H, β-H, γ-H), 3.49 (m, 1 H, β-H, γ-H), 3.56 (m, 1 H, β-
H, γ-H), 3.67 (m, 1 H, β-H, γ-H), 3.70Ϫ3.82 (m, 2 H, β-H, γ-H),
3.88 (m, 1 H, β-H, γ-H), 4.03Ϫ4.11 (m, 1 H, γ-H), 4.18 (m, 1
H, α-H-Lys), 4.23Ϫ4.33 (m, 1 H, γ-H), 4.35Ϫ4.45 (m, 2 H, γ-H),
4.45Ϫ4.53 (m, 1 H, γ-H), 4.53Ϫ4.62 (m, 1 H, γ-H), 4.78Ϫ4.88 (m,
1 H, γ-H), 6.58Ϫ6.62 (m, 2 H, 7-H), 6.63 (d, J ϭ 3 Hz, 1 H, 7-H),
6.64 (d, J ϭ 3 Hz, 1 H, 7-H), 6.76 (d, J ϭ 3 Hz, 1 H, 7-H), 7.06
(m, 2 H, 8-H), 7.13 (d, J ϭ 3 Hz, 1 H, 8-H), 7.21 (d, J ϭ 4 Hz, 1
H, 8-H), 7.35 (d, J ϭ 4 Hz, 1 H, 8-H), 8.12 (m, 2 H, 2-H), 8.16
(m, 3 H, 2-H).
H-[(R)-β-Hal7CA]4-(S)-Lys-NH2 (20): The synthesis followed
the general method for SPPS. Ϫ ESI MS; m/z (%): 1036.6 (5) [M
ϩ Na]ϩ, 508.1 (23) [M ϩ 2 H]2ϩ, 339.2 (100) [M ϩ 3 H]3ϩ. Ϫ
HPLC: 22.1 min, gradient 5Ϫ25% acetonitrile in 30 min. Ϫ 1H
NMR (500 MHz, D2O): δ ϭ 1.29Ϫ1.50 (m, 2 H, H-Lys),
1.55Ϫ1.88 (m, 4 H, H-Lys), 2.13Ϫ2.25 (m, 1 H, α-H), 2.25Ϫ2.45
(m, 4 H, α-H), 2.48Ϫ2.65 (m, 2 H, α-H), 2.68Ϫ2.80 (m, 1 H, α-H),
2.95 (t, J ϭ 7 Hz, 2 H, ε-H-Lys), 3.41 (m, 1 H, β-H), 3.52 (m, 1
H, β-H), 3.66 (m, 1 H, β-H), 3.77 (m, 1 H, β-H), 3.80Ϫ3.95 (m, 3
H, γ-H), 4.05Ϫ4.21 (m, 2 H, α-H-Lys, γ-H), 4.22Ϫ4.32 (m, 1 H, γ-
H), 4.32Ϫ4.49 (m, 2 H, H-γ), 4.50Ϫ4.60 (m, 1 H, γ-H), 4.77Ϫ4.95
(m, 1 H, γ-H), 6.60Ϫ6.75 (m, 3 H, 7-H), 6.80 (d, J ϭ 3 Hz, 1 H,
7-H), 7.11 (d, J ϭ 3 Hz, 1 H, 8-H), 7.15 (d, J ϭ 3 Hz, 1 H, 8-H),
7.22 (d, J ϭ 3 Hz, 1 H, 8-H), 7.56 (d, J ϭ 3 Hz, 1 H, 8-H), 8.17
(s, 1 H, 2-H), 8.19 (s, 1 H, 2-H), 8.20 (s, 1 H, 2-H), 8.21 (s, 1 H,
2-H).
