Total synthesis of jadomycins B, S, T, and ILEVS1080

Article abstract of DOI:10.1002/chem.201301297

Sweetening up jadomycin A: The first total synthesis of jadomycins B, S, T, and ILEVS1080 has been achieved, featuring construction of the unique 8H-benz[b]oxazolo[3,3-f]phenanthridine skeleton by biomimetic condensation of a quinone aldehyde with amino acid sodium salts and elaboration of the glycosides by Mitsunobu condensation (see figure). Copyright

Full text of DOI:10.1002/chem.201301297

COMMUNICATION
DOI: 10.1002/chem.201301297
Total Synthesis of Jadomycins B, S, T, and ILEVS1080
Xiaoyu Yang and Biao Yu*[a]
Jadomycins are a unique family of angucycline antibiotics
methyl)-1,3-oxazolidin-5-one moiety, could not be accessed;
instead the corresponding 1,3-oxazolidin-4-carboxylic acid
derivatives were detected. These results suggest that the
proposed structures of these two jadomycin congeners
should be revised accordingly (Figure 1). Herein, we report
the first total syntheses of jadomycins B, S, T, and
ILEVS1080 (1–4).
featuring an 8H-benzo[b]phenanthridine skeleton fused with
an amino acid unit.[1] With the exception of jadomycin A (7,
which is the aglycon of jadomycin B (1))[2] and ILEVS1080
(2, which bears a 6-deoxy-a-l-altrose group),[3] all of the ja-
domycin congeners bear an a-l-digitoxose unit at C12-OH.
The biosynthesis of the jadomycins is intriguing, involving
an oxidative cleavage of the an-
gucycline B ring[4] and subse-
quent nonenzymatic incorpora-
tion of the amino acid residue.[5]
In fact, all of the jadomycins
are produced upon feeding
Streptomyces venezuelae under
stress conditions with an amino
acid (including non-natural
amino acids) as the nitrogen
source.[2,3,5] These compounds
show potent antitumor activi-
Figure 1. Structures of jadomycin B and ILEVS1080 and the proposed and revised structures of jadomycin S
and T.
ties,[5d–h] which may be due to
their ability to inhibit the
Aurora-B kinase[6] and induce
Inspired by biosynthetic transformations, OꢀDoherty et al.
DNA cleavage.[7] Antibacterial activity and activity against
yeast are also reported for jadomycins.[2c,5h,8] The sugar unit
is found to be critically important to these activities, since
no activity has been observed for the aglycon
jadomycin A.[2c,5,8] The side chain of the amino acid unit sig-
nificantly modulates the activities.[5d–h,7a] Thus, jadomycins S
(3) and T (4), embedded with l-serine and l-threonine, re-
spectively, are among the most active antitumor
congeners.[5e]
Chemical synthesis of jadomycin A was achieved in 2010
by OꢀDoherty and co-workers, but installation of an l-digi-
toxose unit onto the relevant aglycons did not succeed
under various conditions.[9] Recently, Ishikawa et al. report-
ed the synthesis of jadomycin A together with a few of the
jadomycin aglycons.[10] They found that the proposed agly-
cons of jadomycins S and T, which contain the 4-(hydroxy-
condensed a 2-(2-benzyl aldehyde)-1,4-naphthoquinone (5)
with tert-butyl isoleucinate to obtain jadomycin A (7)
through a cascade of imine formation, 6p-electrocyclic ring
closure, hydration and air oxidation, cleavage of the
tert-butyl group (upon treatment with CF3COOH), and oxa-
zolone-ring formation.[9] Alternatively, Ishikawa et al. intro-
duced the amino acid unit onto a 2-[2-(hydroxymethyl)aryl]-
1,4-naphthoquinone moiety through Michael addition and
air oxidation. Subsequent oxidation of the benzyl alcohol
and spontaneous isoquinoline–oxazolidone formation, fol-
lowed by deprotection, furnished jadomycin A and the other
jadomycin aglycons.[10b]
Herein, we describe a protecting-group-free biomimetic
synthesis of jadomycin A (7) through direct condensation of
naphthoquinone aldehyde 5 with sodium l-isoleucinate 6
(Table 1). Thus, a solution of aldehyde 5[9,11,12] in toluene
was charged with solid 6; a red precipitate appeared slowly
over 2 d. Washing a solution of the resulting solid in EtOAc
with 1n HCl, in an effort to remove excess 6, resulted in a
dark-green solution. Chromatography of the concentrated
residue on silica gel provided jadomycin A (7) in 25% yield
(Table 1, entry 1). The same reaction in THF, in which iso-
leucinate 6 has better solubility, proceeded better; the reac-
tion mixture turned purple after 3 h and became dark green
upon treatment with H+ resin for 15 min, and the yield of 7
[a] Dr. X. Yang, Prof. B. Yu
State Key Laboratory of Bioorganic and Natural Products Chemistry
Shanghai Institute of Organic Chemistry
Chinese Academy of Sciences
345 Lingling Road, Shanghai 200032 (P.R. China)
Fax : (+86)21-64166128
E-mail: byu@mail.sioc.ac.cn
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201301297.
