Development of a highly β-selective ribosylation reaction without using neighboring group participation: Total synthesis of (+)-caprazol, a core structure of caprazamycins

Article abstract of DOI:10.1021/jo701699h

(Chemical Equation Presented) Full details of the total synthesis of (+)-caprazol are described. The key elements of our approach include the early stage introduction of the aminoribose in a highly β-selective manner, using the steric hindrance in the tra

Full text of DOI:10.1021/jo701699h

Development of a Highly â-Selective Ribosylation Reaction without
Using Neighboring Group Participation: Total Synthesis of
(+)-Caprazol, a Core Structure of Caprazamycins
Shinpei Hirano, Satoshi Ichikawa,* and Akira Matsuda*
Faculty of Pharmaceutical Sciences, Hokkaido UniVersity, Sapporo 060-0812, Japan
ichikawa@pharm.hokudai.ac.jp; matuda@pharm.hokudai.ac.jp
ReceiVed August 3, 2007
Full details of the total synthesis of (+)-caprazol are described. The key elements of our approach include
the early stage introduction of the aminoribose in a highly â-selective manner, using the steric hindrance
in the transition state and the construction of the diazepanone by a modified intramolecular reductive
amination. The 5′-C-glycyluridine derivative 9, which was prepared stereoselectively via Sharpless
asymmetric aminohydroxylation, was ribosylated with 2,3-O-alkylidene ribofuranosyl donors. It was
revealed that increasing the size of the alkyl substituents of the acetal unit resulted in improving the
stereoselectivity of the anomeric position, and the desired ribosides 21b (1′′-â) and 22b (1′′-R) were
obtained in 80% yield (21b/22b ) 24.0/1) when the ribosyl fluoride 16 possessing a more sterically
hindered 3-pentylidene group was used. The origin of the stereoselectivity of the ribosylation was also
discussed. Construction of the diazepanone system was optimized with the model aldehyde 37, and the
desired diazepanone 38 was obtained in 88% yield via two-step reaction sequence including catalytic
hydrogenation followed by hydride reduction. Application of this method to the aldehyde 44 successfully
afforded the diazepanone derivatives 45 and 46, functional group manipulation of which completed the
total synthesis of (+)-caprazol.
Introduction
culosis strains (MIC ) 3.13 µg/mL), and exhibit no significant
toxicity in mice. With such excellent biological properties, CPZs
are expected to become promising leads for the development
of antituberculosis agents with a novel mode of action.
Caprazol (Figure 2, 5) is a deacylated CPZs whose stereo-
chemical structure (5′S,6′S,2′′′S,3′′′S) was recently revealed
through X-ray crystal analysis1b and confirmed by total syn-
Caprazamycins (CPZs) (Figure 1, 1) were isolated from a
culture broth of the Actinomycete strain Streptomyces sp.
MK730-62F2 in 20031 and represent the newest members of a
class of naturally occurring 6′-N-alkyl-5′-â-O-aminoribosyl-
glycyluridine antibiotics including liposidomycins2 (LPMs, 2),
muraymycins3 (MRYs, 3), and FR-9004934 (4), which have been
shown to exhibit excellent antimicrobial activity against Gram-
positive bacteria. In particular, the CPZs have shown excellent
antimycobacterial activity in vitro against drug-susceptible (MIC
) 3.13 µg/mL) and multi-drug-resistant Mycobacterium tuber-
(2) (a) Isono, K.; Uramoto, M.; Kusakabe, H.; Kimura, K.; Izaki, K.;
Nelson, C. C.; McCloskey, J. A. J. Antibiot., 1985, 38, 1617-1621. (b)
Ubukata, M.; Kimura, K.; Isono, K.; Nelson, C. C.; Gregson, J. M.;
McCloskey, J. A. J. Org. Chem. 1992, 57, 6392-6403. (c) Ubukata, M.;
Isono, K. J. Am. Chem. Soc., 1988, 110, 4416-4417. (d) Kimura, K.; Ikeda,
Y.; Kagami, S.; Yoshihama, M.; Suzuki, K.; Osada, H.; Isono, K. J. Antibiot.
1998, 51, 1099-1104. (e) Muroi, M.; Kimura, K.; Osada, H.; Inukai, M.;
Takatsuki, A. J. Antibiot. 1997, 50, 103-104. (f) Kimura, K.; Ikeda, Y.;
Kagami, S.; Yoshihama, M.; Ubukata, M.; Esumi, Y.; Osada, H.; Isono,
K. J. Antibiot. 1998, 51, 647-654. (g) Kimura, K.; Kagami, S.; Ikeda, Y.;
Takahashi, H.; Yoshihama, M.; Kusakabe, H.; Osada, H.; Isono, K. J.
Antibiot. 1998, 51, 640-646. (h) Kimura, K.; Miyata, N.; Kawanishi, G.;
Kamio, Y.; Izaki, K.; Isono, K. Agric. Biol. Chem. 1989, 53, 1811-1815.
(3) McDonald, L. A.; Barbieri, L. R.; Carter, G. T.; Lenoy, E.; Lotvin,
J.; Petersen, P. J.; Siegel, M. M.; Singh, G.; Williamson, R. T. J. Am. Chem.
Soc. 2002, 124, 10260-10261.
* Address correspondence to this author. S.I.: phone (+81) 11-706-3228,
fax (+81) 11-706-4980. A.M.: phone (+81) 11-706-3763, fax (+81) 11-706-
4980.
(1) (a) Igarashi, M.; Nakagawa, N.; Doi, N.; Hattori, S.; Naganawa, H.;
Hamada, M. J. Antibiot. 2003, 56, 580-583. (b) Igarashi, M.; Takahashi,
Y.; Shitara, T.; Nakamura, H.; Naganawa, H.; Miyake, T.; Akamatsu, Y.
J. Antibiot. 2005, 58, 327-337. (c) Takeuchi, T.; Igarashi, M.; Naganawa,
H.; Hamada, M. JP 2003012687, 2001. (d) Miyake, M.; Igarashi, M.;
Shidara, T.; Takahashi, Y. WO 2004067544 2004.
10.1021/jo701699h CCC: $37.00 © 2007 American Chemical Society
Published on Web 11/21/2007
9936
J. Org. Chem. 2007, 72, 9936-9946

Total Synthesis of (+)-Caprazol
FIGURE 1. Structures of nucleoside antibiotics possessing the 6′-N-alkyl-5′-O-aminoribosyl-glycyluridine.
5′-, 6′-, 2′′′-, and 3′′′-positions, have been addressed previously
by the construction of the characteristic diazepanone system.
One of the major difficulties posed by the previous synthetic
work is the introduction of the 5-aminoribose moiety after
constructing the uridyldiazepanone moiety. The tertiary amine
contained in the diazepanone structure inhibits the usual
ribosylation promoted by Lewis acids,14 and the 5′-hydroxyl
group is presumed to be in a highly sterically hindered position.15
In view of these observations, we selected to set up the 5′-â-
O-aminoribosyl-glycyluridine structure 6 prior to the construc-
tion of the diazepanone ring (Figure 2). From a medicinal
chemical point of view, this strategy would also be suitable for
examining a general structure-activity relationship and for
synthesizing novel analogues because the 5′-â-O-aminoribosyl-
glycyluridine structure 6 is predicted to be a pharmacophore of
this class of natural products. However, a key issue associated
FIGURE 2. Synthetic strategy of caprazol (5).
thesis.5 The CPZs and the LPMs consist of a uridine, an
aminoribose, and a characteristic diazepanone, rendering them
intriguing, challenging synthetic targets.6-13 Most of the syn-
thetic problems, based on the stereochemical assignment at the
(4) (a) Ochi, K.; Ezaki, M.; Morita, I.; Komori, T.; Kohsaka, M. ER
0333177, 1989, A2. (b) Yoshida, Y.; Yamanaka, H.; Sakane, K. JP H05-
78385 1993.
(5) Hirano, S.; Ichikawa, S.; Matsuda, A. Angew. Chem., Int. Ed. 2005,
44, 1854-1856.
(11) (a) Gravier-Pelletier, C.; Milla, M.; Merrer, Y. L.; Depezay, J. Eur.
J. Org. Chem. 2001, 3089-3096. (b) Gravier-Pelletier, C.; Dumas, J.;
Merrer, Y. L.; Depezay, J. C. J. Carbohydr. Chem. 1992, 11, 969-998. (c)
Merrer, Y. L.; Gravier-Pelletier, C.; Gerrouache, M.; Depezay, J. C.
Tetrahedron Lett. 1998, 39, 385-388. (d) Gravier-Pelletier, C.; Ginisty,
M.; Merrer, Y. L. Tetrahedron: Asymmetry 2004, 15, 189-193. (e) Ginisty,
M.; Gravier-Pelletier, C.; Merrer, Y. L. Tetrahedron: Asymmetry 2006, 17,
142-150.
(6) Spada, M. R.; Ubukata, M.; Isono, K. Heterocycles 1992, 34, 1147-
1167.
(7) (a) Knapp, S.; Morriello, G. J.; Doss, G. A. Org. Lett. 2002, 4, 603-
606. (b) Knapp, S.; Morriello, G. J.; Nandan, S. R.; Emge, T. J.; Doss, G.
A.; Mosley, R. T.; Chen, L. J. Org. Chem. 2001, 66, 5822-5831. (c) Knapp,
S.; Nandan, S.; Resnick, L. Tetrahedron Lett. 1992, 33, 5485-5486. (d)
Knapp, S.; Morriello, G. J.; Doss, G. A. Tetrahedron Lett. 2002, 43, 5797-
5800.
(12) (a) Drouillat, B.; Poupardin, O.; Bourdreux, Y.; Greck, C. Tetra-
hedron Lett. 2003, 44, 2781-2783. (b) Bourdreux, Y.; Drouillat, B.; Greck,
C. Lett. Org. Chem. 2006, 3, 368-370.
(8) Nakajima, N.; Isobe, T.; Irisa, S.; Ubukata, M. Heterocycles, 2003,
59, 107-113.
(13) Moore, W. J.; Luzzio, F. A. Tetrahedron Lett. 1995, 36, 6599-
6602.
(14) Introduction of a uracil by the Vorbruggen method following
construction of ribosyl diazepanone was reported to have some drawbacks.
See ref. 7a.
(15) Ichikawa, S.; Shuto, S.; Matsuda, A. J. Am. Chem. Soc. 1999, 121,
10270-10280.
(9) (a) Kim, K. S.; Ahn, Y. H. Tetrahedron: Asymmetry 1998, 9, 3601-
3605. (b) Kim, K. S.; Cho, I. H.; Ahn, Y. H.; Park, J. I. J. Chem. Soc.,
Perkin Trans. 1 1995, 1783-1785. (c) Kim, K. S.; Cheong, C. S.; Hahn, J.