H-[(S)-β-HalA-(S)-β-Hal7CA]3-(R)-Lys-NH2 (21): The syn-
thesis followed the general method for SPPS. Ϫ ESI MS; m/z (%):
1451.6 (34) [M ϩ H]ϩ, 726.7 (100) [M ϩ 2 H]2ϩ. Ϫ HPLC: 12.3
min, gradient 10Ϫ20% acetonitrile in 30 min. Ϫ 1H NMR (500
MHz, D2O): δ ϭ 1.33Ϫ1.50 (m, 2 H, H-Lys), 1.59Ϫ1.82 (m, 4 H,
H-Lys), 2.16Ϫ2.49 (m, 8 H, α-H), 2.50Ϫ2.60 (m, 2 H, α-H),
2.68Ϫ2.80 (m, 2 H, α-H), 2.95 (t, J ϭ 7 Hz, 2 H, ε-H-Lys), 3.58
(m, 1 H, β-H, γ-H), 3.62Ϫ3.80 (m, 3 H, β-H, γ-H), 3.81Ϫ4.10 (m,
6 H, β-H, γ-H), 4.17 (m, α-H-Lys), 4.19Ϫ4.32 (m, 2 H, β-H, γ-H),
4.33Ϫ4.70 (m, 7 H, β-H, γ-H), 6.60Ϫ6.71 (m, 3 H, 7-H 7CA), 7.09
(d, J ϭ 3 Hz, 1 H, 8-H 7CA), 7.19 (d, J ϭ 3 Hz, 1 H, 8-H 7CA),
7.31 (d, J ϭ 3 Hz, 1 H, 8-H, 7CA), 8.05 (s, 2 H, 2-H, 2-H 7CA, 8-
H), 8.11 (s, 2 H, 2-H, 2-H 7CA, 8-H), 8.14 (s, 1 H, 2-H, 2-H 7CA,
8-H), 8.16 (s, 1 H, 2-H, 2-H 7CA, 8-H), 8.21 (s, 3 H, 2-H, 2-H 7CA,
8-H).
(R)-N-tert-Butoxycarbonyl-γ-(7-carba-9-adeninyl)-β-homoala-
nine (ent-2): (R)-N-Boc-γ-(6-azido-7-carba-9-purinyl)-β-homoala-
nine benzyl ester (ent-12) (1.17 g, 2.59 mmol) was dissolved in a
mixture of dioxane (40 ml), water (25 ml), and acetic acid (2 ml)
and reduced by Pd on charcoal (225 mg, 5% Pd on charcoal con-
taining 50.5% water) within 2 d. The precipitated nucleo amino
acid was separated from the charcoal by dissolving in acetonitrile/
methanol (3:2) and subsequent filtration. The solvent was evapo-
rated and the product isolated from the brown residue by crystalli-
zation with water/acetonitrile as a white solid giving 661 mg (76%,
> 99% e.e.). The analytical data of ent-2 and 2 were identical except
20
for [α]589 ϭ Ϫ38.8 (c ϭ 0.9, DMSO).
General Method for SPPS of β-PNA: Oligomerization was per-
formed as a solid-phase peptide synthesis on a 4-methylbenzhydryl-
amine-(MBHA-)polystyrene resin (50 mg, 15.45 mmol) loaded with
(R)-lysine(Z)-OH (Z ϭ benzyloxycarbonyl; 0.309 mmol/g) in a
small column. For each coupling step an excess of four equivalents
of N-tert-butoxycarbonyl-γ-(9-purinyl)-β-homoalanine (77.25
mmol) was used and activated by O-(7-azabenzotriazol-1-yl)-
1,1,3,3-tetramethyluronium hexafluorophosphate (HATU; 26.40
mg, 69.53 mmol) and N,N-diisopropylethylamine (26.3 µl, 154.5
mmol) in DMF (600 µl). After swelling of the resin for 20 min the
following procedure was repeated for every nucleo amino acid unit:
(1) deprotection twice, for 3 min with trifluoracetic acid/m-cresol
(95:5; 2 ml); (2) washing five times each with CH2Cl2/DMF (1:1; 2
ml) and pyridine (2 ml); (3) coupling step, 40Ϫ50 min gentle mov-
ing of the reaction column; (4) washing three times each with
CH2Cl2/DMF (1:1; 2 ml), DMF/piperidine (95:5; 2 ml), and
CH2Cl2/DMF (1:1; 2 ml). Finally, the β-PNA was washed twice
with TFA (2 ml) and cleaved from the solid support within 1 h
using 1.6 ml trifluoracetic acid/trifluormethanesulfonic acid/m-cre-
sol (8:1:1). The dark brown solution was concentrated to 400 µl,
and the β-PNA precipitated with diethyl ether (5 ml) as a white
solid. The β-PNA was separated using a centrifuge, followed by
purification with HPLC (RP-C18). The yield of each coupling step
was estimated from HPLC to be higher than 97%.
