Total syntheses of a conformationally locked North-type methanocarba puromycin analogue and a dinucleotide derivative

Article abstract of DOI:10.1002/chem.200802629

An original synthetic approach for the first synthesis of an enantiopure methanocarba puromycin (3'-α-aminoacylamino-3'-deoxyadenosine) analogue and its cytidine dinucleotide derivative is described. Each compound is conformationally locked in a North-typ

Full text of DOI:10.1002/chem.200802629

DOI: 10.1002/chem.200802629
Total Syntheses of a Conformationally Locked North-Type Methanocarba
Puromycin Analogue and a Dinucleotide Derivative
Benoꢀt Y. Michel and Peter Strazewski*[a]
Abstract: An original synthetic ap-
proach for the first synthesis of an
enantiopure methanocarba puromycin
(3’-a-aminoacylamino-3’-deoxyadeno-
sine) analogue and its cytidine dinucle-
otide derivative is described. Each
compound is conformationally locked
in a North-type pucker and exhibits
both a pseudoaxial hydroxy group and
a pseudoequatorial aminoacyl group.
The syntheses were accomplished from
d-ribose in 18 and 19 steps, respective-
ly, with key steps being a ring-closing
metathesis, a Luche reduction, a Sim-
mons–Smith cyclopropanation, a Mit-
sunobu coupling, a Mattocks bromoa-
cetylation, a regioselective and a ste-
reoselective nucleophilic substitution, a
chemoselective phosphoramidite cou-
pling and a Staudinger–Vilarrasa cou-
pling. Both molecules are being tested
for peptidyl transfer efficiency in ribo-
somes for comparison with the peptidyl
transfer kinetics of natural puromycin
and other natural and synthetic riboso-
mal A site substrates.
Keywords: antibiotics · carbocycles ·
nucleosides
ribosomes
· protein synthesis ·
Introduction
Puromycin (1, Pm, Figure 1), a natural antibiotic nucleoside,
was isolated from Streptomyces alboniger in 1952 by Porter
et al.[1] It has been intensively used to investigate protein
biosynthesis and to clear up understanding of its mecha-
nism.[2] Its structural similarity to the 3’ terminal 3’-O-ami-
noacyl adenylate moiety that is conserved in all aminoacyl-
tRNA explains its activity in the ribosomal A site. Puromy-
cin inhibits protein synthesis by transfer of the nascent poly-
peptide chain to its a-amino group.[3] Unfortunately, puro-
mycin is cytotoxic, and so we wish to develop new analogues
with similar nascent peptide-accepting properties but lower
toxicity.
Figure 1. Puromycin (1) and its carbocyclic locked North-type methano-
carba analogue 2.
Puromycin analogues are currently sought-after com-
pounds for investigation of the ribosomal catalysis of the
peptidyl transfer (PT) reaction. Simpler model compounds
have been used by Wolfenden and colleagues to compare
the kinetics of uncatalyzed bimolecular ester aminolysis in
aqueous solution with those catalyzed by the ribosome’s
peptidyl transferase. When this aminolysis was compared to
the kinetics of ribosome-catalyzed transfer of the N-formyl-
methionyl group to puromycin (PT from P site-bound fMet-
tRNAi to A site-bound Pm), a modest 3.5 to 30 million-fold
rate acceleration could be deduced and—owing to the virtu-
ally absent temperature dependence of PT in Arrhenius
plots—was attributed to a purely entropic effect provided
by the peptidyl transferase and proposed to originate either
from precise substrate alignment[4a–c] or, as observed there-
after in MD simulations, from the conservation of solvation
entropy of the nucleophilic a-amino group between an as-
sumed reaction ground state and the transition state.[4d,e] To
conclude from Arrhenius plots, despite their linearity, that
[a] Dr. B. Y. Michel, Prof. P. Strazewski
UMR 5246, CNRS, Laboratoire de Synthꢀse de Biomolꢁcules
Universitꢁ Claude Bernard Lyon 1, Villeurbanne
Universitꢁ de Lyon, 69622 Lyon (France)
E-mail: strazewski@univ-lyon1.fr
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802629 and from the author.
6244
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 6244 – 6257

FULL PAPER
¼
very low apparent TDS values indicate pure entropic catal-
sized four new locked N-MC 2’-deoxypuromycin analogues
(3, 4, 5, and 6, Figure 2) in which enantiopure N-MC and S-
MC bicycloACTHNGUETRNNU[G 3.1.0]hexane ring systems substituted for puro-
mycin’s d-ribofuranose moiety in order to mimic North-type
and South-type puckers, respectively, that would present
precisely positioned 3’ substituents (nucleoside numbering)
and be unable to switch into the other pucker regime.[13]
ysis may be tempting but is likely to be misleading, because
¼
¼
¼
both TDS and DH are derived from DG (through lnk
and 1/T) and necessarily compensate one another such that
relatively low measured DG values result: 16.5 kcalmolꢀ1
¼
for the above PT reaction, in comparison with 22.2–23.5 kcal
molꢀ1 for the uncatalyzed ester aminolysis (both at 258C).
More recently, similar independent studies in the Ehrenberg
and the Rodnina laboratories revealed faster PT kinetics,
¼
and hence even lower catalyzed free energy barriers: DG
(258C)=15.1, 13.9, and 12.6 kcalmolꢀ1. In those cases, pep-
tide-accepting puromycin had been replaced by phenylalan-
[4f]
yl-tRNA (PT kinetics from fMet-tRNAi to Phe-tRNAPhe
)
and the donating N-formylmethionyl group by two different
N-formyl dipeptides (PT kinetics from fMet-Phe-tRNAPhe or
fMet-Arg-tRNAArg to Pm).[4g] These results suggest that ri-
bosomal rate accelerations with respect to uncatalyzed ester
aminolyses may amount to more than 1010-fold. The molecu-
lar reasons for such efficient catalysis remain unknown. We
wished to study the precise structural and functional prereq-
uisites for ribosomal nascent-peptide acceptors more care-
fully.[5] To this end we set out to synthesize conformationally
restricted puromycin analogues, in order to gain more in-
sight into the nature of ribosomal catalysis of PT, as well as
to find new lead compounds that might be used as novel an-
tibiotics.
Two conformations adoptable by a nucleosidic furanose
ring, usually referred to as North-type and South-type, are
well-known to play a dominant role in the mechanisms of
action of a number of antibiotic nucleosides. Inspired by
crystallographic results for neplanocin C,[6] Marquez concep-
tualized and synthesized methanocarba (MC) nucleosides
Figure 2. Locked N-MC and S-MC 2’-deoxypuromycin analogues. The
aminoacyl side chain is pseudoequatorially oriented in the N-MC 2’-de-
oxypuromycin analogues 3 and 4[7w] and pseudoaxially oriented in the S-
MC 2’-deoxypuromycin analogues 5 and 6.
Crystallographic studies suggest that the 2’-hydroxy group
of puromycin—and thus, equally, of the 3’-terminal A76 resi-
due of any natural 3’-aminoacyl tRNA—may be important
for good binding to the ribosomal A site if pseudoaxially
oriented.[14a] After PT and translocation of the new peptidyl-
tRNA into the P site,[14b] the 2’-hydroxy group of the same
residue A76 switches to a pseudoequatorial orientation (pur-
omycin and analogues are not translocated after PT). The
immediately preceding cytidine residue C75, also present in
all tRNAs, is base-paired in the A and P sites with highly
conserved ribosomal guanosine residues. Puromycin that is
linked to 3’-cytidylate might bind more stably to the A site
and/or further accelerate PT, resulting in more efficient in-
hibition of protein synthesis. We therefore designed two pur-
omycin analogues in which the d-ribofuranose moiety would
incorporating a conformationally rigid bicycloACHTNUTRGNE[NUG 3.1.0]hexane
scaffold (Figure 2), to allow optimal mimicry of the North-
type (N) and South-type (S) conformations of a ribofura-
nose pucker.[7] The North-type and South-type pucker
ranges usually preferred by ribofuranosidic nucleosides are
C3’-endo (3E) and C2’-endo (2E), respectively. This trans-
lates into pseudorotational angles of P ꢁ 158 and 1658
(ꢂ158), respectively.[8] The replacement of C4’ O4’ and
be replaced by an enantiopure N-MC bicycloACTHGNUTERNNU[G 3.1.0]hexane
ꢀ
ꢀ
O4’ C1’ by a fused cyclopropane moiety in a trans configu-
ring system that is unable to switch into another pucker and
contains a pseudoaxially locked 2’-hydroxy function (nucleo-
side numbering), to enhance A site recognition further.
Here we report on the synthesis of the locked ribo-puromy-
cin analogue 2 (Figure 1) and its 5’-O-(3’-cytidylate) deriva-
tive 36 (Scheme 8). All locked analogues—2, 3, 4, 5, 6 and
36—are currently being tested for peptidyl transfer efficien-
cy in ribosomes and the findings are being compared with
the enzymological results obtained for natural puromycin
(1) and other natural and unnatural A site substrates.
ration with respect to the nucleobase constrains the cyclo-
pentane rings of these carbobicyclic nucleosides in boat con-
formations that result in C2’-exo (2E) puckers for the North-
type isomers and C3’-exo (3E) for the South-type isomers (P
ꢁ ꢀ188 and 1988, respectively). Other rigid bicyclic ana-
logues that are locked into one of the two puckers through
an oxymethylene bridge between C4’ and either C3’ or C2’
(nucleoside numbering) exhibit conformations even more
similar to those of the parent analogues.[9–12] However, the
Marquez analogues are the only ones that still maintain the
natural functions at C2’ and C3’, which is of prime impor-
tance for our purpose. The pseudoaxial/-equatorial position-
ing of their 2’- and 3’-substituents is very close to that of the
mimicked ribonucleosides: North=2’-pseudoaxial/3’-pseu-
doequatorial, South=2’-pseudoequatorial/3’-pseudoaxial. In
a collaboration with the Marquez group we recently synthe-
For the retrosynthetic analysis we began with the cleavage
of the peptide bond to the N-MC bicycloACTHNUTRGNEUNG[3.1.0]hexane scaf-
fold (Scheme 1). The amino acid side chain can be easily in-
troduced by Staudinger–Vilarrasa coupling, for which we
had previously optimized the experimental conditions. An
earlier comparative study of the difference in reactivity be-
tween the iminophosphorane generated in situ and the nu-
Chem. Eur. J. 2009, 15, 6244 – 6257
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6245

P. Strazewski and B. Y. Michel
Scheme 2. Synthetic pathway leading to bicyclic alcohols 16 and OP(ꢂ)2/
3/4.
with the six-times-dearer Grubbs II catalyst.[17] The both
regio- and stereoselective Luche reduction[22] and the subse-
quent stereoselective Simmons–Smith cyclopropanation[7d,23]
were carried out on enantiopure 14 by use of the same pro-
tocol as employed earlier for racemic OP(ꢂ)0 and gave al-
cohol 16 as a single diastereoisomer and in high yield.
OP(ꢂ)2 was isomerized in acetone containing catalytic
amounts of p-TsOH into OP(ꢂ)3, which was separated
from its regioisomer through flash chromatography over
silica gel and then reduced with sodium in liquid ammonia
to give OP(ꢂ)4 in 50% yield. We managed to grow mono-
crystals of the last two compounds. Their X-ray structures
(CCDC-711880 and -711881, respectively) served to corrob-
orate their diastereoisomeric identities and to provide us
with additional information on the puckering of the cyclo-
pentane moiety (see Discussion).
We initially wished to introduce the nucleobase through a
Mitsunobu coupling with 6-chloropurine as the nucleophile.
We expected to take advantage of the lability of the chlorine
atom by replacing it with miscellaneous amines at some
later stage to provide us with several nucleobase variants
differently substituted at C6. However, any acidic treatment
tested for the removal of the acetonide and the silyl protect-
ing group from 17 concomitantly hydrolyzed the chlorine
and led to N-MC inosine derivative 18 (Scheme 3), which
had already been synthesized by Marquez and Jacobson,
who showed that it was a weaker binder to A3 AR (an ade-
nosine receptor subtype) than the parent inosine, which is
possibly the endogenous ligand.[7p]
Scheme 1. Retrosynthetic pathway leading to target compound 2.
cleobase’s 6-amino group towards the activated ester during
the amino acid coupling showed that, to improve the yield
of peptide bond formation, N6-amidine protection should be
chemoselectively introduced beforehand.[15] The replace-
ment of the 3’-hydroxy group by an azido group with reten-
tion of configuration and without changes at C2’ and C5’
constitutes one of the major difficulties of this total synthe-
sis. We planned to introduce the azido function through an
SN2 reaction on a pseudoaxial 3’-halogen function. The ap-
propriate xylo-derivative can be obtained through a Mat-
tocks bromoacetylation reaction.[16] A Mitsunobu coupling
allows for stereoselective introduction of the adenine
moiety in protected form. To lock the scaffold with the help
of a MC ring, a diastereoselective Simmons–Smith cyclopro-
panation directed by the 1a-hydroxy group seems appropri-
ate. This alcohol function can be obtained by diastereoselec-
tive Luche reduction of the functionalized cyclopentenone
14 (Scheme 2), which had been described in the literature[17]
and optimized during our previous studies.[18]
Results
The nine-step synthetic pathway to 16 began with d-ribose
(=7, Scheme 2) and profited from a very-well worked-out
route published by Jeong and colleagues[19] and used by
others.[20] The only significant change that we introduced
was in the RCM reaction from 12 to 13, which was carried
out in the presence of the Fꢃrstner catalyst[21] (Neolyst di-
chlorideTM) and exhibited results similar to those obtained
In order to circumvent hypoxanthine protection problems
downstream, we turned to the Garner procedure,[24] in which
N6,N6-bis-Boc adenine (21) can be synthesized in two easily
scalable steps. Adenine (19) was per-protected as its
9,N6,N6-tris-tert-butoxycarbonyl carbamate 20, from which
6246
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 6244 – 6257