Chem. Eur. J. 2013, 19, 8431 – 8434
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8431

Table 1. Protecting-group-free biomimetic synthesis of jadomycin A (7).
quinone precursor was successful, leading finally to a carba-
sugar analogue of jadomycin B.[9] Previously, we have care-
fully studied the glycosylation of a similar angucycline
phenol in landomycins and solved the problem by SN2 sub-
stitution of a glycosyl a-iodide with the angucycline pheno-
late.[13] Thus, we attempted the glycosylation of 7 with
3,4-di-O-acetyl-l-digitoxosyl iodide, which was prepared in
situ from the corresponding acetate 8 (TMSI, CH2Cl2,
08C),[12,14,15] in the presence of KHMDS and [18]C-6 in
THF. The reaction led to a complex mixture of products
(Table 2, entry 1); detection of glycal derivatives in the mix-
ture indicated the vulnerability of the digitoxosyl iodide to
decomposition under the reaction conditions. In fact, the
stability of the glycosyl iodide (which contains an electron-
withdrawing substituent at C6) has been found to be crucial
to the successful glycosylation in the landomycin synthe-
sis.[13] We therefore attempted the glycosylation with the
more stable glycosyl bromide (derived in situ from acetate 8
with TMSBr).[12] Indeed, the desired glycoside 11 was iso-
Entry Conditions
Yield [%]
1
2
3
4
5
6
1) toluene, RT, 2 d; 2) 1n HCl
1) THF, RT, 3 h; 2) H+ resin
25
51
1) THF/MeOH or THF/H2O, RT; 2) H + resin complex mixture
1) DMF, RT; 2) H+ resin
complex mixture
34
50
1) CH2Cl2/H2O, RT; 2) H + resin
1) CHCl3/H2O, 45 8C; 2) H+ resin
(d.r.=7:1 at C3a) was improved to 51% (Table 1, entry 2).
More polar solvents such as THF/MeOH, THF/H2O, and
DMF were also used for the condensation reaction; howev-
er, these reactions resulted in complex mixtures of products
(Table 1, entries 3 and 4).
AHCTUNGTREGUNlNN ated, albeit in only a moderate yield of 25% with no
a/b selectivity (Table 2, entry 2).
Mitsunobu condensation of 7 with 3,4-di-O-acetyl-l-digi-
toxose (9) was also examined.[12,14] Under conventional con-
ditions (PPh3 (2 equiv), DEAD (2.5 equiv), 4 ꢂ MS, tol-
uene),[16] the reaction at 08C led to a complex mixture of
products; fortunately, at À788C, the reaction proceeded
smoothly to furnish the desired glycoside 11 in 45% yield in
favor of the a anomer (a/b=3:1; separable on silica gel)
with 43% of the starting material (7) being recovered. The
yield of 11 was further increased to 64% (with 31% of 7 re-
covered) and the a/b ratio to 6:1 when the amount of donor
9 was increased to 3.0 equivalents (Table 2, entry 5). Apply-
ing these optimized conditions to the coupling of 7 with
2,3,4-tri-O-acetyl-6-deoxy-l-altrose (10)[12,17] provided the
desired a-glycoside 12 in a satisfactory 78% yield without
To mimic a biogenetic environment, we carried out the
condensation reaction under biphasic conditions. Although
the reaction in CH2Cl2/H2O at RT produced jadomycin A
(34%), the reaction in CHCl3/H2O at 458C led to a better
yield (50%; Table 1, entries 5 and 6). The less polar solvent
was found to be crucial to the condensation reaction in that
formation of an intramolecular hydrogen bond between
C7-OH and the quinone in 5 forces the aldehyde residue to
take the orientation necessary for the cascade cyclization re-
action.[9] The red intermediate before treatment with acid
1
showed a pair of H NMR spectroscopic singlets at d=8.05
and 8.10 ppm, supporting the previous proposal that
aldimine A
(in
a
Z/E
mixture),[5a] rather than the cor-
responding amine (derived
from Michael addition),[4a] is in-
volved in the biogenesis of ja-
domycins.