H-[(S)-β-Hal7CA]6-(R)-Lys-NH2 (18): The synthesis followed
the general method for SPPS. Ϫ ESI MS; m/z (%): 1470.6 (12) [M
ϩ Na]ϩ, 725.1 (100) [M ϩ 2 H]2ϩ. Ϫ HPLC: 27.0 min, gradient
1
5Ϫ20% acetonitrile in 30 min. Ϫ H NMR (500 MHz, D2O): δ ϭ
1.32Ϫ1.49 (m, 2 H, H-Lys), 1.58Ϫ1.83 (m, 4 H, H-Lys), 2.12Ϫ2.25
(m, 2 H, α-H), 2.26Ϫ2.46 (m, 7 H, α-H), 2.47Ϫ2.59 (m, 2 H, α-H),
General Method for UV Melting Curves and CD Spectroscopy:
2.69Ϫ2.79 (m, 1 H, α-H), 2.95 (t, J ϭ 8 Hz, 2 H, ε-H-Lys), The oligomers (2Ϫ5 µ) were dissolved in an Na2HPO4/H3PO4
3.28Ϫ3.43 (m, 3 H, β-H, γ-H), 3.54Ϫ3.62 (m, 1 H, β-H, γ-H), 3.60 buffer (pH ϭ 7.0, 0.01 ) containing NaCl (0.1 ) and placed in a
(m, 3 H, β-H, γ-H), 3.78Ϫ3.87 (m, 2 H, β-H, γ-H), 3.88Ϫ3.96 (m, UV cell (10 mm). For UV melting curves the following temperature
1 H, β-H, γ-H), 4.10Ϫ4.19 (m, 1 H, γ-H), 4.19 (m, 1 H, α-H-Lys), program was used to control the heating block: 90°C Ǟ Ϫ2°C (90
4.23Ϫ4.35 (m, 1 H, γ-H), 4.36Ϫ4.55 (m, 4 H, γ-H), 4.55Ϫ4.62 (1 min) Ǟ Ϫ2°C (10 min) Ǟ 90°C (190 min) Ǟ Ϫ2°C (190 min) Ǟ
H, γ-H), 4.78Ϫ4.89 (m, 1 H, γ-H), 6.60Ϫ6.75 (m, 5 H, 7-H), 6.89 Ϫ2°C (10 min) Ǟ 90°C (190 min) Ǟ Ϫ2°C (190 min). The tem-
(d, J ϭ 4 Hz, 1 H, 7-H), 7.09 (m, 2 H, 8-H), 7.12 (d , J ϭ 3 Hz, 1 perature for UV spectra was measured directly in the solution and
834
Eur. J. Org. Chem. 1998, 827Ϫ835

β-Homoalanyl-PNA: A Special Case of β-Peptides with β-Sheet-Like Backbone Conformation
FULL PAPER
[8f]
N. Katagiri, C. Kaneko, Synth. Commun. 1993, 23, 1295. Ϫ
for CD spectra the temperature of the heating block was deter-
mined.
M. R. Harnden, A. Parkin, M. J. Parratt, R. M. Perkins, J.
[8g]
Med. Chem. 1993, 36, 1343. Ϫ
A. Bouali, D. F. Ewing, G.
Mackenzie, Nucleosides Nucleotides 1994, 13, 491.
G. Kokotos, Synthesis 1990, 299.
[9]
[10] [10a]
[10b]
[1]
[1a] U. Koert, Angew. Chem. 1997, 109, 1922; Angew. Chem. Int.
J. Davoll, J. Am. Chem. Soc. 1951, 73, 3174. Ϫ
J. Da-
[1b]
voll, J. Chem. Soc. 1960, 131.