Puromycin Analogues
FULL PAPER
ished to observe, under all tested reaction conditions that
would allow 23 to react at all, that at least 1–3% v/v water
initially had to be added, resulting in the exclusive forma-
tion of the 2’-O-acetyl-3’,5’-dibromo derivative 24, for which
two zero coupling constants—for H2’–H3’ and H1’–H2’—in-
dicated two vicinal trans relationships, and hence the desired
xylo configuration. Under the applied reaction conditions
(i.e., after more water had been added to drive the reaction
to completion), the primary hydroxy group of 23 apparently
did not survive the increasing acidity of the medium and un-
derwent, in part after orthoesterification, a substitution reac-
tion with the released HBr (Scheme 5).
Scheme 3. Synthesis of the N-MC inosine 18.
21 was regioselectively generated under mildly basic condi-
tions and in very good yield. Unlike 6-chloropurine, 21 is
highly soluble in THF. We carried out the Mitsunobu cou-
pling with 21, and it proceeded cleanly, rapidly and almost
quantitatively, most probably thanks to the high solubility of
the nucleophile (Scheme 4).[18] Acidic treatment of 22 pro-
vided the cyclopropane-fused carbocyclic adenosine 23 in
good yields.[7d]
Scheme 5. Proposed mechanism of the Mattocks dibromoacetylation re-
action based on refs. [25a,b] and our observation that 2’-OH was less re-
active than 3’-OH (nucleoside numbering, B=nucleobase). Attack on
C5’ probably occurs only late during the reaction, when more water is
added to solubilize all starting material and the acidity of HBr thus mark-
edly increases. In the original studies on adenosine, no late addition of
water was needed for the completion of the reaction (to give the corre-
sponding trans-monobromoacetate); were it necessary, this would cleave
the glycosidic bond due to overacidification (unpublished observations).
Scheme 4. Synthesis of dibromo analogue 24.
We then wished to functionalize the 3’ position (nucleo-
side numbering) without affecting the 2’- and 5’-hydroxy
groups, in order to continue with the introduction of an
azide function for subsequent coupling with the amino acid
through a Staudinger–Vilarrasa reaction. We therefore en-
visaged a reaction that would invert the 3’-hydroxy moiety
to provide a function that would serve as a leaving group
for the following step: a SN2 reaction with inorganic azide.
The Mattocks bromoacetylation seemed useful for our pur-
pose,[16] because it had been successfully applied for the ste-
reospecific synthesis of 2’,3’-ribo-epoxides both by others[25]
and by ourselves.[15] This reaction would be expected to con-
vert vicinal cis-diols into vicinal trans-bromoacetates. In the
case of the N-MC analogue 23,[25g] however, we were aston-
As a consequence, we decided to protect this alcohol
function beforehand as a benzyl ether, which should resist
harsh acidic conditions. We restarted from 22, which was de-
silylated with TBAF (Scheme 6), and the resulting primary
alcohol 25 was benzylated to give 26. Acidic treatment
cleaved the acetonide and the Boc groups to afford the vici-
nal cis-diol 27, and another Mattocks reaction was attempt-
ed. Astonishingly, we obtained an even more unexpected
and hitherto unprecedented result in the formation of a
xylo-configured dibromoacetate, the stereo- and regioiso-
meric identity of which could be ascertained by NMR spec-
troscopy through two zero coupling constants for H2’–H3’
and H1’–H2’ (desired stereoisomer), together with a 3J
cross-peak between the carbonyl C atom of the acetate
Chem. Eur. J. 2009, 15, 6244 – 6257
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6247

P. Strazewski and B. Y. Michel
to give m/z 226 and 454/456, respectively, from which a
second fragmentation again produced mainly m/z 226 and
fragments from which ketene, AcOH and HBr had cleaved
(m/z 412/414, 394/396, and 374). NMR spectroscopic investi-
gation of 28 identified an N-benzyl compound. Unlike the
1
well-resolved H NMR doublets for OCH2Ph in 26 and 27,
1
the corresponding methylene H resonance in 28 was broad,
due to a geminal quadrupole 14NC1H2Ph coupling. This
methylene carbon signal could not be found either in
13C NMR or in DEPT spectra, again most probably due to
an efficient quadrupole 14N13CH2Ph broadening. With the
1
help of a clear H–13C HSQC cross-peak this 13C atom was
found to resonate at ~44 ppm, which was confirmed by
HMBC through a weak vicinal coupling between 13CH2 at
~44 ppm and the ortho-1H in the phenyl group (see Support-
ing Information).
Protection of the primary alcohol for this bromoacetyla-
tion reaction thus appeared to have become pointless. It was
more judicious to reconsider the previously obtained com-
pound 24 and to profit from the difference in reactivity be-
tween the two bromo groups. Actually, this 3’,5’-dibromodi-
deoxy-2’-acetate derivative, with its primary and secondary
leaving groups, offered the potential for regioselective intro-
duction of different nucleophiles. After reference to studies
by Mander,[26] the use of CsOAc enabled us to substitute the
primary bromide to provide 29 in high yields (Scheme 7).
We were successful in crystallizing this diacetylated deriva-
tive from a CHCl3/MeOH mixture. The X-ray structure
(Figure 3, below; see Discussion section) confirmed the
regio- and stereochemical dispositions of C2’ and C3’ origi-
nally deduced from NMR spectroscopic analyses.
Scheme 6. Benzyl migration during the Mattocks bromoacetylation of 27.
function and H2’ through HMBC (desired regioisomer; see
the Supporting Information). However, the 5’-ether group
had been cleaved to yield 28, in which the benzyl group had
migrated to the adenine moiety. An MS·MS analysis on 28
showed that an initial fragmentation of the central dibromo
isotopomeric molecular ion at m/z 543/536/538 resulted in
the formation of benzyladenine and the elimination of HBr
Scheme 7. Synthesis of the N-MC puromycin analogue 2.
6248
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 6244 – 6257

Puromycin Analogues
FULL PAPER
Scheme 8. Synthesis of the locked cytidine puromycin analogue dinucleotide 36.
During the SN2 reaction between inorganic azide and the
pseudoaxial C3’-bromide (nucleoside numbering) we en-
countered problems, already raised by Rꢁglier,[27] involving
anchimeric assistance by the O2’-acetate group. We there-
fore decided to cleave both ester functions beforehand, first-
ly under mildly basic conditions. Because we had consistent-
ly been able to transform the resulting 3’-bromo-2’,5’-diol
into the corresponding ribo-2’,3’-epoxide MC analogue, we
turned to an acidolysis of 29 by treatment with aqueous HCl
in warm MeOH and obtained the completely deprotected
xylo-derivative 30 (Scheme 7).[28] Now the SN2 reaction with
NaN3 in DMF furnished satisfactory yields of 31. The forma-
tion of the ribo-epoxide (25%) could nevertheless not be
prevented. Recently[15] we showed that it was beneficial first
to protect the adenine N6 group as a dibutyl formamidine
(dbf),[29a] in order to increase the yield of the subsequent
amino acid coupling. The main advantages of dbf protection
are its facile and efficient preparation and cleavage[29] and,
most importantly, that it can be used, unlike a number of
other common amine-protecting groups, in the presence of
numerous hydroxy groups, as well as of many organic func-
tions including azido functionalities.[30] The N6-protected
azido diol 32 was chemoselectively[31] coupled to the N-pro-
tected l-amino acid oxybenzotriazolyl ester Fmoc-Tyr(Me)-
OBt (prepared in situ) by a Staudinger–Vilarrasa procedure,
the conditions of which (reduction with Me3P, facile separa-
tion of Me3PO) had been optimized previously.[13,15] The
coupled product 33 was completely deprotected by treat-
ment with ethanolic MeNH2, to furnish target compound 2.
A final lyophilization from H2O/TFA at pH 2.2 led to the
more water-soluble 2·TFA salt as a white, fluffy solid.
For the synthesis of dinucleotide analogue 36 (Scheme 8)
we restarted from the dbf-protected azido diol 32, which
was chemoselectively coupled with a commercial cytidine
phosphoramidite derivative by a standard solution proto-
col.[32] Subsequent in situ treatment with aqueous methanol-
ic I2 in the presence of pyridine immediately oxidized the in-
termediate phosphite to the desired P-diastereoisomeric
phosphotriesters 34. The Staudinger–Vilarrasa procedure
was applied to 34 and afforded 35 in satisfying yields. Final-
ly, a three-step, one-pot deprotection procedure[33] provided
the zwitterionic target aminoacyl dinucleotide N-MC ana-
logue 36. It is noteworthy that this highly polar compound
could be efficiently purified through preparative normal-
phase silica TLC with the highly polar eluent iPrOH/NH3/
H2O.
Discussion
A comparison of the characteristic torsion and pseudorota-
tional angles in the X-ray structures of the 3’b-bromo-2’,5’-
diacetate derivative 29 (Figure 3) and of other similar
bicycloACHTNUTRGNE[NUG 3.1.0]hexane derivatives on the one hand, with those
in puromycin on the other, revealed that the conformation
ꢀ
around C1’ N9 (nucleoside numbering: the “glycosidic
bond”), when present, always lies in the anti regime, where-
ꢀ
as the one around C5’ C4’ depends on the steric access be-
tween C5’ and the nucleobase: usually gauche+, although
gaucheꢀ in the xylo analogue 29. Both torsion angles are
likely to be potentially influenced by crystal packing forces
as well. More importantly, the pucker of the cyclopentane
moiety always points to North-type 2’-exo (2E), but with a
Chem. Eur. J. 2009, 15, 6244 – 6257
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6249

P. Strazewski and B. Y. Michel
Figure 4. Main close through-space NOESY cross-correlations that sup-
port the stereochemistry of N-MC puromycin analogue 2 (nucleoside
numbering).
Figure 3. X-ray structure of the diacetylated 3’-bromo compound 29
(CCDC-711879). It is characterized through the following conformational
parameters, as obtained from the Pseudo-Rotational Online Service and
Interactive Tool PROSIT (http://cactus.nci.nih.gov/prosit): P=ꢀ20.48
(339.68: 2E), nmax =16.08, c=ꢀ149.88 (purine anti), g=ꢀ69.68 (O5’–C3’
gaucheꢀ, nucleoside numbering). The same parameters for puromycin
(CCDC code PURMYC10): P=3.98 (3E), nmax =38.38, c=ꢀ159.38
(purine anti), g=54.08 (O5’–C3’ gauche+); for a 3’-azido-2’,3’-dideoxy
(nucleoside numbering) N-MC adenosine analogue (CCDC code FAD-
SAF)[7w]: P=ꢀ20.38 (339.78: 2E), nmax =30.58, c=ꢀ160.58 (purine anti),
g=ꢀ69.98 (O5’–C3’ gauche+, nucleoside numbering); for OP(ꢂ)3
(CCDC-711880): P=ꢀ20.48 (339.68: 2E), nmax =14.08, g=49.58 (O5–C3
gauche+, carbohydrate numbering); for OP(ꢂ)4 (CCDC-711881): P=
ꢀ14.78 (345.38: 2E), nmax =24.18, g=61.38 (O5–C3 gauche+, carbohydrate
numbering). For the definitions of P, nmax, c, and g see PROSIT or
ref. [8].
However, according to the X-ray structure of 29 (CCDC-
711879) the unexpected cross-peaks may point to shorter
through-space distances (3.69 ꢄ for H8···H1’ and 3.36 ꢄ for
ꢀ
H8···H2’) reflecting a slightly different C1’ N9 torsion angle
(“glycosidic bond”), which indicates that in dissolved 2 this
torsion angle (c) is closer to ꢀ149.88, as in solid 29, than to
ꢀ160.58, as in the solid 3’-azido-2’,3’-dideoxy N-MC ana-
logue (FASDAF). Neither the unexpected NOESY cross-
correlations nor the missing ones have any important bear-
ing on the relative configuration at C2’ and C3’.
The ensemble that constitutes the NMR spectroscopic evi-
dence for the expected configuration of the target com-
pound is based on the following observations.
quite strongly varying amplitude as characterized by nmax. In
general, the more pronounced a North-type amplitude is,
the “more equatorial” the orientation of the 3’a substituent
becomes. The highest amplitude (i.e., the closest to that of
natural puromycin: nmax =38.38) was observed in the 3’-
azido-2’,3’-dideoxy N-MC adenosine analogue that we had
published earlier (nmax =30.58),[7w] followed by the 1a,2-O,O-
isopropylidene derivative OP(ꢂ)4 (nmax =24.18), the 3’b-
bromo-2’,5’-diacetyl derivative 29 (nmax =16.08) and the 5-O-
A) The scalar vicinal coupling constant for H2’–H3’ is rela-
3
tively large: J=6.0 Hz for 2, 6.9 Hz for 2·TFA, 6.5 Hz
for azide 32, and 6.6 Hz for azide 31, as for all ribo-type
configured compounds 8 to 18 before the Mattocks bro-
moacetylation (corresponding 3J=5.4–7.5 Hz), whereas
the C3’-inverted xylo-configured bromo derivatives 24,
28, 29, and 30 all showed zero coupling constants—these
H2’ and H3’ resonances were all singlets.
benzyl-1a,2-O,O-isopropylidene derivative OP(ꢂ)3 (nmax
=
14.08): see the legend to Figure 3.
The relative configurations of C2’ and C3’ in the target
compounds were ascertained through a systematic compari-
son between the NOESY cross-correlation signals of aque-
ous 2·TFA and the corresponding H···H distances as eluci-
dated from three crystal structures: those of (nucleoside
numbering) the 3’-azido-2’,3’-dideoxy N-MC adenosine ana-
logue[7w] (CCDC code FADSAF) and 29 (CCDC-711879)
for all distances except those relating to amino acid side
chains and of crystalline puromycin dihydrochloride (CCDC
code PURMYC10) for H···H distances within the amino
acid side chain only (see detailed table in the Supporting In-
formation). Out of 24 observed significant and weak
NOESY cross-correlations, only one unexpected significant
correlation and one unexpected weak correlation were
found according to the through-space distances as elucidated
from FADSAF (3.91 ꢄ for H8···H1’ and 4.07 ꢄ for H8···H2’,
respectively), whereas two weak cross-correlations were
somewhat expected from FADSAF but missing: 3.5 ꢄ for
H5A···H6’B and 3.7 ꢄ for H5’B···H6A, respectively (Figure 4).
B) Because of through-space distances well above 4 ꢄ, we
expect absent H5’···H2’ NOESY/ROESY cross-peaks
(nucleoside numbering) for both ribo- and xylo-config-
ured diastereoisomers. Likewise, because of the close
through-space distances from H5’···H4’ and from cis- or
trans-H2’···H3’, consistently lying between 2.12 and
2.64 ꢄ, together with, in MC nucleosides, longer
H5’···cylopropyl proton distances of 2.5–4.0 ꢄ, for both
ribo- and xylo-configured diastereoisomers we would
expect prominent cross-peaks for close contacts in the
former and at least a few weaker cross-peaks for longer
distances in the latter, all of which is experimentally
confirmed in the NOESY spectrum of 2·TFA, as well as
in the ROESY spectrum of a recently published synthet-
ic xylofuranosyl puromycin analogue (see Supporting In-
formation).[15] In the NOESY spectrum of 2·TFA, how-
ever, we found three clear cross-correlations that would
be expected in a ribo-configured diastereoisomer but
not (or much less so) in a xylo-configured one: one
6250
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 6244 – 6257

