Synthesis and enzymic hydrolysis of oligoribonucleotides incorporating 3-deazaguanosine: The importance of the nitrogen-3 atom of single conserved guanosine residues on the catalytic activity of the hammerhead ribozyme

Article abstract of DOI:10.1002/hlca.200390222

Four base-modified hammerhead ribozyme/substrate complexes were constructed in which single guanosine (1) residues were replaced by 3-deazaguanosine (2) in the positions G₅, G₈, G _L2.1, and G₁₂. The base-modifie

Full text of DOI:10.1002/hlca.200390222

2726
Helvetica Chimica Acta Vol. 86 (2003)
Synthesis and Enzymic Hydrolysis of Oligoribonucleotides Incorporating
3-Deazaguanosine: The Importance of the Nitrogen-3 Atom of Single
Conserved Guanosine Residues on the Catalytic Activity of the Hammerhead
Ribozyme
by Frank Seela*a), Harald Debelaka), Lori Andrewsb), and Leonid Beigelmanb)
a) Laboratorium f¸r Organische und Bioorganische Chemie, Institut f¸r Chemie, Universit‰t Osnabr¸ck,
Barbarastr. 7, D-49069 Osnabr¸ck
b) Department of Chemistry and Biochemistry, Ribozyme Pharmaceuticals Inc., 2950 Wilderness Place,
Boulder, CO 80301, USA
Four base-modified hammerhead ribozyme/substrate complexes were constructed in which single
guanosine (1) residues were replaced by 3-deazaguanosine (2) in the positions G5, G8, GL2.1, and G12. The
base-modified ribozyme complexes were prepared by solid-phase synthesis of oligoribonucleotides employing
the novel phosphoramidite 3 derived from 2. Phosphoramidite 3 carried a phenoxyacetyl group at the amino
function and a diphenylcarbamoyl residue at the oxo group of the nucleobase. The 2'-hydroxy group was
blocked with a triisopropylsilyl residue. Kinetic analysis of the phosphodiester hydrolysis showed a moderate
decrease of the ribozyme catalytic activity when the residues G5 or G8 were replaced by 3-deazaguanosine and a
200-fold decrease when G12 was substituted. A 6-fold catalytic increase occurred when 3-deazaguanosine was
replacing GL2.1 in the loop region. The data indicate that the N(3) atom of compound 2, in particular at position
G12 is critical for the ribozyme activity.
Introduction. The hammerhead ribozyme is one of the smallest catalytic RNAs. It
has a small catalytic core consisting of ten conserved nucleotides together with
interchangeable recognition arms to achieve the sequence-specific cleavage of any
target RNA [1 3]. As a result, the transesterification reaction generates two products,
a 2',3'-cyclic phosphate and a 5'-terminal hydroxy group at the cleavage site. The
cleavage site was originally defined by a NUH pattern, where N can be any nucleoside
and H the cleavage site of the nucleotide can be A, U, or C with the most-efficient
cleavage occurring at the GUC triplet. This pattern was defined later as the NHH rule,
since other triplets such as GAC and GCC can be also cleaved by the hammerhead
ribozyme [4]. The hammerhead catalytic activity requires divalent metal ions such as
Mg2‡ or high concentrations of monovalent cations [5]. The precise molecular
mechanism of action is not yet defined despite the fact that several X-ray structures of
the hammerhead ribozyme with non-cleavable substrates [6 8] and crystallographic
studies with trapped intermediates resembling the transition state were performed [9].
The general structure is consistent with the solution studies performed by FRET
analysis [10 12]. However, significant Mg2‡-dependent conformational mobility of the
ribozyme has been detected by time-resolved fluorescence measurements and NMR
[13][14]. The structural modification with modified nucleotide analogs identified a
number of important functional atoms or groups being involved in the hammerhead
catalysis (reviewed in [15][16]). To probe the importance of the N(3) atom of the

Helvetica Chimica Acta Vol. 86 (2003)
2727
conserved guanosine (1) residues of the core region as well as those of the GA platform
in the GNRA tetraloop, nucleoside 1 was replaced by 3-deazaguanosine1) (2) thus
eliminating specific proton-acceptor sites that may be involved in critical interactions
mediating the folding and the catalysis by the hammerhead ribozyme.
To accomplish this replacement within the hammerhead-ribozyme complex, the 3-
deazaguanosine phosphoramidite 3 was synthesized and employed in solid-phase
synthesis for the single nucleoside modification of particular guanosine residues. The
catalytic activity of the native and the mutated hammerhead-ribozyme complexes was
compared and demonstrated the importance of the N(3) position of the G12 residue for
the hammerhead catalysis. Surprisingly, the replacement of guanosine (1) by 3-
deazaguanosine (2) within the loop sequence (GL2.1) resulted in a significant increase in
the catalytic activity of the hammerhead ribozyme.
Results and Discussion. 1. Monomers. Oligo-2'-deoxyribonucleotides containing
3-deazaguanine have been already synthesized with H-phosphonate of 3-deaza-2'-
deoxyguanosine [17]. Alternatively, phosphoramidite chemistry was employed with
appropriately functionalized imidazole nucleosides as precursors, which were con-
verted later to the 3-deazaguanine with ammonia containing oligonucleotide on the
polymeric level [18]. Oligoribonucleotides incorporating 3-deazaguanosine (2) are
unknown. For this synthesis, the phosphoramidite building block 3 of the nucleoside 2
was prepared that allows site-specific incorporation of a 3-deazaguanosine moiety into
any position of an oligoribonucleotide by standard solid-phase chemistry.
The 3-deazaguanosine (2) has already been prepared in 1975 by the Robins group
[19][20]. This route used ammonia for the six-membered-ring formation. In an
alternative route, hydrazine was employed [21]. Minakawa and Matsuda reported a
synthesis that started from AICA-riboside via 5-ethynyl-1-(b-d-ribofuranosyl)-1H-im
idazole-4-carboxamide or -4-carbonitrile [22]. We followed the synthetic route of
Robins and co-workers. Vorbr¸ggen glycosylation of the trimethylsilyl derivative of
methyl 5(4)-(cyanomethyl)-1H-imidazole-4(5)-carboxylate (4) with 1-O-acetyl-2,3,5-
tri-O-benzoyl-b-d-ribofuranose (5) in MeCN in the presence of tin(IV) chloride
afforded the intermediate 6 (Scheme 1). Treatment of 6 with saturated methanolic
ammonia furnished nucleoside 2 [20].
1
)
Purine numbering is used throughout the Results and Discussion section; for systematic numbering, see
Exper. Part.

