Stereochemical assignment and first synthesis of the core of miharamycin antibiotics

Article abstract of DOI:10.1002/chem.200801826

The relative configuration at C-6′ of nucleoside antibiotic miharamycin A has been elucidated by NMR spectroscopy and proved to be S. The total synthesis of miharamycin B has also been investigated, which has led to the unprecedented construction of its core. The bicyclic sugar moiety has been elaborated by means of a SmI₂-based keto-alkyne coupling. Elongation of its C-6 position towards a bicyclic sugar amino acid and conversion into a suitable glycosyl donor enabled efficient N-glycosylation with 2-aminopurine to take place to afford the nucleosidic part of miharamycin B. Final peptide coupling with arginine afforded the skeleton of miharamycin B. Unfortunately, attempts to deprotect this scaffold failed to afford the complex nucleoside antibiotic.

Full text of DOI:10.1002/chem.200801826

DOI: 10.1002/chem.200801826
Stereochemical Assignment and First Synthesis of the Core of Miharamycin
Antibiotics
Filipa Marcelo,[a, b] Jesffls JimØnez-Barbero,[c] JØrôme Marrot,[d] AmØlia P. Rauter,*[b]
Pierre Sinaþ,*[a] and Yves BlØriot*[a]
Abstract: The relative configuration at
C-6’ of nucleoside antibiotic miharamy-
cin A has been elucidated by NMR
spectroscopy and proved to be S. The
total synthesis of miharamycin B has
also been investigated, which has led to
the unprecedented construction of its
core. The bicyclic sugar moiety has
been elaborated by means of a SmI2-
based keto–alkyne coupling. Elonga-
tion of its C-6 position towards a bicy-
clic sugar amino acid and conversion
into a suitable glycosyl donor enabled
efficient N-glycosylation with 2-amino-
purine to take place to afford the nu-
cleosidic part of miharamycin B. Final
peptide coupling with arginine afforded
the skeleton of miharamycin B. Un-
fortunately, attempts to deprotect this
scaffold failed to afford the complex
nucleoside antibiotic.
Keywords: antibiotics
·
natural
products · nucleosides · structure
elucidation · total synthesis
Introduction
still notable targets that have been approached but not yet
synthesized. Amongst them, miharamycins A and B, isolated
forty years ago from Streptomyces miharaensis SF-489, dis-
play strong activity against the rice blast disease caused by
Pyricularia oryzae.[4] The structure of miharamycins was par-
tially elucidated twenty-five years ago with the help of ex-
tensive spectroscopic and chemical degradation studies.[5]
Complex nucleoside antibiotics comprise an extensive array
of natural products that combine the structural features of
nucleosides, higher monosaccharides, disaccharides, peptides
and lipids.[1] They exhibit a variety of biological activities in-
cluding antitumor, antiviral, antibacterial and antifungal
properties.[1,2] While an increasing number of compounds of
this class has been accessed by total synthesis,[3] there are
[a] F. Marcelo, Prof. P. Sinaþ, Dr. Y. BlØriot
UMR 7611, Equipe Glycochimie, UPMC-Paris 6, 4
place Jussieu, Paris, 75005 (France)
Fax : (+33)1-4427-5504
E-mail: pierre.sinay@upmc.fr
yves.bleriot@upmc.fr
[b] F. Marcelo, Prof. A. P. Rauter
Carbohydrate Chemistry Group, CQB-FCUL
Departamento de Química e Bioquímica
Faculdade de CiÞncias da Universidade de Lisboa
1749-016 Lisbon (Portugal)
Miharamycins were found to be made up of an unusual
higher monosaccharide component, appended to an N-ter-
minal amino acid residue at C-6’, and a purine nucleobase
connected at N9. However, the absolute configuration at C-
6’ for both antibiotics remains undetermined. Although con-
siderable effort has been devoted to the total synthesis of
miharamycins because of their unique and complex struc-
ture,[6] none of these attempts has yet been successful.
Fax : (+351)21-7500088
E-mail: aprauter@fc.ul.pt
[c] Prof. J. JimØnez-Barbero
Centro de Investigaciones Biológicas
CSIC, 28040 Madrid (Spain)
[d] J. Marrot
Institut Lavoisier, UniversitØ de Versailles-Saint-Quentin
UMR 8180, 78035 Versailles (France)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200801826.
10066
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 10066 – 10073

FULL PAPER
Results and Discussion
droxyl and amino groups as benzyl ethers and O-benzyl car-
bamates respectively should ensure a clean final deprotec-
tion step under rather neutral conditions.
As part of a program exploiting samarium(II) iodide
methodology[8] to access and elucidate the framework of
natural products displaying unusual sugar-like moieties,[9] we
first investigated the synthesis of the bicyclic sugar moiety
of miharamycin (Scheme 2). This bicyclic sugar moiety can
Determination of the relative configuration at C-6’ of mihar-
amycin A: A prerequisite to the synthetic work was the con-
firmation of the relative structure of miharamycins and the
determination of their configuration at C-6’. This was inves-
tigated by NMR spectroscopy, combining complementary
standard 2D methods at 500 and 700 MHz (see the Support-
1
ing Information). The assignment of the H and 13C NMR
spectroscopic resonance signals collected for miharamycin A
supported the structure proposed by Seto.[5] Moreover, three
key features allowed the non-ambiguous deduction of an S
configuration for C-6’: a small JH5’H6’ value (1.8 Hz, gauche-
type major orientation), a large JH5’C7’ value (>8 Hz, anti-
type major orientation, C-7’ being the carboxylic carbon
atom) and the existence of a H-1’/H-2’’ NOE (associated
distance ca. 4 Š). These three experimental observations
can only be accommodated by the S configuration of the C-
6’ stereogenic centre, along with the existence of a major
conformer in which the lateral chain is folded towards the
aminopurine moiety (see the Supporting Information).[7] Al-
though no sample of miharamycin B has been available to
us for NMR spectroscopic studies, it is highly probable that
miharamycins A and B share the same S configuration at C-
6’ since both antibiotics have been isolated from the same
microorganism and S-configured natural amino acids are
dominant in members of the peptidyl nucleoside family of
antibiotics.[1]
Scheme 2. Synthesis of azido ester 10. a) SmI2 (0.1m in THF), HMPA,
tBuOH, THF, 94%; b) O3, À788C, Me2S, CH2Cl2 then NaBH4, EtOH,
08C, 84% over two steps; c) BnBr, NaH, DMF, 95%; d) LiAlH4/AlCl3,
Et2O/CH2Cl2 2:1, 508C, 88%; e) (COCl)2, DMSO, Et3N, CH2Cl2, À788C
then CH2=CHMgBr, THF, À788C, 65% over 2 steps; f) O3, À788C,
Me2S, CH2Cl2 then NaClO2, NaH2PO4·H2O, tBuOH/H2O/2-methyl-but-2-
ene 2:2:1 then MeI, KHCO3, DMF, 65% over three steps; g) Tf2O, pyri-
dine, CH2Cl2, À788C then NaN3, DMF, 82% over 2 steps.
Chemical synthesis: We then concentrated our efforts to-
wards the total synthesis of miharamycin B, which incorpo-
rates an easily available l-arginine moiety. The synthetic
challenge of miharamycin B stands in its unique elongated
bicyclic sugar moiety and in the presence of an unusual 2-
aminopurine nucleobase, which suggests the four disconnec-
tions listed in Scheme 1. Tactics harvested in previously
completed complex nucleoside antibiotics syntheses usually
place the sugar moiety construction early in the synthesis
and the N-glycosylation step before the peptide coupling
due to the Lewis basicity of the amide bond that can some-
times interfere with glycosylation conditions.[1] We, there-
fore, relied on the following strategy: 1) construction of the
bicyclic carbohydrate unit, 2) elaboration of the amino acid
at C-6’, 3) nucleoside synthesis and 4) peptide coupling fol-
lowed by final deprotection (Scheme 1). Protection of hy-
be considered the equivalent of a ring-opened 3’-keto com-
pound and should be accessible by cyclisation of the corre-
sponding a-propargyloxy ketone 3. When treated with SmI2
in the presence of hexamethylphosphoramide (HMPA) and
tBuOH, ketone 3 smoothly afforded the desired exoalkene 4
through a 5-exo-dig ketyl–alkyne cyclisation.[10] Other ap-
proaches to the bicyclic carbohydrate core of miharamycins
have also been reported more recently.[11] Further, ozonoly-
sis of the olefin followed by NaBH4 reduction, occurring ex-
clusively from the exo face of the bicyclic system, furnished
the desired kinetic diol, which was then benzylated to afford
the tricyclic acetal 5. Construction of the sugar amino acid
core was then studied. Beside non-carbohydrate-based ap-
proaches that condense Garnerꢁs aldehyde as a masked
amino acid,[12] several methods are available to homologate
the C-6 position of a sugar pyranoside and construct the
amino acid. Although stereocontrolled ethynylation of a dia-
ldosugar has been reported,[13] we focused on a chain exten-
sion methodology, which used the vinyl group as a synthetic
equivalent of the carboxylic acid functionality; this led to a
lower degree of stereocontrol to obtain both epimers at C-6’
for further SAR studies. Regioselective benzylidene opening
yielded primary alcohol 6, which was further oxidized under
Scheme 1. Key disconnections planned for the synthesis of the miharamy-
cin B framework.