Puromycin Analogues
FULL PAPER
prominent cross-peak each for both H8···H3’b and
H5’B···H3’b (nucleoside numbering) and a weaker but
still significant cross-peak for H5A···H3’b—through-
space distances that, according to FASDAF, correspond
to 3.17, 2.96 and 3.84 ꢄ, respectively. Indeed, unlike in
the case of the NOESY spectrum of 2·TFA, in the
ROESY spectrum of the xylofuranosyl puromycin ana-
logue we found no cross-peak at all for H8···H3’a, and
cross-correlations for the two H5’··· H3’a distances were
barely above noise (see the Supporting Information).
On the other hand, prominent ROESY cross-peaks
were observed for cis-H1’a···H3’a in the xylofuranosyl
puromycin analogue (trans H1’a···H3’b distance from
FASDAF: 4.0 ꢄ), being almost as intense as the cross-
peak for cis-H1’a···H4’a (expected distance 3.15 ꢄ), as
well as cross-peaks between the adenine H2 and amino-
acyl protons HCa and HCmeta, which were all absent in
the NOESY spectrum of 2·TFA. Close H2···HCa and
H2···HCmeta distances suggest that, unlike in 2, the xy-
lofuranosyl puromycin analogue favors a syn conforma-
tion in the glycosidic bond.
enzymological results obtained with natural puromycin and
natural A site substrates (i.e., 3’-aminoacyl transfer RNAs).
Experimental Section
General methods: All non-aqueous reactions were performed in oven-
dried glassware under nitrogen. The synthetic intermediates were co-
evaporated twice with toluene beforehand and dried in vacuo before use.
All chemical reagents were obtained from commercial sources and were
used as supplied. The reactions were monitored by thin-layer chromatog-
raphy (TLC, Merck silica gel 60 F254 plates) and visualized by UV radia-
tion (254 nm), by spraying with 5% ethanolic H2SO4 then variously soak-
ing in ethanolic ninhydrin (20%) for the free amines, in ethanolic naph-
thoresorcinol (2%) for the ribose intermediates, or in phosphomolybdic
acid in ethanol for the carbocyclic derivatives, and by subsequent warm-
ing with a heat gun. Column chromatography was performed with flash
silica gel (10–63 mm). All the solvents used were purchased over molecu-
lar sieves and were then stored under nitrogen prior to use. Molecular
sieves were dried by microwave activation before use. All NMR spectra
were recorded on Bruker ALS (300 MHz), DRX (300 MHz), and DRX
(500 MHz) spectrometers. 1H NMR (300 and 500 MHz), 13C NMR (75
and 125 MHz, recorded with complete proton decoupling), and 31P NMR
(121.5 MHz, recorded with complete proton decoupling) spectra were ob-
tained with samples dissolved in CDCl3, CD3OD, or [D6]DMSO, with the
residual solvent signals as internal references: 7.26 ppm for CHCl3,
Figure 4 summarizes the most decisive cross-correlations
1
3.31 ppm for CD2HOD, 2.50 ppm for (CD3)
N
observed in 2 that support nothing other than a ribo-type
configuration. The complex H NMR spectrum of dinucleo-
periments, and 77.0 ppm for CDCl3, 49.0 ppm for CD3OD, 39.4 ppm for
(CD3)2S(O) for 13C NMR experiments. 31P NMR experiments were refer-
enced to H3PO4 as an external standard (0.00 ppm). Chemical shifts (d)
are given in ppm to the nearest 0.01 (1H, 31P) or 0.1 ppm (13C). The cou-
pling constants (J) are given in Hertz (Hz). The signals are reported as
follows: (s=singlet, d=doublet, t=triplet, q=quartet, m=multiplet,
br=broad signal, ps=pseudo). Assignments of 1H and 13C NMR signals
were achieved with the help of D/H exchange, COSY, HSQC, HMBC,
and DEPT experiments. High-resolution mass spectrometry was conduct-
ed with a FINIGAN MAT 95 spectrometer with EI or ESI ionization
techniques. Melting points (Mps) are uncorrected and were measured
with a Bꢃchi melting point apparatus. Supplementary data associated
with this article (the experimental protocols for the synthesis of inter-
mediates 8 through 14, the 1H NMR, DEPT, 13C NMR, COSY, 1H–13C
HSQC and (in part) 1H–13C HMBC spectra of all compounds, as well as
a detailed distance analysis of 2 by NOESY and ROESY) can be consult-
ed in the Supporting Information. Systematic non-nucleoside nomencla-
ture is used below, including for heterocycles and carbocycles.
1
tide 36 showed all characteristic fragments in the expected
proportions (i.e., the cyclopropyl, methoxy, Ca- and aromat-
ic protons of the O-methyltyrosyl side chain, cytidine’s
anomeric proton, both adenine and cytosine heterocycles),
whereas the 31P NMR spectrum showed a single resonance,
and high-resolution mass spectrometry gave the expected
exact molecular mass.
Conclusions
We have accomplished the total syntheses of two potential
protein synthesis inhibitors—the locked North-type metha-
nocarba puromycin analogue 2 and its dinucleotide deriva-
tive 36—in their enantiopure forms through an 18- and a 19-
step pathway in 16 and 4.7% overall yields, respectively.
The following key steps were utilized, the last five of which
were elaborated during the synthesis along with a few non-
evident protection/deprotection protocols: ring-opening
Wittig reaction, Swern oxidation (stereodestructive), vinyl
Grignard addition (stereoselective), ring-closing metathesis,
chromo-oxidative rearrangement (stereodestructive), Luche
reduction (stereoselective and regioselective), Simmons–
Smith cyclopropanation (stereoselective), Mitsunobu cou-
pling (stereoselective), Mattocks bromoacetylation (stereo-
selective and regioselective), two nucleophilic substitutions
with acetate (regioselective) and azide (stereoselective), two
Staudinger–Vilarrasa (chemoselective) amino acid couplings,
and a nucleosidyl phosphoramidite coupling (chemoselec-
tive). Both analogues, together with several others,[13,31,34]
are being tested in an in vitro assay for ribosome-catalyzed
peptidyl transfer and the findings are being compared with
AHCTUNGTERG(NNUN 1S,4R,5S)-3-(tert-Butyldiphenyl)silyloxymethyl-4,5-dioxy-O,O-isopropy-
lidene-2-cyclopenten-1-ol (15): Sodium borohydride (220 mg, 5.82 mmol)
was added portionwise to a solution of the cyclopentenone 14 (1.585 g,
3.75 mmol) and ceriumACTHNUTRGNE(NUG III) chloride heptahydrate (1.198 g, 3.15 mmol) in
methanol (7 mL), with the temperature maintained between 08C and
58C. After 30 min, acetic acid was carefully added to adjust to pH 5.
Water (7 mL) was added and the reaction mixture was extracted with
ether. The organic layer was washed with brine and dried over anhydrous
magnesium sulfate. The solvents were removed under reduced pressure
and the residue was purified by silica gel column chromatography (cyclo-
hexane/EtOAc 9:1) to give the allylic alcohol 15 as a colorless oil (1.61 g,
99%). C25H32O4Si (424.60). Rf =0.25 (cyclohexane/EtOAc 7:1); 1H NMR
(300 MHz, CDCl3): d= 1.07 (s, 9H; SiCACHTNUTRGNENUG(CH3)3), 1.33 (s, 3H; CH3), 1.35
(s, 3H; CH3), 2.68 (d, 3J =9.9 Hz, 1H; OH), 4.28 (d, 2J =15.2 Hz, 1H;
CHHOSi), 4.38 (d, 2J =15.2 Hz, 1H; CHHOSi), 4.53–4.59 (m, 1H; H1),
4.75 (pst, 3J =4.8, 5.9 Hz, 1H; H5), 4.86 (d, 3J =5.9 Hz, 1H; H4), 5.84
3
(d, J =1.8 Hz, 1H; H2), 7.33–7.46 (m, 6H; m-HPh, p-HPh), 7.65–7.69 ppm
(m, 4H; o-HPh); 13C NMR (75 MHz, CDCl3): d=19.2 (SiC
ACHTUNGTRENNUNG
(CH3), 26.8 (SiCCAHTUNGTRENNUNG
82.8 (C4), 112.4 (CMe2), 127.7 (4C; m-CSiPh), 129.3 (C2), 129.7 (4C; p-
CSiPh), 133.3 (2C; i-CSiPh), 135.5 (4C; o-CSiPh), 145.2 ppm (C3); HRMS
(ESI+): m/z: calcd for C25H32O4SiNa: 447.1968 [M+Na]+; found:
447.1966.
Chem. Eur. J. 2009, 15, 6244 – 6257
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6251

P. Strazewski and B. Y. Michel
(1R,2R,3S,4S,5S)-1-(tert-Butyldiphenyl)silyloxymethyl-2,3-dioxy-O,O-
isopropylidenebicyclo[3.1.0]hexan-4-ol (16): Diethylzinc (3.61 mL, 1.0m
fied by column chromatography on silica gel with elution with EtOAc/
MeOH/H2O (8:1:0.5, 7:1:0.5, 6:1:0.5) to yield, after lyophilization, the to-
tally deprotected product 18 as a white solid (45 mg, 86%). C12H14N4O4
(278.26). Rf =0.33 (EtOAc/MeOH/H2O 6:1:0.5); 1H NMR (CD3OD,
300 MHz): d=0.76 (ddd, 2J=5.3 Hz, 3J=8.7 Hz, 4J=1.2 Hz, 1H; H6’A),
1.53 (dd, 2J=5.3 Hz, 3J=4.1 Hz, 1H; H6’B), 1.62 (ddd, 3J=4.1, 8.7 Hz,
4J=0.9 Hz, 1H; H5’), 3.31 (d, 2J=11.7 Hz, 1H; CHHOH), 3.90 (d, 3J=
6.6 Hz, 1H; H3’), 4.24 (d, 2J=11.7 Hz, 1H; CHHOH), 4.76 (dd, 3J=
6.6 Hz, 4J=1.2 Hz, 1H; H2’), 4.91 (s, 1H; H1’), 8.07 (s, 1H; H8),
8.51 ppm (s, 1H; H2); HRMS (ESI+): m/z: calcd for C12H15N4O4Na:
301.0913 [M+Na]+; found: 301.0910.
ACHTUNGTRENNUNG
in hexanes) was added dropwise under nitrogen at ꢀ188C to a stirred so-
lution of allylic alcohol 15 (1.394 g, 3.28 mmol) in dichloromethane
(18 mL). The reaction mixture was stirred at this temperature for 15 min.
A solution of CH2I2 (308 mL, 3.78 mmol) in dichloromethane (2.4 mL)
was added to the mixture and stirring was continued for 15 min at
ꢀ188C. These two diethylzinc (3.61 mL, 1.0m in hexanes) and CH2I2
(308 mL, 3.78 mmol) addition steps were repeated once again and the so-
lution was finally allowed to warm to room temperature and stirred over-
night. The reaction mixture was quenched with a saturated NH4Cl solu-
tion. The organic layer was extracted three times with CH2Cl2, dried over
MgSO4, and filtered. The solvents were removed in vacuo and the resi-
due was purified by column chromatography on silica gel with elution
with cyclohexane/EtOAc (6:1, 5:1, 4:1, 3:1) to afford the carbobicyclic
derivative 16 as a colorless oil (1.206 g, 84%). C26H34O4Si (438.63). Rf =
N6,N6,9-Tris-(tert-butoxycarbonyl)adenine
(20):
Boc2O
(2.00 g,
8.88 mmol) was added to a stirred suspension of adenine (19, 300 mg,
2.22 mmol) and DMAP (82 mg, 0.67 mmol) in THF (11 mL). After
20 min, the solution had become pale yellow. The reaction mixture was
stirred overnight at room temperature. The volatiles were then removed
under reduced pressure and the residue was purified by column chroma-
tography on silica gel with elution with cyclohexane/EtOAc (9:1, 4:1, 7:3)
to give the tris-protected compound 20 as a colorless oil (840 mg, 87%).
C20H29N5O6 (435.47). Rf =0.48 (cyclohexane/EtOAc 7:3); 1H NMR
(300 MHz, CDCl3): d=1.36 (s, 18H; N6Boc2), 1.65 (s, 9H; N9Boc), 8.46
(s, 1H; H8), 8.94 ppm (s, 1H; H2); 13C NMR (75 MHz, CDCl3): d=27.5,
1
0.27 (cyclohexane/EtOAc 5:1); H NMR (300 MHz, CDCl3): d=0.55 (dd,
3
2
2J=5.1 Hz, J=8.7 Hz, 1H; H6A), 1.07 (s, 9H; SiC
G
3
5.1 Hz, J = 4.8 Hz, 1H; H6B), 1.32 (s, 3H; CH3), 1.55 (s, 3H; CH3), 1.62
(quin. “ddd”, 3J =4.8, 8.7 Hz, 1H; H5), 2.42 (brs, 1H; OH), 3.30 (d, 2J
2
=10.8 Hz, 1H; CHHOSi), 4.09 (d, J =10.8 Hz, 1H; CHHOSi), 4.47 (pst,
3J=4.8, 6.6 Hz, 1H; H4), 4.55 (pst, J=6.6, 6.9 Hz, 1H; H3), 5.02 (d, J=
3
6.9 Hz, 1H; H2), 7.35–7.46 (m, 6H; m-HPh, p-HPh), 7.62–7.67 ppm (m,
27.7 (N(C(O)OCACTHNUTRGENN(UG CH3)3)), 83.7, 87.3 (N(C(O)OCCAHTUNTGERN(NUGN CH3)3)), 129.4 (C5),
143.0 (C8), 145.4 (N(C(O)OtBu), 151.0 (C4), 152.3 (C6), 149.8
(N(C(O)OtBu)2), 153.9 ppm (C2).
4H; o-HPh); 13C NMR (75 MHz, CDCl3): d=10.3 (C6), 19.2 (SiC
ACHTUNGTRENNUNG
24.6, 26.1 (2C; CH3), 26.8 (3C; SiCACTHUNGTRENNNUG
(CH2OSi), 71.0 (C4), 79.6 (C3), 81.1 (C2), 112.7 (CMe2), 127.6 (4C; m-
CSiPh), 129.7 (2C; p-CSiPh), 133.5, 133.6 (2C; i-CSiPh), 135.5 ppm (4C; o-
CSiPh); HRMS (ESI+): m/z: calcd for C26H34O4SiNa: 461.2124 [M+Na]+;
found: 461.2121.
N6,N6-Bis-(tert-butoxycarbonyl)adenine (21):
A saturated solution of
NaHCO3 (35 mL) was added to a stirred solution of tris-Boc-adenine (20,
3.080 g, 7.07 mmol) in MeOH (70 mL) and the reaction mixture became
cloudy. The solution was then warmed to 508C for 1 h 15 min. When the
conversion was quantitative according to TLC, MeOH was removed
under reduced pressure. Water (70 mL) was added and the aqueous layer
was extracted three times with CHCl3. The organic layer was dried over
anhydrous magnesium sulfate and filtered. The solvents were removed
under reduced pressure and the residue was purified by column chroma-
tography on silica gel with elution with cyclohexane/EtOAc (3:7, 1:4, 1:9,
0:1) to give the bis-protected compound 21 (2.21 g, 93%) as a white
solid. C15H21N5O4 (335.36). Rf =0.36 (EtOAc); m.p. 149–1508C;[24]
1H NMR (300 MHz, CDCl3): d= 1.36 (s, 18H; N6Boc2), 8.63 (s, 1H; H8),
8.81 (s, 1H; H2), 13.68 ppm (brs, 1H; NH); 13C NMR (75 MHz, CDCl3):
(1’R,2’R,3’S,4’R,5’S)-9-[1’-(tert-Butyldiphenyl)silyloxymethyl-2’,3’-dioxy-
O,O-isopropylidenebicycloACTHNUGTRNEUNG[3.1.0]hex-4’-yl]-6-chloropurine (17): Diiso-
propyl azodicarboxylate (1.09 mL, 5.27 mmol) was added dropwise at
08C to a stirred solution of triphenylphosphine (1.405 g, 5.27 mmol) in
THF (13 mL) and the yellow reaction mixture was stirred at this temper-
ature for 30 min. A solution of the carbobicyclic alcohol 16 (1.006 g,
2.29 mmol) in THF (13 mL), previously co-evaporated with toluene (3ꢅ
5 mL), was added and the reaction mixture was stirred at 08C for 10 min.
The cooling bath was then removed and the yellow solution was stirred
for 30 min at room temperature. 6-Chloropurine (822 mg, 5.27 mmol)
was added and the reaction mixture was stirred overnight at room tem-
perature. The solution was filtered and the volatiles were concentrated in
vacuo. The residue was purified by column chromatography on silica gel
with elution with cyclohexane/EtOAc (5:1, 4:1, 3:1, 1:1). The coupling
product was still contaminated with a large amount of N,N’-diisopropyl
hydrazine. To eliminate it, the compound was dissolved in a minimum
volume of EtOAc and was then added dropwise to a large volume of pe-
troleum ether. The solution was put in the freezer at ꢀ228C for 2 h and
then filtered, and eventually the volatiles were removed under reduced
pressure to yield 17 as a colorless oil (1.081 g, 82%). C31H35ClN4O3Si
(575.17). Rf =0.35 (cyclohexane/EtOAc 3:1); 1H NMR (300 MHz,
d=27.2 (N(C(O)OC
149.9 (N(C(O)OtBu)2), 151.4 ppm (C2).
(1’R,2’R,3’S,4’R,5’S)-9-[1’-(tert-Butyldiphenyl)silyloxymethyl-2’,3’-dioxy-
O,O-isopropylidenebicyclo
[3.1.0]hex-4’-yl]-N6,N6-bis-(tert-butoxycarbo-
nyl)adenine (22): Diisopropyl azodicarboxylate (1.81 mL, 8.77 mmol) was
added dropwise at 08C to stirred solution of triphenylphosphine
ACHTUNTGNERN(GU CH3)3)2), 83.1 (N(C(O)OCACHUTNGTNER(NUGN CH3)3)2), 145.6 (C8),
AHCTUNGTRENNUNG
a
(2.336 g, 8.77 mmol) in THF (20 mL) and the yellow reaction mixture
was stirred at this temperature for 30 min. After that, a solution of the
carbobicyclic alcohol 16 (1.673 g, 3.81 mmol) in THF (21 mL), previously
co-evaporated with toluene (3ꢅ5 mL), was added and the reaction mix-
ture was stirred at 08C for 10 min. The cooling bath was removed and
the yellow solution was stirred for 30 min at room temperature. Com-
pound 21 (3.163 g, 8.77 mmol) was added and the solution had become
clear after 2 min. The reaction mixture was stirred overnight at room
temperature. The volatiles were removed under reduced pressure and the
residue was purified by column chromatography on silica gel with elution
with cyclohexane/EtOAc (6:1, 4:1, 3:1). The coupling product was still
contaminated with a large amount of the diisopropyl hydrazine. To elimi-
nate it, the compound was dissolved in a minimum volume of EtOAc,
and was then added dropwise to a large volume of petroleum ether. The
solution was put in the freezer at ꢀ228C for 2 h and then filtered, and
eventually the volatiles were concentrated in vacuo to yield 22 as a color-
less oil (2.736 g, 95%). C41H53N5O7Si (755.97). Rf =0.36 (cyclohexane/
EtOAc 5:1); 1H NMR (300 MHz, CDCl3): d=0.98 (dd, 2J=6.0 Hz, 3J=
2
3
4
CDCl3): d=1.02 (ddd, J=5.7 Hz, J=9.3 Hz, J=1.1 Hz, 1H; H6’A), 1.08
(s, 9H; SiCACHTUNGTRENNUNG
(CH3)3), 1.16 (dd, 2J=5.7 Hz, 3J=4.2 Hz, 1H; H6’B), 1.23 (s,
3H; CH3), 1.54 (s, 3H; CH3), 1.65 (ddd, 3J=4.2, 9.3 Hz, 4J=1.4 Hz, 1H;
H5’), 3.63 (d, 2J=10.8 Hz, 1H; CHHOSi), 4.25 (d, 2J=10.8 Hz, 1H;
CHHOSi), 4.60 (dd, 3J=7.2 Hz, 4J=1.4 Hz, 1H; H3’), 5.12 (s, 1H; H4’),
5.28 (dd, 3J=6.9 Hz, 4J=1.1 Hz, 1H; H2’), 7.29–7.44 (m, 6H; m-HPh, p-
3
HPh), 7.58–7.63 (m, J=7.5 Hz, 4H; o-HPh), 8.43 (s, 1H; H8), 8.66 ppm (s,
1H; H2); 13C NMR (75 MHz, CDCl3): d=12.3 (C6’), 19.2 (SiC
ACHTUNGTRENNUNG
24.2, 25.9 (2C; CH3), 26.9 (3C; SiCACTHUNGTRENNNUG
(C4’), 64.4 (CH2OSi), 81.1 (C2’), 88.8 (C3’), 112.5 (CMe2), 127.7, 127.7
(4C; m-CSiPh), 129.8 (2C; p-CSiPh), 131.8 (C5), 132.8, 132.9 (2C; i-CSiPh),
135.5, 135.5 (4C; o-CSiPh), 143.7 (C8), 151.0 (C4), 151.1 (C6), 152.0 ppm
(C2); HRMS (ESI+): m/z: calcd for C31H36ClN4O3Si: 575.2245 [M+H]+;
found: 575.2249.
8.7 Hz, 1H; H6’A), 1.09 (s, 9H; SiCACHTUNTGRENUNG(CH3)3), 1.16–1.20 (m, 1H; H6’B), 1.24
(1’R,2’R,3’S,4’R,5’S)-9-(2’,3’-Dihydroxy-1’-hydroxymethylbicyclo-
(s, 3H; CH3), 1.48 (s, 18H; NBoc2), 1.54 (s, 3H; CH3), 1.65–1.72 (m, 1H;
H5’), 3.57 (d, 2J=10.8 Hz, 1H; CHHOSi), 4.27 (d, 2J=10.8 Hz, 1H;
CHHOSi), 4.58 (d, 3J=6.9 Hz, 1H; H3’), 5.16 (s, 1H; H4’), 5.29 (d, 3J=
6.9 Hz, 1H; H2’), 7.32–7.44 (m, 6H; m-HPh, p-HPh), 7.62 (d, 3J=7.5 Hz,
ACHTUNGTRENNUNG[3.1.0]hexan-4’-yl)hypoxanthine (18): Trifluoroacetic acid (330 mL) was
added to a stirred solution of 17 (107 mg, 0.186 mmol) in ClCH2CH2Cl
(1 mL), and the reaction mixture was kept at 408C overnight. The solu-
tion was then co-evaporated twice with toluene and the residue was puri-
6252
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 6244 – 6257