2728
Helvetica Chimica Acta Vol. 86 (2003)
Scheme 1
i) HMDS (hexamethyldisilazane), (NH4)2SO4, 4 h, reflux, SnCl4, MeCN, 3 h, r.t. ii) NH3/MeOH, 48 h, 1208,
autoclave. DMTrˆ (MeO)2 Tr ˆ (4-MeOC6H4)PhC.
Compound 2 is more p-electron-rich than the parent guanosine. This is apparent
from the pKa values of protonation and deprotonation of 2, which are 2.7 (N(7)) and
12.3 (N(1)), respectively [23]. The parent guanosine (1) shows pKa values of 1.9 and 9.2
[23]. The lipophilic character of 2 is decreased over that of 1 as it is found for 7-
deazaguanosine and demonstrated by the chromatographic mobility on a reversed-
phase (RP-18) column. The corresponding 2'-deoxyribonucleosides show similar
behavior (Fig. 1,a). Differences of 1 and 2 are observed in their CDspectra ( Fig. 1,b).
While the positive lobes observed between 215 and 220 nm are similar for 1 and its 3-
deaza analogue 2, only the latter shows a significant negative trough at 270 nm.
Fig. 1. a) HPLC Profiles of c3G (2) and c3Gd in comparison to their unmodified counterparts G (1) and dG (RP-
18 column, 0.5 ml minÀ1, 0.1m (Et3NH)OAc buffer with 5% MeCN). b) CD Spectra of c3G (2) and G (1) (in bi-
distilled H2O with 1 mm nucleoside concentration).
Earlier, it was observed for 3-deaza-2'-deoxyguanosine that regular acyl protecting
groups used for the 2-amino function are too stable to be used for oligonucleotide
synthesis (Table 1). Only the phenoxyacetyl(pac)-protected nucleoside or the (di-
methylamino)methylidene (m2fa) derivative can be considered for further reactions

Helvetica Chimica Acta Vol. 86 (2003)
2729
[24][25]. However, when this latter group was used [17], problems appeared during the
preparation of the phosphoramidite, which resulted from the reactivity of the 6-oxo-
group [26]. Thus, the 6-oxo-group was protected. For this purpose, the diphenylcarba-
moyl (dpc; Ph2NC(ˆO)) residue was chosen, which was already used successfully as an
oxo-protecting group in the case of guanosine [27][28] and 2'-deoxyisoguanosine [29].
Prior to the 6-oxo group protection, the phenoxyacetyl (pac) group [30] was introduced
to block the amino function (2 ! 7; see Scheme 2). Next, it was established whether
introduction of the dpc group (7 ! 8) or that of the 4,4'-dimethoxytrityl ((MeO)2Tr )
residue (7 ! 9) would be the most-effective step to take next. As the latter route gave
the better yields, the (MeO)2Tr derivative 9 was prepared and derivatized further.
Scheme 2
i) 1. HMDS, DMF; 2. pac-Cl, pyridine, r.t., 3 h. ii) dpc-Cl, (i-Pr)2EtN, pyridine, r.t., 6 h. iii) (MeO)2Tr-Cl,
pyridine, r.t., 4 h. iv) (i-Pr)3SiCl, AgNO3, pyridine/THF, r.t., 20 h. DMTr ˆ (MeO)2 Tr.
Table 1. Half-Life Values (t) of Amino-Protected 3-Deazaguanine Derivatives, Measured Spectrophotometri-
cally in 25% Aq. Ammonia Solutiona)
t [min]
l [nm]
pac2c3G (7)
pac2Gd [30]
ibu2c3Gd [17]
ibu2Gd [24]
m2fa2c3Gd [17]
m2fa2Gd [25]
190b), 4c)
15d)
322
500b)
112b)
28b)
300
300
320
300
19b)
a) pac ˆ phenoxyacetyl, c ˆ deaza, d ˆ 2'-deoxy, ibu ˆ isobutyryl ˆ Me2CHCO, m2fa ˆ (dimethylamino)meth-
ylene. b) Measured at 408. c) Measured at 658 (40% MeNH2/H2O). d) Measured at 208.
To assure the applicability of the protecting groups during oligoribonucleotide
synthesis, the deprotection of compounds 7 and 8 was studied spectrophotometrically in
25% ammonia or 40% aqueous methylamine solution. The overall half-life value for

2730
Helvetica Chimica Acta Vol. 86 (2003)
deprotection of 8 (for both protecting groups) was 4.9 min in 40% MeNH2/H2O at 658.
To follow the deprotection stepwise, the reaction was performed at 208 and monitored
at 322 nm. According to the pH-dependent UV spectra of 3-deazaguanosine (2)
(Fig. 2,a) and of those of the derivatives 7 and 8 (Fig. 2,b), the independent removal of
the dpc residue can be followed at this wavelength. Thus, for compound 8, the half-life
of the dpc removal was found to be only 4 min, while that of the pac group was much
longer. The rather long half-life time observed for the pac residue of compound 7 was
not unexpected as anion formation stabilizes the pac residue. Also the amino group of
3-deazaguanosine is more basic than that of guanosine. A similar observation has been
made for 3-deaza-2'-deoxyadenosine [31].
Fig. 2. a) UV Spectra of 3-deazaguanosine (2) at pH 1.0, 7.0, and 13.0. b) UV Spectra (MeOH) of the protected
nucleosides 7 and 8 compared with that of compound 2.
Due to the reactivity of the 6-oxo group of 9, it was not possible to protect the sugar
2'-hydroxy group with the triisopropylsilyl residue without affecting the oxo group of
the nucleobase. For instance, silylation of 9 afforded the 6-oxo-protected derivative 10
in moderate yield (Scheme 2). Consequently, the 6-oxo function had to be protected
before manipulations of the sugar residue took place. Reaction of compound 9 with
diphenylcarbamoyl chloride (dpc-Cl) gave the dpc derivative 11 (61% yield) together
with the bis-dpc compound 12 (13%) (Scheme 3). Afterwards, silylation of compound
11 with (i-Pr)3SiCl was performed in the presence of AgNO3 as it was described for
other nucleosides [32][33]. The 2'-O-silyl derivative 13 was isolated in 51% yield, and
the 3'-O-isomer 14 in 20%. Phosphitylation of compound 13 under standard conditions
[34] resulted in formation of phosphoramidite 3 (75%). All new compounds were
characterized by 1H-, 13C-, and 31P-NMR spectra (see Table 2 and Exper. Part) as well as
by elemental analyses. From the NMR spectra of the dpc derivatives, it is apparent that
this residue is linked to the 6-oxo group and not to the N(1) atom. Due to
aromatization, all 13C-NMR signals of the dpc derivative 8 are shifted significantly in
comparison to the mono-protected compound 7. A similar observation was made for
other O-dpc-protected guanine nucleosides [27 29].
As the conformational equilibrium of the sugar moiety is controlled by the
electronic properties of the base, changes are expected when a 3-deazaguanine replaces
guanine within a nucleoside. The sugar conformation (N vs. S and around the
C(4')ÀC(5') bond) of the nucleosides 1 and 2 as well as of dG, and c3Gd for comparison