Ed. Engl. 1997, 36, 1836. Ϫ
S. Borman, Chem. Eng. News
[11]
[1c]
M. Ho, J. K. K. Chung, N. Tang, Tetrahedron Lett. 1993, 34,
1997, June 16, 32. Ϫ
B. L. Iverson, Nature 1997, 385, 113.
[2] [2a]
6513.
D. Seebach, M. Overhand, F. N. M. Kühnle, B. Martinoni,
[12] [12a]
F. R. Benson, T. W. Hartzel, E. A. Otten, J. Am. Chem.
L. Oberer, U. Hommel, H. Widmer, Helv. Chim. Acta 1996, 79,
[12b]
[2b]
Soc. 1954, 76, 1858. Ϫ
Z. Kazimierczuk, H. B. Cottam, G.
913. Ϫ
D. Seebach, P. E. Ciceri, M. Overhand, B. Jaun, D.
R. Revankar, R. K. Robins, J. Am. Chem. Soc. 1994, 106, 6379.
The N-9 constitution of the isolated regioisomer was proved by
HMQC and HMBC NMR experiments [H. Kessler, S. Seip in
Two-Dimensional NMR Specroscopy, 2nd ed. (Eds.: W. R.
Croasmun, R. M. K. Carlson), VCH, Weinheim, 1994, p. 619].
C. Temple, C. L. Kussner, J. A. Montgomery, J. Org. Chem.
1966, 31, 2210.
(R)-Lysine amide was introduced for solubility reasons. The (R)
configuration was chosen by analogy to α-alanyl-PNA provid-
ing all side chains oriented on the same side of the backbone
in the extended conformation.
Rigo, L. Oberer, U. Hommel, R. Amstutz, H. Widmer, Helv.
[13]
Chim. Acta 1996, 79, 2043. Ϫ [2c] J. L. Matthews, M. Overhand,
F. N. M. Kühnle, P. E. Ciceri, D. Seebach, Liebigs Ann. 1997,
[2d]
1371. Ϫ
T. Hintermann, D. Seebach, Chimia 1997, 51, 244.
[2e]
Ϫ
G. Guichard, D. Seebach, Chimia 1997, 51, 315. Ϫ [2f] X.
[14]
[15]
Daura, W. F. van Gunsteren, D. Rigo, B. Jaun, D. Seebach,
[2g]
Chem. Eur. J. 1997, 3, 1410. Ϫ
T. Hintermann, D. Seebach,
Synlett 1997, 437. Ϫ [2h] D. Seebach, K. Gademann, J. V. Schrei-
ber, J. L. Matthews, T. Hintermann, B. Jaun, Helv. Chim. Acta
1997, 80, 2033. Ϫ [2i] D. Seebach, J. L. Matthews, J. Chem. Soc.,
Chem. Commun. 1997, 2015.
[16] [16a]
[3]
[4]
[3a] D. H. Appella, L. A. Christianson, I. L. Karle, D. R. Powell,
W. Saenger, Principles of Nucleic Acid Structure, Springer
Verlag, New York, 1983. Ϫ [16b] I. Berger, C. Kang, A. Fredian,
R. Ratliff, R. Moyzis, A. Rich, Nat. Struct. Biol. 1995, 2, 416.
[3b]
S. H. Gellman, J. Am. Chem. Soc. 1996, 118, 13071. Ϫ
D.
H. Appella, L. A. Christianson, D. A. Klein, D. R. Powell, X.
Huang, J. J. Barchi Jr., S. H. Gellman, Nature 1997, 387, 381.
S. Marqusee, R. L. Baldwin, Proc. Natl. Acad. Sci. USA 1987,
84, 8898.
Ϫ
[16c] K. J. Baejens, H. L. De Bondt, A. Pardi, S. R. Holbrook,
Proc. Natl. Acad. Sci. USA 1996, 93, 12851.
[17] [17a] J. Hunziker, H.-J. Roth, M. Böhringer, A. Giger, U. Dieder-
ichsen, M. Göbel, R. Krishnan, B. Jaun, C. Leumann, A.