Puromycin Analogues
FULL PAPER
4H; o-HPh), 8.46 (s, 1H; H8), 8.81 ppm (s, 1H; H2); 13C NMR (75 MHz,
(C2), 152.1 ppm (C6); HRMS (ESI+): m/z: calcd for C25H36N5O7:
518.2615 [M+H]+; found: 518.2611.
CDCl3): d=12.5 (C6’), 19.1 (SiC
SiC(CH3)3), 27.8 (6C; N(C(O)OC
(C4’), 64.7 (CH2OSi), 80.9 (C2’), 83.7 (2C; N(C(O)OCAHCTUNGTRENNUNG
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
(1’R,2’R,3’S,4’R,5’S)-9-(1’-Benzyloxymethyl-2’,3’-dioxy-O,O-isopropyli-
dene-bicycloACTHNUTRGNEUNG
[3.1.0]hex-4’-yl)-N6,N6-bis-(tert-butoxycarbonyl)adenine (26):
(C3’), 112.4 (CMe2), 127.8 (4C; m-CSiPh), 128.8 (C5), 129.8 (2C; p-CSiPh),
132.7, 132.9 (2C; i-CSiPh), 135.4, 135.5 (4C; o-CSiPh), 143.1 (C8), 150.3
(C4), 150.5 (2C; N(C(O)OtBu)2), 152.1 (C2), 152.5 ppm (C6); HRMS
(ESI+): m/z: calcd for C41H54N5O7Si: 756.3793 [M+H]+; found: 756.3792.
NaH (14 mg, 0.57 mmol) was cautiously added to a stirred solution of al-
cohol 25 (259 mg, 0.514 mmol) in THF (2 mL). The reaction mixture was
stirred for 20 min at room temperature, and next nBu4I (20 mg,
0.051 mmol) and BnBr (69 mL, 0.57 mmol) were added. After 2 h, the re-
action mixture was quenched with a saturated NH4Cl solution. The or-
ganic layer was extracted three times with EtOAc, dried over MgSO4,
and filtered. The volatiles were removed in vacuo and the residue was
purified by column chromatography on silica gel with elution with cyclo-
hexane/EtOAc (3:1, 2:1, 1:1, 1:2) to afford the benzylated alcohol 26 as a
colorless oil (219 mg, 70%). C32H41N5O7 (607.70). Rf =0.50 (cyclohexane/
EtOAc 1:1); 1H NMR (300 MHz, CDCl3): d=1.06 (ddd, 2J=5.9 Hz, 3J=
9.3 Hz, 4J=1.2 Hz, 1H; H6’A), 1.30 (dd, 2J=5.9 Hz, 3J=5.1 Hz, 1H;
H6’B), 1.42 (s, 9H; tBu), 1.49 (s, 9H; tBu), 1.55 (s, 3H; CH3), 1.59 (s, 3H;
CH3), 1.77 (ddd, 3J=5.1, 9.3 Hz, 4J=1.2 Hz, 1H; H5’), 4.14 (d, 2J=
11.7 Hz, 1H; CHHOBn), 4.46 (d, 2J=11.7 Hz, 1H; CHHOBn), 4.61 (d,
3J=7.4 Hz, 4J=1.2 Hz, 1H; H3’), 5.09 (s, 1H; H4’), 5.28 (s, 2H; CH2Ph),
5.32 (dd, 3J=7.2 Hz, 4J=1.2 Hz, 1H; H2’), 7.16–7.29 (m, 3H; p-HPh, m-
HPh), 7.35–7.42 (m, 2H; o-HPh), 8.20 (s, 1H; H8), 8.76 ppm (s, 1H; H2);
C32H42N5O7: 608.3084 [M+H]+; found: 608.3080.
(1’R,2’R,3’S,4’R,5’S)-9-(2’,3’-Dihydroxy-1’-hydroxymethylbicyclo-
ACHTUNGTRENNUNG[3.1.0]hexan-4’-yl)adenine (23): Compound 22 (1.342 g, 1.78 mmol) was
dissolved in a solution of MeOH/TFA (3.7:2.2 mL) and the reaction mix-
ture was warmed at 658C for 1 d. The solution was then co-evaporated
twice with ClCH2CH2Cl and the residue was purified by column chroma-
tography on silica gel with elution with EtOAc/MeOH/H2O (8:1:0.5,
7:1:0.5, 6:1:0.5) to yield, after lyophilization, the totally deprotected prod-
uct 23 as a white solid (480 mg, 97%). C12H15N5O3 (277.28). Rf =0.32
(EtOAc/MeOH/H2O 5:1:0.5); m.p. 129–1308C; 1H NMR ([D6]DMSO,
300 MHz): d=0.65 (dd, 2J=4.5 Hz, 3J=8.7 Hz, 1H; H6’A), 1.30 (pst
“dd”, 2J=4.5 Hz, 3J=3.9 Hz, 1H; H6’B), 1.49 (dd, 3J=3.9, 8.7 Hz, 1H;
H5’), 3.13 (d, 2J=11.7 Hz, 1H; CHHOH), 3.69 (d, 3J=6.3 Hz, 1H; H3’),
4.03 (d, 2J=11.7 Hz, 1H; CHHOH), 4.53 (d, 3J=6.3 Hz, 1H; H2’), 4.70
(s, 1H; H1’), 8.13 (s, 1H; H2), 8.38 ppm (s, 1H; H8); 13C NMR
([D6]DMSO, 75 MHz): d=12.2 (C6’), 23.6 (C5’), 36.9 (C1’), 62.0 (C4’),
63.1 (CH2OH), 71.0 (C2’), 76.9 (C3’), 119.2 (C5), 140.2 (C8), 149.4 (C4),
153.2 (C2), 156.3 ppm (C6); HRMS (ESI+): m/z: calcd for C12H16N5O3:
278.1253 [M+H]+; found: 278.1251.
(1’R,2’R,3’S,4’R,5’S)-9-(1’-Benzyloxymethyl-2’,3’-dihydroxybicyclo-
AHCTUNGERTG[NNUN 3.1.0]hex-4’-yl)adenine (27): Compound 26 (185 mg, 0.304 mmol) was
dissolved in MeOH/TFA (1.2:1.0 mL v/v) and the reaction mixture was
warmed to 558C for 10 h. The solution was co-evaporated twice with
ClCH2CH2Cl and the resulting residue was purified by column chroma-
tography on silica gel with elution with EtOAc/MeOH/H2O (1:1:0, 8:1:1,
7:1:0.5, 6:1:0.5) to yield 27 as a opaque, colorless resin (107 mg, 96%).
C19H21N5O3 (367.40). Rf =0.42 (EtOAc/MeOH/H2O 8:1:1); 0.82 (ddd,
(1’R,2’S,3’S,4’R,5’S)-9-(3’-Acetoxy-1’-bromomethyl-2’-bromobicyclo-
ACHTUNGTRENNUNG[3.1.0]hexan-4’-yl)adenine (24): H2O (100 mL) was added to a stirred sus-
pension of 23 (100 mg, 0.36 mmol) in acetonitrile (8 mL), followed by a-
acetoxyisobutyryl bromide (375 mL, 2.52 mmol), and the reaction mixture
was stirred at room temperature for 45 min. More H2O (200 mL) was
added and the solution was stirred for another 15 min. The reaction mix-
ture was quenched with a saturated NaHCO3 solution. The organic layer
was extracted three times with EtOAc, dried over MgSO4, and filtered.
The volatiles were removed in vacuo and the residue was purified by
column chromatography on silica gel with elution with EtOAc/MeOH/
H2O (10:1:0.5, 8:1:0.5, 6:1:0.5) to afford compound 24 as a white solid
(141 mg, 88%). C14H15Br2N5O2 (445.11). Rf =0.51 (EtOAc/MeOH/H2O
10:1:0.5); 1H NMR ([D6]DMSO, 300 MHz): d=1.33–1.36 (m, 1H; H6’A),
1.45 (dd, 2J=5.9 Hz, 3J=9.0 Hz, 1H; H6’B), 2.13 (s, 3H; OC(O)CH3),
2.67 (dd, 3J=3.9, 9.0 Hz, 1H; H5’), 3.65 (d, 2J=10.5 Hz, 1H; CHHBr),
4.36 (d, 2J=10.5 Hz, 1H; CHHBr), 4.40 (s, 1H; H2’), 4.91 (s, 1H; H4’),
5.38 (s, 1H; H3’), 7.29 (s, 2H; NH2), 8.14 (s, 1H; H2), 8.34 ppm (s, 1H;
H8); 13C NMR ([D6]DMSO, 75 MHz): d=17.2 (C6’), 20.6 (OC(O)CH3),
31.6 (C5’), 36.3 (C1’), 40.0 (CH2Br), 55.7 (C2’), 60.2 (C4’), 84.3 (C3’),
118.8 (C5), 139.2 (C8), 148.9 (C4), 151.9 (C2), 155.4 (C6), 168.7 ppm
(OC(O)CH3); HRMS (ESI+): m/z: calcd for C14H16Br2N5O2: 443.9671
[M+H]+; found: 443.9668.
3
4
2
2J=5.4 Hz, J=8.4 Hz, J=0.9 Hz, 1H; H6’A), 1.55 (pst “dd”, J=5.4 Hz,
3J=3.6 Hz, 1H; H6’B), 1.66 (dd, 3J=3.6, 8.4 Hz, 1H; H5’), 3.35 (d, 2J=
11.7 Hz, 1H; CHHOBn), 4.03 (d, 3J=6.9 Hz, 1H; H3’), 4.22 (d, 2J=
3
11.7 Hz, 1H; CHHOBn), 4.87 (d, J=6.9 Hz, 1H; H2’), 4.91 (s, 1H; H4’),
5.03 (s, 2H; CH2Ph), 7.00–7.33 (m, 3H; p-HPh, m-HPh), 7.34–7.42 (m, 2H;
o-HPh), 7.86 (s, 1H; H8), 8.32 ppm (s, 1H; H2); C19H22N5O3: 368.1723
[M+H]+; found: 368.1721.
(1’R,2’S,3’S,4’R,5’S)-9-(3’-Acetoxy-1’-bromomethyl-2’-bromobicyclo-
AHCTUNGTRENNUNG
[3.1.0]hex-4’-yl)-N6-benzyladenine (28): H2O (40 mL) was added to a
stirred suspension of 27 (54 mg, 0.147 mmol) in acetonitrile (1.5 mL), fol-
lowed by a-acetoxyisobutyryl bromide (153 mL, 1.029 mmol), and the re-
action mixture was stirred at room temperature for 45 min. More H2O
(130 mL) was added and the solution was stirred for another 15 min. The
reaction was quenched with a saturated NaHCO3 solution. The organic
layer was extracted with three times with EtOAc, dried over anhydrous
magnesium sulfate, and filtered. The volatiles were removed in vacuo
and the residue was purified by column chromatography on silica gel
with elution with cyclohexane/EtOAc (10:1:0.5, 8:1:0.5, 6:1:0.5) to afford
(1’R,2’R,3’S,4’R,5’S)-9-(1’-Hydroxymethyl-2’,3’-dioxy-O,O-isopropylidene-
the dibrominated compound 29 as
a whitish oil (49 mg, 62%).
C21H21Br2N5O2 (535.23). Rf =0.50 (EtOAc/cyclohexane 9:1); 1H NMR
bicyclo
N
(25):
2
3
TBAF in THF (1m, 1.58 mL, 1.58 mmol) was added dropwise to a stirred
solution of silyl ether 24 (1.046 mg, 1.37 mmol) in THF (15 mL). Initially
slightly yellow, the reaction mixture darkened to deep orange and was
stirred at ambient temperature for 30 min. The volatiles were removed in
vacuo and the residue was purified by column chromatography on silica
gel with elution with cyclohexane/EtOAc (3:1, 2:1, 1:1, 0:1) to afford al-
cohol 25 as a white foam (709 mg, 99%). C25H35N5O7 (517.57). Rf =0.63
(EtOAc); 1H NMR (300 MHz, CDCl3): d=0.95 (dd, 2J=6.3 Hz, 3J=
8.7 Hz, 1H; H6’A), 1.13–1.18 (m, 1H; H6’B), 1.18 (s, 3H; CH3), 1.41 (s,
(CDCl3, 500 MHz): d=1.35 (dd, J=6.5 Hz, J=8.5 Hz, 1H; H6’A), 1.39–
1.44 (m, 1H; H6’B), 2.15 (s, 3H; OC(O)CH3), 2.21 (dd, 3J=3.5, 8.5 Hz,
1H; H5’), 3.26 (d, 2J=10.5 Hz, 1H; CHHBr), 4.34 (d, 2J=10.5 Hz, 1H;
CHHBr), 4.37 (s, 1H; H2’), 4.88 (brs, 2H; CH2Ph), 5.08 (s, 1H; H4’),
3
5.54 (s, 1H; H3’), 6.60 (brs, 1H; NHBn), 7.27 (d, J=7.0 Hz, 1H; p-HPh),
3
3
7.33 (t “dd”, J=7.0, 7.5 Hz, 2H; m-HPh), 7.38 (d, J=7.5 Hz, 2H; o-HPh),
8.13 (s, 1H; H8), 8.40 ppm (s, 1H; H2); 13C NMR (CDCl3, 125 MHz): d=
17.9 (C6’), 20.8 (OC(O)CH3), 33.1 (C5’), 37.4 (C1’), 37.7 (CH2Br), ~44
(brs NCH2Ph, from HSQC and HMBC only), 54.9 (C2’), 60.6 (C4’), 86.0
(C3’), 119.5 (C5), 127.4 (2C; p-CPh), 127.6 (2C; o-CPh), 128.6 (2C; m-
CPh), 138.4 (C8), 138.4 (i-CPh), 153.3 (2C; C6, C4), 154.7 (C2), 168.6 ppm
(OC(O)CH3); MS (ESI+) m/z 538/536/534 [M+H]+, 456/454 [MꢀBr]+,
414/412 [MꢀBrꢀCH2CO]+, 396/394 [MꢀAcOH]+, 374 [Mꢀ2Br]+, 226.1
[(PhCH2)adenine+2H]+; HRMS (ESI+): m/z: calcd for C21H22Br2N5O2:
534.0140 [M+H]+; found: 534.0139.
3
18H; NBoc2), 1.48 (s, 3H; CH3), 1.71 (dd, J=3.9, 8.7 Hz, 1H; H5’), 3.31
2
2
(d, J=11.6 Hz, 1H; CHHOH), 4.19 (d, J=11.6 Hz, 1H; CHHOH), 4.29
(brs, 1H; OH), 4.56 (d, 3J=7.4 Hz, 1H; H3’), 4.93 (s, 1H; H4’), 5.47 (d,
3J=7.4 Hz, 1H; H2’), 8.24 (s, 1H; H8), 8.78 ppm (s, 1H; H2); 13C NMR
(75 MHz, CDCl3): d=13.9 (C6’), 23.9, 25.7 (2C; CH3), 27.6 (6C;
N(C(O)OC
80.6 (C2’), 83.8 (2C; N(C(O)OC
129.4 (C5), 144.5 (C8), 150.3 (2C; N(C(O)OtBu)2), 150.7 (C4), 151.5
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
(1’R,2’S,3’S,4’R,5’S)-9-(3’-Acetoxy-1’-acetoxymethyl-2’-bromobicyclo-
AHCTUNGERTG[NNUN 3.1.0]hexan-4’-yl)adenine (29): Compound 24 (100 mg, 0.224 mmol) was
Chem. Eur. J. 2009, 15, 6244 – 6257
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6253