Chem. Eur. J. 2008, 14, 10066 – 10073ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10067

Y. BlØriot, P. Sinaþ, A. P. Rauter et al.
Swern conditions to give the aldehyde; this was then directly
engaged in the alkylation step with vinyl magnesium bro-
mide to afford the diastereomeric allylic alcohols (S)-7 and
(R)-8 in a 3:2 ratio. The newly established C-6’ S configura-
tion of miharamycin A and the required double inversion at
C-6 to introduce the amino acid moiety prompted us to start
from the S-configured alcohol 7. Ozonolysis of allylic alco-
hol 7 and further oxidation to the carboxylic acid with
NaClO2 followed by esterification with iodomethane yielded
the C-6 (R)-methyl ester 9. Triflation of the free OH group
and subsequent displacement with sodium azide provided
the desired C-6 (S)-azido ester 10 as an oil (Scheme 2).
To firmly establish the stereochemistry of the azido ester
10 at C-6 (Scheme 3), the synthesis of the corresponding tri-
acetylated azido ester was envisioned for X-ray crystallogra-
phy purposes. Swapping hydroxyl protecting groups while
keeping the azido group as a masked amine required the ini-
tial removal of the benzyl groups in the presence of BCl3 at
low temperature.[14] Unfortunately, under these conditions, a
fused bistetrahydrofuran derivative A was isolated as the
sole product, the structure of which was confirmed by X-ray
crystallography (Figure 1).[15]
Figure 1. X-ray structure of the fused tetrahydrofuran derivative A.
yield.[17] In a first approach, N-glycosylation with 2-amino-
purine under classical Vorbrüggen conditions[18] that require
persilylation of the nucleobase was studied. Unfortunately,
all attempts to glycosylate 2-aminopurine as its bis-silylated
N-acetyl derivative by using Czernecki[6a] and Garnerꢁs[6b]
procedures (SnCl4, (CH2Cl)2/CH3CN, reflux and trimethyl-
silyl trifluoromethanesulfonate (TMSOTf), (CH2Cl)2, reflux,
respectively) reported with glucose peracetate failed and led
either to decomposition or recovery of the glycosyl donor
with no trace of the expected nucleoside.[19] Hence, the
more reactive 6-chloro-2-aminopurine was used as a masked
2-aminopurine. Coupling of the bis(trimethylsilyl) N-acetyl
derivative of 6-chloro-2-aminopurine 12 with the glycosyl
donor 11 under Vorbrüggenꢁs conditions was examined
(Scheme 4, Table 1).
Glycosylation of purines is rarely regiospecific and usually
produces mixtures of N9 and N7 products despite numerous
studies aimed at finding ways to maximize N9 glycosyla-
tion.[21] While Garner[6b] devised conditions leading mainly
to the thermodynamic N9 regioisomer starting from glucose
Scheme 3. Proposed mechanism for the skeleton rearrangement of azi-
doester 10. a) BCl3 (1m in CH2Cl2), CH2Cl2, À78 to 08C, 65%.
This skeleton rearrangement could be tentatively ex-
plained by a BCl3-induced endocyclic cleavage of the pyra-
nosidic ring to form an acyclic oxonium ion followed by an
intramolecular nucleophilic attack of the hydroxyl group at
the 4-position leading to an inversion of configuration of the
anomeric methoxy group. This problematic ring contraction
should be addressed because the key N-glycosylation step
requires similar Lewis acidic conditions. The azido ester
moiety of compound 10 was not implicated in this transfor-
mation and its S configuration was confirmed at this point.
Introduction of 2-aminopurine, an atypical nucleobase,[16]
was then investigated and required preliminary conversion
of the sugar amino acid precursor into a suitable glycosyl
donor. Acetolysis of the anomeric methoxy group in azido
ester 10 was optimised (Ac2O, 5% conc. H2SO4 in AcOH,
low temperature) to minimize pyranosidic ring contraction
and led to the corresponding glycosyl donor 11 in 39%
Scheme 4. N-glycosylation step. a) Ac2O, H2SO4 5% in AcOH, À20 to
08C, 39%; b) 12, TMSOTf, solvent and temperature (see Table 1).
10068
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 10066 – 10073

First Synthesis of the Core of Miharamycin Antibiotics
FULL PAPER
Table 1. TMSOTf-mediated N-glycosylation of persilylated N2-acetyl 6-
chloropurine 12 with glycosyl donor 11.
Entry[a] Solvent
T [8C] Reaction t [h] 14/13
(N7/N9)[b]
Yield [%]
AHCTREUNG
1
2
3
4
5
6
7
CH3CN
65
31:0
38:1
32:1
16
16
4
55
53
43
A
G
85
85
85
110
85
R
2:1
40
–
32
[c]
[c]
A
–
PhCH3
PhCH3
1:4[d]
4
1:358
[a] All reactions were carried out on a 0.05 mmol scale of sugar substrate
and yielded b-anomers exclusively. [b] The regioselectivity (N7/N9 ratio)
of the N-glycosylation was determined by HMBC.[20] [c] Decomposition
of the nucleobase and the glycosyl donor. [d] Deacetylation of the purine
nucleobase was observed.
Scheme 5. Synthesis of the skeleton of miharamycin B. a) H2, 10% Pd/C,
Et3N, EtOAc; b) 16, isobutyl chloroformate, Et3N, THF, À208C, 70%
over 2 steps.
peracetate, the same conditions applied to the perbenzylated
glycosyl donor 11 afforded the kinetic N7 regioisomer 14 as
the major product (Table 1, entries 3and 4). Since pyranosi-
dic ring contraction was observed either during BCl3-medi-
ated deprotection of benzyl ethers or during acetolysis of
the corresponding peracetylated methyl glycoside, benzyl
groups had to be kept and glycosylation conditions (solvent,
temperature) had to be finely tuned to favour the N9 regio-
isomer. Non-polar solvents and high reaction temperatures
usually lead to the thermodynamic N9 compound as the
major regioisomer,[22] but such harsh conditions (prolonged
reaction times, temperatures above 1008C) resulted in de-
composition of the nucleobase and of the glycosyl donor 11.
Finally, glycosylation performed with TMSOTf in toluene at
858C for 4 h gave the best yield and N9/N7 ratio of the corre-
sponding b-N-nucleoside 13 (entry 7). Noteworthy is the ex-
clusive b-nucleoside formation observed in all cases, which,
in the absence of neighbouring-group participation, can be
tentatively explained by the steric hindrance of the a face of
the pyranosidic ring due to the presence of the bulky tetra-
hydrofuran ring. We next moved to the peptidyl coupling
step that required preliminary reduction of the azido group
in nucleoside 13. This reduction step was further exploited
to also remove the chlorine atom on the nucleobase and
concomitantly unmask the 2-aminopurine moiety. Optimised
hydrogenation (Pd/C) of 13
Unfortunately, removal of the protecting groups to afford
miharamycin B proved to be highly problematic (Scheme 6).