Helvetica Chimica Acta Vol. 86 (2003)
2731
Scheme 3
i) dpc-Cl, (i-Pr)2EtN, pyridine, r.t., 4 h. ii) (i-Pr)3SiCl, AgNO3, pyridine/THF, r.t., 20 h.
Table 2. 13C-NMR Chemical Shifts of the 3-Deazaguanosine Derivatives 2, 7 9, and 11 14a)
b
c
)
)
C(2)
C(6)
C(3)
C(7)
C(4)d)
C(7a)
C(5)d)
C(3a)
C(6)
C(4)
C(8)
C(2)
C(1')
C(2')
C(3')
C(4')
C(5')
2
7
8
147.6
139.6
143.4
139.9
143.6
143.8
143.7
143.7
70.5
81.8
94.6
81.6
94.7
94.5
94.3
94.8
123.0
128.1
128.7
128.2
128.7
128.8
128.6
128.7
142.4
136.7
142.8
136.8
142.9
143.5
143.4
142.9
156.5
155.9
146.8
155.9
146.8
146.8
146.9
146.8
136.6
139.1
144.1
138.3
144.8
144.7
144.8
144.7
88.1
88.2
88.6
88.6
89.0
86.7
90.0
88.9
73.8
74.2
73.9
73.7
73.5
76.8
75.5
73.1
70.1
70.3
70.3
70.3
70.1
68.6
70.5
72.2
85.3
85.7
85.9
81.6
83.5
83.4
84.0
84.4
61.2
61.2
61.2
63.5
63.6
63.0
63.4
63.1
9
11
12
13
14
a) Spectra were measured in (D6)DMSO. b) Purine numbering. c) Systematic numbering. d) Tentative.
3
1
was determined from the vicinal J(H,H) coupling constants of the H-NMR spectra
measured in D2O (Table 3) by applying the PSEUROT program package (version 6.2)
[35][36] or according to Westhof et al. [37]. The 3-deazaguanosine shows the same
preference of the anti-conformation as guanosine (68%) does [38]. Table 3 indicates
that no severe changes are occurring between the modified and the parent nucleosides.
2. Hammerhead-Ribozymes Substrate Complexes Containing Single 3-Deazagua-
nosine Substitutions. 2.1. Synthesis. To probe the importance of the N(3) atom of the
conserved guanosine (1) residues within the catalytic core of the hammerhead-
ribozyme, individual guanosine residues were replaced by 3-deazaguanosine (2) in the
53 nucleotide unit of the ribozyme substrate complex 15 ¥ 16 (see Table 4). In addition,

2732
Helvetica Chimica Acta Vol. 86 (2003)
Table 3. 1H-NMR Coupling Constants of the Sugar Protons, the Calculated Pseudorotational Parameters, and the
a
Rotational Equilibrium about the C(4')ÀC(5') Bond of G (1), dG, c3G (2), and c3Gd
)
3J(1',2'a) 3J(1',2'b) 3J(2'a,3') 3J(2'b,3') 3J(3',4') 3J(4',5'a) 3J(4',5'b) %N %S %g(‡)g %gt %g(À)g
G (1) 5.80
dG 7.30
c3G (2) 5.65
c3Gd
6.80
5.00
6.30
5.00
6.75
4.00
3.20
4.40
3.70
3.00
3.60
2.70
3.85
4.20
4.70
4.40
5.15
36 64 65
29 71 53
39 61 66
35 65 45
24 11
30 17
6.50
6.40
3.60
3.90
26
8
35 20
a) Measured in D2O at 308; r.m.s. ꢀ 0.4 for all calculations; DJmax ꢀ 0.5 Hz.
a non-conserved residue in the GAAA loop was substituted as well (Fig. 3). Each of
these oligonucleotides contains a single substitution at positions G5, G8, G12, or GL2.1
.
The oligonucleotide synthesis was performed on a solid support employing the
phosphoramidite 3. The coupling conditions were according to a protocol applied for
the phosphoramidite of 1-deazaadenosine (0.1m phosphoramidite, 0.1m 5-(ethylthio)-
1H-tetrazole, coupling time 450 s). The coupling efficiency of 3 was 90 93%, which is
slightly below that of the unmodified building blocks but still in an acceptable range for
an efficient assembly of the target oligonucleotides 17 20 (see Table 4; cf. also Exper.
Part and Fig. 3). The choice of the protecting groups for compound 3 proved to be
adequate, since complete deblocking was achieved with 40% MeNH2/H2O at 658
(15 min). The MALDI-TOF-MS analysis was performed with the deprotected
‡
oligonucleotides. The M data were in accordance with the calculated values.
Fig. 3. Structure of the hammerhead RNA complex 15/16 showing the sites of the 3-deazaguanosine
incorporation and the relative cleavage rates

Helvetica Chimica Acta Vol. 86 (2003)
2733
Table 4. Oligoribonucleotide Sequences of the Enzymes 16 20 and the Substrate 15
5'-[32P]p(C-A-G-G-G-A-U-U-A-A-U-G-G-A-G-A-U)-3' (15)
5'-(U-C-U-C-C-A-U-C-U-G-A-U-G-A-G-G-C-C-G-A-A-A-G-G-C-C-G-A-A-A-A-U-C-C-C-U)-3' (16)
5'-(U-C-U-C-C-A-U-C-U-G-A-U-2-A-G-G-C-C-G-A-A-A-G-G-C-C-G-A-A-A-A-U-C-C-C-U)-3' (17)
5'-(U-C-U-C-C-A-U-C-U-G-A-U-G-A-G-G-C-C-G-A-A-A-G-G-C-C-2-A-A-A-A-U-C-C-C-U)-3' (18)
5'-(U-C-U-C-C-A-U-C-U-2-A-U-G-A-G-G-C-C-G-A-A-A-G-G-C-C-G-A-A-A-A-U-C-C-C-U)-3' (19)
5'-(U-C-U-C-C-A-U-C-U-G-A-U-G-A-G-G-C-C-2-A-A-A-G-G-C-C-G-A-A-A-A-U-C-C-C-U)-3' (20)
2.2. Ribozyme Activity. To probe the importance of the N(3) atom of the conserved
guanosine residues on the catalytic activity of the hammerhead-ribozyme complex, the
cleavage reaction of the native 53 oligonucleotide was compared with that of the
modified complexes containing single guanosine replacements by compound 2
(Table 4). The 3-deazaguanosine residues can participate in Watson-Crick or Hoogs-
teen base pairing, with the exception that the H-bonding involving N(3) is eliminated
(Fig. 4). In cases where N(3) is participating in specific interaction(s) being necessary
for an efficient cleavage, a significant reduction in the catalytic efficiency can be
expected. Fig. 4 shows the potential donor and acceptor positions for the modified
nucleoside as well as the electron densities of 3-deazaguanine (2) as well as of guanine
(1). The electron densities were calculated on the HF/6-31G* level using HyperChem 5
(Hypercube Inc., FL, USA). The data indicate that 1 and 2 differ in their electron
densities at the atoms located at the positions 2, 3, and 4 [39]. According to this, the H-
bonding via N(3) is eliminated, and the 2-amino group becomes more basic (i.e., a less-
efficient proton donor).
Fig. 4. a) Proton acceptor/donor pattern of 3-deazaguanosine (2) and b) electron densities (in electron units) of
3-deazaguanine and guanine determined by ab initio calculations at the 6-31G* level
2.3. Guanine ! 3-Deazaguanine Substitutions in Domain II. The results of several
X-ray analyses demonstrated that the hammerhead ribozyme is composed of two
structural domains: a −uridine turn× (domain I; C3 to A6) and a tandem G ¥ A mismatch
(domain II; U7 to A9 and G12 to A14) located at the junction of three helices [6] (cf.
Fig. 3). The domain II is strictly conserved, and its Mg2‡ complex pulls stems II and III
into the alignment required for catalysis. Two symmetric metal-binding sites associated
with domain II were identified by biochemical methods as well as by NMR and X-ray
analysis [11][40 46]. These include the less-defined A13 site [42] and the site A9 to