S.; Park, J. I. Bull. Korean Chem. Soc. 1997, 18, 465-467.
(10) Sarabia, F.; Martin-Ortiz, L.; Lopez-Herrera, F. J. Org. Lett. 2003,
5, 3927-3930.
J. Org. Chem, Vol. 72, No. 26, 2007 9937

Hirano et al.
SCHEME 1. Preparation of 5′-C-Glycyluridine 9
with the construction of the 5′-â-O-aminoribosyl-glycyluridine
structure needs to be addressed. It was reported that the
elimination of the fatty acyloxy group of the diazepanone moiety
of the LPMs occurred under basic conditions because they
contain â-heterosubstituted carboxyl moieties.2b Caprazol (5)
as well as the CPZs would also be sensitive to similar basic
conditions. A general method exists for the construction of
â-glycosides, using a glycosyl donor protected with a 2-O-acyl
group, via a neighboring group participation.16 However,
because a 2-O-acyl group is usually deprotected under basic
conditions, this strategy would not be suitable for the synthesis
of 5. Compared to the neighboring group participation strategy,
little attention has been focused on alternative methods to
synthesize â-ribosides.17,18 The anomeric effect has been used
in pyranose chemistry to construct glycosyl bonds but due to
the weak anomeric effect of furanoses, anomeric selectivities
would be difficult to control.19 Furthermore, furanosides are
inherently flexible and can adopt both twist and envelope
conformations, which can interconvert via pseudorotational
itineraries.20,2121 As a result, furanoses can glycosylate through
several different transition states, potentially compromising
anomeric selectivities.22 Considering the inherent nature of
furanoses, we planned to lock the conformation of the ribo-
furanose by introducing a cyclic acetal protecting group at the
2,3-hydroxyl groups, that could be deprotected under acidic
conditions, thereby controlling the â-selectivity via the steric
hindrance of the protecting group installed on the R-face of the
ribofuranose. Recently we have completed the total synthesis
of (+)-caprazol utilizing the above-mentioned concept.5 Herein,
we provide full details of â-selective ribosylation reaction
utilizing a steric hindrance installed on the R-face of a ribosyl
donor and the total synthesis of (+)-caprazol.
Control of the â-Selectivity in a Ribosylation Reaction
Utilizing a Steric Hindrance Installed on the R-Face of a
Ribosyl Donor. Our initial objective was to construct a
â-ribofuranoside utilizing the steric hindrance installed on the
R-face of a ribosyl donor to permit direct access to the predicted
pharmacophore of the CPZs 6. The 5′-C-glycyluridine derivative
9, which is a ribosyl acceptor, was prepared as shown in Scheme
1. Oxidation of 2′,3′-O-isopropylideneuridine (7)23 with IBX
(3 equiv, MeCN, 80 °C)24 followed by a two-carbon elongation
with Ph3PdCHCO2Me (1.2 equiv, CH2Cl2, -20 °C) and a BOM
protection of the N-3-position of the uracil moiety by mild
biphasic conditions (1.5 equiv of BOMCl, 4 equiv of Na2CO3,
0.05 equiv of Bu4NI, CH2Cl2-H2O, 66% overall) provided
compound 8 (trans/cis ) 37/1). The best reagent for the
oxidation of 7, furnishing the best quality of the corresponding
aldehyde among other oxidation methods tested (i.e., Swern
conditions, Pfitzner-Moffatt conditions, PDC, Dess-Martin
periodinane), was IBX. Since the 4′-position of 8 became more
acidic when the R,â-unsaturated ester functionality at the 5′-
position was introduced, the use of the usual BOM protection
conditions (i.e., Et3N or iPr2NEt, CH2Cl2, 0 °C) resulted in near
complete epimerization at the 4′-position. When a Sharpless
aminohydroxylation25 of 8 was carried out in the absence of
chiral ligands, no diastereoselectivity was observed and com-
pounds 9 and 10 (a ratio of 9/10 ) 40/60) were obtained in
low yield. The aminohydroxylation with (DHQD)2AQN as a
chiral ligand26-28 (15 mol % of K2[Os2(OH)4], 15 mol % of
(DHQD)2AQN, 3 equiv of benzyl carbamate, 2.6 equiv of
NaOH, PrOH-H2O, 15 °C, 52%) afforded 9 as the major
diastereomer, with the ratio of 9/10 being 86/14. The regiose-
lectivity of the aminohydroxylation reaction was relatively high
compared with that reported in the literature29 and only trace
amounts of the corresponding â-amino-R-hydroxy derivatives
were observed. The stereochemistry of the 5′-position of
compounds 9 and 10 was determined to be the S and R
configuration by using the modified Mosher method30 with the
amide derivatives 11 and 12, respectively, which were prepared
by deprotection of the Cbz protecting group of either 9 or 10
followed by acylation of the liberated amine with 2-methoxy-
2-trifluoromethylphenylacetic acid. We also found that the
(16) Jung, K.; Muller, M.; Schmidt, R. R. Chem. ReV. 2000, 100, 4423-
4442.
(17) (a) Uchiro, H.; Mukaiyama, T. Chem. Lett. 1996, 271-272. (b)
Mukaiyama, T.; Matsubara, K.; Hora, M. Synthesis 1994, 1368-1373. (c)
Uchiro, H.; Mukaiyama, T. Chem. Lett. 1996, 79-80. (d) Mukaiyama, T.;
Shimpuku, T.; Takashima, T.; Kobayashi, S. Chem. Lett. 1989, 145-148.
(18) Scott, R. W.; Fox, D. E.; Williams, D. K. Tetrahedron Lett. 2000,
41, 8207-8210.
(19) Lowary, T. L. Curr. Opin. Chem. Biol. 2003, 7, 749-756.
(20) Ellervik, U.; Magnusson, G. J. Am. Chem. Soc. 1994, 116, 2340-
2347.
(25) Rudolph, J.; Sennhenn, P. C.; Vlaar, C. P.; Sharpless, K. B. Angew.
Chem., Int. Ed. Engl. 1996, 35, 2810-2817.
(26) O’Brlen, P.; Osborne, S. A.; Parkaer, D. D. Tetrahedron Lett. 1998,
39, 4099-4102.
(27) Tao, B.; Schlingloft, G.; Sharpless, K. B. Tetrahedron Lett. 1998,
39. 2507-2510.
(28) Morgan, A. J.; Masse, C. E.; Panek, J. S. Org. Lett. 1999, 1, 1949-
1952.
(21) Houseknecht, J. B.; Lowary, T. L.; Hadad, C. M. J. Phys. Chem.
2003, 107, 5763-5777.
(22) Zhu, X.; Kawatkar, S.; Rao, Y.; Boons, G.-J. J. Am. Chem. Soc.
2006, 128, 11948-11957.
(23) Asakura, J.; Matsubara, Y.; Yoshihara, M. J. Carbohydr. Chem.
1996, 15, 231-239.
(29) Hoye, T. R.; Rennere, M. K. J. Org. Chem. 1996, 61, 2056-2064.
(30) Kusumi, T.; Fukushima, T.; Ohtani, I.; Kakisawa, H. Tetrahedron
Lett. 1991, 32, 2939-2942.
(24) More, J. D.; Finny, N. S. Org. Lett. 2002, 4, 3001-3003.
9938 J. Org. Chem., Vol. 72, No. 26, 2007

Total Synthesis of (+)-Caprazol
with excellent â-selectivity (21b/22b ) 24.0/1) when activation
was conducted with BF3‚OEt2 at 0 °C (entry 9). Crystallization
of the mixture gave pure 21b.
We further investigated the effect of the 5-O-substituent of
ribofuranosyl donor 18-2031 and expanded the â-selective
O-ribosylation reaction with other nucleophiles. The ribosylation
was examined with ribosyl donors possessing a variety of
substituents at the 5-position, using N-Cbz-L-threonine methyl
ester as a nucleophile, to understand the effect of these
substituents on the stereoselectivity of the O-ribosylation
reaction, and the results are summarized in Table 2. When the
5-azidoribosyl fluoride 16 was activated with BF3‚OEt2 at
-30 °C, the corresponding ribosides 23a and 24a were obtained
in 95% yield with good stereoselectivity at the anomeric position
(23a/24a ) 97/3, entry 1). When a 5-O-alkyl-substituted ribosyl
fluoride such as a methyl derivative 18 and a more sterically
hindered benzyl derivative 19 were used, the corresponding
ribosides were obtained in 99% and 94% yield with good
stereoselectivity, respectively (entries 2 and 3). The observed
high â-selectivity was not only the case for the azide group,
and was revealed to be independent from the 5-O-alkyl
substituent. On the other hand, the stereoselectivity was slightly
decreased to 91/9 when a 5-O-acetylribosyl donor 20 was used
(entry 4). Presumably, a neighboring group participation of the
acetyl group to the oxocarbenium intermediate at the â-face
and the rate of SN2 displacement to the intermediate would be
increased although the stereoselectivity was still a practical level.
Next, O-ribosylations with other nucleophiles, using the
5-azidoribosyl fluoride 16 as a donor, were examined as shown
in Table 3. Reaction with N-Cbz-L-serine methyl ester, which
has a primary alcohol, gave 25 in 58% with good selectivity
(â/R ) 95/5, entry 1). Ribosylation with the primary alcohol
of methyl 2,3,5-tri-O-benzyl-R-D-glucoside, however, resulted
in a reduced stereoselectivity (â/R ) 82/18, entry 2). When other
secondary alcohols such as choresterol and sterically more
hindered menthol were used as nucleophiles, the stereoselectivity
was moderate (â/R ) 81/19-89/11, entries 3 and 4) compared
to the excellent selectivity with N-Cbz-L-threonine methyl ester,
which has also a secondary alcohol. In conjunction with the
results observed in the reactions with the primary alcohols,
alcohols at a side chain of the amino acid derivatives, which
have an electron-withdrawing group and are unreactive, exhibit
the better â-selectivity. These results suggested that the stereo-
selectivity largely depends on the reactivity of the alcohol, but
not the steric hindrance.
FIGURE 3. Structure of ribosyl donors used in this study.
aminohydroxylation of 8 was sensitive to the amount of chiral
ligand and to the reaction temperature. When a reduced amount
of (DHQD)2AQN was used or when the reaction was conducted
at higher temperatures, the yields and diastereoselectivity of 9
were dramatically decreased.