While hydrogenolysis (Pd/C) of the Z groups present on the
l-arginine moiety seemed to occur smoothly, the benzyl
groups present on the bicycle proved reluctant to undergo
hydrogenolysis under these conditions. Deprotection of
model compounds was thus explored. Successful hydroge-
nolysis under similar conditions of nucleobase-free bicyclic
sugar 10 and sugar amino acid 18 was observed to afford tri-
hydroxylated structures 19 and 20, respectively, but the nu-
cleoside 21 was totally inert to hydrogenolysis performed
with 10% Pd/C, Pd black, Pearlman catalyst at 30 psi or cat-
alytic hydrogen transfer, a behaviour also reported with
other benzylated nucleosides.[25] Nevertheless, hydrogenoly-
sis of nucleoside 21 under pressure (400 psi) with 10% Pd/C
afforded the fully deprotected nucleoside 22 after acetamide
and ester hydrolysis. When these hydrogenolysis conditions
(H2, 400 psi, 10% Pd/C, AcOH, 8 h) were applied to pro-
tected miharamycin 17, removal of the Z group on the argi-
nine moiety was observed as well as removal of all but one
of the benzyl groups. Iteration of this process to convert the
remaining monobenzylated compound led to decomposition
of the fully debenzylated molecule.
cleanly afforded the tribenzylat-
ed amino ester 15, which was di-
rectly engaged in the peptide
coupling reaction without purifi-
cation. While classical peptidyl
coupling conditions (1,3-dicyclo-
hexylcarbodiimide, 1-hydroxy-
benzotriazole) gave unsatisfac-
tory yields, activation of the N-
benzyloxycarbonyl-protected l-
arginine 16[23] with isobutyl
chloroformate[24] resulted in the
formation of the core of mihara-
mycin
B 17 in 70% yield
Scheme 6. Debenzylation attempts on model compounds. a) H2 (30 psi), 10% Pd/C, glacial AcOH, 2.5 h; b) H2
(Scheme 5).
(400 psi), 10% Pd/C, glacial AcOH, 8 h then NH4OH/MeOH 1:1, 608C, 24 h, 86%.
Chem. Eur. J. 2008, 14, 10066 – 10073ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10069

Y. BlØriot, P. Sinaþ, A. P. Rauter et al.
layer was further extracted with CH2Cl2 (3”150 mL). The organic layers
were combined, washed with brine, dried (MgSO4), filtered and the sol-
vent removed under reduced pressure. The residue obtained was purified
by flash chromatography (cyclohexane/EtOAc 2:1) to yield the alkene 4
(2.1 g, 95%) as a white solid. Rf =0.40 (cyclohexane/EtOAc 1:1); m.p.
125–1278C (ether/cyclohexane); [a]D =+151.0 (c=1.0 in CHCl3);
Conclusion
We have confirmed the structure of miharamycin A and elu-
cidated for the first time its stereochemistry at C-6’. NMR
spectroscopic conformational analysis further suggested a
folding of the arginine appendage above the sugar ring to-
wards the 2-aminopurine nucleobase. Total synthesis of mi-
haramycin B was also undertaken and has led to the first
synthesis of the core of miharamycins 20, including construc-
tion of its unique bicyclic sugar amino acid moiety as well as
the regio- and stereoselective introduction of the atypical 2-
aminopurine nucleobase. Unfortunately, final deprotection
of this core proved reluctant to all conditions used. Never-
theless, this work has revealed and solved unexpected syn-
thetic problems (ring contraction, N-glycosylation) arising
from the unique structure of the miharamycins. The results
presented herein could be helpful for the construction of re-
lated complex nucleoside natural products, such as amipuri-
mycin,[26] a synthetic target under investigation,[27] but still to
be reached by total synthesis.
1H NMR (400 MHz, CDCl3, 258C, TMS): d=7.55–7.41 (m, 5H;
H
arom.), 5.87 (t, 1H, J8’a,9a =J8’a,9b =2.5 Hz; H-8’a), 5.58 (s, 1H; H-7), 5.10
(t, 1H, J8’b,9a =J8’b,9b =2.5 Hz; H-8’b), 4.84 (d, 1H; J1,2 =5.5 Hz; H-1), 4.72
(m, 2H; H-9a, H-9b), 4.36 (dd, 1H; J6a,5 =5.0, J6a,6b =10.0 Hz; H-6a), 4.16
(d, 1H, J2,1 =5.5 Hz; H-2), 4.08 (d, 1H, J4,5 =10.0 Hz; H-4), 3.95 (dt,
J
5,6a =5.0 Hz, 1H, J5,4 =J5,6b =10.0 Hz; H-5), 3.77 (t, 1H, J6b,5 =J6b,6a =
10.0 Hz; H-6b), 3.41 (s, 3H; OCH3), 2.60 ppm (s, 1H; OH); 13C NMR
(100 MHz, CDCl3, 258C): d=147.5 (C-8), 137.2 (C arom., quat.), 129.2,
128.4, 126.2 (5CH arom.), 108.4 (C-8’), 102.6 (C-7), 100.1 (C-1), 83.7 (C-
2), 82.8 (C-4), 76.3 (C-3), 73.2 (C-9), 69.2 (C-6), 59.7 (C-5), 55.5 ppm
(OCH3); elemental analysis calcd (%) for C17H20O6: C 63.74, H 6.29;
found: C 63.48, H 6.29.
Bicycle 5: Alkene
4 (0.75 g, 2.34 mmol) was dissolved in CH2Cl2
(120 mL) and cooled to À788C (note: this reaction cannot be performed
on a large scale (above 1 g) since monitoring of the reaction (detection
of the blue coloration) is difficult and results in a partial reaction or over-
oxidation of the product). Ozone was bubbled through the reaction mix-
ture until a blue coloration persisted (typically 8 min). Dimethylsulfide
(0.05 mL) was added and the mixture was allowed to warm to room tem-
perature over a period of 1 h. The solvent was then removed under re-
duced pressure and the crude ketone was used in the next step without
further purification. Sodium borohydride (98 mg, 2.58 mmol) was added
carefully to a solution of the crude ketone in ethanol (50 mL) at 08C.
The reaction mixture was allowed to reach room temperature. After 1 h,
TLC analysis indicated the complete consumption of starting material
and the formation of a single compound. Methanol (50 mL) was added,
the solvent was removed under reduced pressure and the residue was co-
evaporated with methanol (3”25 mL). Purification by flash column chro-
matography (cyclohexane/EtOAc 2:1 1:1) afforded the corresponding
diol (0.64 g, 84% over two steps) as a white crystalline solid. Rf =0.20
(cyclohexane/EtOAc 1:1); m.p. 167–1698C (CH2Cl2/cyclohexane); [a]D =
+94 (c=1.0 in CHCl3); 1H NMR (400 MHz, CDCl3, 258C, TMS): d=
7.58–7.41 (m, 5H; H arom.), 5.55 (s, 1H; H-7), 4.92 (d, 1H, J1,2 =5.5 Hz;
Experimental Section
NMR spectroscopic studies: NMR spectroscopic experiments in D2O so-
lution were carried out at 500 and 700 MHz and at 288 and 298 K. A con-
centration of ca. 18 mm of miharamycin A was used. A complete assign-
ment of the 1H NMR resonance signals of miharamycin A was achieved
on the basis of TOCSY, HMQC and HMBC experiments. NMR spectro-
scopic experiments were also performed in a mixture of H2O/D2O 90:10
to detect the NH resonance. In this case, a sample concentration ca. of
7 mm was used and the experiments were recorded at 500 MHz, 290 K
and with the Watergate pulse sequence for water suppression.
H-1), 4.42 (dd, 1H, J9a,8 =4.5, J9a,9b =9.5 Hz; H-9a), 4.40 (d, 1H, JOH,8
11.5 Hz; OH), 4.35 (dd, 1H, J6a,5 =5.0, J6a,6b =10.0 Hz; H-6a), 4.25 (dt,
1H, J5,6a =5.0, J5,4 =J5,6b =10.0 Hz; H-5), 4.21 (dd, 1H, J8,9a =4.5, J8,OH
11.5 Hz; H-8), 4.10 (d, 1H, J2,1 =5.5 Hz; H-2), 3.99 (d, 1H, J9b,9a =9.5 Hz;
H-9b), 3.92 (d, 1H, J4,5 =10.0 Hz; H-4), 3.68 (t, 1H, J6b,5 =J6b,6a =10.0 Hz;
H-6b), 3.58 (s, 3H; OCH3), 2.63ppm (s, 1H; OH); 13C NMR (100 MHz,
CDCl3, 258C): d=137.0 (C arom., quat.), 129.5, 128.4, 126.5 (5CH
arom.), 103.0 (C-7), 99.2 (C-1), 83.3 (C-4), 88.9 (C-2), 79.9 (C-3), 78.5 (C-
9), 77.1 (C-8), 69.9 (C-6), 60.5 (C-5), 55.8 ppm (OCH3); elemental analy-
sis calcd (%) for C16H20O7: C 59.25, H 6.22; found: C 59.19, H 6.22.