Table 2. Glycosylation of jadomycin A and synthesis of jadomycin B (1) and ILEVS1080 (2).[a]
Installation of an l-digitoxose
unit onto the poorly nucleophil-
ic
C12-OH
group
in
jadomycin A (or its precursors)
has been tried by OꢀDoherty
et al. under a variety of Mitsu-
nobu conditions, as well as
under the Schmidt glycosylation
conditions and Pd-catalyzed
glycosylation conditions with a
pyranone derivative as the
donor.[9] Unfortunately, no gly-
cosylation product could be iso-
lated. However, the Mitsunobu
condensation of a cyclitol deriv-
Entry Donor
([equiv])
Conditions
Results
(yield [%], a/b)
G
ACHTUNGTRENNUNG
1
2
3
4
5
6
8 (1.5)
8 (1.5)
9 (1.5)
9 (1.5)
9 (3)
1) 8, TMSI, CH2Cl2, 08C; 2) KHMDS, [18]C-6, 4 ꢂ MS, THF, 08C
1) 8, TMSBr, CH2Cl2, 08C; 2) KHMDS, [18]C-6, 4 ꢂ MS, THF, 08C 11 (25, 1:1)
PPh3, DEAD, 4 ꢂ MS, toluene, 08C
PPh3, DEAD, 4 ꢂ MS, toluene, À78C
PPh3, DEAD, 4 ꢂ MS, toluene, À788C
PPh3, DEAD, 4 ꢂ MS, toluene, À708C
complex mixture
complex mixture
11 (45, 3:1); 7 (43)
11 (64, 6:1); 7 (31)
12 (78, 1:0)
10 (3)
[a] TMS=trimethylsilyl, KHMDS=potassium hexamethyldisilazide, [18]C-6=1,4,7,10,13,16-hexaoxacycloocta-
ative with a 2-aryl-1,4-naphtho- decane ([18]Crown-6), DEAD=diethyl azodicarboxylate, MS=molecular sieves.
8432
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 8431 – 8434

Jadomycins B, S, T, and ILEVS1080
COMMUNICATION
detection of the b anomer. In contrast, the rigid l-digitoxose
donors that are installed with 3,4-O-carbonate or 3,4-O-iso-
propylidene could not be coupled with 7 under Mitsunobu
conditions, as reported by OꢀDoherty et al.[9] Finally, remov-
al of the acetyl groups on glycosides 11a (3aS/3aR=3:2)
and 12 (d.r.=1:1) with NaOMe afforded jadomycin B (1,
3aS/3aR=3:2) and ILEVS1080 (2, d.r.=3:2), respectively, in
high yields. The analytical data for the synthetic compounds
1 and 2 are in full agreement with those reported for the
natural products.[3,5a,12]
Considering the ambiguity in the structure of jadomycin S,
we devised an unambiguous synthesis by condensing naph-
thoquinone aldehyde 5 with sodium O-TBS-l-serinate 13
(Scheme 1). Thus, treatment of 5 and 13 under the opti-
mized conditions (Table 1, entry 2) gave 1,3-oxazolidin-5-
one 14 in 59% yield (d.r.=1:1). Coupling of 14 with l-digi-
toxose donor 9 under the previously optimized Mitsunobu
conditions provided a complex mixture due to the unexpect-
ed cleavage of the TBS group. Reducing the amounts of 9
(1.2 equiv) and PPh3/DEAD (1.2 equiv) in the reaction led
to the desired glycoside 15 in 28% yield (a/b=8:1; d.r.=
3:1). Removal of the acetyl
out a synthesis directed specifically towards the revised
structure (Scheme 2). Thus, l-serine allyl ester 17 was used
in the condensation reaction with aldehyde 5, to provide the
1,3-oxazolidine 19 in good yield (64%). In this case, only
one isomer was isolated, which was determined to be the S
isomer at C3a based on the absence of a NOSEY signal be-
tween H-1 and H-3a. Mitsunobu glycosylation of 19 with
l-digitoxose 9 led to glycoside 21 in good yield (62%; a/b=