2734
Helvetica Chimica Acta Vol. 86 (2003)
G10.1, where functional groups involved in coordination of metal were characterized in
detail [44 46]. According to the X-ray data, N(3) of the guanosine residue G12 is
involved in H-bonding with the 6-amino group of the adenosine residue A9 (see Fig. 5)
The same type of interaction is observed for the residues G8 and A13, although in this
case, OÀC(4') of the furanose moiety of G8 is participating in an additional H-bond
with the 6-amino group of A13, while in the G12 ¥ A9 pair, the 2'-hydroxy group of G12
serves as another H-bond donor.
Fig. 5. Hydrogen-bonding pattern of domain II of the hammerhead-ribozyme complex, with donor and acceptor
sites of guanine and adenine and the sugar moieties. The pattern follows X-ray data described recently [44][45].
Interestingly, the G ! c3G substitution of G8 and G12 resulted in different outcomes.
In the case of the G8 substitution, the decrease of the cleavage of the ribozyme
substrate complex is only 10-fold compared to 200-fold for the G12 substitution (Fig. 3).
This most likely indicates that N(3) of the guanosine residue G12 is involved in H-
bonding being critical for the hammerhead catalysis, while N(3) of G8 is less critical.
The odd base pair G8 ¥ A13 observed in the ground state of the ribozyme substrate
complex identified by X-ray analysis is probably rearranged during the transition
towards the productive complex in which N(3)(G8) ¥¥¥ H2N(A13) H-bonds are not
critical for the activity or are compensated by new interactions. It is also notable that
the G8 adjacent to U7 is the beginning of the flexible uridine turn. This may account for

Helvetica Chimica Acta Vol. 86 (2003)
2735
the tolerance in geometry of the conserved G8 ¥ A13 pair. A significantly larger effect of
3-deazaguanine substitution is found for the G12 ¥ A9 pair. That might be related to a
precise positioning of adjacent metal-binding sites at A9 to G10.1. This is implicated by a
direct participation in the cleavage reaction [46].
Our results on differential sensitivity of G12 ¥ A9 and G8 ¥ A13 base pairs toward 3-
deazaguanine substitutions are in agreement with those of McLaughlin and co-workers
[47] as well as with those of Seela and co-workers [48] regarding the pair G8 ¥ A13. In
contrast to the results of McLaughlin and co-workers, which obtained no change when
A9 was replaced by 7-deazaadenosine, we observed 200-fold-reduced activity when
guanine was replaced by 3-deazaaguanine in the A9 ¥ G12 interaction. The relative
insignificance of the potential H-bonds in the G8 ¥ A13 pair correlates well with the
finding for adenosine ! 7-deazaadenosine substitution [47][48] as well as with those
for the replacement of guanosine by 3-deazaguanosine described in the present work.
The results obtained for the substitution of G12 by 3-deazaguanosine and of A9 by 7-
deazaadenosine in the A9 ¥ G12 pair probably indicate that the H-bond of N(3) with the
6-amino group which is found for the ground state by X-ray analysis (Fig. 5) is absent
during transition to the active complex. The N(3) of the G12 residue might find another
binding partner during this transition.
2.4. Guanine ! 3-Deazaguanine Substitution of G5. The replacement of the G5
residue in the domain I and its role in hammerhead catalysis is a subject of continuing
controversy. Any substitution of the base moiety of this nucleobase residue by
hypoxanthine, 2-aminopurine, purine, isoguanine, xanthine [49], and 7-deazaguanine
[47][48] resulted in significant (>1000 fold) loss of the cleavage activity in the
ribozyme substrate complex. According to the first two X-ray structures [44][45], this
residue interacts only with solvent molecules. Our data, observed for the substitution of
G5 with 3-deazaguanosine, results in only a modest (ninefold) decrease of the cleavage
rate (cf. Fig. 3), placing 3-deazaguanosine into the rare category of non-invasive
modification as it was observed for 7-deazaguanosine. It also confirms that the Watson-
Crick pairing site of G5 rather than the Hoogsteen site of this guanosine residue is
involved in interactions crucial for the activity of the hammerhead catalysis. The
recently identified metal-binding site near G5 [50], with potential outer-sphere
coordination to the Watson-Crick site of G5, supports this finding.
2.5. Guanine ! 3-Deazaguanine Substitution of GL2.1. An unexpected result was
observed when 3-deazaguanosine was replacing GL2.1 in the GAAA tetra loop (cf.
Fig. 3). A sixfold increase in cleavage rate was observed. To the best of our knowledge,
this is only the second example of enhancement of the hammerhead ribozyme activity
produced by a modification [51]. We have demonstrated recently that the GNRA tetra
loop is not optimal for the hammerhead catalysis. The GUUA loop confers increased
cleavage rates for this type of ribozyme [52]. At the same time, 1-deazaadenosine at
position L2.3 of the GAAA part has no effect on catalysis [53]. It is possible that GL2.1 is
indirectly involved in a conformational change from the ground state to the productive
ribozyme substrate complex. Additional modifications at this position are needed to
confirm this idea.
Conclusions.
An efficient synthesis for ribozyme substrate constructs is
described based on the novel phosphoramidite 3 of 3-deazaguanosine (2), protected