P. Strazewski and B. Y. Michel
dissolved in acetonitrile at 758C. After the mixture had cooled to room
temperature, CsOAc (272 mg, 1.35 mmol), molecular sieves (3 ꢄ,
250 mg), and 18-crown-6 (120 mg, 0.450 mmol) were added, and the reac-
tion mixture was again warmed to 758C. After 1 h, the precipitation of a
white solid had occurred. The reaction mixture was still stirred for 2 h at
the same temperature, and was then filtered and rinsed with ethyl ace-
tate. The volatiles were removed under reduced pressure and the residue
was dissolved in EtOAc and water. The organic layer was extracted three
times with EtOAc, dried over MgSO4, and filtered, and the volatiles were
concentrated in vacuo. The residue was purified by column chromatogra-
phy on silica gel with elution with EtOAc/MeOH/H2O (10:1:0.5, 8:1:0.5,
6:1:0.5) to provide 29 as a white solid (81 mg, 85%). C16H18BrN5O4
di-n-butyl formamide) to give a clear colorless or pale yellow oil that can
be safely stored under nitrogen in the cold. C11H25NO2 (203.1885).
1H NMR (300 MHz, CDCl3): d= 0.89 (t, 3J=7.2 Hz, 6H;
N(CH2CH2CH2CH3)2), 1.30 (q, 3J=7.2 Hz, 4H; N(CH2CH2CH2CH3)2),
1.40 (q, 3J=7.2 Hz, 4H; N(CH2CH2CH2CH3)2), 2.59 (t, 3J=7.2 Hz, 4H;
N(CH2CH2CH2CH3)2), 3.30 (s, 6H; (OCH3)2), 4.51 ppm (s, 1H; NCH-
A
13C NMR
(75 MHz,
CDCl3):
d=13.9
(2C;
N(CH2CH2CH2CH3)2), 20.4 (2C; N(CH2CH2CH2CH3)2), 30.9 (2C;
N(CH2CH2CH2CH3)2), 47.0 (2C; N(CH2CH2CH2CH3)2), 53.6 (2C;
(OCH3)2), 112.6 ppm (NCH
(1’S,2’R,3’S,4’R,5’S)-9-(2’-Azido-3’-hydroxy-5’-hydroxymethylbicyclo-
[3.1.0]hex-4’-yl)-N6-[di-(n-butyl)aminomethinyl]adenine (32): N,N-Di-n-
ACHTUNGTRNE(NUNG OCH3)2).
AHCTUNGTRENNUNG
1
(424.25). Rf =0.46 (EtOAc/MeOH/H2O 12:1:0.5). H NMR (CD3OD with
butylformamide dimethylacetal (31 mg, 0.152 mmol) was added dropwise
at ambient temperature to a stirred solution of the azide 31 (54 mg,
0.189 mmol) in MeOH (1 mL). The reaction mixture was stirred for 1 h
and heated with a heat gun for a few seconds every 15 min. The volatiles
were concentrated in vacuo and the residue was purified by column chro-
matography on silica gel with elution with EtOAc/toluene/MeOH (2:1:0,
4:1:0, 6:1:0, 8:1:0, 8:1:0.5) to afford 32 as an opaque oil (30 mg, 90%).
C21H31N9O2 (441.53). Rf =0.60 (EtOAc/Tol/MeOH 8:1:1); 1H NMR
(CD3OD, 300 MHz): d=0.95–1.01 (m, 1H; H6’A), 0.975/0.983 (2ꢅt, 3J=
a few drops of CDCl3, 300 MHz): d=0.98–1.03 (m, 2H; H6’A, H6’B), 1.71
(s, 3H; 1’-OC(O)CH3), 1.75 (s, 3H; 3’-OC(O)CH3), 1.91 (dd, 3J=4.5,
2
8.1 Hz, 1H; H5’), 3.88 (d, J=11.7 Hz, 1H; CHHOAc), 3.96 (s, 1H; H2’),
2
4.35 (d, J=11.7 Hz, 1H; CHHOAc), 4.57 (s, 1H; H4’), 5.08 (s, 1H; H3’),
7.81 (s, 1H; H2), 7.99 ppm (s, 1H; H8); 13C NMR (CD3OD with a few
drops of CDCl3, 75 MHz): d=16.3 (C6’), 20.9, 21.0 (2C; OC(O)CH3),
27.9 (C5’), 34.8 (C1’), 55.0 (C2’), 60.8 (C4’), 67.8 (CH2OAc), 86.9 (C3’),
119.7 (C5), 140.2 (C8), 149.7 (C4), 153.4 (C2), 156.6 (C6), 169.9 (3’-
OC(O)CH3), 172.1 ppm (1’-CH2OC(O)CH3); HRMS (ESI+): m/z: calcd
for C16H19BrN5O4: 424.0620 [M+H]+; found: 424.0618.
7.2 Hz,
2ꢅ3H;
N(CH2CH2CH2CH3)2),
1.33–1.48
(m,
5H;
N(CH2CH2CH2CH3)2, H5’), 1.63–1.75 (m, 4H; N(CH2CH2CH2CH3)2),
1.77 (dd, 2J=4.2 Hz, 3J=9.0 Hz, 1H; H6’B), 3.41 (d, 2J=12.0 Hz, 1H;
CHHOH), 3.48 (t, 3J=7.2 Hz, 2H; N(CH2CH2CH2CH3)2), 3.69 (t, 3J=
(1’R,2’S,3’S,4’R,5’S)-9-(2’-Bromo-3’-hydroxy-1’-hydroxymethylbicyclo-
ACHTUNGTRENNUNG[3.1.0]hex-4’-yl)adenine (30): Aqueous HCl (37%, 1 mL) was added
3
7.7 Hz, 2H; N(CH2CH2CH2CH3)2), 4.13 (d, J=6.5 Hz, 1H; H3’), 4.21 (d,
dropwise to a stirred solution of 29 (102 mg, 0.240 mmol) in MeOH
(3 mL) and the reaction mixture was warmed to 508C for 8 h. The reac-
tion mixture was co-evaporated three times with toluene and the residue
was purified by column chromatography on silica gel with elution with
EtOAc/MeOH/H2O (12:1:0.5, 10:1:0.5, 8:1:0.5, 5:1:0.5) to yield 30 as a
white solid (75 mg, 92%). C12H14N5O2 (340.18). Rf =0.53 (EtOAc/MeOH/
3J=12.0 Hz, 1H; CHHOH), 4.28 (d, 3J=6.5 Hz, 1H; H2’), 4.94 (s, 1H;
H4’), 8.43 (s, 1H; H2), 8.61 (s, 1H; H8), 8.96 ppm (s, 1H; N=CHNBu2);
13C NMR (CD3OD, 75 MHz): d=13.1 (C6’), 14.1, 14.3 (2C;
N(CH2CH2CH2CH3)2), 20.8, 21.2 (2C; N(CH2CH2CH2CH3)2), 23.9 (C5’),
30.4, 32.1 (2C; N(CH2CH2CH2CH3)2), 36.3 (C1’), 46.5, 53.1 (2C;
N(CH2CH2CH2CH3)2), 63.0 (C2’), 63.9 (C4’), 64.5 (CH2OH), 79.2 (C3’),
126.7 (C5), 142.4 (C8), 152.1 (C4), 153.3 (C2), 160.2 (N=CHNBu2),
161.5 ppm (C6); IR (CH2Cl2): n˜ =2117 cmꢀ1 (N3 st); HRMS (ESI+): m/z:
calcd for C21H32N9O2: 442.267 [M+H]+; found: 442.2674.
1
H2O 5:1:0.5); H NMR (CD3OD, 300 MHz): d=1.23–1.41 (m, 1H; H6’A),
1.50–1.64 (m, 1H; H6’B), 2.17–2.36 (m, 1H; H5’), 3.65 (d, 2J=11.4 Hz,
1H; CHHOH), 4.35 (d, 2J=11.4 Hz, 1H; CHHOH), 4.35 (s, 1H; H2’),
4.64 (s, 1H; H3’), 4.88 (s, 1H; H4’), 8.47 (s, 1H; H2), 8.62 ppm (s, 1H;
H8); 13C NMR (CD3OD, 75 MHz): d=16.1 (C6’), 27.0 (C5’), 38.1 (C1’),
60.0 (C2’), 64.4 (C4’), 65.7 (CH2OH), 87.7 (C3’), 120.0 (C5), 144.4 (C8),
145.1 (C2), 149.9 (C4), 151.6 ppm (C6); HRMS (ESI+): m/z: calcd for
C12H15N5O2: 340.0409 [M+H]+; found: 340.0408.
(1’S,2’R,3’R,4’R,5’S)-9-{2’-[N-(Fluoren-9-yl)methoxycarbonyl-O-methyl-l-
tyrosylamino]-3’-hydroxy-5’-hydroxymethylbicycloACTHNUTRGNEUNG
[3.1.0]hex-4’-yl}-N6-[di-
(n-butyl)aminomethinyl]adenine (33): A mixture of N-Fmoc-O-Me-l-Tyr
(36 mg, 0.088 mmol) and HOBT (14 mg, 0.088 mmol) was co-evaporated
three times with anhydrous THF (2 mL). The residue was dissolved in
THF (1 mL) and the solution was cooled to 08C under nitrogen for
10 min. Diisopropyl carbodiimide (13 mL, 0.814 mmol) was added and the
reaction mixture was stirred for 15 min at the same temperature. Mean-
while, Me3P (1m in THF, 146 mL, 0.146 mmol) was added to a solution of
azide 32 (30 mg, 0.068 mmol) in THF (1 mL), and the mixture was stirred
for 5 min at room temperature. The amino acid solution was allowed to
warm to room temperature over 5 min and was then added to the imino-
phosphorane solution. The reaction mixture was stirred at room tempera-
ture overnight. The volatiles were evaporated under reduced pressure
and co-evaporated from CHCl3 (5 mL), dissolved in EtOAc (30 mL), and
extracted with saturated NaHCO3 (15 mL). The organic layer was ex-
tracted twice with EtOAc and washed with H2O (2ꢅ10 mL), dried
(MgSO4), filtered, and concentrated in vacuo. The residue was purified
by column chromatography on silica gel with elution with EtOAc/tolu-
ene/MeOH (2:1:0, 4:1:0, 6:1:0, 8:1:0, 8:1:0.5, 8:1:0.75, 8:1:1) to give 33 as
a colorless oil (44 mg, 81%). C46H54N8O6 (814.97). Rf =0.38 (EtOAc/Tol/
MeOH 8:1:0.75); 1H NMR (CD3OD, 300 MHz): d=0.66 (pst “dd”, 2J=
5.7 Hz, 3J=9.6 Hz, 1H; H6’A), 0.955/0.960 (2ꢅt, 3J=7.2 Hz, 2ꢅ3H;
N(CH2CH2CH2CH3)2), 1.32–1.46 (m, 5H; N(CH2CH2CH2CH3)2, H6’B),
1.59–1.73 (m, 5H; N(CH2CH2CH2CH3)2, H5’), 2.85 (dd, 2J=13.5 Hz, 3J=
7.8 Hz, 1H; p-MeOPhCHH), 3.03 (dd, 2J=13.5 Hz, 3J=6.8 Hz, 1H; p-
MeOPhCHH), 3.15 (d, 2J=11.7 Hz, 1H; CHHOH), 3.43 (t, 3J=7.1 Hz,
2H; N(CH2CH2CH2CH3)2), 3.61–3.71 (m, 2H; N(CH2CH2CH2CH3)2),
3.69 (s, 3H; OCH3), 3.89–3.98 (m, 1H; H3’), 3.94 (d, 2J=11.7 Hz, 1H;
CHHOH), 4.06 (pst, 3J=6.6, 6.9 Hz, 1H; aliphatic fluorene), 4.20 (pst
(1’R,2’R,3’S,4’R,5’S)-9-(2’-Azido-3’-hydroxy-1’-hydroxymethylbicyclo-
ACHTUNGTRENNUNG[3.1.0]hex-4’-yl)adenine (31): NaN3 (29 mg, 0.440 mmol) in H2O
(200 mL) was added to a stirred solution of 30 (30 mg, 0.088 mmol) in
DMF (1 mL). The reaction mixture was warmed between 75–808C for
3 h. The solution was then co-evaporated with toluene to remove DMF
and the residue was purified by column chromatography on silica gel
with elution with EtOAc/MeOH/H2O (12:1:0.5, 10:1:0.5, 8:1:0.5, 5:1:0.5)
to provide the azido derivative 31 (20 mg, 74%) as a white resin.
C12H14N8O2 (302.29). Rf =0.58 (EtOAc/MeOH/H2O 5:1:0.5); 1H NMR
(CD3OD, 300 MHz): d=0.78–0.88 (m, 1H; H6’A), 1.61 (pstd, 3J=3.9,
8.4 Hz, 4J=0.9 Hz, 1H; H5’), 1.66 (dd, 2J=4.5, 3J=8.4 Hz, 1H; H6’B),
3.31 (d, 2J=11.7 Hz, 1H; CHHOH), 4.01 (d, 3J=6.6 Hz, 1H; H3’), 4.11
(d, 2J=11.7 Hz, 1H; CHHOH), 4.18 (dd, 3J=6.6 Hz, 4J=1.4 Hz, 1H;
H2’), 4.78 (s, 1H; H4’), 8.11 (s, 1H; H2), 8.40 ppm (s, 1H; H8); 13C NMR
(CD3OD, 75 MHz): d=13.1 (C6’), 24.0 (C5’), 36.3 (C1’), 63.0 (C2’), 64.0
(C4’), 64.6 (CH2OH), 79.3 (C3’), 120.4 (C5), 141.0 (C8), 150.2 (C4), 153.7
(C2), 157.5 ppm (C6); HRMS (ESI+): m/z: calcd for C12H15N8O2:
303.1318 [M+H]+; found: 303.1317.
N,N-Di-n-butylformamide
dimethylacetal:
Di-n-butyl
formamide
(50 mL) and fresh dimethyl sulfate (26 mL) were mixed under nitrogen
and heated to reflux (1008C) for 4 h, and were then allowed to cool to
ambient temperature and stirred overnight. The mixture was worked up
with ice-cold absolute MeOH (150 mL) into which sodium (8 g) had
been dissolved previously. After reaching room temperature, the volatiles
were concentrated in vacuo and diethyl ether was added with vigorous
stirring. The precipitate was filtered off and rinsed with more Et2O. The
filtrate was evaporated under reduced pressure and the oily residue was
distilled in vacuo (bp 110–1208C under oil pump vacuum: early fractions
contain N,N-di-n-butylformamide dimethyl acetal, late fractions contain
3
3
“dd”, J=7.5 Hz, 2H; CH2-fluorene), 4.42 (pst, J=7.2, 7.8 Hz, 1H; Ha),
4.84- 4.93 (m, 1H; H2’), 4.90 (s, 1H; H4’), 6.79 (d, 3J=8.4 Hz, 2H; m-
HPhOMe), 7.16 (d, 3J=8.4 Hz, 2H; o-HPhOMe), 7.21 (pst, 3J=6.6, 7.5 Hz,
2H; m2-Hfluorene), 7.28 (pst, 3J=7.2, 7.5 Hz, 2H; p3-Hfluorene), 7.50 (2ꢅd,
6254
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 6244 – 6257