We next examined the ribosylation of the 5′-hydroxyl group
of 9 with a variety of ribosyl donors31 listed in Figure 3, and
the results are summarized in Table 1. First, we screened the
leaving group of ribosyl donors protected with an isopropylidene
group at the 2- and 3-positions. During the course of synthetic
studies on nucleoside antibiotics by other groups32,33 and by
ours15,34 it was revealed that the glycosylation or the introduction
of a substituent at the sterically encumbered 5′-hydroxyl group
of nucleoside derivatives proved quite difficult. Therefore, our
initial attempts focused on finding glycosyl donors known to
be highly reactive, for example, the sulfoxide 13 and the
trichloroacetimidate 14. When 13 was activated with Tf2O in
CH2Cl2 at -40 °C,34 the corresponding ribosides 21a and 22a
were obtained in 80% yield;36 however, no â-selectivity was
observed (entry 1). The ribosylation with 14 by using TfOH as
an activator also resulted in no â-selectivity (entry 2).37 On the
other hand, when the ribosyl fluoride 15 was activated with
BF3‚OEt238,39 at 0 °C, the ribosides 21a and 22a were obtained
in 72% yield with moderate stereoselectivity at the anomeric
position (21a/22a ) 2.8/1, entry 3). Lower temperature did not
affect the selectivity (entry 4). When other activators, including
AgOTf/Cp2HfCl2,40 AgOTf/SnCl2,41 or AgClO4/SnCl2, were
used, the stereoselectivity did not improve (entries 5-8). Further
exploration with the ribosyl fluoride 16 possessing a more
sterically hindered 3-pentylidene group afforded the desired 21b
During ribosylation, an oxocarbenium intermediate is formed
by the activation of the fluoride. Oxocarbenium ions have a
double-bond character between the endocyclic oxygen and
anomeric carbon atoms. As a result, oxocarbenium ions of
D-ribofuranosides can adopt either of two possible low-energy
conformations, in which C-3 is either above or below the C-2-
(31) Preparation of donors can be found in the Supporting Information.
(32) Beader, J. R.; Dewis, M. L.; Whiting, D. A. J. Chem. Soc., Perkin
Trans. 1 1995, 227-233.
3
C-1-O-C-4 plane, the E conformer or the E3 conformer,
(33) Knapp, S.; Nandan, S. R. J. Org. Chem. 1994, 59, 281-283.
(34) Ichikawa, S.; Matsuda, A. Nucleoside, Nucleotides Nucleic Acids
2005, 24, 319-329.
respectively.42 To obtain more insight into the factors responsible
for the high â-selectivity with the 2,3-pentylidene protected
ribosyl donors, we optimized these conformers for the two
different 5-O-methylribofuranosyl oxocarbenium ions having
the isopropylidene (Figure 4a) and the 3-pentylidene protected
(35) Kahne, D.; Walker, S.; Cheng, Y.; Engen, D. V. J. Am. Chem. Soc.
1989, 111, 6881-6882.
(36) The stereochemistry of the 1′′-position was confirmed by NOESY
correlations between H-1′′ and H-4′′ for the â-riboside and the typical small
coupling constant (J1,2 ) 0-1.6 Hz) for the 2,3-O-alkylidene protected
1
ribofuranoside observed in the H NMR spectrum.
(37) Schmidt, R. R.; Michel, J. Angew. Chem., Int. Ed. 1980, 19, 731-
732.
(40) Suzuki, K.; Maeta, H.; Matsumoto, T.; Tsuchihashi, G. Tetrahedron
Lett. 1988, 29, 3571-3574.
(38) Nicolaou, K. C.; Chucholoeski, A.; Dolle, R. E.; Randall, J. L. J.
Chem. Soc., Chem. Commun. 1984, 1155-1158.
(39) Kunz, H.; Waldmann, H. J. Chem. Soc., Chem. Commun. 1985,
638-640.
(41) (a) Ito, Y.; Ogawa, T. Tetrahedron Lett. 1988, 29, 3987-3990. (b)
Ogawa, T.; Takahashi, Y. Carbohydr. Res. 1985, 138, 5-9.
(42) Larsen, C. H.; Ridgway, B. H.; Shaw, J. T.; Smith, D. M.; Woerpel,
K. A. J. Am. Chem. Soc. 2005, 127, 10879-10884.
J. Org. Chem, Vol. 72, No. 26, 2007 9939

Hirano et al.
TABLE 1. Ribosylation Reactions of 5′-C-Glycyluridine 9
entry
donor
activator (equiv)
Tf2O (2.0)
temp, °C
yield, %
ratio (21/22)
1
2
3
4
5
6
7
8
9
13
14
15
15
15
15
15
15
16
16
17
-40
-15
0
80
79
72
78
trace
quant.
60
40
80
1.2/1
1.1/1
2.8/1
2.7/1
TfOH (1.0)
BF3‚OEt2 (1.5)
BF3‚OEt2 (1.2)
-30
TMSOTf (1.0)
0
AgOTf (1.5)/Cp2HfCl2 (1.5)
AgOTf (1.5)/SnCl2 (1.5)
AgClO4 (1.5)/SnCl2 (1.5)
BF3‚OEt2 (1.5)
AgOTf (1.5)/Cp2HfCl2 (1.5)
BF3‚OEt2 (1.5)
-40
0
1.2/1
2.0/1
2.6/1
24.0/1
8.1/1
2.4/1
0
0
10
11
-40
quant.
75
0
TABLE 2. Effect of a Substituent at the 5-Position on the
Stereoselectivity
TABLE 3. Nucleophilic Substitution Reaction of
3-Pentylidene-Protected Ribosyl Fluoride
entry
donor
R
yield, %
ratio (23/24)
1
2
3
4
16
18
19
20
N3
95
99
94
96
97/3
96/4
96/4
91/9
OMe
OBn
OAc
ribofuranosides (Figure 4b) by density functional theory (DFT)
quantum mechanical calculations at the BL3LYP/6-31G**
level.43 From these results, we calculated the E3 conformer to
be lower in energy than the 3E conformer, where the 3-oxy group
was orientated in the pseudoaxial position, in both cases.44,4545
Two potentially favorable modes of nucleophilic attack are
possible for the oxocarbenium ions, and these modes are
governed by both ground-state effects and transition-state effects,
issues which have been extensively studied by Woerpel et al.45
For the oxocarbenium ion in the E3 conformer, nucleophilic
attack from the R-face is favored by inside attack in the ground
state to give the R-riboside. However, nucleophilic attack from
the R-face of 29 or 30 would suffer significant steric interactions
from one of the alkyl groups of the cyclic ketal moiety in its
transition state. Thus, the approach of the nucleophile is
diminished resulting in poor selectivity when the oxocarbenium
ion 29 with the smaller methyl substituents of the isopropylidene
group is used. Increasing the size of the alkyl substituents such
as ethyl groups in the 3-pentylidene group would result in severe
steric repulsion on the R-face of 30, leading to outside attack
with complete reversal of stereoselectivity. This mechanism was
further supported by the following experiments. Ribosylation
of 9 with the cyclopentylidene protected ribofuranosyl donor
17, which has the same number of carbon atoms as the
3-pentylidene group, gave the corresponding ribosides with
reduced selectivity (21c/22c ) 2.4/1, entry 11 in Table 1). Thus,
the cyclopentylidene group, held away from the course of the
approaching nucleophile, did not exert severe steric repulsion
(43) (a) Becke, A. D. Phys. ReV. 1998, 38, 3098-3100. (b) Becke, A.
D. Phys. ReV. 1998, 37, 785-789. (c) Becke, A. D. J. Phys. Chem. 1993,
98, 1372-1377.
(44) (a) Chang, G.; Guida, W. C.; Still, W. C. J. Am. Chem. Soc. 1989,
111, 4379-4386. (b) Saunders, M.; Houk, K. N.; Wu, Y. D.; Still, W. C.;
Lipton, M.; Chang, G.; Guida, W. C. J. Am. Chem. Soc. 1990, 112, 1419-
1427. (c) Halgren, T. A. J. Comput. Chem. 1999, 20, 720-729. (d) Halgren,
T. A. J. Comput. Chem. 1999, 20, 730-748.
(45) Chamberland, S.; Ziller, J. W.; Woerpel, K. A. J. Am. Chem. Soc.
2005, 127, 5322-5323.
9940 J. Org. Chem., Vol. 72, No. 26, 2007

Total Synthesis of (+)-Caprazol
FIGURE 4. Optimized geometries (B3LYP/6-31G**) of the E3 Conformers of the oxacarbenium ions of (a) 2,3-O-isopropylidene-5-O-methyl-
D-ribofuranose and (b) 2,3-O-3-pentylidene-5-O-methyl-D-ribofuranose
SCHEME 2. Preparation of N-Methyl-2-amino-4-pentene-1,3-diol Derivative 37
on the R-face thereby permitting inside attack, which results in
a poor selectivity, similar to that with the isopropylidene-
protected ribofuranosides. This would suggest that the steric
effect in the transition state is largely responsible for the
excellent â-selectivity seen with the ribosyl donor 16 in our
study. Because of the inherent nature of the E3 conformation,
the terminal methyl substituent of the 3-pentylidene group on
the R-face is oriented toward the anomeric carbon atom and
not toward the C-4 carbon atom, an orientation that would also
help to prevent the approach of the nucleophile from the R-face
of 30. This could be a possible explanation for the large disparity
in the anomeric selectivity observed between the isopropylidene-
and the 3-pentylidene-protected ribofuranosides. It should be
noted that a decrease in â-selectivity was also observed under
conditions where a trifluoromethanesulfonate ion was present
(AgOTf/Cp2HfCl2, 21b/22b ) 8.1/1, entry 10). Presumably a
â-O-trifluoromethanesulfonyl riboside intermediate47 was formed,
followed by SN2 attack of the alcohol to give the undesired
R-riboside 22b.
on 21b. Optimization of the reductive amination was examined
by using the model compound 37, which was prepared as shown
in Scheme 2.48 The secondary alcohol of the 2-amino-4-pentene-
1,3-diol derivative 31,49 prepared from D-serine in 7 steps, was
protected with a TBS group to give 32 (1.2 equiv of TBSCl,
3.6 equiv of imidazole, DMF, 89%), which was successively
N-methylated (2.5 equiv of MeI, 2 equiv of NaH, DMF) and
deprotected (HCl/AcOEt). During the deprotection step of the
isopropylidene group, the TBS group was partially deprotected.
Therefore, the crude mixture was re-treated with TBSCl and
(46) Sonntag, L. S.; Schweizer, S.; Ochsenfeld, C.; Wennemers, H. J.
Am. Chem. Soc. 2006, 128, 14697-14703.
(47) Crich, D.; Sun, S. J. Am. Chem. Soc. 1997, 119, 11217-11223.