2736
Helvetica Chimica Acta Vol. 86 (2003)
at the 2-amino group of the base with a phenoxyacetyl residue and at the 6-oxo function
with a diphenylcarbamoyl group. Single replacements of guanosine residues G8, G12,
G5, and GL2.1 by 3-deazaguanosine (2) and analysis of the kinetic data established the
following: i) a differential sensitivity of the G12 ¥ A9 and G8 ¥ A13 base pairs towards 3-
deazaguanosine substitution and the importance of the base pairing mediated by the
N(3) atom of G12 for hammerhead catalysis, ii) the confirmation that the Watson-Crick
pairing site of G5 is involved in interactions important for catalysis, whereas N(3) is not,
and iii) a surprising increase of the ribozyme substrate-complex cleavage rate when
GL2.1 was substituted by 3-deazaguanosine. The significant decrease of the catalytic
activity of the ribozyme substrate complex, in particular the almost complete loss
when the G12 residue was replaced, is in accordance with the X-ray structure.
We thank Mr. Yang He and Dr. H. Rosemeyer for the NMR spectra, and Mrs. E. Feiling for excellent
technical assistance. Financial support by the Roche Diagnostics GmbH, Penzberg, Germany, is gratefully
acknowledged.
Experimental Part
General. Chemicals were purchased from Aldrich, Sigma, or Fluka (Sigma-Aldrich Chemie GmbH,
Deisenhofen, Germany). The solvents were of laboratory grade. TLC: silica gel 60 F254 aluminium sheets (0.2-
mm layer; Merck, Darmstadt, Germany). Column flash chromatography (FC): silica gel 60 (Merck, Darmstadt,
Germany); at 0.4 bar (4 ¥ 104 Pa). M.p.: B¸chi-SMP-20 apparatus (B¸chi, Flawil, Switzerland); uncorrected. UV
Spectra: U-3200 spectrometer (Hitachi, Tokyo, Japan); lmax (e) in nm. NMR Spectra: Avance-DPX-250 or
AMX-500 spectrometer (Bruker, Karlsruhe, Germany); 250.13 (1H) and 500.14 MHz (1H; 125.13 MHz for 13C),
resp.; chemical shifts d in ppm rel. to SiMe4 as internal standard; J values in Hz. The elemental analyses were
performed by Mikroanalytisches Laboratorium Beller, Gˆttingen, Germany.
Protecting-Group Stability. The stability of the protecting groups was determined spectrophotometrically
with a U-3200 spectrometer (Hitachi, Japan), heated with an RC-6-CP thermostat (Lauda, Germany) at a
defined temperature. UV-Overlay spectra of compounds 7 and 8 were measured, followed by time scans at that
wavelength where the highest absorbance change was observed.
1,5-Dihydro-6-[(phenoxyacetyl)amino]-1-(b-d-ribofuranosyl)-4H-imidazo[4,5-c]pyridin-4-one (7). To a
suspension of 3-deazaguanosine (2; 1.5 g, 5.3 mmol) in dry DMF (50 ml), 1,1,1,3,3,3-hexamethyldisilazane
(9 ml, 42.7 mmol) was added, and the mixture was stirred for 30 min at r.t. The resulting clear soln. was
evaporated to give an oil, which was dissolved in anh. pyridine (50 ml). Phenoxyacetyl chloride (0.88 ml,
6.4 mmol) was added under stirring. After 3 h, the soln. was cooled (ice bath), diluted with H2O (6 ml), and kept
for 15 min. Then, the mixture was treated with 25% aq. NH3 soln. (6 ml), stirred for another 15 min, and
concentrated to 10 ml. H2O (150 ml) was added, the aq. soln. washed with CH2Cl2 (50 ml, twice), the org. phase
evaporated, and the residue co-evaporated with toluene (50 ml) and MeOH (50 ml) and then applied to FC
(silica gel, column 10 Â 3 cm, CH2Cl2/MeOH 4 :1). The content of the main zone was crystallized from EtOH: 7
(1.6 g, 72%). Colorless crystals. M.p. 190 1928. Rf (CH2Cl2/MeOH 8 :2) 0.2. UV (MeOH): 268 (13200), 274
(13100), 302 (13200). 1H-NMR ((D6)DMSO): 3.62 (m, 2 HÀC(5')); 3.97 (m, HÀC(4')); 4.09 (m, HÀC(3'));
4.32 (m, HÀC(2')); 4.75 (s, CH2); 5.10 (br. s, OHÀC(5')); 5.33 (br. s, OHÀC(3')); 5.58 (br. s, OHÀC(2')); 5.67
(d, 3J ˆ 6.3, HÀC(1')); 6.98 7.07 (m, HÀC(7), 3 arom. H); 7.35 (t, 3J ˆ 7.7, 2 arom. H); 8.23 (s, HÀC(2)); 10.27
(br. s, NH), 11.33 (br. s, NH). Anal. calc. for C19H20N4O7 (416.38): C 54.81, H 4.84, N 13.46; found: C 54.89, H
4.91, N 13.39.
4-[(Diphenylcarbamoyl)oxy]-N6-(phenoxyacetyl)-1-(b-d-ribofuranosyl)-1H-imidazo[4,5-c]pyridin-6-
amine (8). Compound 7 (200 mg, 0.48 mmol) was dried by co-evaporation with anh. pyridine (5 ml) and
suspended in dry pyridine (5 ml). To this suspension, diphenylcarbamic chloride (250 mg, 1.1 mmol) and (i-
Pr)2EtN (180 ml, 1.3 mmol) were added, and the mixture was stirred for 6 h at r.t. This soln. was poured into 5%
aq. NaHCO3 soln. (20 ml), the org. phase extracted with CH2Cl2 (20 ml, twice), the combined org. phase dried
(Na2SO4) and evaporated, and the residue adsorbed on silica gel and applied to FC (silica gel, column 10 Â 2 cm,