Eschenmoser, Helv. Chim. Acta 1993, 76, 259. Ϫ [17b] K. Groeb-
ke, J. Hunziker, W. Fraser, L. Peng, U. Diederichsen, K. Zim-
[5] [5a]
U. Diederichsen, Angew. Chem. 1996, 108, 458; Angew.
[5b]
Chem. Int. Ed. Engl. 1996, 35, 445 Ϫ
U. Diederichsen, An-
gew. Chem. 1997, 109, 1966; Angew. Chem. Int. Ed. Engl. 1997,
36, 1886. Ϫ [5c] U. Diederichsen, Bioorg. Med. Chem. Lett. 1997,
7, 1743. Ϫ [5d] U. Diederichsen, Bioorg. Med. Chem. Lett. 1998,
8, 145. Ϫ [5e] U. Diederichsen, H. W. Schmitt, Tetrahedron Lett.
1996, 37, 475.
mermann, A. Holzner, C. Leumann, A. Eschenmoser, Helv.
[17c]
Chim. Acta 1998, 81, 375. Ϫ
S. Pitsch, S. Wendeborn, B.
Jaun, A. Eschenmoser, Helv. Chim. Acta 1993, 76, 2161.
U. Diederichsen, unpublished results.
[18]
[19]
[6]
[7]
The double-strand formation of 7-carbaadeninyl tetramer 20
was much slower than the others when cooled to 0°C. Therefore
the UV melting curve was measured with a different tempera-
ture program [80°C Ǟ Ϫ2°C (60 min) Ǟ Ϫ2°C (270 min) Ǟ
80°C (300 min) Ǟ Ϫ2°C (180 min) Ǟ Ϫ2°C (270 min) Ǟ 80°C
(300 min) Ǟ Ϫ2°C (180 min)].
U. Diederichsen, H. W. Schmitt, Angew. Chem. 1998, 110, 312;
Angew. Chem. Int. Ed. 1998, 37, 302.
[7a] E. Juaristi, D. Quintana, J. Escalante, Aldrichim. Acta 1994,
[7b]
[7c]
27, 3. Ϫ
D. C. Cole, Tetrahedron Lett. 1994, 50, 9517. Ϫ
[7d]
G. Cardillo, C. Tomasini, Chem. Soc. Rev. 1996, 117. Ϫ
J.
Podlech, D. Seebach, Angew. Chem. 1995, 107, 507; Angew.
[20] [20a]
[20b]
Chem. Int. Ed. Engl. 1995, 34, 471. Ϫ [7e] J. Podlech, D. Seebach,
V. Prelog, H. Gerlach, Helv. Chim. Acta 1964, 47, 2288. Ϫ
H. Gerlach, J. A. Owtschinnikow, V. Prelog, Helv. Chim.
[7f]
Liebigs Ann. 1995, 1217. Ϫ
N. N. Romanova, A. G. Gravis,
Y. G. Bundel, Russ. Chem. Rev. 1996, 65, 1083.
Acta 1964, 47, 2294. Ϫ [20c] M. M. Shemyakin, Y. A. Ovchinni-
[8] [8a]
C. G. Overberger, J. Y. Chang, Tetrahedron Lett. 1989, 30,
kov, V. T. Ivanov, Angew. Chem. 1969, 81, 523, Angew. Chem.
[8b]
[20d]
51. Ϫ
2787. Ϫ
Lett. 1991, 32, 7029. Ϫ
Benner, Nucleosides Nucleotides 1992, 11, 1257. Ϫ [8e] A. Toyota,
M. L. Peterson, R. Vince, J. Med. Chem. 1991, 34,
Int. Ed. Engl. 1969, 8, 492. Ϫ
Acc. Chem. Res. 1993, 26, 266.
M. Chorev, M. Goodman,
[97410]
[8c]
T. F. Jenny, N. Previsani, S. A. Benner, Tetrahedron
[8d]
T. F. Jenny, K. C. Schneider, S. A.
Eur. J. Org. Chem. 1998, 827Ϫ835
835