=
General methods: Melting points were determined with a Büchi B-510
capillary apparatus and are uncorrected. Optical rotations were measured
at (20Æ2)8C with a Perkin–Elmer Model 241 digital polarimeter, by
using a 10 cm, 1 mL cell. Mass spectra (CI (ammonia) and FAB) were
obtained with a JMS-700 spectrometer. Elemental analysis were per-
formed by the service d’analyse de l’UniversitØ Pierre et Marie Curie,
75252 Paris cedex 05 (France). 1H NMR spectra were recorded at
400 MHz with a Brüker DRX 400 for solutions in CDCl3, [D6]DMSO or
D2O at room temperature. Assignments were confirmed by COSY ex-
periments. Multiplicity is indicated as follows: s (singlet), d (doublet), t
(triplet), dd (doublet of doublets), brs (broad singlet) etc. 13C NMR spec-
tra were recorded at 100.6 MHz with a Brüker DRX 400 spectrometer.
Assignments were confirmed by the J-mod technique, HMQC and
HMBC. Reactions were monitored by TLC on a precoated plate of Silica
Gel 60 F254 (layer thickness 0.2 mm; E. Merck, Darmstadt, Germany)
and detection by charring with H2SO4 10% in EtOH or with 0.2% w/v
cerium sulfate and 5% ammonium molybdate in 2m H2SO4. Flash
column chromatography was performed on silica gel 60 (230–400 mesh,
E. Merck).
=
Sodium hydride (2.30 g, 45.35 mmol, 60% w/w) was added in portions to
a solution of the diol (3.68 g, 11.35 mmol) in anhydrous DMF (40 mL) at
08C whilst stirring. After 30 min, BnBr (8.10 mL, 68.09 mmol) was added
at 08C to the solution, which was then stirred for a further 2.5 h at room
temperature. MeOH (30 mL) was added and the mixture was concentrat-
ed under reduced pressure. The residue was then diluted with ether
(200 mL) and water (80 mL). The aqueous layer was extracted with ether
(3”150 mL), the organic layers were combined, dried (MgSO4), filtered
and concentrated under reduced pressure. Purification by flash chroma-
tography (cyclohexane/EtOAc 5:1) afforded the benzylated compound 5
(5.45 g, 95%) as a colourless oil. Rf =0.40 (cyclohexane/EtOAc 3:1);
[a]D =+73.6 (c=1.0 in CHCl3); 1H NMR (400 MHz, CDCl3, 258C,
TMS): d=7.48–7.24 (m, 15H; H arom.), 5.48 (s, 1H; H-7), 4.94 (d, 1H,
Tricyclic exoalkene 4: A suspension of samarium(II) iodide (170 mL of a
0.1m solution in THF, 17.28 mmol), HMPA (13.8 mL, 79.49 mmol) and
tBuOH (2.2 mL, 22.80 mmol) were stirred at room temperature under
argon. A solution of ketone 3 (2.2 g, 6.91 mmol) in dry THF (100 mL)
was added by syringe and the resulting mixture was stirred at room tem-
perature for 45 min under argon. At this point, TLC analysis indicated
the complete consumption of starting material and the formation of a
single compound. Dilute HCl (20 mL, 0.1m) was added and the reaction
mixture was then filtered through a Celite plug. The reaction mixture
was concentrated under reduced pressure and the residue obtained was
then diluted with CH2Cl2 (250 mL) and water (75 mL). The aqueous
J
1,2 =5.8 Hz; H-1), 4.85 (d, 1H, J=12.2 Hz; CHPh), 4.77 (d, 1H, J=
12.3Hz; CHPh), 4.75 (d, 1H, J=12.2 Hz; CHPh), 4.66 (d, 1H, J=
12.3Hz; CHPh), 4.55 (dt, 1H, 5,6a =5.1, J5,4 =J5,6b =10.8 Hz; H-5), 4.38
J
(dd, 1H, J8,9a =2.1, J 8,9b =4.9 Hz; H-8), 4.36–4.28 (m, 3H; H-6a, H-9a, H-
9b), 4.27 (d, 1H; J2,1 =5.8 Hz; H-2), 4.11 (d, 1H, J4,5 =10.8 Hz; H-4), 3.67
(t, 1H, J6b,5 =J6b,6a =10.8 Hz; H-6b), 3.46 ppm (s, 3H; OCH3); 13C NMR
(100 MHz, CDCl3, 258C): d=138.8, 138.6, 137.6 (C arom., quat.), 129.1–
10070
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 10066 – 10073

First Synthesis of the Core of Miharamycin Antibiotics
FULL PAPER
126.2 (15CH arom.), 102.1 (C-7), 99.5 (C-1), 86.0 (C-3), 82.6 (C-2), 81.9
(C-4), 81.5 (C-8), 75.6 (C-9), 72.6 (CH2Ph), 70.0 (C-6), 67.0 (CH2Ph), 59.3
(C-5), 55.2 ppm (OCH3); HRMS (CIMS): m/z: calcd for C30H36O7N:
522.2486 [M+NH4]+; found: 522.2480; elemental analysis calcd (%) for
C30H32O7: C 71.41, H 6.39; found: C 71.27, H 6.48.
elution afforded the allylic alcohol (6R)-8 (268 mg, 26%) as a colourless
oil. Rf =0.59 (cyclohexane/EtOAc 1:1); [a]D =+94.2 (c=10 in CHCl3);
1H NMR (400 MHz, CDCl3, 258C, TMS): d=7.39–7.27 (m, 15H; H
arom.), 6.00 (appddd, 1H, J7,6 =6.3, J7,8a =10.4, J7,8b =17.0 Hz; H-7), 5.35
(dt, 1H, J8a,8b =J8a,6 =1.2, J8a,7 =17.0 Hz; H-8a), 5.26 (dt, 1H, J8b,8a =J8b,6
=
1.5, J8b,7 =10.4 Hz; H-8b), 4.97 (d, 1H, J1,2 =5.8 Hz; H-1), 4.73(d, 1H, J=
10.9 Hz; CHPh), 4.68 (s, 2H; CH2Ph), 4.66 (d, 1H, J=12.0 Hz; CHPh),
4.57 (d, 1H, J=12.0 Hz; CHPh), 4.48 (d, 1H, J2,1 =5.8 Hz; H-2), 4.46–
4.42 (m, 2H; H-4, H-6), 4.22 (dd, 1H, J10b,9 =1.7, J10b,10a =8.7 Hz; H-10b),
4.08 (dd, 1H, J5,6 =2.3, J5,4 =11.7 Hz; H-5), 3.44 (s, 3H; OCH3), 2.68 ppm
(d, 1H, JOH,6 =5.1 Hz; OH); 13C NMR (100 MHz, CDCl3, 258C): d=
138.7, 138.1, 138.0, (C arom., quat.), 136.4 (C-7), 128.4–127.0 (15CH
arom.), 116.8 (C-8), 98.9 (C-1), 89.9 (C-3), 85.3 (C-9), 77.6 (C-2), 76.5 (C-
5), 75.2 (C-10), 74.3(CH 2Ph), 73.6 (C-4 or C-6), 72.9 (CH2Ph), 70.1 (C-4
or C-6), 65.0 (CH2Ph), 55.0 ppm (OCH3); HRMS (CIMS): m/z: calcd for
C32H40O7N: 550.2799 [M+NH4]+; found: 550.2801; elemental analysis
calcd (%) for C32H36O7: C 71.98, H 6.81; found: C 71.90, H 6.63.