Helvetica Chimica Acta Vol. 86 (2003)
2737
CH2Cl2/MeOH 9 :1): 8 (115 mg, 39%). Pale yellow powder. Rf (CH2Cl2/MeOH 9 :1) 0.5. UV (MeOH): 228
1
(43100), 269 (17700), 275 (sh, 16700). H-NMR ((D6)DMSO): 3.63 (m, 2 HÀC(5')); 3.99 (m, HÀC(4')); 4.10
(m, HÀC(3')); 4.39 (m, HÀC(2')); 4.81 (s, CH2); 5.09 (t, 3J ˆ 5.2, OHÀC(5')); 5.32 (d, 3J ˆ 4.7, OHÀC(3')); 5.61
(d, 3J ˆ 6.6, OHÀC(2')); 5.84 (d, 3J ˆ 6.3, HÀC(1')); 6.93 7.47 (m, 15 arom. H); 8.28 (s, HÀC(7)); 8.61
(s, HÀC(2)); 10.73 (br. s, NH). Anal. calc. for C32H29N5O8 (611.60): C 62.84, H 4.78, N 11.45; found: C 62.93, H
4.78, N 11.24.
1-{5'-O-[Bis(4-methoxyphenyl)phenylmethyl]-b-d-ribofuranosyl}-1,5-dihydro-6-[(phenoxyacetyl)amino]-
4H-imidazo[4,5-c]pyridin-4-one (9). Compound 7 (840 mg, 2.0 mmol) was dried by co-evaporation with anh.
pyridine (20 ml) and dissolved in dry pyridine (20 ml) under gentle heating. To this soln., (MeO)2Tr-Cl (815 mg,
2.4 mmol) was added under stirring, and the reaction was continued for 4 h at r.t. Then, the mixture was
quenched by the addition of MeOH (5 ml), stirred for further 30 min, and then concentrated to half of its
volume. The soln. was diluted with CH2Cl2 (100 ml), washed with 5% aq. NaHCO3 soln. (50 ml, twice) and brine
(50 ml), dried (Na2SO4) and evaporated and the residue co-evaporated with toluene (2 Â 30 ml) and purified by
FC (silica gel, column 4 Â 10 cm, CH2Cl2/MeOH 9 :1): 9 (1.04 g, 72%). Colorless solid. Rf (CH2Cl2/MeOH 9 :1)
0.4. UV (MeOH): 270 (sh, 12100), 275 (12800). 1H-NMR ((D6)DMSO): 3.19 (m, 2 HÀC(5')); 3.71 (s, 2 MeO);
4.12 (m, HÀC(3'), HÀC(4')); 4.78 (m, HÀC(2')); 4.73 (s, CH2); 5.33 (d, 3J ˆ 5.0, OHÀC(3')); 5.73 (d, 3J ˆ 4.1,
OHÀC(2')); 5.75 (d, 3J ˆ 3.8, HÀC(1')); 6.81 7.38 (m, 19 arom. H, HÀC(7)); 8.13 (s, HÀC(2)); 10.17 (br.
s, NH); 11.32 (br. s, NH). Anal. calc. for C40H38N4O9 (718.75): C 66.84, H 5.33, N 7.80; found: C 66.89, H 5.61, N
7.76.
1-{5'-O-[Bis(4-methoxyphenyl)phenylmethyl]-b-d-ribofuranosyl}-N6-(phenoxyacetyl)-4-{[tris(1-methyl-
ethyl)silyl]oxy}-1H-imidazo[4,5-c]pyridin-6-amine (10). To a soln. of 9 (210 mg, 0.29 mmol) in anh. pyridine
(3 ml), AgNO3 (74 mg, 0.44 mmol) was added under stirring at r.t. After 10 min, a soln. of (i-Pr)3SiCl (66 ml,
0.31 mmol) in dry THF (4 ml) was added with stirring under exclusion of light and moisture. The reaction was
continued for 6 h, and another portion of (i-Pr)3SiCl (33 ml, 0.15 mmol) was added. Stirring was continued for
another 16 h. The AgCl was filtered off, the filtrate diluted with 5% aq. NaHCO3 soln. (5 ml), the aq. phase
extracted with CH2Cl2 (10 ml, twice), the combined org. phase washed with H2O (10 ml) and brine (20 ml),
dried (Na2SO4), and evaporated, and the residue applied to FC (silica gel, column 4 Â 10 cm, CH2Cl2/acetone
7:3): 10 (61 mg, 23%). Colorless foam. Rf (CH2Cl2/MeOH 9 :1) 0.4. UV (MeOH): 233 (sh, 37000), 269 (25300),
275 (25900), 302 (21600). 1H-NMR ((D6)DMSO): 1.09 (d, 3J ˆ 7.5, 3 Me2CH), 1.49 (d, 3J ˆ 7.5, 3 Me2CH), 3.20
(m, 2 HÀC(5')); 3.71 (s, 2 MeO); 4.10 (m, HÀC(3'), HÀC(4')); 4.51 (s, HÀC(2')); 4.81 (s, CH2), 5.32 (d, 3J ˆ
3.8, OHÀC(3')); 5.69 (d, 3J ˆ 5.0, OHÀC(2')); 5.84 (d, 3J ˆ 4.7, HÀC(1')); 6.77 7.36 (m, 18 arom. H); 7.99
(s, HÀC(7)); 8.28 (s, HÀC(2)); 9.59 (br. s, NH). Anal. calc. for C49H58N4O9Si (875.09): C 67.25, H 6.68, N 6.40;
found: C 67.27, H 6.62, N 6.44.
1-{5'-O-[Bis(4-methoxyphenyl)phenylmethyl]-b-d-ribofuranosyl}-4-[(diphenylcarbamoyl)oxy]-N6-(phen-
oxyacetyl)-1H-imidazo[4,5-c]pyridin-6-amine (11) and 1-{5'-O-[Bis(4-methoxyphenyl)phenylmethyl]-2'-O-(di-
phenylcarbamoyl)-b-d-ribofuranosyl}-4-[(diphenylcarbamoyl)oxy]-N6-(phenoxyacetyl)-1H-imidazo[4,5-c]pyr-
idin-6-amine (12). To a soln. of 9 (850 mg, 1.18 mmol) in anh. pyridine (15 ml), diphenylcarbamic chloride
(410 mg, 1.8 mmol) and (i-Pr)2EtN (0.3 ml, 2.2 mmol) were added at r.t. The mixture was stirred for 4 h and then
poured into 5% aq. NaHCO3 soln. (15 ml). This soln. was extracted with CH2Cl2 (30 ml, twice), the org. phase
washed with H2O (20 ml) and brine (20 ml), dried (Na2SO4), and evaporated, and the residue purified by FC
(silica gel, column 4 Â 10 cm, CH2Cl2/acetone 4 :1). The slow migrating main zone furnished 11 (650 mg, 61%).
Colorless solid. Rf (CH2Cl2/acetone 4 :1) 0.4. UV (MeOH): 228 (62100), 269 (20700), 274 (sh, 20200). 1H-NMR
((D6)DMSO): 3.23 (m, 2 HÀC(5')); 3.67 (s, 1 MeO); 3.68 (s, 1 MeO); 4.12 (m, HÀC(4'), HÀC(3')); 4.51
(m, HÀC(2')); 4.79 (s, CH2); 5.34 (d, 3J ˆ 5.0, OHÀC(3')); 5.76 (d, 3J ˆ 6.0, OHÀC(2')); 5.93 (d, 3J ˆ 4.4,
HÀC(1')); 6.76 7.48 (m, 28 arom. H); 8.31 (s, HÀC(7)); 8.52 (s, HÀC(2)); 10.77 (br. s, NH). Anal. calc. for
C53H47N5O10 (913.97): C 69.65, H 5.18, N 7.66; found: C 69.56, H 5.16, N 7.70.
The faster-migrating zone gave 12 (165 mg, 13%). Colorless solid. Rf (CH2Cl2/MeOH 9 :1) 0.4. UV
(MeOH): 229 (59400), 269 (18100), 274 (17500). 1H-NMR ((D6)DMSO): 3.26 (m, 2 HÀC(5')); 3.67
(s, 1 MeO); 3.68 (s, 1 MeO); 3.93 (m, HÀC(4')); 4.50 (m, HÀC(3')); 4.79 (s, CH2); 5.53 (t, 3J ˆ 4.4,
HÀC(2')); 5.87 (d, 3J ˆ 5.7, OHÀC(3')); 6.12 (d, 3J ˆ 3.5, HÀC(1')); 6.75 7.48 (m, 38 arom. H); 8.21
(s, HÀC(7)); 8.53 (s, HÀC(2)); 10.79 (br. s, NH). Anal. calc. for C66H56N6O11 (1109.18): C 71.47, H 5.09, N
7.58; found: C 71.67, H 5.19, N 7.35.
1-{5'-O-[Bis(4-methoxyphenyl)phenylmethyl]-2'-O-[tris(1-methylethyl)silyl]-b-d-ribofuranosyl}-4-[(diphen-
ylcarbamoyl)oxy]-N6-(phenoxyacetyl)-1H-imidazo[4,5-c]pyridin-6-amine (13) and 1-{5'-O-[Bis(4-methoxyphen-
yl)phenylmethyl]-3'-O-[tris(1-methylethyl)silyl]-b-d-ribofuranosyl}-4-[(diphenylcarbamoyl)oxy]-N6-(phen-
oxyacetyl)-1H-imidazo[4,5-c]pyridin-6-amine (14). To a soln. of 11 (500 mg, 0.55 mmol) in anh. pyridine (5 ml),