5:1, easily separable). Cleavage of the allyl ester group in
21a with [PdACHTUNTRGNEUNG(PPh3)4] in the presence of dimethyl barbituric
acid gave the corresponding acid in 73% yield,[18] which was
then subjected to the conditions for removal of the acetyl
groups to furnish the desired compound 3 (64%). Similarly,
starting from the condensation of l-threonine allyl ester 18
with aldehyde 5, the revised structure of jadomycin T (4)
was obtained concisely (Scheme 2), the analytical data of
which agree reasonably well with those reported for the nat-
ural product.[5a,12]
In conclusion, the total syntheses of jadomycins B, S, T,
and ILEVS1080 (1–4) have been achieved, in that the chal-
lenging glycosylation of jadomycin aglycons was realized for
groups on 15 led to 16 (82%;
d.r.=1:1), which was then sub-
jected to cleavage of the TBS
group to furnish the proposed
structure of jadomycin S. Treat-
ment of 16 with tetra-n-butyl-
ACHTUNGTRENNUNGammonium fluoride (TBAF) or
TBAF/HOAc led to decomposi-
tion of the oxazolone ring, as
indicated by the color change
from purple to yellow; however,
with 3HF·Et3N in MeOH, the
final product 3 was obtained in
good yield (73%), the analyti-
cal data of which match reason-
ably well with those reported Scheme 1. Synthesis of the proposed structure of jadomycin S (TBS=tert-butyldimethylsilyl).
for natural jadomycin S.[5a,12]
The structure of 3 was assigned
to be the 1,3-oxazolidin-4-car-
boxylic acid based on the facts
that 1) the H-3a NMR spectro-
scopic signal of 3 appears at
5.73 ppm, whereas for 16 it ap-
pears at 6.56 and 6.37 ppm;
2) the
IR
absorption
at
1621 cmÀ1 in 3 is in accord with
a
carboxylic group, whereas
that at 1808 cmÀ1 for 16 implies
an ester group; and 3) com-
pound 3 is obtained as a single
isomer from
a
1:1 C3a
diastereoACHTUNGTRENNUNGisomeric mixture of
16.
To further confirm the struc-
ture of jadomycin S, we carried Scheme 2. Synthesis of the revised jadomycins S (3) and T (4) structures.
Chem. Eur. J. 2013, 19, 8431 – 8434
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8433

B. Yu and X. Yang
Kharel, C. Morris, T. Downey, Y. Hou, T. S. Bugni, J. Rohr, J. Am.
Chem. Soc. 2012, 134, 18181; f) K. Fan, G. Pan, X. Peng, J. Zheng,
W. Gao, J. Wang, W. Wang, Y. Li, K. Yang, Chem. Biol. 2012, 19,
1381.
the first time by resorting to Mitsunobu condensation with
sugar lactols bearing the proper protecting-group patterns.
The protecting-group-free biomimetic condensation of a
2-aryl-1,4-naphthoquinone (i.e., 5) with the sodium salts of
amino acids (e.g., 6) to furnish the unique
8H-benz[b]oxazoloACHTUNGTRENNUNG[3,3-f]phenanthridine jadomycin aglycons
is also hightlighted. This synthesis has unambiguously con-
firmed that the originally assigned structures of jadomy-
[5] a) U. Rix, J. T. Zheng, L. L. R. Rix, L. Greenwell, K. Yang, J. Rohr,
J. Am. Chem. Soc. 2004, 126, 4496; b) D. L. Jakeman, S. Farrel, W.
Young, R. J. Doucet, S. C. Timmons, Bioorg. Med. Chem. Lett. 2005,
15, 1447; c) D. L. Jakeman, C. L. Graham, T. R. Reid, Bioorg. Med.