(48) Initial attempts to construct the diazepanone via deprotection of the
Cbz group in a model compound such as A followed by reductive amination
of the aldehyde, both promoted by catalytic hydrogenation with Pd/C, were
unsuccessful because of difficulty in hydrogenating the corresponding cyclic
imine. Additional forcing conditions under medium pressure gave a 5,6-
dihydrouridine derivative B due to over-reduction
Total Synthesis of (+)-Caprazol. After selectively preparing
the key aminoriboside 21b, we directed our attention to the
construction of the characteristic 7-membered diazepanone, the
system on which most of the previous synthetic studies for the
LPMs have been focused.6-13 An intramolecular reductive
amination strategy with an amino aldehyde derivative seemed
to be an efficient way of constructing the diazepanone system,
a strategy we decided to use to install the diazepanone moiety
.
(49) (a) Garner, P.; Park, J. M. J. Org. Chem. 1988, 53, 2979-2984.
(b) Garner, P.; Park, J. M. J. Org. Chem., 1987, 52, 2361-2364. (c) Garner,
P.; Park, J. M. Org. Synth. 1991, 70, 18-28.
J. Org. Chem, Vol. 72, No. 26, 2007 9941

Hirano et al.
TABLE 4. Formation of a Diazepanone by Intramolecular
Reductive Amination
FIGURE 5. Key NOESY peaks for diazepanone 46.
the solvent in the catalytic hydrogenation of the Cbz group was
also examined to minimize the silyl group deprotection.52 The
use of a bulkier alcohol as solvent also was expected to minimize
the formation of the hemiacetal. Hydrogenolysis of the Cbz
group of 37 catalyzed by Pd/C in EtOH followed by hydride
reduction with NaBH3CN (2 equiv, AcOH, AcOEt, 46% overall,
entry 3) improved the yield of the desired 38. A better yield
was obtained with NaBH(OAc)3 (3 equiv, AcOH, AcOEt, 60%
overall, entry 4). However, cleavage of the silyl ether in 38
was still observed in the Cbz-deprotection step when EtOH was
yield,
%
entry
conditions A
conditions B (equiv)
1
2
3
4
5
H2, Pd/C, MeOH
H2, Pd/C, MeOH NaBH3CN (2.0), AcOH, EtOAc
H2, Pd/C, EtOH
H2, Pd/C, EtOH
H2, Pd/C, iPrOH
trace
25
46
60
88
NaBH3CN (2.0), AcOH, EtOAc
NaBH(OAc)3 (3.0), AcOH, EtOAc
NaBH(OAc)3 (4.0), AcOH, EtOAc
i
used as solvent. The use of the more hydrophobic PrOH
imidazole in DMF to afford the secondary amine derivative 33
(83% overall). The amine 33 was coupled with a commercially
available N-Cbz-L-threonine 34 (1.5 equiv, 3 equiv of 3-(di-
ethoxyphosphoryloxy)-1,2,3-benzotriazin-4(3H)-one (DEPBT),50
3 equiv of NaHCO3, THF, 0 °C, 65%) to give the amide 35
without any racemization. As will be mentioned in the synthesis
of 43, other coupling conditions such as EDCI and HOBt gave
unsatisfactory results. Protection of the secondary alcohol of
35 with a TBS group (2 equiv of TBSCl, 4 equiv of imidazole,
DMF) followed by oxidative cleavage of the terminal olefin by
ozonolysis provided the aldehyde 37 (O3, CH2Cl2, -78 °C, 60%
overall), a precursor to the cyclization. In a manner similar to
previous studies, construction of the diazepanone via deprotec-
tion of the Cbz group in 37 followed by reductive amination of
the amino aldehyde 39, both promoted by catalytic hydrogena-
tion of the Cbz and imino groups with Pd/C, was examined in
MeOH as the solvent (Table 4, entry 1). However, only a trace
amount of the desired diazepanone 38 was obtained. Careful
analysis revealed several drawbacks in the reaction, the first
one being that formation of a cyclic imine 40, observed on TLC
and isolable, was disfavored at equilibrium with the linear amino
aldehyde 39, indicating the presence of a possible conforma-
tional disadvantage in our system. The hemiacetal formation
of 39 with MeOH, also observed on TLC (hexane:AcOEt )
5:1), might explain why the equilibrium favored the linear amino
aldehyde 39. The slow reaction rate of the imine reduction by
hydrogenation presents another problem as does the removal
of the TBS protecting group during the course of the reaction
process promoted by catalytic hydrogenation in MeOH.51 With
these drawbacks in mind, we modified the conditions in our
system to construct the diazepanone. Namely, each of the above
troublesome steps in the intramolecular reductive amination
process might be overcome by using hydride reagents after
deprotection of the Cbz group on the amino group. Hydride
reduction is much more reactive than hydrogenation of imines;
moreover, dehydration of 39 would also be accelerated by the
use of hydride reagents to give 40. Decreasing the polarity of
improved the yield of 38 up to 88% without extensive
desilylation.
Having established the optimized method for construction of
the diazepanone structure in our model system, the total
synthesis of caprazol (5) was then undertaken (Scheme 3). The
azide group in 21b was reduced by Staudinger’s conditions (3
equiv of PPh3, 5 equiv of H2O, benzene-THF, 50 °C) to the
corresponding amine, which was sequentially protected with a
Boc group to give 41 (2 equiv of Boc2O, 2 equiv of NaHCO3,
95% overall). Saponification of the methyl ester in 41 proved
to be troublesome. Extensive efforts to obtain the acid 42 have
been conducted for hydrolysis of 41 under other conditions (i.e.,
LiOH, THF-MeOH-H2O; NaOH, CH2Cl2-H2O; lipase, aque-
ous MeCN) and demethylation (PhSH, Cs2CO3, DMF), and
resulted in decomposition of 41 to give a mixture of products
containing an N-Boc-5-amino-2,3-O-(3-pentylidene)-ribose and
a uracil. It was suggested that the acidic 6′-proton was prone to
release the aminoribosyloxy group by â-elimination. In addition,
the â-elimination would be thermodynamically favored because
steric hindrance would be reduced by releasing the sterically
encumbered aminoribose moiety installed at the O5′-position.
Therefore, the desired carboxylic acid 42 was obtained in only
50% yield when 41 was treated with Ba(OH)2 in aqueous THF.
Coupling of 42 with the secondary amine 33 with use of DEPBT
(4 equiv, 1.5 equiv of 33, 4 equiv of NaHCO3, THF, 0 °C, 65%)
gave the amide 43 without any racemization at the 6′-position.
Unsatisfactory results were obtained when other coupling
conditions (i.e., EDCI and HOBt, EDCI and HOAt) were used
for the preparation of 43. Under these conditions, starting
materials remained unreacted in the reaction mixture, and only
a trace amount of 43 was obtained. To convert the terminal
olefin of 43 to an aldehyde, ozonolysis was first tried. However,
oxidation of the double bond at the uracil moiety was favored.
Alternatively, 43 was treated with OsO4 (0.5 mol %, 2.5 equiv
of N-methylmorpholine N-oxide, acetone-H2O), and the result-
ing mixture of the diastereomeric diols was oxidatively cleaved
to provide the aldehyde 44 (2.7 equiv of NaIO4, acetone-
phosphate buffer (pH 7), 60% overall) to afford the precursor
for the cyclization. Then, the optimized reductive amination
condition was applied to 44 (H2, Pd black, iPrOH, followed by
(50) Jiang, H.; Li, X.; Fan, Y.; Ye, C.; Romoff, T.; Goodman, M. Org.
Lett. 1999, 1, 91-93.
(51) Desilylation during Pd-catalyzed hydrogenation reactions was
reported. See: Sajiki, H.; Ikawa, T.; Hattori, K.; Hirota, K. Chem. Commun.
2003, 654-655.
(52) Dess, D. B.; Martin, J. C. J. Org. Chem. 1988, 48, 4155-4156.
9942 J. Org. Chem., Vol. 72, No. 26, 2007

Total Synthesis of (+)-Caprazol
SCHEME 3. Total Synthesis of (+)-Caprazol
SCHEME 4. Reaction Mechanism of Formation of 46
4 equiv of NaBH(OAc)3, AcOH, AcOEt), and the desired
diazepanone 45 was obtained in 24% yield along with its
N-methylated derivative 46, which was the desired product of
the next step of the synthesis, in 34% yield. The structure of
46 was confirmed by several NMR measurements including
HMBC and HMQC. In the diazepanone moiety, the substituents
at the 2′′′ and 3′′′ positions are placed in the pseudoaxial
orientation as determined by NOESY experiments, where
correlations were observed between H-6′, H-1′′′, and H-4′′′
(Figure 5). It was presumed that the methyl source in 46 was
formaldehyde arising upon deprotection of the BOM protecting
group at the N-3 position. Thus, the formaldehyde generated in
situ reacted with the newly formed secondary amine of the
diazepanone 45 to form the corresponding exo-iminium ion,
which was further reduced with NaBH(OAc)3 to afford 46
(Scheme 4). Compound 46 could also be prepared from 45
((CHO)n, 4 equiv of NaBH(OAc)3, AcOH, AcOEt, 65%). At
this stage, the remaining steps were functional group manipula-
tions and deprotections. Treatment of 46 with NH4F (20 equiv,
MeOH, 60%) resulted in selective deprotection of the TBDPS
protecting group at the primary hydroxyl to give 47, which was
transformed to the carboxylic acid 48 by a two-step sequence
with Dess-Martin periodinane52 (3.4 equiv, CH2Cl2) and
NaClO2 oxidation53 (3.5 equiv, 1 equiv of NaH2PO4‚2H2O, 4.5
equiv of 2-methyl-2-butene, tBuOH-H2O, 56% overall). Oxida-
tion of the tertiary amine in the diazepanone was a side reaction
that reduced the chemical yield of 48. Finally, a global
deprotection of 48 (40% aqueous HF, MeCN, 50%) provided
(+)-(5), [R]25D +23.8 (c 0.24, DMSO) [lit.1b [R]19D +28 (c 0.5,
DMSO)], the properties of which were identical in all respects
with those reported for the natural material.1b
Conclusion. In summary, we have described the total
synthesis of (+)-caprazol. The key elements of the approach
include the early stage introduction of the aminoribose in a
highly â-selective manner, using steric hindrance in the transi-
tion state and construction of the diazepanone by a modified
intramolecular reductive amination. For the ribosylation, it was
revealed that increasing the size of the alkyl substituents of the
acetal unit resulted in severe steric repulsion on the R-face of
the riboside and led to an unusual outside attack to provide
â-ribosides with a complete reversal of stereoselectivity. This
method can be considered as an alternative for the construction
of â-ribosides without neighboring group participation. Con-
struction of the diazepanone structure was optimized and the
reductive amination with a hydride reagent proved to be
effective.
(53) Andres, J. M.; Elena, N. D.; Pedrosa, R. Tetrahedron 2000, 56,
1523-1531.