2738
Helvetica Chimica Acta Vol. 86 (2003)
AgNO3 (140 mg, 0.83 mmol) was added under stirring at r.t. A soln. of (i-Pr)3SiCl (125 ml, 0.58 mmol) in dry
THF (8 ml) was introduced after 5 min under exclusion of light and moisture. The reaction was continued for
8 h, before a second portion of (i-Pr)3SiCl (62 ml, 0.29 mmol) was added. After 12 h, the AgCl was filtered off,
the filtrate diluted with 5% aq. NaHCO3 soln. (10 ml), the aq. phase extracted with CH2Cl2 (20 ml, twice), the
combined org. phase washed with H2O (20 ml) and brine (30 ml), dried (Na2SO4), and evaporated, and the
residue applied to FC (silica gel, column 4 Â 10 cm, CH2Cl2/AcOEt/petroleum ether 3 :1 :1). The faster
migrating zone yielded 13 (300 mg, 51%). Colorless foam. Rf (CH2Cl2/AcOEt/petroleumether 3 :1 :1) 0.4. UV
1
(MeOH): 229 (59900), 269 (20100), 274 (sh, 19600). H-NMR ((D6)DMSO): 0.81 0.91 (m, 3 Me2CH2); 3.28
(m, 2 HÀC(5')); 3.69 (s, 2 MeO); 4.14 (m, HÀC(3'), HÀC(4')); 4.69 (t, 3J ˆ 4.2, HÀC(2')); 4.77 (s, CH2); 5.33
(d, 3J ˆ 6.0, OHÀC(3')); 5.98 (d, 3J ˆ 4.4, HÀC(1')); 6.79 7.47 (m, 28 arom. H); 8.28 (s, HÀC(7)); 8.54
(s, HÀC(2)); 10.74 (br. s, NH). Anal. calc. for C62H67N5O10Si (1070.31): C 69.57, H 6.31, N 6.54; found: C 69.35,
H 6.15, N, 6.55.
The slower migrating zone gave isomer 14 (118 mg, 20%). Colorless foam. Rf (CH2Cl2/AcOEt/petroleum-
ether 3 :1:1) 0.45. UV (MeOH): 229 (61900), 269 (20600), 274 (sh, 20200). 1H-NMR ((D6)DMSO): 0.95 0.98
(m, 3 Me2CH2); 3.25 (m, 2 HÀC(5')); 3.68 (s, 2 MeO); 4.12 (m, HÀC(4')); 4.28 (m, HÀC(3')); 4.55
(m, HÀC(2')); 4.78 (s, CH2); 5.66 (d, 3J ˆ 6.3, OHÀC(2')); 5.91 (d, 3J ˆ 4.7, HÀC(1')); 6.78 7.48 (m, 28 arom.
H); 8.31 (s, HÀC(7)); 8.53 (s, HÀC(2)); 10.77 (br. s, NH). Anal. calc. for C62H67N5O10Si (1070.31): C 69.57, H
6.31, N 6.54; found: C 69.65, H 6.38, N 6.56.
1-{5'-O-[Bis(4-methoxyphenyl)phenylmethyl]-2'-O-[tris(1-methylethyl)silyl]-b-d-ribofuranosyl}-4-[(diphen-
ylcarbamoyl)oxy]-N6-(phenoxyacetyl)-1H-imidazo[4,5-c]pyridin-6-amine 3'-(2-Cyanoethyl Diisopropylphos-
phoramidite) (3). A stirred soln. of 13 (300 mg, 0.28 mmol) in anh. CH2Cl2 (9 ml) was pre-flushed with Ar
and treated with (i-Pr)2EtN (0.14 ml, 1.0 mmol) followed by 2-cyanoethyl diisopropylphosphoramidochloridite
(0.19 ml, 0.85 mmol) at r.t. The reaction was monitored by TLC (CH2Cl2/acetone 98 :2). After 4 h, the mixture
was diluted with CH2Cl2 (40 ml), washed with 5% aq. NaHCO3 soln. (20 ml, twice) and brine (20 ml), dried
(Na2SO4), and evaporated and the residue co-evaporated with CH2Cl2 (20 ml) and purified by FC (silica gel,
column 3 Â 10 cm, CH2Cl2/acetone 98 :2): 3 (267 mg, 75%). Colorless foam. Rf (CH2Cl2/acetone 98 :2): Rf 0.45,
0.5. 31P-NMR (CDCl3): 149.9, 153.3.
Oligonucleotides: Synthesis and Properties. The oligoribonucleotides were synthesized on polystyrene
supports (ABI) on an Applied-Biosystems-394-DNA/RNA synthesizer (Foster City, CA, USA). The synthesis
and the deprotection of the oligonucleotides 15 20 (Table 4) were performed as described previously [53][54].
The final oligonucleotides were purified by anion-exchange HPLC and characterized by MALDI-TOF-MS .The
measured masses for the enzyme 16 and for the modified enzymes 17 20 is in accordance with the calculated
values.
Substrate-Cleavage Assay. Ribozyme oligonucleotides and 5'-end 32P-labeled substrates were heated
separately in the reaction buffer at 958 for 1 min, quenched on ice and then equilibrated to the final reaction
temp. (378) prior to starting the reactions. Reactions were carried out with an excess of the enzyme and were
initiated by mixing equal volumes (20 ml) of the substrate (final concentration < 1 nm) and ribozyme (1 mm) in
50 mm Mes buffer, pH 6.5, 10 mm MgCl2. Aliquots (4 ml) were removed at various times (5 s to 2 h), quenched in
8 ml of formamide loading buffer (95% formamide, 20 mm EDTA), and loaded onto a 15% polyacrylamide TBE
(89 mm tris-borate, 2 mm EDTA) gel containing 7m urea. The fraction of substrate and product present at each
time point was determined by quantification of scanned images with a Molecular Dynamics PhosphorImager.
The ribozyme-cleavage rates were obtained from the plots of the fraction of the substrate remaining vs. time
using a nonlinear, least-squares fit to a double-exponential curve (KaleidaGraph, Synergy Software, Reading,
PA, USA). The initial-rate section of the curve represented 80 90% of the total reaction; thus, the observed
cleavage rates (kobs) were taken from the rate constant for the first exponential. Relative rates of cleavage (krel
)
were calculated by dividing the observed cleavage rate by the cleavage rate of unmodified ribozyme
(0.52 minÀ1). The total extent of cleavage was always > 70%.
REFERENCES
[1] K. R. Birikh, P. A. Heaton, F. Eckstein, Eur. J. Biochem. 1997, 245, 1.
[2] D. M. Lilley, Curr. Opin. Struct. Biol. 1999, 9, 330.
[3] C. Carola, F. Eckstein, Curr. Opin. Chem. Biol. 1999, 3, 274.
[4] A. R. Kore, N. K. Vaish, U. Kutzke, F. Eckstein, Nucleic Acids Res. 1998, 26, 4116.
[5] J. B. Murray, A. A. Seyhan, N. G. Walter, J. M. Burke, W. G. Scott, Chem. Biol. 1998, 5, 587.