Puromycin Analogues
FULL PAPER
3J=6.6, 6.9 Hz, 2H; o1-Hfluorene), 7.68 (2ꢅd, 3J=7.2, 7.5 Hz, 2H; m4-
Hfluorene), 8.41 (s, 1H; H2), 8.71 (s, 1H; H8), 8.93 ppm (s, 1H; N=
CHNBu2); 13C NMR (CD3OD, 75 MHz): d=12.9 (C6’), 14.5, 14.7 (2C;
N(CH2CH2CH2CH3)2), 21.2, 21.6 (2C; N(CH2CH2CH2CH3)2), 25.1 (C5’),
30.8, 32.4 (2C; N(CH2CH2CH2CH3)2), 37.1 (C1’), 38.9 (p-MeOPhCH2),
46.8 (N(CH2CH2CH2CH3)2), 48.7 (aliphatic fluorene), 53.5 (2C;
N(CH2CH2CH2CH3)2, C2’), 56.1 (OMe), 58.5 (Ca), 64.2 (C4’), 64.6
(CH2OH), 68.4 (CH2 fluorene), 77.2 (C3’), 115.3 (2C; m-CPhOMe), 121.3 (2C;
m4-Cfluorene), 126.6 (2C; o1-Cfluorene), 127.1 (C5), 128.5 (2C; m2-Cfluorene),
129.1 (2C; p3-Cfluorene), 130.7 (i-CPhOMe), 131.8 (2C; o-CPhOMe), 142.8 (C8),
142.9 (2C; o5-Cfluorene), 145.5 (2C; i-Cfluorene), 152.3 (C4), 153.6 (C2), 158.6
(C6), 160.4 (p-CPhOMe), 160.6 (N=CHNBu2), 161.9 (RC(O)OCH2 fluorene),
175.1 ppm (2’-NHC(O)R); IR (CH2Cl2): n˜ =2103 cmꢀ1 (N3 st); HRMS
(ESI+): m/z: calcd for C46H55N8O6: 815.4245 [M+H]+; found: 815.4248.
5.4 Hz, 2H), 2.87 (t, J=5.4 Hz, 2H), 2.99 (t, J=4.8 Hz, 1H), 3.20 (q, J=
7.2 Hz, 2H), 3.35 (s, 1H), 3.37–3.46 (m, 4H), 3.60–3.71 (m, 4H), 3.71–
3.81 (m, 12H), 4.14–4.39 (m, 8H), 4.39–4.57 (m, 2H), 4.59–4.71 (m, 2H),
4.94–5.08 (m, 2H), 5.90 (s, 2H), 6.74–6.94 (m, 10H), 7.07 (dd, J=5.1,
7.2 Hz, 2H), 7.20–7.36 (m, 16H), 7.36–7.46 (m, 4H), 8.17 (s, 1H), 8.23 (s,
1H), 8.38–8.50 (m, 4H), 9.02 (s, 1H), 9.06 ppm (s, 1H); 13C NMR
(CD3OD, 75 MHz): d=ꢀ4.7, ꢀ4.2, 9.3, 13.9, 14.1, 14.3, 14.5, 19.0, 20.3,
20.4, 20.8, 20.9, 21.2, 24.6, 24.9, 172.9, 25.0, 26.4, 26.5, 29.3, 30.4, 32.0,
34.2, 34.3, 34.3, 34.4, 46.4, 47.9, 53.1, 53.2, 55.8, 61.5, 62.4, 62.4, 64.0, 64.1,
64.2, 64.3, 64.5, 64.6, 64.6, 72.6, 76.1, 76.1, 76.4, 79.1, 82.3, 82.4, 88.7, 88.7,
92.1, 98.2, 114.3, 114.4, 118.5, 128.3, 128.3, 129.1, 129.1, 129.5, 131.5,
131.5, 136.4, 136.4, 142.0, 142.1, 145.6, 145.6, 145.7, 152.2, 152.3, 153.3,
157.7, 160.3, 160.4, 161.3, 161.4, 164.3, 172.8 ppm; 31P NMR (CD3OD,
121.5 MHz): d=ꢀ0.66, 0.01 ppm (2ꢅs, diast.); IR (CH2Cl2): n˜ =
2105 cmꢀ1 (N3 st); HRMS (ESI+): m/z: calcd for C62H81N13O12PSi:
1258.5635 [M+H]+; found: 1258.5638.
(1’S,2’R,3’R,4’R,5’S)-9-[2’-(O-Methyl)-l-tyrosylamino-3’-hydroxy-5’-
hydroxymethylbicycloACHTUNGTRENNUNG[3.1.0]hex-4’-yl]adenine (2): Compound 33 (42 mg,
0.051 mmol) was dissolved in 33% CH3NH2/EtOH (5 mL). The reaction
mixture was stirred at ambient temperature in a closed vessel for 4 h.
The volatiles were concentrated in vacuo and co-evaporated from CHCl3
(2ꢅ5 mL). The oily residue was purified by column chromatography on
silica gel with elution with EtOAc/MeOH/H2O (8:1:0.5, 6:1:0.5, 4:1:0.5)
to yield, after evaporation, the target compound 2 as a fluffy solid
(21 mg, 91%), which after lyophilization from H2O/TFA (pH 2.2) gave
the more water-soluble 2·TFA salt. C22H27N7O4 (453.49). Rf =0.28
(EtOAc/MeOH/H2O 5:1:0.75); 1H NMR (2, CD3OD, 500 MHz): d=0.73
(pst “dd”, 2J=4.5 Hz, 3J=8.0 Hz, 1H; H6’A), 1.48 (pst “dd”, 2J=4.5 Hz,
3J=3.5 Hz, 1H; H6’B), 1.70 (dd, 3J=3.5, 8.0 Hz, 1H; H5’), 2.91 (dd, 2J=
13.0 Hz, 3J=6.8 Hz, 1H; p-MeOPhCHH), 3.03 (dd, 2J=13.0 Hz, 3J=
7.0 Hz, 1H; p-MeOPhCHH), 3.16 (d, 2J=12.0 Hz, 1H; CHHOH), 3.78
(s, 3H; OCH3), 3.83–3.90 (m, 1H; Ha), 3.88 (d, 2J=12.0 Hz, 1H;
CHHOH), 3.98 (d, 3J=6.0 Hz, 1H; H3’), 4.80–4.93 (br, >2H; H2’, H4’),
6.90 (d, 3J=8.0 Hz, 2H; m-HPhOMe), 7.20 (d, 3J=8.0 Hz, 2H; o-HPhOMe),
8.21 (s, 1H; H2), 8.60 ppm (s, 1H; H8); 1H NMR (2·TFA, D2O,
300 MHz): d=0.70 (ddd, 2J=5.7 Hz, 3J=8.7 Hz, 4J=1.2 Hz, 1H; H6A),
1.12 (dd, 2J=5.7 Hz, 3J=4.2 Hz, 1H; H6’B), 1.68 (ddd, 3J=4.2, 8.7 Hz,
4J=1.2 Hz, 1H; H5’), 2.99 (dd, 2J=13.7 Hz, 3J=9.6 Hz, 1H; p-
MeOPhCHH), 3.02 (d, 2J=12.3 Hz, 1H; CHHOH), 3.16 (dd, 2J=
13.7 Hz, 3J=6.0 Hz, 1H; p-MeOPhCHH), 3.42 (d, 2J=12.3 Hz, 1H;
CHHOH), 3.75 (s, 3H; OCH3), 4.06 (d, 3J=6.9 Hz, 1H; H3’), 4.16 (dd,
(1’S,2’R,3’R,4’R,5’S)-9-{5’-[N-Acetyl-5’-O-dimethoxytrityl-2’-O-(tert-bu-
tyldimethyl)silylcytid-3’-yl]-(2-cyanoethyl)phosphoryloxy-2’-[N-(fluoren-
9-ylmethoxycarbonyl)-O-methyl-l-tyrosyl]amino-3’-
hydroxymethylbicycloACHTNUTGRNEUNG
[3.1.0]hex-4’-yl}-N6-[di-(n-butyl)aminomethinyl]-
adenine (35): A mixture of N-Fmoc-O-Me-l-Tyr (14 mg, 0.033 mmol)
and HOBT (5 mg, 0.033 mmol) was co-evaporated three times with anhy-
drous THF (2 mL). The residue was dissolved in THF (1 mL) and the so-
lution was cooled to 08C under nitrogen for 10 min. Diisopropyl carbo-
diimide (4.8 mL, 0.031 mmol) was added and the reaction mixture was
stirred for 15 min at the same temperature. Meanwhile, Me3P (1m in
THF, 53 mL, 0.053 mmol) was added to a solution of azide 34 (27 mg,
0.215 mmol) in THF (1 mL), and the mixture was stirred for 5 min at
room temperature. The amino acid solution was allowed to warm to
room temperature over 5 min and then added to the iminophosphorane
solution. The reaction mixture was stirred at room temperature over-
night. The volatiles were evaporated under reduced pressure and co-
evaporated from CHCl3 (5 mL), dissolved in EtOAc (30 mL), and ex-
tracted with saturated NaHCO3 (15 mL). The organic layer was extracted
twice with EtOAc and washed with H2O (2ꢅ10 mL), dried (MgSO4), fil-
tered, and concentrated in vacuo. The residue was purified by column
chromatography on silica gel with elution with EtOAc/toluene/MeOH,
(10:1:1, 8:1:1, 6:1:1) and EtOAc/MeOH/H2O (10:1:0.5) to give 35 as an
off-white solid (27 mg, 77%). C87H103N12O16PSi (1631.88). Rf =0.59
(EtOAc/MeOH/H2O 10:1:0.5); 1H NMR (CD3OD, 500 MHz): d=0.05–
0.24 (m, 12H), 0.79–1.00 (m, 32H), 1.19–1.46 (m, 10H), 1.47–1.74 (m,
10H), 1.91 (s, 3H), 1.99–2.05 (m, 3H), 2.09–2.21 (m, 6H), 2.66–2.75 (m,
3H), 2.76–2.93 (m, 3H), 2.99–3.12 (m, 3H), 3.19 (s, 12H), 3.35 (s, 6H),
3.37–3.48 (m, 4H), 3.49–3.59 (m, 2H), 4.03–4.17 (m, 5H), 4.17–4.29 (m,
3H), 4.29–4.40 (m, 3H), 4.45–4.70 (m, 4H), 4.94–4.99 (m, 2H), 5.33 (t,
J=5.0 Hz, 1H), 5.55–5.62 (m, 1H), 5.85–5.93 (m, 1H), 5.94–5.99 (m,
1H), 6.71–6.93 (m, 12H), 7.00–7.47 (m, 32H), 7.47–7.60 (m, 5H), 7.63–
7.79 (m, 7H), 8.16–8.28 (m, 2H), 8.30–8.48 (m, 7H), 8.53–8.57 (m, 1H),
9.04 (s, 1H), 9.10 ppm (s, 1H); 13C NMR (CD3OD, 125 MHz): d=ꢀ4.4,
ꢀ4.5, 14.0, 14.2, 14.3, 14.5, 19.0, 19.1, 20.3, 20.8, 21.2, 22.7, 23.2, 23.5, 23.6,
24.6, 26.4, 26.6, 26.9, 27.4, 28.1, 29.4, 30.4, 30.4, 30.6, 30.8, 30.9, 32.0, 33.1,
35.2, 36.6, 38.4, 42.7, 46.3, 46.4, 47.7, 49.9, 53.1, 53.2, 55.8, 55.8, 55.9, 55.9,
57.8, 58.2, 61.6, 63.9, 64.5, 64.7, 68.1, 71.9, 76.5, 88.2, 88.4, 88.6, 88.7, 91.2,
92.0, 97.9, 98.1, 98.2, 98.2, 108.6, 114.3, 114.4, 114.8, 114.9, 118.6, 120.9,
126.2, 126.2, 128.1, 128.7, 129.0, 129.1, 129.1, 129.4, 129.6, 130.5, 130.9,
131.3, 131.4, 131.5, 131.5, 136.4, 136.4, 136.5, 136.7, 136.8, 138.5, 142.5,
145.2, 145.7, 145.8, 146.0, 146.1, 152.1, 153.4, 157.4, 157.8, 158.2, 159.9,
179.3, 160.3, 160.4, 161.5, 164.3, 172.9, 172.9 ppm; 31P NMR (CD3OD,
121.5 MHz): d=ꢀ0.68, 0.01 ppm (2ꢅs, diast.); HRMS (ESI+): m/z:
calcd for C87H103N12O16PSi: 1631.7200 [M+H]+; found: 1631.7201.
3
3J=6.0, 9.6 Hz, 1H; Ha), 4.62 (d, J=6.9 Hz, 1H; H2’), 4.87 (s, 1H; H4’),
6.93 (d, 3J=8.7 Hz, 2H; m-HPhOMe), 7.18 (d, 3J=8.7 Hz, 2H; o-HPhOMe),
8.27 (s, 1H; H2), 8.43 ppm (s, 1H; H8); 13C NMR (2, CD3OD, 125 MHz):
d=12.6 (C6’), 24.7 (C5’), 36.6 (p-MeOPhCH2), 40.2 (C1’), 53.2 (C2’), 55.8
(OMe), 56.9 (Ca), 64.1 (CH2OH), 64.1 (C4’), 76.9 (C3’), 115.3 (2C; m-
C
PhOMe), 120.3 (C5), 129.4 (i-CPhOMe), 131.5 (2C; o-CPhOMe), 141.0 (C8),
150.1 (C4), 153.7 (C2), 157.4 (C6), 160.4 (p-CPhOMe), 174.3 ppm (2’-
NHC(O)R’); HRMS (ESI+): m/z: calcd for C22H28N7O4: 454.2203
[M+H]+; found: 454.2199.
(1’S,2’R,3’S,4’R,5’S)-9-{5’-[N-Acetyl-5’-O-dimethoxytrityl-2’-O-(tert-butyl-
dimethyl)silylcytid-3’-yl]-(2-cyanoethylphosphoryl)oxymethyl-2’-azido-3’-
hydroxy-bicycloACHTUNGTRENNUNG
[3.1.0]hex-4’-yl}-N6-[di-(n-butyl)aminomethinyl]adenine
(34): Compound 32 (49 mg, 0.111 mmol) and 5-ethylthiotetrazole (22 mg,
0.166 mmol) were co-evaporated three times with anhydrous toluene
(5 mL) and dissolved in anhydrous CH3CN (750 mL). After the addition
of commercial N4-acetyl-5’-O-dimethoxytrityl-2’-O-tert-butyldimethylsilyl-
cytid-3’-yl-(2-cyanoethyl)-N,N-diisopropylphosphoramidite
(122 mg,
0.122 mmol) in anhydrous CH3CN (750 mL), the solution was stirred at
room temperature for 1 h, followed by the addition of I2/THF/pyridine/
H2O solution (0.02m, 6.65 mL, 0.133 mmol I2). After 10 min, the reaction
mixture was poured into EtOAc (20 mL), and extracted with NaHSO3
(0.2m, 2ꢅ8 mL) and brine (10 mL). The organic layer was dried over an-
hydrous MgSO4, filtered, and concentrated in vacuo. The residue was pu-
rified by column chromatography on silica gel with elution with EtOAc/
MeOH/H2O (14:1:0.5, 12:1:0.5, 10:1:0.5) to provide recovered starting
material 32 (17 mg, 35%), as well as the dinucleotide derivative 34
(32 mg, 35%), as a colorless oil. C62H80N13O12PSi (1258.44). Rf =0.53
(EtOAc/MeOH/H2O 10:1:0.5); 1H NMR (CD3OD, 300 MHz): d=0.13–
0.26 (m, 12H), 0.80–1.02 (m, 32H), 1.18–1.46 (m, 10H), 1.55–1.75 (m,
8H), 1.77–1.86 (m, 2H), 2.01 (s, 3H), 2.08–2.19 (m, 7H), 2.72 (t, J=
(1’S,2’R,3’R,4’R,5’S)-9-{5’-[(Cytid-3’-yl)-phosphoryloxy]-2’-(O-methyl-l-
tyrosyl)amino-3’-hydroxymethylbicycloACTHNUGTRNEUNG[3.1.0]hex-4’-yl}adenine (36): Di-
nucleotide 35 (27 mg, 0.016 mmol) was dissolved in 33% CH3NH2/EtOH
(3 mL). The reaction mixture was stirred at room temperature for 7 h in
a closed vessel. The solution was concentrated under reduced pressure
and co-evaporated from CHCl3 (2ꢅ3 mL). Ammonium fluoride (2.5 mg,
0.066 mmol) was added to a stirred solution of the residue in MeOH
(1.5 mL). The reaction mixture was warmed to 50–558C for 2 h, and
monitored by TLC. The volatiles were removed in vacuo and the result-
Chem. Eur. J. 2009, 15, 6244 – 6257
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6255