J. Org. Chem, Vol. 72, No. 26, 2007 9943

Hirano et al.
phenyl), 7.12 (d, 1H, H-6, J6,5 ) 7.7 Hz), 5.74 (d, 1H, H-5, J5,6
)
Experimental Section
7.7 Hz), 5.58 (d, 1H, NH, JNH,6′ ) 8.6 Hz), 5.47 (d, 1H, NCH2-
OBn, J ) 9.9 Hz), 5.40 (d, 1H, NCH2OBn, J ) 9.9 Hz), 5.35 (s,
1H, H-1′), 5.13 (m, 3H, H-3′, CO2CH2Ph), 5.08 (dd, 1H, H-2′, J2′,1′
) 1.6 Hz, J2′,3′ ) 6.7 Hz), 4.66 (m, 2H, NCH2OCH2Ph), 4.58 (d,
1H, H-4′, J4′,5′ ) 9.2 Hz), 4.42 (m, 1H, H-5′), 4.15 (m, 1H, H-6′),
3.71 (s, 3H, COCH3), 3.26 (br s, 1H, OH), 1.53 (s, 3H, acetonide),
1.31 (s, 3H, acetonide); 13C NMR (CDCl3, 125 MHz) δ 171.2,
162.3, 157.0, 151.0, 142.2, 137.7, 136.2, 128.5, 128.3, 128.2, 128.0,
127.8, 127.6, 114.3, 102.3, 97.8, 87.8, 84.0, 81.2, 72.4, 71.5, 70.4,
67.2, 55.6, 52.7, 27.1, 25.1, 11.4; FABMS-HR (NBA) calcd for
C31H36N3O11 626.2350, found 626.2368.
Methyl (E)-1-(3-Benzyloxymethyluracil-1-yl)-5,6-dideoxy-2,3-
O-isopropylidene-â-D-ribo-5-ene-heptofuranuronate (8). A solu-
tion of 2′,3′-O-isopropylideneuridine (10.0 g, 41.0 mmol) and IBX
(28.6 g, 103 mmol) in MeCN (400 mL) was heated at 80 °C for
1 h. The reaction mixture was cooled in an ice bath, then the white
precipitates were filtered off. The filtrate was concentrated in vacuo.
The residue in CH2Cl2 (400 mL) was cooled to -20 °C, to which
a solution of Ph3PdCHCO2Me (16.4 g, 49.2 mmol) in CH2Cl2 (100
mL) was added dropwise. The reaction mixture was stirred at
-20 °C for 1 h. The mixture was diluted with AcOEt, then washed
with H2O and saturated aqueous NaCl. The organic phase was dried
(Na2SO4), filtered, and concentrated in vacuo. The residue in CH2-
Cl2 (300 mL) was treated with BOMCl (8.5 mL, 61.5 mmol), Bu4-
NI (756 mg, 2.05 mmol), and Na2CO3 (17.4 g, 164 mmol) in H2O
(200 mL). The resulting biphasic layers were vigorously stirred at
room temperature for 12 h. The organic phase was diluted with
AcOEt, then washed with H2O and saturated aqueous NaCl. The
organic phase was dried (Na2SO4), filtered, and concentrated in
vacuo. The residue was purified by a silica gel column (15 × 20
Methyl 5-O-[5-Azido-5-deoxy-2,3-O-(3-pentylidene)-â-D-ri-
bofuranosyl]-6-benzyloxycarbonylamino-1-(3-benzyloxymethy-
luracil-1-yl)-6-deoxy-2,3-O-isopropylidene-â-D-glycero-L-talo-
heptofuranuronate (21b) and Methyl 5-O-[5-Azido-5-deoxy-2,3-
O-(3-pentylidene)-R-D-ribofuranosyl]-6-benzyloxycarbonylamino-
1-(3-benzyloxymethyluracil-1-yl)-6-deoxy-2,3-O-isopropylidene-
â-D-glycero-L-talo-heptofuranuronate (22b). A mixture of 9 (20.0
mg, 0.032 mmol), 16 (11.8 mg, 0.048 mmol), and MS4A (100 mg)
in CH2Cl2 (1 mL) was stirred at -30 °C for 15 min. BF3‚OEt2 (1.2
µL, 0.01 mmol) was added five times at 1 h intervals. The reaction
mixture was stirred for 5 h in total. Saturated aqueous NaHCO3 (5
mL) was added and the mixture was extracted with AcOEt. The
organic phase was washed with saturated aqueous NaCl, dried (Na2-
SO4), filtered, and concentrated in vacuo. The residue was purified
by a silica gel column (6 × 15 cm, 33% AcOEt-hexane) to afford
cm, 25-45% AcOEt-hexane) to afford 8 (12.4 g, 66%) as a
1
colorless syrup: [R]22 +52.0 (c 1.66, CHCl3); H NMR (CDCl3,
D
500 MHz) δ 7.32 (m, 5H, phenyl), 7.13 (d, 1H, H-6, J6,5 ) 8.1
Hz), 7.02 (dd, 1H, H-5′, J5′,6′ ) 15.6 Hz, J5′,4′ ) 5.6 Hz), 6.04 (dd,
1H, H-6′, J6′,5′ ) 15.6 Hz, J6′,4′ ) 1.5 Hz), 5.76 (d, 1H, H-5, J5,6
)
8.1 Hz), 5.57 (d, 1H, H-1′, J1′,2′ ) 1.3 Hz), 5.48 (d, 1H, NCH2-
OBn, J ) 9.8 Hz), 5.43 (d, 1H, NCH2OBn, J ) 9.8 Hz), 5.04 (dd,
1H, H-2′, J2′,1′ ) 1.3 Hz, J2′,3′ ) 6.4 Hz), 4.87 (dd, 1H, H-3′, J3′,2′
) 6.4 Hz, J3′,4′ ) 1.1 Hz), 4.70 (s, 2H, NCH2OCH2Ph), 4.66 (ddd,
1H, H-4′, J4′,3′ ) 1.1 Hz, J4′,5′ ) 5.6 Hz, J4′,6′ ) 1.5 Hz), 3.70 (s,
21b (20.9 mg, 77%) and 22b (0.8 mg, 3%) each as a white foam.
3H, COCH3), 1.58 (s, 3H, acetonide), 1.36 (s, 3H, acetonide); 13
C
1
Data for 21b: [R]22 +20.2 (c 1.36, CHCl3); H NMR (CDCl3,
D
NMR (CDCl3, 125 MHz) δ 166.1, 162.4, 150.6, 143.7, 141.1, 137.9,
128.3, 127.7, 127.6, 122.3, 114.7, 102.4, 96.0, 86.9, 84.6, 84.0,
72.3, 70.3, 51.7, 27.1, 25.3; FABMS-HR (NBA) calcd for
C23H27N2O8 459.1767, found 459.1756.
500 MHz) δ 7.38-7.28 (m, 11H, phenyl, H-6), 5.79 (d, 1H, NH,
JNH,6′ ) 9.7 Hz), 5.71 (d, 1H, H-5, J5,6 ) 8.1 Hz), 5.61 (s, 1H,
H-1′), 5.48 (d, 1H, NCH2OBn, J ) 9.7 Hz), 5.43 (d, 1H, NCH2-
OBn, J ) 9.7 Hz), 5.22 (d, 1H, CO2CH2Ph, J ) 12.1 Hz), 5.12 (s,
1H, H-1′′), 5.05 (d, 1H, CO2CH2Ph, J ) 12.1 Hz), 4.81 (m, 2H,
H-2′, H-3′), 4.70 (s, 2H, NCH2OCH2Ph), 4.65 (d, 1H, H-6′, J6′,NH
Methyl 6-Benzyloxycarbonylamino-1-(3-benzyloxymethylu-
racil-1-yl)-6-deoxy-2,3-O-isopropylidene-â-D-glycero-L-talo-hep-
tofuranuronate (9) and Methyl 6-Benzyloxycarbonylamino-1-
(3-benzyloxymethyluracil-1-yl)-6-deoxy-2,3-O-isopropylidene-R-
L-glycero-D-allo-heptofuranuronate (10). tert-Butyl hypochlorite
(1.13 mL, 10.0 mmol) was added to a solution of benzyl carbamate
(1.49 g, 9.87 mmol) in aqueous NaOH (0.6 M, 14 mL) and n-PrOH
(16.5 mL) at 15 °C, and the mixture was stirred for 15 min, then
allowed to reach room temperature. A solution of [DHQD]2AQN
(422 mg, 0.49 mmol) in PrOH (8.3 mL), a solution of 8 (1.50 g,
3.28 mmol) in PrOH (8.3 mL), and a solution of K2OsO2(OH)4
(180 mg, 0.498 mmol) were sequentially added to the mixture. The
resulting mixture was stirred at room temperature for 2 h. After
addition of saturated aqueous Na2S2O3 (100 mL), the reaction
mixture was extracted with AcOEt. The combined organic phases
were washed with saturated aqueous NaCl, dried (Na2SO4), filtered,
and concentrated in vacuo. The residue was purified by a silica gel
column (6 × 15 cm, 40-45% AcOEt-hexane) to afford 9
(1.08 g, 52%) as a white foam and 10 (175 mg, 9%) as a white
foam. Data for 9; [R]22D +34.4 (c 1.03, CHCl3); 1H NMR (CDCl3,
) 9.7 Hz), 4.56 (m, 2H, H-2′′, H-3′′), 4.46 (d, 1H, H-5′, J5′,4′
)
7.0 Hz), 4.22 (m, 2H, H-4′, H-4′′), 3.74 (s, 3H, COCH3), 3.38 (dd,
1H, H-5′′a, J5′′a,5′′b ) 12.7 Hz, J5′′a,4′′ ) 5.6 Hz), 3.32 (dd, 1H, H-5′′b,
J5′′b,5′′a ) 12.7 Hz, J5′′b,4′′ ) 7.9 Hz), 1.59 (m, 2H, CH2CH3), 1.48
(m, 5H, CH2CH3, acetonide), 1.32 (s, 3H, acetonide), 0.83 (m, 6H,
CH2CH3 × 2); 13C NMR (CDCl3, 125 MHz) δ 170.5, 162.4, 156.3,
150.9, 140.7, 137.9, 136.3, 128.5, 128.3, 128.2, 127.7, 117.3, 114.9,
113.4, 102.2, 94.6, 86.8, 86.0, 85.4, 83.9, 81.9, 80.8, 78.8, 72.4,
70.4, 67.2, 54.6, 53.2, 52.8, 29.3, 28.9, 27.1, 25.4, 8.4, 7.4; FABMS-
HR (NBA) calcd for C41H51N6O14 851.3463, found 851.3447. Data
for 22b: [R]22D +14.9 (c 1.13, CHCl3); 1H NMR (CDCl3, 500 MHz)
δ 7.66 (d, 1H, H-6, J6,5 ) 7.9 Hz), 7.37-7.24 (m, 10H, phenyl),
5.97 (s, 1H, H-1′), 5.87 (d, 1H, NH, JNH,6′ ) 8.0 Hz), 5.73 (d, 1H,
H-5, J5,6 ) 7.9 Hz), 5.48 (d, 1H, NCH2OBn, J ) 9.7 Hz), 5.44 (d,
1H,NCH2OBn, J ) 9.7 Hz), 5.17 (d, 1H, CO2CH2Ph, J ) 12.3
Hz), 5.14 (br s, 1H, H-1′′), 5.07 (d, 1H, CO2CH2Ph, J ) 12.3 Hz),
4.87 (dd, 1H, H-3′, J3′,2′ ) 6.2 Hz, J3′,4′ ) 3.7 Hz), 4.70 (s, 2H,
NCH2OCH2Ph), 4.63 (d, 1H, H-6′, J6′,NH ) 8.0 Hz), 4.59 (m, 2H,
H-2′, H-2′′), 4.53 (dd, 1H, H-3′′, J3′′,2′′ ) 6.8 Hz, J3′′,4′′ ) 3.0 Hz),
4.44 (m, 1H, H-4′), 4.32 (m, 1H, H-5′), 4.22 (m, 1H, H-4′′), 3.73
(s, 3H, COCH3), 3.46 (m, 1H, H-5′′a), 3.35 (m, 1H, H-5′′b), 1.56
(s, 3H, acetonide), 1.49 (s, 3H, acetonide), 1.33 (s, 3H, acetonide),
1.32 (s, 3H, acetonide); 13C NMR (CDCl3, 125 MHz) δ 170.7,
162.5, 156.0, 150.9, 139.9, 137.9, 136.2, 128.5, 128.3, 128.2, 128.0,
127.7, 127.6, 115.5, 114.7, 103.2, 102.4, 91.6, 84.8, 83.6, 80.9,
80.8, 80.6, 80.2, 78.0, 77.3, 72.3, 70.4, 67.2, 54.6, 52.8, 52.5, 29.7,
27.2, 26.2, 25.4, 25.3; FABMS-HR (NBA) calcd for C39H47N6O14
823.3150, found 823.3167.