Chem. Lett. 2005, 15, 5280; d) J. T. Zheng, U. Rix, L. Zhao, C. Mat-
tingly, V. Adams, Q. Chen, J. Rohr, K. Q. Yang, J. Antibiot. 2005, 58,
405; e) C. N. Borissow, C. L. Graham, R. T. Syvitski, T. R. Reid, J.
Blay, D. L. Jakeman, ChemBioChem 2007, 8, 1198; f) S. N. Dupuis,
T. Veinot, S. M. A. Monro, S. E. Douglas, R. T. Syvitski, K. B. Goral-
ski, S. A. McFarland, D. L. Jakeman, J. Nat. Prod. 2011, 74, 2420;
g) K. Fan, X. Zhang, H. Liu, H. Han, Y. Luo, Q. Wang, M. Geng, K.
Yang, J. Antibiot. 2012, 65, 449; h) S. N. Dupuis, A. W. Robertson, T.
Veinot, S. M. A. Monro, S. E. Douglas, R. T. Syvitski, K. B. Goralski,
S. A. McFarland, D. L. Jakeman, Chem. Sci. 2012, 3, 1640.
[6] D. H. Fu, W. Jiang, J. T. Zheng, G. Y. Zhao, Y. Li, H. Yi, Z. R. Li,
J. D. Jiang, K. Q. Yang, Y. Wang, S. Y. Si, Mol. Cancer Ther. 2008, 7,
2386.
cins S and
T should be revised from containing a
4-(hydroxylmethyl)-1,3-oxazolidin-5-one unit to containing a
1,3-oxazolidin-4-carboxylic acid unit. This structural revision
explains the fact that jadomycins S and T are the jadomycin
congeners which occur as single isomers, with no epimeriza-
tion occurring at the aminal C3a, supporting the conclusion
that the carbonyl group of the oxazolone unit is essential for
ring opening of jadomycins.[19] The unique carboxylic acid
side residue in jadomycins S and T also rationalizes their
stronger antitumor activities than the other congeners. The
present concise approach could be easily adapted to the syn-
thesis of various jadomycin congeners, thus in-depth studies
on their activities is now a feasible task.
[7] a) K. M. Cottreau, C. Spence, J. R. Wentzell, C. L. Graham, C. N.
Borissow, D. L. Jakeman, S. A. McFarland, Org. Lett. 2010, 12, 1172;
b) S. M. A. Monro, K. M. Cottreau, C. Spencer, J. R. Wentzell, C. L.
Graham, C. N. Borissow, D. L. Jakeman, S. A. McFarland, Bioorg.
Med. Chem. 2011, 19, 3357.
[8] D. L. Jakeman, S. Bandi, C. L. Graham, T. R. Reid, J. R. Wentzell,
S. E. Douglas, Antimicrob. Agents Chemother. 2009, 53, 1245.
[9] M. Shan, E. U. Sharif, G. A. OꢀDoherty, Angew. Chem. 2010, 122,
9682; Angew. Chem. Int. Ed. 2010, 49, 9492.
Acknowledgements
[10] a) Y. Akagi, S. Yamada, N. Etomi, T. Kumamoto, W. Nakanishi, T.
Ishikawa, Tetrahedron Lett. 2010, 51, 1338; b) T. Tajima, Y. Akagi,
T. Kumamoto, N. Suzuki, T. Ishikawa, Tetrahedron Lett. 2012, 53,
383.
[11] Aldehyde 5 was prepared from 3-hydroxy-5-methyl benzoic acid
and 1,5-dihydroxynaphthalene in 12 steps with 5% overall yield; for
the major references, see: a) ꢃ. de Frutos, A. M. Echavarren, Tetra-
hedron Lett. 1996, 37, 8953; b) ꢃ. de Frutos, C. Atienza, A. M. Echa-
varren, Eur. J. Org. Chem. 2001, 163.