Alcohol 6: LiAlH4 (1.88 g, 49.54 mmol) was carefully added in three por-
tions to a solution of compound 5 (5.0 g, 9.91 mmol) in a CH2Cl2/Et2O
mixture 1:1 (100 mL) at 08C under argon. After 10 min, the reaction mix-
ture was warmed to 508C, and a solution of AlCl3 (3.96 g, 29.73 mmol) in
Et2O (50 mL) was added dropwise under argon. After the reaction mix-
ture had been stirred for 2.5 h, TLC analysis revealed complete consump-
tion of the starting material. The reaction mixture was cooled to 08C and
quenched by slow addition of EtOAc followed by water. The organic
layer was separated, dried (MgSO4), filtered and concentrated under re-
duced pressure. Purification by flash column chromatography (cyclohex-
ane/EtOAc 3:1 then 1:1) afforded alcohol 6 (4.38 g, 88%) as a colourless
oil. Rf =0.30 (cyclohexane/EtOAc 1:1); [a]D =+123.2 (c=1.0 in CHCl3);
1H NMR (400 MHz, CDCl3, 258C, TMS): d=7.38–7.27 (m, 15H; H
arom.), 4.94 (d, 1H, J1,2 =5.9 Hz; H-1), 4.95 (d, 1H, J=11.3Hz; CHPh),
4.78 (d, 1H, J=11.3Hz; CHPh), 4.67 (d, 1H, J=12.2 Hz; CHPh), 4.66
(d, 1H, J=11.9 Hz; CHPh), 4.63(d, 1H, J=11.9 Hz; CHPh), 4.57 (d,
1H, J=12.2 Hz; CHPh), 4.38–4.30 (m, 4H; H-2, H-5, H-7, H-8a), 4.23
(dd, 1H, J8b,7 =6.1, J8b,8a =13.7 Hz; H-8b), 4.17 (d, 1H, J4,5 =10.1 Hz; H-
4), 3.90 (ddd, 1H, J6a,5 =2.9, J6a,OH =4.9, J6a,6b =11.6 Hz; H-6a), 3.83 (ddd,
1H, J6b,5 =3.7, J6b,OH =7.9, J6b,6a =11.6 Hz; H-6b), 3.44 (s, 3H; OCH3),
Hydroxy ester 9: Allylic alcohol (6S)-7 (0.80 g, 1.50 mmol) was dissolved
in CH2Cl2 (120 mL) and cooled to À788C. Ozone was bubbled through
the reaction mixture until
a blue colour persisted. Dimethylsulfide
(0.1 mL) was added and the reaction mixture was allowed to reach room
temperature. The solvent was evaporated and the crude aldehyde was
used in the next step without further purification. NaH2PO4·H2O (2.69 g,
19.5 mmol) and NaClO2 (2.44 g, 27 mmol) were added to a solution of
the crude aldehyde in 2-methyl-but-2-ene (3mL), tBuOH (6 mL) and
water (6 mL). After stirring overnight at room temperature, the reaction
mixture was diluted with EtOAc (50 mL) and water (20 mL). The aque-
ous layer was extracted with EtOAc (3”50 mL). The organic layers were
combined, dried (MgSO4), filtered and concentrated. The crude carboxyl-
ic acid obtained was directly engaged in the next step. To a solution of
the crude carboxylic acid in DMF (8 mL) was added KHCO3 (1.50 g,
15 mmol) followed by the slow addition of methyl iodide (0.79 mL,
12.75 mmol). After the reaction mixture had been stirred overnight, the
solvent was removed and the crude residue diluted with Et2O (80 mL)
and water (30 mL). The aqueous layer was extracted with ether (3”
80 mL) and the organic layers were combined, dried (MgSO4), filtered
and concentrated under reduced pressure. Purification by flash chroma-
tography (cyclohexane/EtOAc 3:1) afforded the methyl ester (6R)-9
(550 mg, 65% over three steps) as a colourless oil; Rf =0.25 (cyclohex-
ane/EtOAc 2:1); [a]D =+80.5 (c=1.0 in CHCl3); 1H NMR (400 MHz,
CDCl3, 258C, TMS): d=7.41–7.32 (m, 15H; H arom.), 5.03 (d, 1H, J=
11.2 Hz; CHPh), 4.96 (d, 1H, J1,2 =5.8 Hz; H-1), 4.86 (d, 1H, J=11.2 Hz;
CHPh), 4.82 (dd, 1H, J5,6 =1.6, J5,4 =10.1 Hz; H-5), 4.75 (d, 1H, J=
11.9 Hz; CHPh), 4.69 (s, 2H; CH2Ph), 4.65–4.59 (m, 2H; H-6, CHPh),
4.41 (d, 1H, J4,5 =10.1 Hz; H-4), 4.39–4.35 (m, 3H; H-2, H-8, H-9a),
4.28–4.24 (m, 1H; H-9a), 3.37 (s, 3H; OCH3), 3.12 (s, 3H; OCH3),
1.83ppm (dd, 1H,
J
OH,6a =4.9, JOH,6b=7.9 Hz; OH); 13C NMR (100 MHz,
CDCl3, 258C): d=139.0, 139.0, 138.2 (C arom., quat.), 128.4–127.0
(15CH arom.), 99.1 (C-1), 89.6 (C-3), 85.0 (C-7), 78.3(C-2), 75.3(C-8),
74.8 (C-4), 74.5 (CH2Ph), 72.8 (CH2Ph), 68.3(C-5), 64.8 (CH 2Ph), 62.2
(C-6), 55.0 ppm (OCH3); HRMS (CIMS): m/z: calcd for C30H38O7N:
524.2643[ M+NH4]+; found: 524.2640; elemental analysis calcd (%) for
C30H34O7: C 71.13, H 6.76; found: C 70.85, H 7.01.
Allylic alcohols 7 and 8: Anhydrous DMSO (0.82 mL, 11.61 mmol) was
added dropwise to
a stirred solution of oxalyl chloride (0.84 mL,
9.67 mmol) in anhydrous CH2Cl2 (30 mL) at À788C under argon. After
15 min, a solution of alcohol 6 (0.98 g, 1.93mmol) in anhydrous CH 2Cl2
(20 mL) was added dropwise. After 1 h, dry Et3N (2.16 mL, 15.48 mmol)
was added and the reaction mixture was allowed to reach room tempera-
ture. After 1.5 h, water was added (20 mL) and the aqueous layer was ex-
tracted with CH2Cl2 (3”80 mL). The organic layers were combined,
dried (MgSO4), filtered and concentrated under reduced pressure. The
crude aldehyde was coevaporated with toluene and used without further
purification. Crude aldehyde was dissolved in dry THF (10 mL) and the
solution cooled to À788C under argon. Vinyl magnesium bromide
(9.65 mL, 9.65 mmol, 1m in THF) was added dropwise to the solution
and the reaction mixture was allowed to reach room temperature. After
1 h, the reaction mixture was quenched by the slow addition of a saturat-
ed aqueous NH4Cl solution (10 mL) at 08C and was then diluted with
ether (50 mL). The organic layer was separated, and the aqueous layer
extracted with ether (3”50 mL). The organic layers were combined,
dried over MgSO4, filtered and concentrated. Purification by flash
column chromatography (cyclohexane/EtOAc 3:1 then 2:1) afforded the
allylic alcohol (6S)-7 (402 mg, 39%) as a colourless oil. Rf =0.66 (cyclo-
hexane/EtOAc 1:1); [a]D =+101.5 (c=10 in CHCl3); 1H NMR
(400 MHz, CDCl3, 258C, TMS): d=7.38–7.29 (m, 15H; H arom.), 5.96
3.12 ppm (d, 1H,
J
OH,6 =7.6 Hz; OH); 13C NMR (100 MHz, CDCl3,
258C): d=173.6 (C=O), 138.9, 138.5, 138.1 (C arom., quat.), 128.3,
À126.9 (15CH arom.), 99.4 (C-1), 89.6 (C-3), 85.0 (C-2 or C-8), 77.8 (C-2
or C-8), 75.1 (C-9), 74.6 (CH2Ph), 74.2 (C-4), 72.7 (CH2Ph), 69.3(C-5
and C-6), 64.7 (CH2Ph), 54.9 (OCH3), 52.2 ppm (OCH3); HRMS
(CIMS): m/z: calcd for C32H40O9N: 582.2698 [M+NH4]+; found:
582.2694; elemental analysis calcd (%) for C32H36O9: C 68.07, H 6.43;
found: C 68.27, H 6.58.