500 MHz) δ 7.33-7.24 (m, 10H, phenyl), 7.15 (d, 1H, H-6, J6,5
)
8.1 Hz), 5.72 (d, 1H, H-5, J5,6 ) 8.1 Hz), 5.59 (br s, 1H, NH),
5.46-5.41 (m, 3H, NCH2OBn, H-1′), 5.12 (d, 1H, CO2CH2Ph, J
) 11.1 Hz), 5.06 (d, 1H, CO2CH2Ph, J ) 11.1 Hz), 4.95 (m, 2H,
H-2′, H-3′), 4.68 (m, 2H, NCH2OCH2Ph), 4.53 (m, 1H, H-6′), 4.26
(m, 2H, H-4′, H-5′), 3.72 (s, 3H, COCH3), 3.72 (br s, 1H, OH),
1.54 (s, 3H, acetonide), 1.34 (s, 3H, acetonide); 13C NMR (CDCl3,
125 MHz) δ 170.7, 162.2, 141.6, 151.2, 141.6, 137.7, 136.1, 128.5,
128.3, 128.2, 128.1, 127.8, 127.6, 114.8, 102.5, 97.0, 86.2, 82.8,
81.4, 72.4, 71.5, 70.4, 67.5, 56.5, 52.7, 27.2, 25.3, 12.0; FABMS-
LR m/z 626 (MH+); FABMS-HR (NBA) calcd for C31H36N3O11
626.2356, found 626.2340. Data for 10: [R]22 +24.9 (c 1.74,
D
CHCl3); 1H NMR (CDCl3, 500 MHz) δ 7.36-7.25 (m, 10H,
9944 J. Org. Chem., Vol. 72, No. 26, 2007

Total Synthesis of (+)-Caprazol
Methyl 5-O-[5-tert-Butoxycarbonylamino-5-deoxy-2,3-O-(3-
pentylidene)-â-D-ribofuranosyl]-6-deoxy-6-benzyloxycarbony-
lamino-1-(3-benzyloxymethyluracil-1-yl)-2,3-O-isopropylidene-
â-D-glycero-L-talo-heptofuranuronate (41). A solution of 21b (712
mg, 0.839 mmol) and Ph3P (660 mg, 2.52 mmol) in benzene-
THF (1:1, 8 mL) and H2O (755 µL, 41.1 mmol) was heated at
50 °C for 12 h. The reaction mixture was allowed to warm to room
temperature, to which (Boc)2O (389 µL, 1.68 mmol) and NaHCO3
(141 mg, 1.68 mmol) were added. The resulting mixture was stirred
at room temperature for 1 h and partitioned between AcOEt and
H2O. The organic phase was washed with saturated aqueous NaCl,
dried (Na2SO4), filtered, and concentrated in vacuo. The residue
was purified by a silica gel column (5 × 20 cm, 37-40% AcOEt-
hexane) to afford 41 (736 mg, 95% in 2 steps) as a white foam:
[R]23D +31.2 (c 0.63, CHCl3); 1H NMR (CDCl3, 500 MHz) δ 7.38-
7.27 (m, 10H, phenyl), 7.22 (d, 1H, H-6, J6,5 ) 8.0 Hz), 5.73 (d,
1H, H-5, J5,6 ) 8.0 Hz), 5.56 (d, 1H, NCH2OBn, J ) 9.8 Hz),
5.18 (m, 1H, NH), 5.12 (d, 1H, CO2CH2Ph, J ) 12.3 Hz), 5.06 (s,
1H, H-1′′), 5.02 (d, 1H, CO2CH2Ph, J ) 12.3 Hz), 4.97 (d, 1H,
129.6, 128.5, 128.4, 128.4, 128.3, 127.8, 127.7, 127.6, 117.5, 117.1,
116.7, 116.3, 115.2, 114.8, 112.0, 110.1, 102.6, 102.4, 92.7, 86.7,
86.2, 85.8, 84.8, 84.0, 83.5, 82.0, 81.8, 80.6, 79.2, 78.9, 78.3, 73.3,
73.0, 72.2, 70.5, 70.3, 67.1, 63.2, 60.8, 60.5, 50.9, 43.4, 43.2, 29.7,
29.5, 29.3, 29.0, 28.4, 27.2, 27.2, 26.9, 26.7, 25.7, 25.6, 25.5, 25.3,
19.0, 18.8, 17.8, 17.8, 8.3, 7.5, 7.4, -4.0, -4.2, -5.2, -5.3;
FABMS-HR (NBA) calcd for C73H102N5O17Si2 1376.6807, found
1376.6800.
N-[(1R,2S)-2-tert-Butyldimethylsiloxy-1-tert-butyldiphenylsi-
loxymethyl-3-oxopropyl]-N-methyl-6-benzyloxycarbonylamino-
1-(3-benzyloxymethyluracil-1-yl)-5-O-[5-tert-butoxycarbonylamino-
5-deoxy-2,3-O-(3-pentylidene)-â-D-ribofuranosyl]-6-deoxy-2,3-
O-isopropylidene-â-D-glycero-L-talo-heptofuranuronamide (44).
A solution of 43 (310 mg, 0.23 mmol) and NMO (121 mg, 0.56
mmol) in acetone-H2O (4:1, 2.5 mL) was treated with OsO4 (5.0
mg/mL t-BuOH solution, 500 µL), and the resulting reaction
mixture was stirred at room temperature for 24 h. After addition
of saturated aqueous Na2S2O3 (10 mL), the mixture was extracted
with AcOEt. The organic phase was washed with saturated aqueous
NaCl, dried (Na2SO4), filtered, and concentrated in vacuo. The
residue was purified by a short silica gel column (3 × 5 cm, 33%
AcOEt-hexane, then 2% MeOH, 50% AcOEt-hexane), and
fractions containing the diol were combined and concentrated in
vacuo. A solution of the diol in acetone-phosphate buffer (4:1,
pH 7.5, 3 mL) was treated with NaIO4 (123 mg, 0.61 mmol) at
room temperature for 12 h. After addition of saturated aqueous
Na2S2O3 (10 mL), the mixture was extracted with AcOEt. The
organic phase was washed with saturated aqueous NaCl, dried (Na2-
SO4), filtered, and concentrated in vacuo. The residue was purified
by a silica gel column (3 × 10 cm, 3% acetone-CHCl3) to afford
44 (189 mg, 60%) as a white foam: [R]21D -19.8 (c 0.34, CHCl3);
1H NMR (CDCl3, 500 MHz, 4:1 mixture of rotamers) δ 9.70 (s,
0.3H, H-1′′′), 9.56 (s, 1.0H, H-1′′′), 7.64-7.59 (m, 5.0H, phenyl),
7.42-7.20 (m, 21.3H), 6.07 (br s, 1.0H), 5.89 (br s, 0.3H, H-1′),
5.76 (m, 1.3H), 5.68-5.66 (m, 1.9H), 5.48 (d, 0.3H, NCH2OBn, J
) 9.7 Hz), 5.44 (d, 0.3H, NCH2OBn, J ) 9.7 Hz), 5.37 (d, 1.0H,
NCH2OBn, J ) 9.6 Hz), 5.28 (d, 1.0H, NCH2OBn, J ) 9.6 Hz),
5.19 (m, 1.3H), 5.12 (m, 0.3H), 4.96-4.92 (m, 2.5H), 4.90-4.75
(m, 2.2H), 4.70-4.35 (m, 7.6H), 4.32-4.18 (m, 2.0H), 4.14-4.08
(m, 2.3H), 4.01-3.98 (m, 1.3H), 3.85 (m, 1.0H), 3.67 (m, 0.3H),
3.25-3.16 (m, 1.3H), 3.10 (m, 4.0H, CONCH3), 2.86 (s, 0.7H,
CONCH3), 1.60-1.40 (m, 20H), 1.31 (m, 5.0H), 1.00 (s, 9H, tert-
butyl), 0.98 (s, 3.0H), 0.93 (s, 3.0H), 0.87 (s, 9.0H, tert-butyl),
0.85-0.78 (m, 8H), 0.10 (s, 0.7H, SiCH3), 0.07 (s, 0.7H, SiCH3),
0.03 (s, 3.0H, SiCH3), -0.05 (s, 3.0H, SiCH3); 13C NMR (CDCl3,
125 MHz) δ 199.9, 162.3, 156.2, 150.9, 140.7, 137.8, 136.2, 135.5,
135.4, 132.7, 132.5, 129.8, 128.4, 128.2, 128.1, 127.9, 127.8, 127.6,
127.6, 116.7, 114.6, 112.7, 102.4, 86.8, 85.9, 84.3, 82.0, 80.8, 79.7,
79.1, 72.2, 70.4, 70.1, 67.2, 60.0, 51.5, 43.2, 29.3, 28.8, 28.4, 27.2,
27.1, 26.6, 26.5, 25.8, 25.7, 25.3, 19.0, 18.0, 17.9, 8.4, 7.3, 7.2,
-4.2, -4.6, -5.3, -5.4; FABMS-HR (NBA) calcd for C72H100N5O18-
Si2 1378.6602, found 1378.6580.