This work was financially supported by the NSFC (20932009 and
20921091) and the Ministry of Science and Technology of China
(2012CB822102). We are grateful to Prof. Keqian Yang and Keqiang Fan
for providing crude jadomycins B and S for HPLC comparison and to
Prof. J. Rohr for a valuable comment on the revision of the structures of
jadomycins S and T.
Keywords: amino acids
· antibiotics · glycosylation ·
[12] See the Supporting Information for details.
[13] a) X. Yang, B. Fu, B. Yu, J. Am. Chem. Soc. 2011, 133, 12433; b) X.
Yang, P. Wang, B. Yu, Chem. Rec. 2013, 13, 70.
jadomycins · Mitsunobu condensation · natural products ·
total synthesis
[14] The l-digitoxose derivative 9 was prepared from l-rhamnose in
9 steps and 12% overall yield; acetylation of 9 gave 8; for the major
reference, see: S. C. Timmons, D. L. Jakeman, Carbohydr. Res. 2007,
342, 2695.
[15] S. N. Lam, J. Gervay-Hague, Org. Lett. 2003, 5, 4219.
[16] a) G. Grynkiewicz, Carbohydr. Res. 1977, 53, C11; b) P. J. Garegg, T.
Iversen, T. Norberg, Carbohydr. Res. 1979, 73, 313; c) W. R. Roush,
X. F. Lin, J. Org. Chem. 1991, 56, 5740; d) W. R. Roush, X. F. Lin, J.
Am. Chem. Soc. 1995, 117, 2236.
[17] 6-Deoxy-l-altrose triacetate 10 was prepared from methyl a-l-rham-
nopyranoside in 9 steps and 12% overall yield; for the major refer-
ences, see: a) S. Ramesh, R. W. Franck, Tetrahedron: Asymmetry
1990, 1, 137; b) O. Han, H.-w. Liu, Tetrahedron Lett. 1987, 28, 1073.
[18] H. Kunz, J. Marz, Synlett 1992, 591.
[1] a) E. U. Sharif, G. A. OꢀDoherty, Eur. J. Org. Chem. 2012, 2095;
b) M. K. Kharel, P. Pahari, M. D. Shepherd, N. Tibrewal, S. E. Nybo,
K. A. Shaaban, J. Rohr, Nat. Prod. Rep. 2012, 29, 264; c) M. K.
Kharel, J. Rohr, Curr. Opin. Chem. Biol. 2012, 16, 150.
[2] a) S. W. Ayer, A. G. McInnes, P. Thibault, J. A. Walter, Tetrahedron
Lett. 1991, 32, 6301; b) J. L. Doull, S. W. Ayer, A. K. Singh, P. Thil-
bault, J. Antibiot. 1993, 46, 869; c) J. L. Doull, A. K. Singh, M.
Hoare, S. W. Ayer, J. Ind. Microbiol. 1994, 13, 120.
[3] D. L. Jakeman, C. N. Borissow, C. L. Graham, S. C. Timmons, T. R.
Reid, R. T. Syvitski, Chem. Commun. 2006, 3738.
[4] a) K. Yang, L. Han, S. W. Ayer, L. C. Vining, Microbiology 1996,
142, 123; b) U. Rix, C. Wang, Y. Chen, F. M. Lipata, L. L. R. Rix,
L. M. Greenwell, L. C. Vining, K. Yang, J. Rohr, ChemBioChem
2005, 6, 838; c) Y.-H. Chen, C. C. Wang, L. Greenwell, U. Rix, D.
Hoffmeister, L. C. Vining, J. Rohr, K. Q. Yang, J. Biol. Chem. 2005,
280, 22508; d) M. K. Kharel, L. Zhu, T. Liu, J. Rohr, J. Am. Chem.
Soc. 2007, 129, 3780; e) N. Tibrewal, P. Pahari, G. Wang, M. K.
[19] R. T. Syvitski, C. N. Borissow, C. L. Grahamand, D. L. Jakeman,
Org. Lett. 2006, 8, 697.
Received: April 6, 2013
Published online: May 13, 2013
8434
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 8431 – 8434