Azido ester 10: Dry pyridine (1.40 mL, 17.28 mmol) was added to a solu-
tion of methyl ester (6R)-9 (610 mg, 1.08 mmol) in dry CH2Cl2 (18 mL),
followed by slow addition of Tf2O (1.45 mL, 8.64 mmol) at À788C under
argon. After the reaction mixture had been stirred for 90 min at room
temperature, water (10 mL) and CH2Cl2 (20 mL) were added. The organ-
ic layer was separated, dried (MgSO4), filtered and concentrated under
reduced pressure. The crude triflate was directly engaged in the next
step. Sodium azide (420 mg, 6.48 mmol) was added to a solution of the
crude triflate in anhydrous DMF (12 mL) at room temperature. After
stirring overnight, the reaction mixture was concentrated under reduced
pressure. The residue was then diluted with ether (80 mL) and water
(30 mL). The aqueous layer was extracted with ether (3”80 mL) and the
organic layers were combined, dried (MgSO4), filtered and concentrated
under reduced pressure. Purification by flash chromatography (cyclohex-
ane/EtOAc 4:1) afforded azido ester (6S)-10 (522 mg, 82% over two
(ddd, 1H; J7,6 =5.1, J7,8b =10.6, J7,8a =17.3Hz; H-7), 5.33(dt, 1H, J8a,8b
8a,6 =1.5, 8a,7 =17.3Hz; H-8a), 5.23 (dt, 1H, 8b,8a =J8b,6 =1.5, J8b,7
10.6 Hz; H-8b), 4.99 (d, 1H, J=11.2 Hz; CHPh), 4.94 (d, 1H, J1,2
=
=
=
J
J
J
6.0 Hz; H-1), 4.82 (d, 1H, J=11.2 Hz; CHPh), 4.67 (d, 1H, J=11.8 Hz;
CHPh), 4.66 (s, 2H; CH2Ph), 4.58 (d, 1H, J=11.8 Hz; CHPh), 4.49–4.45
(m, 1H; H-6), 4.37 (dd, 1H, J5,6 =1.4, J5,4 =10.2 Hz; H-5), 4.34–4.19 (m,
5H; H-2, H-4, H-9, H-10a, H-10b), 4.01 (d, 1H, JOH,6 =9.1 Hz; OH),
3.38 ppm (s, 3H; OCH3); 13C NMR (100 MHz, CDCl3, 258C): d=139.0
(C arom., quat.), 138.8 (C-7), 138.6, 138.3 (C arom., quat.), 128.4–127.1
(15CH arom.), 115.3(C-8), 99.3(C-1), 90.0 (C-3), 85.3(C-9), 78.1 (C-2),
75.3(C-10), 75.0 (C-5), 74.2 (CH 2Ph), 72.9 (CH2Ph), 70.5 (C-6), 70.0 (C-
4), 64.8 (CH2Ph), 55.1 ppm (OCH3); HRMS (CIMS): m/z: calcd for
C32H40O7N: 550.2799 [M+NH4]+; found: 550.2803; elemental analysis
calcd (%) for C32H36O7: C 71.98, H 6.81; found: C 72.16, H 6.81; further
Chem. Eur. J. 2008, 14, 10066 – 10073ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10071

Y. BlØriot, P. Sinaþ, A. P. Rauter et al.
steps) as a colourless oil; Rf =0.58 (cyclohexane/EtOAc 2:1); [a]D =+
89.3( c=1.0 in CHCl3); 1H NMR (400 MHz, CDCl3, 258C, TMS): d=
7.42–7.29 (m, 15H; H arom.), 5.02 (d, 1H, J1,2 =5.7 Hz; H-1), 5.01 (dd,
1H, J5,6 =2.5, J5,4 =10.3Hz; H-5), 4.97 (d, 1H, J=10.0 Hz; CHPh), 4.74–
4.67 (m, 3H; CH2Ph, CHPh), 4.68 (d, 1H, J=12.3Hz; CHPh), 4.53 (d,
1H, J=12.3 Hz; CHPh), 4.43–4.41 (m, 2H; H-2, H-6), 4.38 (dd, 1H,
1H; NH), 7.30–7.19 (m, 15H; H arom.), 6.08 (d, 1H, J1’,2’ =8.9 Hz; H-1’),
4.95 (d, 1H, J=9.9 Hz; CHPh), 4.86 (brd, 1H, J2’,1’ =8.9 Hz; H-2’), 4.79
(dd, 1H, J5’,6’ =2.3, J5’,4’ =9.8 Hz; H-5’), 4.74 (d, 1H, J=11.2 Hz; CHPh),
4.70 (d, 1H, J=11.2 Hz; CHPh), 4.58 (d, 1H, J=11.3Hz; CHPh), 4.57–
4.55 (m, 2H; H-4’, CHPh), 4.52 (d, 1H, J=11.3Hz; CHPh), 4.37 (d, 1H,
J
6’,5’ =2.3Hz; H-6 ’), 4.27 (d, 1H, J8’,9’a =3.5 Hz; H-8’), 4.45 (dd, 1H, J9’a,8’ =
J
9a,8 =5.5, J9a,9b =8.9 Hz; H-9a), 4.36 (d, 1H, J4,5 =10.3Hz; H-4), 4.31 (dd,
3.5, J9’a,9’b =9.5 Hz; H-9’a), 4.04 (d, 1H, J9’b,9’a =9.5 Hz; H-9’b), 3.36 (s,
3H; OCH3), 2.41 ppm (s, 3H; NHAc); 13C NMR (100 MHz, CDCl3,
258C): d=170.5, 167.2 (2C=O), 152.6 (C-2 or C-6), 152.2 (C-4), 151.7 (C-
2 or C-6), 142.5 (C-8), 137.5, 137.1, 137.0 (C arom., quat.), 128.7–127.4
(15CH arom.), 128.1 (C-5), 89.86 (C-3’), 85.1 (C-8’), 82.8 (C-1’), 77.4 (C-
2’ or C-5’), 76.9 (C-2’ or C-5’), 74.8 (CH2Ph), 74.7 (C-9’), 73.5 (CH2Ph),
72.7 (C-4’), 65.4 (CH2Ph), 63.1 (C-6’), 52.6 (OCH3), 25.1 ppm (NHAc);
HRMS (FAB): m/z: calcd for C38H37O8N8 Na35Cl, C38H37O8N8 Na37Cl:
791.2321 [M+Na]+, 793.2291; found: 791.2328, 793.2301; further elution
(cyclohexane/EtOAc 1:1 then 1:2) afforded the N7 nucleoside 14 (4.5 mg,
15%) as a colourless oil. Rf =0.16 (cyclohexane/EtOAc 1:1); [a]D =+35
(c=0.3in CHCl 3); 1H NMR (400 MHz, CDCl3, 258C, TMS): d=8.45 (s,
1H; H-8), 8.05 (brs, 1H; NH), 7.41–7.30 (m, 15H; H arom.), 6.54 (d, 1H,
1H, J8,9a =5.4, J8,9b =1.7 Hz; H-8), 4.26 (dd, 1H, J9b,8 =1.7, J9b,9a =8.9 Hz;
H-9b), 3.56 (s, 3H; OCH3), 3.29 ppm (s, 3H; OCH3); 13C NMR
(100 MHz, CDCl3, 258C): d=167.8 (C=O), 138.9, 138.0, 137.8 (C arom.,
quat.), 128.4–126.6 (15CH arom.), 99.6 (C-1), 89.8 (C-3), 85.6 (C-8), 77.2
(C-2), 75.3(C-9), 74.4 (CH Ph), 73.7 (C-4), 72.7 (CH2Ph), 68.9 (C-5), 64.5
2
(CH2Ph), 63.1 (C-6), 55.5 (OCH3), 52.0 ppm (OCH3); HRMS (CIMS):
m/z: calcd for C32H39O8N4: 607.2762 [M+NH4]+; found: 607.2761.
Bistetrahydrofuran A: Boron trichloride (3.4 mL, 3.40 mmol, 1m in
CH2Cl2) was added dropwise to a stirred solution of azido ester (6S)-10
(100 mg, 0.17 mmol) in dry CH2Cl2 (4.5 mL) at À788C under argon. The
reaction mixture was allowed to reach 08C over a period of 7 h, and was
then quenched with a solution of CH2Cl2/MeOH 1:1 (12 mL) and concen-
trated under reduced pressure. Purification by flash chromatography (cy-
clohexane/EtOAc 1:2) afforded the azido ester (6S)-11 (35 mg, 65%) as
a white crystalline solid. Rf =0.30 (cyclohexane/EtOAc 1:2); m.p. 828C
(EtOAc); [a]D =À26.7 (c=1.0 in MeOH); 1H NMR (400 MHz, CDCl3,
258C, TMS): d=4.89 (s, 1H; H-1), 4.74 (d, 1H, J5,4 =5.6 Hz; H-5), 4.45
(t, 1H, J8,9a =J8,9b =5.6 Hz; H-8), 4.34–4.30 (m, 2H; H-4, H-6), 4.22 (s,
J
1’,2’ =8.8 Hz; H-1’), 5.05 (d, 1H, J=10.1 Hz; CHPh), 4.87 (dd, 1H, J5’,6’ =
2.2, J5’,4’ =9.7 Hz; H-5’), 4.81 (d, 1H, J=11.0 Hz; CHPh), 4.76 (d, 1H, J=
11.0 Hz; CHPh), 4.66–4.55 (m, 5H; H-2’, H-4’, CHPh, CH2Ph), 4.50 (d,
1H, J6’,5’ =2.2 Hz; H-6’), 4.39 (d, 1H, J8’,9’a =3.3 Hz; H-8’), 4.35 (dd, 1H,
J
9’a,8’ =3.3, J9’a,9’b =9.4 Hz; H-9’a), 4.19 (d, 1H, J9’b,9’a =9.4 Hz; H-9’b), 3.47
(s, 3H; OCH3), 2.63ppm (s, 3H; NHAc); 13C NMR (100 MHz, CDCl3,
258C): d=171.0, 167.1 (2C=O), 163.2 (C-4), 152.6 (C-2 or C-6), 145.8 (C-
8), 143.9 (C-2 or C-6), 137.5, 136.6, 136.8 (C arom. quat.), 128.8–127.4
(15CH arom.), 119.1 (C-5), 90.0 (C-3’), 84.9 (C-8’), 83.0 (C-1’), 77.5 (C-
2’), 77.3(C-5 ’), 74.7 (CH2Ph), 74.4 (C-9’), 73.6 (CH2Ph), 72.7 (C-4’), 65.5
(CH2Ph), 63.1 (C-6’), 52.7 (OCH3), 25.2 ppm (NHAc); HRMS (FAB):
m/z: calcd for C38H37O8N8 Na35Cl, C38H37O8N8 Na37Cl: 791.2321
[M+Na]+, 793.2291; found: 791.2317, 793.2277.