P. Strazewski and B. Y. Michel
ing residue was dissolved in AcOH/H2O 4:1 (1.5 mL). The solution was
stirred at room temperature for 1 h and then lyophilized. The residue
was purified by preparative thin layer chromatography (TLC) plates first
pre-eluted with iPrOH/NH3/H2O (7:2:1) and then loaded and eluted to
yield, after extraction with and lyophilization from the same solvent mix-
ture, target compound 36 as a fluffy solid (10 mg, 81%). C31H39N11O15P
(758.68). Rf =0.23 (iPrOH/NH3/H2O, 7:2:1); 1H NMR (D2O, 300 MHz):
d=0.89–0.97 (m, 1H; H6’A), 1.18–1.22 (m, 1H; H6’B), 1.87–1.92 (m,
1H,), 3.18–3.25 (m, 2H), 3.85 (s, 4H; OCH3), 3.92–3.99 (m, 2H), 4.07–
4.12 (m, 1H; Ha), 4.24–4.31 (m, 4H), 4.51–4.61 (m, 2H), 4.95 (s, 1H;
H4’), 5.85 (d, J=3.9 Hz, 1H; H1’), 5.87 (d, 3J=7.5 Hz, 1H; H5), 7.04 (d,
3J=8.7 Hz, 2H; m-HPhOMe), 7.26 (d, 3J=8.7 Hz, 2H; o-HPhOMe), 7.78 (d,
3J=7.5 Hz, 1H; H6), 8.25 (s, 1H; H2), 8.50 ppm (s, 1H; H8); 31P NMR
(CD3OD, 121.5 MHz): d=0.00 ppm; HRMS (ESI+): m/z: calcd for
C31H40N10O15P: 759.2616 [M+H]+; found: 759.2616.
quez, C. S. Yuan, R. T. Borchardt, Heterocycles 1995, 41, 2651–2656;
e) L. S. Jeong, V. E. Marquez, Tetrahedron Lett. 1996, 37, 2353–
2356; f) V. E. Marquez, M. A. Siddiqui, A. Ezzitouni, P. Russ, J. Y.
Wang, R. W. Wagner, M. D. Matteucci, J. Med. Chem. 1996, 39,
3739–3747; g) M. A. Siddiqui, H. Ford, C. George, V. E. Marquez,
Nucleosides Nucleotides 1996, 15, 235–250; h) A. Ezzitouni, V. E.
Marquez, J. Chem. Soc. Perkin Trans. 1 1997, 1073–1078; i) P. Russ,
A. Ezzitouni, V. E. Marquez, Tetrahedron Lett. 1997, 38, 723–726;
j) L. S. Jeong, G. Buenger, J. J. McCormack, D. A. Cooney, Z. Hao,
V. E. Marquez, J. Med. Chem. 1998, 41, 2572–2578; k) V. E. Mar-
quez, A. Ezzitouni, P. Russ, M. A. Siddiqui, H. Ford, R. J. Feldman,
H. Mitsuya, C. George, J. J. Barchi, J. Am. Chem. Soc. 1998, 120,
2780–2789; l) V. E. Marquez, A. Ezzitouni, P. Russ, M. A. Siddiqui,
H. Ford, R. J. Feldman, H. Mitsuya, C. George, J. J. Barchi, Nucleo-
sides Nucleotides 1998, 17, 1881–1884; m) V. E. Marquez, P. Russ,
R. Alonso, M. A. Siddiqui, S. Hernandez, C. George, M. C. Nicklaus,
F. Dai, H. Ford, Helv. Chim. Acta 1999, 82, 2119–2129; n) V. E.
Marquez, P. Russ, R. Alonso, M. A. Siddiqui, K. J. Shin, C. George,
M. C. Nicklaus, F. Dai, H. Ford, Nucleosides Nucleotides Nucleic
Acids 1999, 18, 521–530; o) H. R. Moon, H. O. Kim, M. W. Chun,
L. S. Jeong, V. E. Marquez, J. Org. Chem. 1999, 64, 4733–4741;
p) K. A. Jacobson, X. D. Ji, A. H. Li, N. Melman, M. A. Siddiqui,
K. J. Shin, V. E. Marquez, R. G. Ravi, J. Med. Chem. 2000, 43, 2196–
2203; q) H. R. Moon, H. Ford, V. E. Marquez, Org. Lett. 2000, 2,
3793–3796; r) E. Nandanan, S. Y. Jang, S. Moro, H. O. Kim, M. A.
Siddiqui, P. Russ, V. E. Marquez, R. Busson, P. Herdewijn, T. K.
Harden, J. L. Boyer, K. A. Jacobson, J. Med. Chem. 2000, 43, 829–
842; s) K. J. Shin, H. R. Moon, C. George, V. E. Marquez, J. Org.
Chem. 2000, 65, 2172–2178; t) K. A. Jacobson, R. G. Ravi, E. Nan-
danan, H. S. Kim, S. Moro, Y. C. Kim, K. Lee, D. Barak, V. E. Mar-
quez, X. D. Ji, Nucleosides Nucleotides Nucleic Acids 2001, 20, 333–
341; u) K. Lee, G. Ravi, X. D. Ji, V. E. Marquez, K. A. Jacobson,
Bioorg. Med. Chem. Lett. 2001, 11, 1333–1337; v) G. Ravi, K. Lee,
X. D. Ji, H. S. Kim, K. A. Soltysiak, V. E. Marquez, K. A. Jacobson,
Bioorg. Med. Chem. Lett. 2001, 11, 2295–2300; w) Y. Choi, C.
George, P. Strazewski, V. E. Marquez, Org. Lett. 2002, 4, 589–592;
x) H. S. Kim, R. G. Ravi, V. E. Marquez, S. Maddileti, A. K. Wihl-
borg, D. Erlinge, M. Malmsjo, J. L. Boyer, T. K. Harden, K. A. Ja-
cobson, J. Med. Chem. 2002, 45, 208–218; y) Y. Yoshimura, H. R.
Moon, Y. Choi, V. E. Marquez, J. Org. Chem. 2002, 67, 5938–5945;
z) L. Zalah, M. Huleihel, E. Manor, A. Konson, H. Ford, V. E. Mar-
quez, D. G. Johns, R. Agbaria, Antiviral Res. 2002, 55, 63–75;
aa) Y. S. Choi, C. George, M. J. Comin, J. J. Barchi, H. S. Kim, K. A.
Jacobson, J. Balzarini, H. Mitsuya, P. L. Boyer, S. H. Hughes, V. E.
Marquez, J. Med. Chem. 2003, 46, 3292–3299; ab) V. E. Marquez, T.
Ben-Kasus, J. J. Barchi, K. M. Green, M. C. Nicklaus, R. Agbaria, J.
Am. Chem. Soc. 2004, 126, 543–549; ac) P. Schelling, M. T. Claus, R.
Johner, V. E. Marquez, G. E. Schulz, L. Scapozza, J. Biol. Chem.
2004, 279, 32832–32838; ad) B. V. Joshi, H. R. Moon, J. C. Fettinger,
V. E. Marquez, K. A. Jacobson, J. Org. Chem. 2005, 70, 439–447;
ae) V. E. Marquez, Y. Choi, M. J. Comin, P. Russ, C. George, M. Hu-
leihel, T. Ben-Kasus, R. Agbaria, J. Am. Chem. Soc. 2005, 127,
15145–15150; af) M. J. Comin, D. A. Parrish, J. R. Deschamps, V. E.
Marquez, Org. Lett. 2006, 8, 705–708; ag) V. E. Marquez, S. H.
Hughes, S. Sei, R. Agbaria, Antiviral Res. 2006, 71, 268–275;
ah) O. R. Ludek, V. E. Marquez, Synthesis 2007, 3451–3460;
ai) V. E. Marquez in Modified Nucleosides in Biochemistry Biotech-
nology and Medicine (Ed.: P. Herdewijn), Wiley-VCH, Weinheim,
2008, Chapter 12, pp. 307–341.
CCDC-711879, CCDC-711880 and CCDC-711881 contain the supplemen-
tary crystallographic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgements
We gratefully acknowledge Denis Bouchu and Christian Duchamps for
mass spectrometric analyses, Bernard Fenet and Christophe Gilbert for
NMR analyses, Victor Marquez for providing us with generous amounts
of OP(ꢂ)0 and OP(ꢀ)0, Oana Pantelimonescu for synthesizing and pre-
paring the monocrystals of OP(ꢂ)3 and OP(ꢂ)4, and Dominique
Luneau and Erwann Jeanneau for the crystallographic structures. We are
indebted to the CIBLE 2006 program of the Rꢁgion Rhꢆne-Alpes for a
substantial financial contribution to our PURBLOQ project, the COST
Action CM0703 on Systems Chemistry (www.cost.esf.org/domains_
actions/cmst/Actions/Systems_Chemistry) and the European project no.
043359 SynthCells (synthcells.org) for ongoing financial support. B.Y.M.
is thankful to the Ministꢀre de la Recherche for his graduate fellowship.
[1] J. N. Porter, R. I. Hewitt, C. W. Hesseltine, G. Krupka, J. A. Lowery,
W. S. Wallace, N. Bohonos, J. H. Williams, Antibiot. Chemother.
1952, 2, 409–410.
[2] a) R. J. Suhadolnik, Nucleoside Antibiotics, Wiley-Interscience, New
York, 1970, pp. 1–50; b) R. J. Suhadolnik, Nucleosides as Biological
Probes, Wiley, New York, 1979, pp. 96–102.
[3] a) M. B. Yarmolinsky, G. L. De La Haba, Proc. Natl. Acad. Sci. USA
1959, 45, 1721–1729; b) D. W. Allen, P. C. Zamecnik, Biochim. Bio-
phys. Acta 1962, 55, 865–874; c) D. Nathans, A. Neidle, Nature
1963, 197, 1076–1077; d) W. Gilbert, J. Mol. Biol. 1963, 6, 389–403;
e) R. H. Symons, R. J. Harris, L. P. Clarke, J. F. Wheldrake, W. H.
Elliott, Biochim. Biophys. Acta Nucleic Acids Protein Synth. 1969,
179, 248–250.
[4] a) A. Sievers, M. Beriner, M. V. Rodnina, R. Wolfenden, Proc. Natl.
Acad. Sci. USA 2004, 101, 7897–7901; b) A. Sievers, M. Beriner,
M. V. Rodnina, R. Wolfenden, Proc. Natl. Acad. Sci. USA 2004, 101,
12397; c) G. K. Schroeder, R. Wolfenden, Biochemistry 2007, 46,
4037–4044; d) S. Trobro, J. ꢄqvist, Proc. Natl. Acad. Sci. USA 2005,
102, 12395–12400; e) S. Trobro, J. ꢄqvist, Biochemistry 2006, 45,
7049–7056; f) M. Johansson, E. Bouakaz, M. Lovmar, M. Ehren-
berg, Mol. Cell 2008, 30, 589–598; g) I. Wohlgemuth, S. Brenner, M.
Beringer, M. V. Rodnina, J. Biol. Chem. 2008, 283, 32229–32235.
[5] B. Y. Michel, K. S. Krishnakumar, M. Johansson, M. Ehrenberg, P.
Strazewski, Nucleic Acids Symp. Ser. 2008, 52, 33–34.
[8] C. Altona, M. J. Sundaralingam, J. Am. Chem. Soc. 1972, 94, 8205–
8212.
[9] a) A. A. Koshkin, S. K. Singh, P. Nielsen, V. K. Rajwanshi, R.
Kumar, M. Meldgaard, C. E. Olsen, J. Wengel, Tetrahedron 1998, 54,
3607–3630; b) A. G. Olsen, V. K. Rajwanshi, C. Nielsen, J. Wengel,
J. Chem. Soc. Perkin Trans. 1 2000, 3610–3614; c) T. Bryld, M. H.
Sorensen, P. Nielsen, T. Koch, C. Nielsen, J. Wengel, J. Chem. Soc.
Perkin Trans. 1 2002, 1655–1662; d) M. Petersen, K. Bondensgaard,
J. Wengel, J. P. Jacobsen, J. Am. Chem. Soc. 2002, 124, 5974–5982;
e) B. Ravindra Babu, N. E. Raunak Poopeiko, M. Juhl, A. D. Bond,
[6] K. Kinoshita, S. Yaginuma, M. Hayashi, K. Nakatsu, Nucleosides
Nucleotides 1985, 4, 661–668.
[7] a) J. B. Rodriguez, V. E. Marquez, M. C. Nicklaus, J. J. Barchi, Tetra-
hedron Lett. 1993, 34, 6233–6236; b) J. B. Rodriguez, V. E. Marquez,
M. C. Nicklaus, H. Mitsuya, J. J. Barchi, J. Med. Chem. 1994, 37,
3389–3399; c) A. Ezzitouni, J. J. Barchi, V. E. Marquez, J. Chem.
Soc. Chem. Commun. 1995, 1345–1346; d) L. S. Jeong, V. E. Mar-
6256
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 6244 – 6257