Helvetica Chimica Acta Vol. 86 (2003)
2739
[6] J. E. Wedekind, D. B. McKay, Annu. Rev. Biophys. Biomol. Struct. 1998, 27, 475.
[7] J. B. Murray, D. P. Terwey, L. Maloney, A. Karpeisky, N. Usman, L. Beigelman, W. G. Scott, Cell 1998, 92,
665.
[8] W. G. Scott, Curr. Opin. Struct. Biol. 1998, 8, 720.
[9] T. Tuschl, C. Gohlke, T. M. Jovin, E. Westhof, F. Eckstein, Science (Washington, D.C.) 1994, 266, 785.
[10] J. B. Murray, H. Szoke, A. Szoke, W. G. Scott, Mol. Cell 2000, 5, 279.
[11] G. S. Bassi, A. I. Murchie, F. Walter, R. M. Clegg, D. M. Lilley, EMBO J. 1997, 16, 7481.
[12] G. S. Bassi, N. E. Mollegaard, A. I. Murchie, D. M. Lilley, Biochemistry 1999, 38, 3345.
[13] M. Menger, F. Eckstein, D. Pˆrschke, Nucleic Acids Res. 2000, 28, 4428.
[14] C. Hammann, D. G. Norman, D. M. Lilley, Proc. Natl. Acad. Sci. 2001, 98, 5503.
[15] D. B. McKay, RNA 1996, 2, 395.
[16] The Ribozyme Webpage http://www.highveld.com/ribozyme.html.
[17] F. Seela, S. Lampe, Helv. Chim. Acta 1991, 74, 1790.
[18] P. D. Cook, O. L. Acevedo, R. S. Andrews, US Patent Application, No. 5,457,191, 1995.
[19] P. D. Cook, R. J. Rousseau, A. M. Mian, R. B. Meyer Jr., P. Dea, G. Ivanovics, D. G. Streeter, J. T.
Witkowski, M. G. Stout, L. N. Simon, R. W. Sidwell, R. K. Robins, J. Am. Chem. Soc. 1975, 97, 2916.
[20] P. D. Cook, R. J. Rousseau, A. M. Mian, P. Dea, R. B. Meyer Jr., R. K. Robins,J. Am. Chem. Soc. 1976, 98,
1492.
[21] G. R. Revankar, P. K. Gupta, A. D. Adams, N. K. Dalley, P. A. McKernan, P. D. Cook, P. G. Canonico,
R. K. Robins, J. Med. Chem. 1984, 27, 1389.
[22] N. Minakawa, A. Matsuda, Tetrahedron 1993, 49, 557.
[23] N. Minakawa, N. Kojima, A. Matsuda, J. Org. Chem. 1999, 64, 7158.
[24] F. Seela, H. J. Driller, Helv. Chim. Acta 1988, 71, 1191.
[25] J. Zemlicka, Collect. Czech. Chem. Commun. 1963, 28, 1060.
[26] K. Mersmann, University of Osnabr¸ck, personal communication.
[27] T. Kamimura, M. Tsuchiya, K. Urakami, K. Koura, M. Sekine, K. Shinozaki, K. Miura, T. Hata, J. Am.
Chem. Soc. 1984, 106, 4552.
[28] T. Kamimura, M. Tsuchiya, K. Koura, M. Sekine, T. Hata, Tetrahedron Lett. 1983, 24, 2775.
[29] F. Seela, C. Wei, Helv. Chim. Acta 1997, 80, 73.
[30] J. C. Schulhof, D. Molko, R. Teoule, Nucleic Acids Res. 1987, 15, 397.
[31] F. Seela, T. Grein, Nucleic Acids Res. 1992, 20, 2297.
[32] N. Usman, K. K. Ogilvie, M.-Y. Jiang, R. J. Cedergren, J. Am. Chem. Soc. 1987, 109, 7845.
[33] G. H. Hakimelahi, Z. A. Proba, K. K. Ogilvie, Tetrahedron Lett. 1981, 22, 4775.
[34] B. C. Froehler, −Protocols of Oligonucleotides and Analogs×, in −Methods in Molecular Biology×, Vol. 20,
Ed. E. S. Agrawal, Humana Press, Tutowa, New Jersey, 1994, p. 33.
[35] J. van Wijk, C. Altona, −PSEUROT 6.2 A Program for the Conformational Analysis of the Five-
membered Rings×, University of Leiden, 1993.
[36] C. A. G. Haasnoot, F. A. A. M. de Leeuw, C. Altona, Tetrahedron 1980, 36, 2783.
[37] E. Westhof, O. Rˆder, I. Croneiss, H. D. L¸demann, Z. Naturforsch. 1975, 30, 131.
[38] H. Rosemeyer, G. Toth, B. Golankiewicz, Z. Kazimierczuk, W. Bourgeois, U. Kretschmer, H.-P. Muth, F.
Seela, J. Org. Chem. 1990, 55, 5784.
[39] S.-P. Jiang, G. Raghunathan, K.-L. Ting, J. C. Xuan, R. L. Jernigan, J. Biomol. Struct. Dyn. 1994, 12, 367.
[40] S. B. Tzokov, I. A. Murray, J. A. Grasby, J. Mol. Biol. 2002, 324, 215.
[41] R. Knoll, R. Bald, J. P. Furste, RNA 1997, 3, 132.
[42] M. R. Hansen, J. P. Simorre, P. Hanson, V. Mokler, L. Bellon, L. Beigelman, A. Pardi, RNA 1999, 5, 1099.
[43] A. Peracchi, L. Beigelman, E. C. Scott, O. C. Uhlenbeck, D. Herschlag, J. Biol. Chem. 1997, 272, 26822; Y.
Takagi, M. Warashina, W. J. Stec, K. Yoshinari, K. Taira, Nucleic Acids Res. 2001, 29, 1815.
[44] H. W. Pley, K. M. Flaherty, D. B. McKay, Nature (London) 1994, 372, 68.
[45] W. G. Scott, J. T. Finch, A. Klug, Cell 1995, 81, 991.
[46] S. Wang, K. Karbstein, A. Peracchi, L. Beigelman, D. Herschlag, Biochemistry 1999, 38, 14363.
[47] D. J. Fu, S. B. Rajur, L. W. McLaughlin, Biochemistry 1993, 32, 10629; D. J. Fu, L. W. McLaughlin,
Biochemistry 1992, 31, 10941.
[48] F. Seela, K. Mersmann, J. A. Grasby, M. J. Gait, Helv. Chim. Acta 1993, 76, 1809.
[49] M. J. Gait, J. A. Grasby, J. Karn, K. Mersmann, C. E. Pritchard, Nucleosides Nucleotides 1995, 14, 1133.
[50] J. B. Murray, D. P. Terwey, L. Maloney, A. Karpeisky, N. Usman, L. Beigelman, W. G. Scott, Cell 1998, 92,
665.

2740
Helvetica Chimica Acta Vol. 86 (2003)
[51] A. B. Burgin Jr., C. Gonzalez, J. Matulic-Adamic, A. Karpeisky, N. Usman, J. A. McSwiggen, L. Beigelman,
Biochemistry 1996, 35, 14090.
[52] A. B. Burgin, L. Beigelman, (Boulder, Colorado, USA), Ribozyme, unpublished results.
[53] F. Seela, H. Debelak, N. Usman, A. Burgin, L. Beigelman, Nucleic Acids Res. 1998, 26, 1010.
[54] F. Wincott, A. DiRenzo, C. Shaffer, S. Grimm, D. Tracz, C. Workman, D. Sweedler, C. Gonzalez, S.
Scaringe, N. Usman, Nucleic Acids Res. 1995, 23, 2677.
Received February 14, 2003