1H; H-2), 4.13(dd, 1H,
J9a,8 =5.6, J9a,9b =9.5 Hz; H-9a), 3.88 (s, 3H;
OCH3), 3.84 (dd, 1H, J9b,8 =5.6, J9b,9a =9.5 Hz; H-9b), 3.46 ppm (s, 3H;
OCH3), 13C NMR (100 MHz, CDCl3, 258C): d=170.0 (C=O), 108.9 (C-
1), 92.9 (C-2), 90.7 (C-3), 79.5 (C-4), 78.8 (C-8), 74.2 (C-5), 74.1 (C-9),
63.4 (C-6), 56.4 (OCH3), 53.3 ppm (OCH3); HRMS (CIMS): m/z: calcd
for C11H21O8N4: 337.1359 [M+NH4]+; found: 337.1343.
Glycosyl donor 11: A solution of sulfuric acid (5% in AcOH, 70 mL) was
added dropwise at À208C to a solution of azido ester (6S)-10 (150 mg,
254 mmol) in acetic anhydride (5 mL). The reaction mixture was stirred
at À208C for 15 min and then neutralized by the slow addition of a
NaHCO3 saturated aqueous solution. The organic layer was separated
and the aqueous layer was extracted with CH2Cl2 (3”50 mL). The organ-
ic layers were combined, dried (MgSO4), filtered and concentrated. Pu-
rification by flash chromatography (cyclohexane/CH2Cl2/EtOAc 8:1:1) af-
forded the glycosyl donor (6S)-11 (61 mg, 39%) as a colourless oil. Rf =
0.44 (cyclohexane/EtOAc/CH2Cl2 4:1:1); [a]D =+23.9 (c=1.0 in CHCl3);
1H NMR (400 MHz, CDCl3, 258C, TMS): d=7.37–7.22 (m, 15H; H
arom.), 6.09 (d, 1H, J1,2 =7.7 Hz; H-1), 4.94 (d, 1H, J=9.7 Hz; CHPh),
4.79 (dd, 1H, J5,6 =2.4, J5,4 =9.7 Hz; H-5), 4.69 (s, 2H; CH2Ph), 4.59–4.56
(m, 3H; CH2Ph, CHPh), 4.49 (d, 1H, J6,5 =2.4 Hz; H-6), 4.41 (d, 1H,
Protected miharamycin 17: 10% Pd/C (18 mg) and Et3N (5 mL) were
added to a solution of the N9 nucleoside 13 (18 mg, 45 mmol) in EtOAc
(1 mL). The solution was degassed three times and the air replaced by
H2. After stirring for 4 h at room temperature, the mixture was filtered
through a Rotilabo Nylon 0.45 mm filter and the solvent evaporated
under reduced pressure. The crude amine 15 was directly engaged in the
next step. Triethylamine (16 mL, 115 mmol) was added at À208C under
argon to a solution of protected l-arginine 16 (53mg, 92 mmol) in anhy-
drous THF (1 mL) followed by slow addition of isobutylchloroformate
(12 mL, 92 mmol). This solution was stirred for 45 min at À208C and a so-
lution of the crude amine in dry THF (1 mL) was added at À208C under
argon. After the reaction mixture had been stirred for 2 h at À208C
under argon, the reaction was quenched by the addition of CH2Cl2/
MeOH 1:1 (1 mL) and the solvent removed under reduced pressure. Pu-
rification by flash chromatography (CH2Cl2/EtOAc/MeOH 8:2:0.1) af-
forded the core of miharamycin B 17 (20 mg, 70% over two steps) as a
colourless oil. Rf =0.24 (CH2Cl2/EtOAc/MeOH 1:1); [a]D =+8 (c=0.2 in
CHCl3); 1H NMR (400 MHz, CDCl3, 258C, TMS): d=9.42, 9.30, 8.96
(brs, 3H; 3”NH), 8.87 (s, 1H; H-6), 8.80 (s, 1H; H-8), 7.40–7.21 (m,
J
4,5 =9.7 Hz; H-4), 4.28 (d, 1H, J2,1 =7.7 Hz; H-2), 4.22–4.20 (m, 2H; H-8,
H-9a), 4.04 (brd, 1H, J9b,9a =9.2 Hz; H-9b), 3.38 (s, 3H; OCH3), 2.19 ppm
(s, 3H; OAc); 13C NMR (100 MHz, CDCl3, 258C): d=167.4, 167.6 (2”
C=O), 137.7, 137.4, 137.3 (C arom., quat.), 128.6–127.5 (15CH arom.),
93.5 (C-1), 90.1 (C-3), 84.6 (C-8), 77.5 (C-2), 74.8 (C-5), 74.6 (C-9), 73.8
(CH2Ph), 72.8 (C-4), 72.6 (CH2Ph), 65.0 (CH2Ph), 63.1 (C-6), 52.3
(OCH3), 21.1 ppm (OAc); HRMS (CIMS): m/z: calcd for C33H39O9N4:
635.2712 [M+NH4]+; found: 635.2719.
30H; H arom.), 6.02 (d, 1H, J1’,2’ =8.7 Hz; H-1’), 5.95 (d, 1H, JNH,H6’
=
8.3Hz; NH), 5.20 (d, 1H, J=12.2 Hz; CHPh), 5.16 (d, 1H, J=12.2 Hz;
CHPh), 5.13–4.97 (m, 7H; H-2’, H-6’, CHPh, 2CH2Ph), 4.79–4.57 (m,
7H; H-4’, H-5’, CHPh, 2CH2Ph), 4.39–2.29 (m, 1H; H-2“), 4.30 (d, 1H,
Nucleoside 13: N,O-bis(trimethylsilyl)acetamide (30 mL, 123 mmol) was
added to a suspension of 2-acetamido-6-chloropurine (13mg, 61 mmol) in
dry 1,2-dichloroethane (1 mL) under argon. The mixture was heated to
808C for 45 min and then evaporated to dryness to afford crude com-
pound 12. This residue was then dissolved in dry toluene (0.5 mL) and a
solution of the glycosyl donor 11 (25 mg, 41 mmol) in dry toluene
(0.5 mL) was added, followed by the slow addition of TMSOTf (59 mL,
328 mmol) under argon. The reaction mixture was stirred at 858C for 4 h,
cooled to room temperature and then diluted with CH2Cl2 (10 mL). The
organic layer was neutralized with a NaHCO3 saturated aqueous solution
(2”5 mL), dried (MgSO4), filtered and concentrated under reduced pres-
sure. Purification by flash chromatography (cyclohexane/EtOAc 3:1 then
2:1) afforded the N9 nucleoside 13 (13.5 mg, 43%) as a colourless oil.