Puromycin Analogues
FULL PAPER
V. S. Parmar, J. Wengel, Eur. J. Org. Chem. 2005, 2297–2321; f) P. J.
Hrdlicka, N. K. Andersen, J. S. Jepsen, F. G. Hansen, K. F. Hasel-
mann, C. Nielsen, J. Wengel, Bioorg. Med. Chem. 2005, 13, 2597–
2621.
[22] a) V. E. Marquez, M. I. Lim, C. K. H. Tseng, A. Markovac, M. A.
Priest, M. S. Khan, B. Kaskar, J. Org. Chem. 1988, 53, 5709–5714;
b) M. I. Lim, V. E. Marquez, Tetrahedron Lett. 1983, 24, 5559–5562;
c) A. B. Smith, R. E. Richmond, J. Am. Chem. Soc. 1983, 105, 575–
585; d) J. L. Luche, J. Am. Chem. Soc. 1978, 100, 2226–2227.
[23] a) K. H. Altmann, R. Kesselring, E. Francotte, G. Rihs, Tetrahedron
Lett. 1994, 35, 2331–2334; b) J. A. Staroscik, B. Rickborn, J. Org.
Chem. 1972, 37, 738–740.
[24] S. Dey, P. Garner, J. Org. Chem. 2000, 65, 7697–7699.
[25] a) S. Greenberg, J. G. Moffatt, J. Am. Chem. Soc. 1973, 95, 4016–
4025; b) A. F. Russell, S. Greenberg, J. G. Moffatt, J. Am. Chem.
Soc. 1973, 95, 4025–4030; c) M. J. Robins, F. Hansske, N. H. Low,
J. I. Park, Tetrahedron Lett. 1984, 25, 367–370; d) F. Hansske, M. J.
Robins, Tetrahedron Lett. 1985, 26, 4295–4298; e) M. C. Samano,
M. J. Robins, Tetrahedron Lett. 1989, 30, 2329–2332; f) M. J. Robins,
R. W. Miles, M. C. Samano, R. L. Kaspar, J. Org. Chem. 2001, 66,
8204–8210; g) Z. G. Gao, S. K. Kim, T. Biadatti, W. Z. Chen, K.
Lee, D. Barak, S. G. Kim, C. R. Johnson, K. A. Jacobson, J. Med.
Chem. 2002, 45, 4471–4484.
[10] a) S. Obika, D. Nanbu, Y. Fari, K. Morio, Y. In, T. Ishida, T. Ima-
nishi, Tetrahedron Lett. 1997, 38, 8735–8738; b) S. Obika, Y. Hari,
K. Morio, T. Imanishi, Tetrahedron Lett. 2000, 41, 215–219; c) S.
Obika, K.-I. Morio, D. Nanbu, Y. Hari, H. Itoh, T. Imanishi, Tetrahe-
dron 2002, 58, 3039–3049; d) S. Obika, T. Uneda, T. Sugimoto, D.
Nanbu, T. Minami, T. Doi, T. Imanishi, Bioorg. Med. Chem. 2001, 9,
1001–1011.
[11] a) M. Ohno, S. Costanzi, H. S. Kim, V. Kempeneers, K. Vastmans, P.
Herdewijn, S. Maddileti, A.-G. Gao, T. K. Harden, K. A. Jacobson,
Bioorg. Med. Chem. 2004, 12, 5619–5630; b) H. S. Kim, K. A. Jacob-
son, Org. Lett. 2003, 5, 1665–1668.
[12] S. A. Kim, H. M. Lee, J. S. Ryu, H. S. Kim, Synlett 2007, 1055–1058.
[13] H. Saneyoshi, B. Y. Michel, Y. Choi, P. Strazewski, V. E. Marquez, J.
Org. Chem. 2008, 73, 9435–9438.
[14] a) P. Nissen, J. Hansen, N. Ban, P. B. Moore, T. A. Steitz, Science
2000, 289, 920–930; b) M. Erlacher, D. N. Wilson, R. Micura, N. Po-
lacek, Chem. Biol. 2008, 15, 485–492.
[26] M. Pour, A. C. Willis, M. Furber, L. N. Mander, Tetrahedron 1998,
54, 13833–13850.
[15] B. Y. Michel, K. S. Krishnakumar, P. Strazewski, Synlett 2008, 2461–
2464.
[27] A. Mitrochkine, F. Eydoux, M. Martres, A. Heumann, M. Rꢁglier,
Tetrahedron: Asymmetry 1995, 6, 59–62.
[16] a) A. R. Mattocks, J. Chem. Soc. 1964 1918–1930; b) A. R. Mat-
tocks, J. Chem. Soc. 1964, 4840–4845.
[17] W. J. Choi, H. R. Moon, H. O. Kim, B. N. Yoo, J. A. Lee, D. H. Shin,
L. S. Jeong, J. Org. Chem. 2004, 69, 2634–2636.
[28] N. Yamamoto, T. Nishikawa, M. Isobe, Synlett 1995, 505–506.
[29] a) B. C. Froehler, M. D. Matteucci, Nucleic Acids Res. 1983, 11,
8031–8036; b) L. J. McBride, R. Kierzek, S. L. Beaucage, M. H. Car-
uthers, J. Am. Chem. Soc. 1986, 108, 2040–2048.
[18] B. Y. Michel, P. Strazewski, Tetrahedron 2007, 63, 9836–9841.
[19] a) H. R. Moon, W. J. Choi, H. O. Kim, L. S. Jeong, Tetrahedron:
Asymmetry 2002, 13, 1189–1193; b) J. A. Lee, H. O. Kim, D. K.
Tosh, H. R. Moon, S. Kim, L. S. Jeong, Org. Lett. 2006, 8, 5081–
5083; c) L. S. Jeong, J. A. Lee, H. O. Kim, D. K. Tosh, H. R. Moon,
S. J. Lee, K. M. Lee, Y. Y. Sheen, M. W. Chun, Nucleosides Nucleo-
tides Nucleic Acids 2007, 26, 1021–1024.
[20] A. Maddaford, T. Guyot, D. Leese, R. Glen, J. Hart, X. Zhang, R.
Fisher, D. S. Middleton, C. L. Doherty, N. N. Smith, D. C. Pryde,
S. C. Sutton, Synlett 2007, 3149–3154.
[30] T. W. Greene, P. G. M. Wuts, Protective Groups in Organic Synthesis,
3rd ed., Wiley, New York, 1999.
[31] A. Charafeddine, H. Chapuis, P. Strazewski, Org. Lett. 2007, 9,
2787–2790.
[32] X-F. Zhu, A. I. Scott, Nucleosides Nucleotides Nucleic Acids 2001,
20, 197–211.
[33] W. Zhang, M. J. Robins, Tetrahedron Lett. 1992, 33, 1177–1180.
[34] a) H. Chapuis, P. Strazewski, Tetrahedron 2006, 62, 12108–12115;
b) A. Charafeddine, P. Strazewski, Chem. Eur. J. 2007, 13, 5566–
5584.
[21] a) A. Fꢃrstner, Angew. Chem. Int. Ed. 2000, 39, 3012–3043; Angew.
Chem. 2000, 112, 3140–3172O. Guth, A. Dꢃffels, G. Seidel, M.
Liebl, B. Gabor, R. Mynott, Chem. Eur. J. 2001, 7, 4811–4820; b) A.
Fꢃrstner, Angew. Chem. Int. Ed. 2000, 39, 3012–3043; Angew.
Chem. 2000, 112, 3140–3172.
Received: December 13, 2008
Revised: February 27, 2009
Published online: May 13, 2009
Chem. Eur. J. 2009, 15, 6244 – 6257
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6257