H-2′, J2′,3′ ) 6.3 Hz), 4.84 (dd, 1H, H-3′, J3′,2′ ) 6.3 Hz, J3′,4′
)
4.4 Hz), 4.70 (s, 2H, NCH2OCH2Ph), 4.67 (m, 1H, H-6′), 4.52 (m,
2H, H-2′′, H-3′′), 4.42 (d, 1H, H-5′, J5′,4′ ) 8.2 Hz), 4.21 (m, 2H,
H-4′, H-4′′), 3.77 (s, 3H, COCH3), 3.20 (m, 1H, H-5′′a), 3.07 (m,
1H, H-5′′b), 1.57-1.44 (m, 4H, CH2CH3 × 2), 1.48 (s, 3H,
acetonide), 1.41 (s, 9H, tert-butyl), 1.33 (s, 3H, acetonide), 0.79-
0.73 (m, 6H, CH2CH3 × 2); 13C NMR (CDCl3, 125 MHz) δ 171.1,
162.4, 156.1, 156.0, 150.8, 141.4, 137.8, 136.1, 128.4, 128.3, 128.2,
127.7, 116.5, 114.7, 112.5, 102.2, 96.1, 87.0, 86.2, 84.1, 82.0, 81.1,
79.8, 79.3, 72.3, 70.3, 67.2, 54.9, 53.0, 43.3, 29.3, 28.8, 28.3, 27.0,
25.3, 8.3, 7.2; FABMS-HR (NBA) calcd for C46H61N4O16 925.4083,
found 925.4064.
N-[(1R,2S)-2-tert-Butyldimethylsiloxy-1-tert-butyldiphenylsi-
loxymethyl-3-butenyl]-N-methyl-6-benzyloxycarbonylamino-1-
(3-benzyloxymethyluracil-1-yl)-5-O-[5-tert-butoxycarbonylamino-
5-deoxy-2,3-O-(3-pentylidene)-â-D-ribofuranosyl]-6-deoxy-2,3-
O-isopropylidene-â-D-glycero-L-talo-heptofuranuronamide (43).
A mixture of barium hydroxide octahydrates (260 mg, 0.824 mmol)
and 41 (558 mg, 0.60 mmol) in THF-H2O (4:1, 10 mL) was stirred
at room temperature for 10 h. The reaction mixture was poured
onto 1 M aqueous HCl and extracted with CHCl3. The organic phase
was dried (Na2SO4), filtered, and concentrated in vacuo. The residue
was purified by a silica gel column (5 × 13 cm, 33% MeOH-
CHCl3) to afford crude 33 (319 mg, ca. 0.35 mmol, 50%) as a
colorless foam. A solution of the crude 42 and 33 (254 mg, 0.527
mmol) in THF (3.5 mL) was treated sequentially with NaHCO3
(118 mg, 1.40 mmol) and DEPBT (420 mg, 1.40 mmol) at 0 °C
for 1 h, then the mixture was allowed to reach room temperature
and stirred for an additional 28 h. The reaction mixture was
partitioned between AcOEt and saturated aqueous NaHCO3. The
organic phase was washed with 1 M aqueous HCl and saturated
aqueous NaCl, dried (Na2SO4), filtered, and concentrated in vacuo.
The residue was purified by a flash silica gel column (8 × 10 cm,
17-33% AcOEt-hexane) to afford 43 (315 mg, 65%) as a white
foam: [R]23D -9.78 (c 0.45, CHCl3); 1H NMR (CDCl3, 500 MHz,
3:1 mixture of the rotamers) δ 7.69-7.56 (m, 5.2H, phenyl), 7.45-
7.23 (m, 22.1H, phenyl, H-6), 6.22 (m, 0.3H), 5.98 (m, 1.0H), 5.92-
5.80 (m, 3.0H), 5.48-5.44 (m, 2.0H, CO2CH2Ph), 5.38 (d, 1.0H,
CO2CH2Ph, J ) 9.6 Hz), 5.28 (m, 0.6H), 5.18-5.09 (m, 3.6H),
5.05-4.98 (m, 3.9H), 4.82-4.76 (m, 1.9H), 4.71-4.68 (m, 4.0H),
4.51 (m, 1.8H), 4.23 (m, 1.0H), 4.32-4.22 (m, 4.0H), 4.16-4.09
(m, 2.6H), 4.09-3.90 (m, 2.7H), 3.62 (t, 0.3H, J ) 10.2 Hz), 3.28
(m, 1.3H), 3.10 (m, 4.0H), 2.79 (s, 0.9H, include CONCH3), 1.71-
1.42 (m, 20.8H, tert-butyl, acetonide, CH2CH3), 1.30 (s, 3H,
acetonide), 1.29 (s, 3H, acetonide), 1.03 (s, 9H, tert-butyl), 0.98
(s, 3H, tert-butyl), 0.82 (m, 7.8H, CH2CH3), 0.73 (s, 9H, tert-butyl),
0.72 (s, 3H, tert-butyl), -0.12 (m, 3.9H, SiCH3), -0.15 (m, 3.9H,
SiCH3); 13C NMR (CDCl3, 125 MHz) δ 171.2, 170.5, 162.4, 162.3,
156.2, 156.2, 151.0, 150.9, 139.9, 139.7, 138.7, 138.1, 137.9, 136.2,
135.7, 135.5, 135.4, 136.2, 133.1, 132.7, 132.6, 129.9, 129.8, 129.7,
Diazepanone (45) and N-Methyldiazepanone (46). A mixture
of 44 (78.3 mg, 0.054 mmol) and Pd black (150 mg) in i-PrOH
(10 mL) was vigorously stirred under H2 atmosphere at room
temperature for 12 h. The catalyst was filtered off through a Celite
pad, and the filtrate was concentrated in vacuo. The residue in
AcOEt (4 mL) was treated with AcOH (40 µL) and NaBH(OAc)3
(45.4 mg, 0.22 mmol), and the reaction mixture was stirred at room
temperature for 12 h. The mixture was partitioned between AcOEt
and saturated aqueous NaHCO3. The organic phase was washed
with saturated aqueous NaCl, dried (Na2SO4), filtered, and con-
centrated in vacuo. The residue was purified by a neutral silica gel
column (1.5 × 13 cm, 33-50% AcOEt-hexane) to afford 45
(eluted with 40% AcOEt-hexane, 20.0 mg, 34% in 2 steps) and
46 (eluted with 50% AcOEt-hexane, 14.4 mg, 24% in 2 steps)
each as a white solid. Data for 45; [R]21 +19.8 (c 0.34, CHCl3);
D
1H NMR (CDCl3, 500 MHz) δ 7.71 (br s, 1H, H-3), 7.67-7.38
(m, 10H, phenyl), 7.21 (d, 1H, H-6, J6,5 ) 8.3 Hz), 6.22 (m, 1H,
J. Org. Chem, Vol. 72, No. 26, 2007 9945

Hirano et al.
NHBoc), 5.65 (d, 1H, H-5, J5,6 ) 8.3 Hz), 5.45 (s, 1H, H-1′), 5.33
(s, 1H, H-1′′), 4.84 (d, 1H, H-2′, J2′,3′ ) 6.1 Hz), 4.63 (d, 1H, H-2′′,
J2′′,3′′ ) 5.9 Hz), 4.54 (dd, 1H, H-3′, J3′,2′ ) 6.1 Hz, J3′,4′ ) 6.1
saturated aqueous Na2S2O3 (2:1, 2 mL), and AcOEt (2 mL) were
added to the mixture, and the resulting biphasic layers were
vigorously stirred at room temperature for 10 min. The organic
phase was washed with saturated aqueous NaCl, dried (Na2SO4),
filtered, and concentrated in vacuo. The residue was dissolved in
t-BuOH-H2O (3:1, 0.5 mL) and treated sequentially with 2-methyl-
2-butene (3.7 µL, 35.1 mmol), NaH2PO4‚2H2O (1.2 mg, 7.8 mmol),
and NaClO2 (3.1 mg, 27.3 mmol). The resulting reaction mixture
was stirred at room temperature for 10 min. After addition of
phosphate buffer (pH 7.5, 2 mL), the mixture was extracted with
CHCl3. The organic phase was washed with saturated aqueous NaCl,
dried (Na2SO4), filtered, and concentrated in vacuo. The residue
was purified by preparative TLC (17% MeOH-CHCl3) to afford
Hz), 4.49 (d, 1H, H-5′, J5′,4′ ) 7.7 Hz), 4.44 (d, 1H, H-3′′, J3′′,2′′
)
5.9 Hz), 4.25 (dd, 1H, H-4′′, J4′′,5′′a ) 8.9 Hz, J4′′,5′′b ) 5.2 Hz),
4.23 (dd, 1H, H-4′, J4′,3 ) 6.1, J4′,5′ ) 7.7 Hz), 3.95 (m, 1H, H-6′′′),
3.80 (m, 2H, H-8′′′a, H-8′′′b), 3.49 (m, 1H, H-7′′′, H-5′′a), 3.15 (s,
1H, H-3′′′), 3.09 (s, 3H, CONCH3), 2.98 (dd, 1H, H-5′′′a, J5′′′a,5′′′b
) 14.7, J5′′′a,6′′′ ) 3.2 Hz), 2.77 (d, 1H, H-5′′′b, J5′′′b,′′′a ) 14.7 Hz),
2.72 (m, 1H, H-5′′b), 1.59 (m, 2H, CH2CH3), 1.45 (m, 2H, CH2-
CH3), 1.40 (s, 9H, tert-butyl), 1.26 (s, 3H, acetonide), 1.23 (s, 3H,
acetonide), 1.05 (s, 9H, tert-butyl), 0.86 (s, 9H, tert-butyl), 0.81
(m, 6H, CH2CH3 × 2), 0.09 (s, 3H, SiCH3), 0.05 (s, 3H, SiCH3);
13C NMR (CDCl3, 125 MHz) δ 173.3, 162.3, 156.1, 149.3, 142.4,
135.5, 135.4, 132.8, 132.7, 130.1, 130.1, 128.0, 128.0, 116.3, 114.8,
110.5, 102.3, 93.8, 87.0, 86.6, 86.4, 84.2, 82.1, 80.1, 80.3, 78.4,
68.5, 66.8, 62.6, 59.9, 50.7, 43.7, 40.3, 29.8, 29.0, 28.5, 27.3, 26.7,
25.7, 25.6, 19.1, 17.8, 8.3, 7.3, -4.8, -4.9; FABMS-LR m/z 1108
48 (3.9 mg, 56% in 2 steps) as a white solid: [R]21 -27.5 (c
D
1
0.39, CHCl3); H NMR (CD3CN, 500 MHz) δ 9.50-9.23 (br s,
1H, CO2H), 7.65 (d, 1H, H-6, J6,5 ) 6.5 Hz), 6.52 (br s, 1H,
NHBoc), 5.80 (s, 1H, H-1′), 5.65 (d, 1H, H-5, J5,6 ) 6.5 Hz), 5.25
(s, 1H, H-1′′), 4.78 (m, 2H), 4.55 (m, 2H), 4.40 (m, 1H), 4.36 (m,
1H), 4.29 (m, 1H), 4.16 (m, 1H), 3.57 (m, 1H), 3.28-3.26 (m,
2H), 3.11 (m, 1H), 3.06 (s, 3H, CONCH3), 2.94 (m, 2H), 2.44 (s,
3H, NCH3), 1.60 (m, 2H, CH2CH3), 1.52 (m, 5H, CH2CH3,
acetonide), 1.36 (s, 9H, tert-butyl), 0.89 (s, 9H, tert-butyl), 0.83
(m, 6H, CH2CH3 × 2), 0.11 (s, 6H, SiCH3 × 2); 13C NMR (CD3-
OD, 100 MHz) δ 173.6, 166.1, 158.4, 151.8, 144.0, 143.9, 117.5,
115.8, 112.7, 102.6, 92.2, 88.3, 87.9, 86.7, 85.3, 83.7, 81.9, 81.6,
80.4, 71.4, 71.2, 64.2, 61.2, 44.3, 39.6, 38.9, 30.8, 30.5, 30.0, 29.0,
27.7, 27.6, 26.4, 25.8, 18.8, 8.8, 7.9, -4.8, -4.9; FABMS-HR
(NBA) calcd for C41H68N5O15Si 898.4481, found 898.4475.