J
8’,9’a =3.2 Hz; H-8’), 4.22 (dd, 1H, J9’a,8’ =3.2, J9’a,9’b =9.4 Hz; H-9’a), 4.04
(d, 1H, J9’b,9’a=9.4 Hz; H-9’b), 3.98–3.75 (m, 2H; H-5”a, H-5“b), 3.53 (s,
3H; OCH3), 2.19 (s, 3H; NHAc), 1.82–1.48 (m, 4H; H-3”a, H-3“b, H-
4”a, H-4“b); 13C NMR (100 MHz, CDCl3, 258C): d=171.8, 171.0, 168.8,
163.5 (4”C=O), 160.8 (C=NH), 156.4, 155.8 (2”C=O), 152.9 (C-2), 152.2
(C-4), 149.8 (C-6), 143.0 (C-8), 137.7, 137.5, 137.0, 136.4, 136.3, 134.5 (C
arom., quat.), 131.1 (C-5), 128.8–127.2 (30CH arom.), 89.9 (C-3’), 84.6
(C-8’), 83.0 (C-1’), 77.5 (C-5’ or C-4’), 76.3(C-2 ’), 74.7 (CH2Ph), 74.2 (C-
9’), 73.6 (C-4’ or C-5’), 73.8 (CH2Ph), 68.9 (CH2Ph), 67.1 (CH2Ph), 66.8
(CH2Ph), 65.2 (CH2Ph), 54.5 (C-2”), 52.6 (OCH3), 52.4 (C-6’), 44.3(C-
5“), 29.7 (C-3” or C-4“), 25.2 (C-3” or C-4“), 24.9 ppm (NHAc); HRMS
(FAB): m/z: calcd for C68H70O15N10Na: 1289.4914 [M+Na]+; found:
1289.4920.
Rf =0.63(cyclohexane/EtOAc 1:1);
[ a]D =+25 (c=0.4 in CHCl3);
1H NMR (400 MHz, CDCl3, 258C, TMS): d=8.05 (s, 1H; H-8), 7.91 (brs,
10072
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 10066 – 10073

First Synthesis of the Core of Miharamycin Antibiotics
FULL PAPER
[13] a) S. Czernecki, J.-M. Valery, J. Carbohydr. Chem. 1988, 7, 151–156;
b) S. Czernecki, S. Horns, J.-M. Valery, J. Org. Chem. 1995, 60, 650–
655.
[14] T. W. Greene, P. G. M. Wuts, Protective Groups in Organic Synthesis,
3rd ed., Wiley, New York, 1999.
[15] CCDC 695607 contains the supplementary 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
This work was supported by the Fundażo para a CiÞncia e Tecnologia
(FCT, project POCI/QUI/59672/2004, Portugal). F.M. thanks the FCT for
a research grant (SFRH/BD/17775/2004). M. Tani (Meiji Seika Kaisha,
Ltd.) is gratefully acknowledged for the gift of miharamycin A. We also
thank R. Abou-Jneid and Dr. A. J. Fairbanks for their contribution to the
project.
[16] a) A.A. Koshkin, J. Org. Chem. 2004, 69, 3711–3718; b) M.A.
OꢁNeill, J.K. Barton, J. Am. Chem. Soc. 2002, 124, 13053–13066;
c) S. P. Parel, C. J. Leumann, Helv. Chim. Acta 2000, 83, 2514–2526.
[17] The same reaction performed on acetylated azidoester 13 only pro-
duced the corresponding ring contracted compound.
[18] a) H. Vorbrüggen, G. Hçfle, Chem. Ber. 1981, 114, 1256–1268; b) H.
Vorbrüggen, K. Krolikiewicz, B. Bennua, Chem. Ber. 1981, 114,
1234–1255; c) H. Vorbrüggen, B. Bennua, Chem. Ber. 1981, 114,
1279–1286.
[1] a) S. Knapp, Chem. Rev. 1995, 95, 1859–1876, and references there-
in; b) P. Garner, in Studies in Natural Product Chemistry, Vol. 1
(Ed.: Att-ur-Raham), Elsevier, Amsterdam, 1988, pp. 397–434.
[2] S. Ichikawa, A. Matsuda, Expert Opin. Ther. Pat. 2007, 17, 487–498.
[3] For some recent syntheses see: S. Ichikawa, A. Matsuda, Nucleo-
sides Nucleotides Nucleic Acids 2005, 24, 319–329.
[4] a) T. Tsuruoka, H. Yumoto, N. Ezaki, T. Niida, Sci. Rep. Meiji Seika
Kasha 1967, 9, 1–4; b) T. Noguchi, Y. Yasuda, T. Niida, T. Shomura,
Ann. Phytopath. Soc. Japan 1968, 34, 323–327.
[19] This lack of reactivity of the 2-aminopurine nucleobase has also
been experienced by other groups. See C. S. Stauffer, A. Datta, J.
Org. Chem. 2008, 73, 4166–4174.
[5] H. Seto, M. Koyama, H. Ogino, T. Tsuruoka, S. Inouye, N. Otake,
Tetrahedron Lett. 1983, 24, 1805–1808.
[20] For the N9 regioisomer, the H-1’of the sugar moiety correlates with
the C-4 of the purine base, while, for the N7 regioisomer, the H-1’
correlates with C-5. See Experimental Section.
[6] a) For synthetic efforts towards the northern part of miharamycins
see: S. Czernecki, S. Franco, S. Horns, J.-M. Valery, Tetrahedron
Lett. 1996, 37, 4003–4006; b) For synthetic efforts towards the
southern part of miharamycins see: P. Garner, J. U. Yoo, R. Saraku,
Tetrahedron 1992, 48, 4259–4270.
[7] The step-by-step protocol with the associated figures is given in the
Supporting Information.
[8] D. J. Edmonds, D. Johnston, D. J. Procter, Chem. Rev. 2004, 104,
3371–3403.
[21] H. Vorbrüggen, C. Ruh-Pohlenz, Org. React. 2000, 55, 1–630.
[22] a) M. J. Robins, R. Zou, F. Hansske, D. Madej, D. L. J Tyrrel, Nu-
cleosides Nucleotides 1989, 8, 725–741; b) M. J. Robins, R. Zou, Z.
Gou, S. F. Wnuk, J. Org. Chem. 1996, 61, 9207–9212.
[23] M. Jetten, A. M. Peters, J. W. F. M. Van Nispen, H. C. J. Ottenheijm,
Tetrahedron Lett. 1991, 32, 6025–6028.
[24] J. R. Vaughan Jr. , J. Am. Chem. Soc. 1951, 73, 3547.
[25] D. C. Johnson II, T. S. Widlanski, Org. Lett. 2004, 6, 4643–4646.
[26] a) T. Iwasa, T. Kishi, K. Matsuura, O. Wakae, J. Antibiot. 1977, 30,
1–10; b) S. Harada, T. Kishi, J. Antibiot. 1977, 30, 11–16.
[27] a) T. Goto, Y. Toya, T. Ohgi, T. Kondo, Tetrahedron Lett. 1982, 23,
1271–1274; b) K. Hara, H. Fujimoto, K. I. Sato, H. Hashimoto, J.
Yoshimura, Carbohydr. Res. 1987, 159, 65–79; c) A. P. Rauter, A. C.
Fernandes, S. Czernecki, J.-M. Valery, J. Org. Chem. 1996, 61, 3594–
3598; d) S. Czernecki, S. Franco, J.-M. Valery, J. Org. Chem. 1997,
62, 4845–4847; e) J. Xue, Z. Guo, J. Carbohydr. Chem. 2008, 27, 51–
69.
[9] a) Y. BlØriot, E. Untersteller, B. Fritz, P. Sinaþ, Chem. Eur. J. 2002,
8, 240–246; b) A. ChØnedØ, P. Pothier, M. Sollogoub, A. J. Fair-
banks, P. Sinaþ, J. Chem. Soc. Chem. Commun. 1995, 1373–1374.
[10] A. J. Fairbanks, P. Sinaþ, Synlett 1995, 277–279.
[11] a) A. P. Rauter, O. Oliveira, T. Canda, E. Leroi, H. Ferreira, M. J.
Ferreira, J.A. Ascenso, J. Carbohydr. Chem. 2002, 21, 257–273;
b) A. Rauter, M. Ferreira, C. Borges, T. Duarte, F. Piedade, M.
Silva, H. Santos, Carbohydr. Res. 2000, 325, 1–15; c) M. Etheve-
Quelquejeu, J. Xie, J.-M. ValØry, Synlett 2004, 1089–1091.
[12] a) G. Casiraghi, L. Colombo, G. Rassu, P. Spanu, J. Org. Chem.
1991, 56, 6523–6527; b) P. Garner, J. U. Yoo, R. Sarabu, V. O. Ken-
nedy, W. J. Youngs, Tetrahedron 1998, 54, 9303–9316; c) P. Bhaket,
C. S. Stauffer, A. Datta, J. Org. Chem. 2004, 69, 8594–8601.
Received: September 3, 2008
Published online: October 2, 2008
Chem. Eur. J. 2008, 14, 10066 – 10073ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10073