Synthetic (+)-Caprazol (5). A solution of 48 (7.4 mg, 8.25
µmol) in MeCN (1.0 mL) was treated with 40% aqueous HF (100
µL), and the resulting mixture was stirred at room temperature for
18 h. After the mixture was neutralized with saturated aqueous
NaHCO3 and concentrated in vacuo, the residue was purified by
C18 reverse-phase HPLC (20 × 250 mm, 100% H2O) to afford
(MH+) FABMS-HR (NBA) calcd for C56H86N5O14Si2 1108.5711,
1
found 1108.5700. Data for 46: [R]21 -17.6 (c 0.67, CHCl3); H
D
NMR (CDCl3, 500 MHz) δ 9.06 (br s, 1H, H-3), 7.66-7.30 (m,
11H, phenyl, H-6), 6.67 (m, 1H, NHBoc), 5.59 (m, 2H, H-5, H-1′),
5.26 (s, 1H, H-1′′), 4.68 (m, 2H, H-2′, H-3′), 4.56 (m, 2H, H-2′′,
H-3′′), 4.33 (d, 1H, H-5′, J5′,6′ ) 7.3 Hz), 4.24-4.14 (m, 2H, H-4′,
H-4′′), 4.01 (m, 1H, H-6′′′), 3.92 (dd, 1H, H-8′′′a, J8′′′a,8′′′b ) 10.4,
J8′′′a,7′′′ ) 7.5 Hz), 3.86 (dd, 1H, H-8′′′b, J8′′′b,8′′′a ) 10.4 Hz, J8′′′b,7′′′
) 6.9 Hz), 3.52-3.48 (m, 2H, H-7′′′, H-3′′′), 3.20 (m, 1H, H-5′′a),
3.12 (m, 1H, H-5′′b), 3.07 (d, 1H, H-5′′′a, J5′′′a,5′′′b ) 14.6 Hz),
2.98 (s, 3H, CONCH3), 2.87 (d, 1H, H-5′′′b, J5′′′b,5′′′a ) 14.6 Hz),
2.41 (s, 3H, NCH3), 1.62 (m, 2H, CH2CH3), 1.52 (m, 2H, CH2-
CH3), 1.48 (s, 3H, acetonide), 1.35 (s, 9H, tert-butyl), 1.26 (s, 3H,
acetonide), 0.99 (s, 9H, tert-butyl), 0.88 (s, 9H, tert-butyl), 0.86-
0.79 (m, 6H, CH2CH3 × 2), 0.07 (s, 3H, SiCH3), 0.06 (s, 3H,
SiCH3); 13C NMR (CDCl3, 125 MHz) δ 170.3, 162.4, 156.4, 149.3,
139.9, 135.3, 135.3, 132.8, 132.6, 130.2, 127.9, 116.8, 114.8, 112.1,
102.4, 89.8, 87.4, 86.7, 85.9, 84.3, 82.7, 79.6, 78.9, 75.8, 69.3, 63.8,
63.2, 59.7, 43.2, 39.0, 37.7, 29.7, 29.4, 28.0, 28.6, 27.3, 26.7, 25.7,
25.6, 19.0, 17.8, 8.4, 7.5, -4.9, -5.0; FABMS-HR (NBA) calcd
for C57H88N5O14Si2 1122.5866, found 1122.5820.
synthetic (+)-caprazol (5, 2.4 mg, 50%) as a white solid: [R]25
D
+23.8 (c 0.24, DMSO); 1H NMR (D2O, 500 MHz) δ 7.77 (d, 1H,
H-6, J6,5 ) 8.1 Hz), 5.81 (d, 1H, H-5, J6,5 ) 8.1 Hz), 5.59 (s, 1H,
H-1′), 5.16 (s, 1H, H-1′′), 4.43 (m, 1H, H-3′′′), 4.38 (d, 1H, H-5′,
J5′,6′′′ ) 9.5 Hz), 4.30 (d, 1H, H-2′, J2′,3′ ) 5.1 Hz), 4.24 (m, 1H,
H-3′′), 4.21-4.18 (m, 2H, H-4′′, H-2′′′), 4.13-4.11 (m, 2H, H-4′,
H-2′′), 4.07 (dd, 1H, H-3′, J3′,2′ ) 5.2 Hz, J3′,4′ ) 7.8 Hz), 3.84 (d,
1H, H-6′′′, J6′′′,5′ ) 9.5 Hz), 3.32 (dd, 1H, H-5′′a, J5′′a,5′′b ) 14.0
Hz, J5′′a,4′′ ) 3.3 Hz), 3.19 (dd, 1H, H-5′′b, J5′′b,5′′a ) 14.0 Hz, J5′′b,4′′
) 4.4 Hz), 3.12 (d, 1H, H-4′′′a, J4′′′a,4′′′b ) 14.7 Hz), 3.06 (s, 3H,
CONCH3), 3.00 (d, 1H, H-4′′′b, J4′′′b,4′′′a ) 14.7 Hz), 2.42 (s, 3H,
NCH3); 13C NMR (D2O, 100 MHz) δ 174.1, 172.7, 167.0, 151.8,
143.0, 111.2, 101.7, 91.9, 82.5, 79.0, 77.7, 75.5, 74.2, 70.6, 70.1,
69.4, 63.6, 59.2, 40.2, 39.4, 37.0; ESIMS-HR (NBA) calcd for
C22H32N5O13 574.1997, found 574.2012.
Alcohol (47). A solution of 46 (10.0 mg, 7.8 mmol) in MeOH
(1 mL) was treated with NH4F (50 mg) at room temperature for
48 h. The mixture was diluted with AcOEt, and the insoluble
materials were filtered off. The filtrate was concentrated in vacuo,
and the residue was purified by preparative TLC (33% AcOEt-
hexane) to afford 47 (5.7 mg, 72%) as a white solid: [R]21D -17.6
1
(c 0.67, CHCl3); H NMR (CD3CN, 500 MHz, 10:1 mixtures of
the conformers) δ 9.00 (br s, 1.0 H, NH-3), 7.56 (d, 1H, H-6, J6,5
) 8.2 Hz), 6.82 (br s, 1H, NHBoc), 5.81 (s, 1H, H-1′), 5.64 (d,
1H, H-5, J5,6 ) 8.2 Hz), 5.27 (s, 1H, H-1′′), 4.80 (m, 2H, H-2′,
H-3′), 4.59 (m, 2H, H-2′′, H-3′′), 4.34 (m, 1H, H-5′), 4.26 (m, 2H,
H-4′, H-4′′), 4.00 (m, 1H, H-6′′), 3.69 (m, 3H, H-3′′′, H-8′′′a,
H-8′′′b), 3.42 (m ,1H, H-7′′′), 3.18 (m, 3H, H-5′′a, H-5′′b, H-5′′′a),
3.04 (s, 3H, CONCH3), 3.02 (br s, 1H, OH), 2.83 (m, 1H, H-5′′′b),
2.41 (s, 3H, NCH3), 1.63 (m, 2H, CH2CH3), 1.52 (m, 5H, CH2-
CH3, acetonide), 1.39 (s, 9H, tert-butyl), 1.31 (s, 3H, acetonide),
0.89 (s, 9H, tert-butyl), 0.83 (m, 6H, CH2CH3 × 2), 0.09 (s, 6H,
SiCH3 × 2); 13C NMR (CD3CN, 125 MHz) δ 163.7, 157.2, 151.1,
142.0, 142.0, 117.1, 115.2, 113.0, 103.1, 90.9, 87.8, 87.5, 87.0,
84.9, 83.2, 80.8, 79.2, 77.8, 70.9, 70.3, 63.8, 62.1, 60.5, 43.9, 38.3,
30.1, 29.6, 28.7, 27.5, 26.1, 25.6, 18.4, 14.3, 11.3, 8.7, 7.7, -4.7,
-4.9; FABMS-HR (NBA) calcd for C41H70N5O14Si 884.4689,
found 884.4678.
Acknowledgment. This work was supported by grants-in-
aid for scientific research from the Ministry of Education. We
thank Dr. Masayuki Igarashi (Bioresources Unit, Bioactive
Molecular Research Group, Microbial Chemistry Research
Center) for providing us with a sample of the natural caprazol
and the coordinate data of its X-ray crystal structure analysis.
We thank Ms. S. Oka (Center for Instrumental Analysis,
Hokkaido University) for measurement of the mass spectra.
Supporting Information Available: Experimental procedures,
1H NMR, and 13C NMR spectra for all new compounds. This
material is available free of charge via the Internet at http://pubs.
acs.org.
Carboxylic Acid (48). A solution of 47 (6.5 mg, 7.8 mmol) in
CH2Cl2 (0.5 mL) was treated with Dess-Martin periodinane (11.4
mg, 26.5 mmol) at 0 °C for 40 min. Saturated aqueous NaHCO3-
JO701699H
9946 J. Org. Chem., Vol. 72, No. 26, 2007