Formal synthesis of tubelactomicins via a transannular Diels-Alder approach

Article abstract of DOI:10.1016/j.tetlet.2007.10.002

A formal synthesis of the antimicrobial tricyclic macrolides, tubelactomicins A and E, featured by a transannular Diels-Alder (TADA) approach, has been explored. The key issue for the transannular cyclization was the synthesis of a 24-membered macrolacton

Full text of DOI:10.1016/j.tetlet.2007.10.002

Available online at www.sciencedirect.com
Tetrahedron Letters 48 (2007) 8442–8448
Formal synthesis of tubelactomicins via a transannular
Diels–Alder approach
Toshimi Anzo, Akari Suzuki, Kiyoto Sawamura, Toru Motozaki, Madoka Hatta,
Ken-ichi Takao and Kin-ichi Tadano*
Department of Applied Chemistry, Keio University, Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan
Received 29 August 2007; revised 27 September 2007; accepted 1 October 2007
Available online 4 October 2007
Abstract—A formal synthesis of the antimicrobial tricyclic macrolides, tubelactomicins A and E, featured by a transannular Diels–
Alder (TADA) approach, has been explored. The key issue for the transannular cyclization was the synthesis of a 24-membered
macrolactone equipped with all the requisite functionalities, which has been achieved using an intramolecular Hiyama cross-coup-
ling strategy. The Hiyama coupling reaction spontaneously triggered off the TADA reaction. From the endo-TADA adduct, formal
syntheses of tubelactomicins A and E were achieved. The 24-membered macrolactone formation was also achieved via an intra-
molecular ring-closing metathesis approach.
Ó 2007 Elsevier Ltd. All rights reserved.
The isolation and structure determination of (+)-tube-
lactomicin A (1) (Scheme 1) were reported by studies
at the Institute of Microbial Chemistry in 2000.1 This
tricyclic 16-membered macrolide 1 showed potent anti-
microbial activity against acid-fast bacteria, including
drug-resistant strains. Following the isolation of 1, the
same group isolated and characterized structurally
similar macrolides, tubelactomicins B (2), D (3), and E
(4).2 These antibiotics 2–4 also showed a broad range
of antimicrobial activity. In 2005, we reported the total
synthesis of 1, thereby establishing the stereochemistry
of the (+)-natural form.3 Tatsuta and co-workers have
also accomplished the total synthesis of 1.4 Recently,
we also accomplished the total syntheses of 2–4 and
established their unknown stereochemistries.5 In the
accomplished total synthesis of (+)-tubelactomicins,
our group and Tatsuta’s group utilized intramolecular
Diels–Alder (IMDA) approaches for the construction
of the octahydronaphthalene parts in 1–4. On the other
hand, it might be possible that the tricyclic structures of
1–4 are constructed by a transannular Diels–Alder
(TADA) reaction of a 24-membered macrolactone
equipped with all the requisite functionalities. Attracted
by this hypothesis, we have explored the total synthesis
of tubelactomicins using this plausible TADA approach
as a key step.6 Herein, we report the total syntheses of 1
and 4, which were realized using the attempted TADA
strategy. Our synthetic approach is summarized in
Scheme 1. As substrates for the synthesis of the key
24-membered macrolactone such as 5, we envisioned
long-chained esters 6, which incorporate with vinylst-
annane, vinylboronate, or vinylsilane and vinyl iodide
functionalities at both terminals. These highly function-
alized esters 6 would be obtained by the Wittig
olefination of a-phosphonopropionyl ester
7 and
(E,E,E)-undeca-6,8,10-trienal (for with M =
9
6
trimethylsilyl = TMS). Ester 7 could be prepared from
the previously reported vinylstannane 8.3b On the other
hand, (trienyl)trimethylsilane 9 could be obtained by
the Horner–Wadsworth–Emmons (HWE) olefination
of the known aldehyde 103a and (E,E)-pentadienyl phos-
phonate 11. Along this synthetic plan, we started the
synthesis of the 24-membered lactone 5.
First, the a-phosphonopropionate 7 was synthesized
from 8 via a two-step manipulation, that is, tin–iodine
exchange, followed by esterificaton with commercially
available 12 (Scheme 2). We initially explored the
HWE olefination of 7 with aldehyde 13,3a a synthetic
intermediate of our previous total synthesis of 1. This
Keywords: Tubelactomicins; Macrolide antibiotics; Transannular
Diels–Alder reaction; Hiyama coupling; Ring-closing olefin metathesis.
*
Corresponding author. Tel./fax: +81 45 566 1571; e-mail: tadano@
applc.keio.ac.jp
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.10.002

T. Anzo et al. / Tetrahedron Letters 48 (2007) 8442–8448
8443
R1
R1
MOMO
HO
Me
Stille coupling
Me
Me
OH
Me
then
macrolactonization
O
O
H
SnBu3
Me
+
I
see refs. 3 and 5
CO2Me
H
R2
R3
H
OH
Tubelactomicins
R2
R3
H
A (1): R1=CO2H, R2=Me, R3=Me
OH
B (2): R1=Me, R2=Me, R3=Me
the octahydronaphthalene parts
prepared by IMDA approaches
D (3): R1=CO2H, R2=CH2OH, R3=Me
E (4): R1=CO2H, R2=Me, R3=CH2OH
The total synthesis of tubelactomicins featured by two IMDA reactions
CO2Me
CO2Me
O
MOMO
Me
MOMO
Me
Me
Me
sp2-sp2
O
O
O
I
M
coupling
Me
TADA
Me
1 and 4
Me
Me
O
O
O
O
Ph
5
6: M=SnBu3, B(OH)2,
Ph
a 24-membered macrolactone
or SiMe3
E-selective
HWE olefination
CO2Me
CO2Me
MOMO
MOMO
Me
Me
OH
Me
Me
O
O
SnBu3
8 see, ref. 3b
I
P(O)(OEt)2
Me
7
+
TBDPSO
OHC
TMS
CHO
TMS
+
Me
P(O)(OEt)2
O
O
Me
O
O
Ph
10 see, ref. 3a
11
Ph
( for 6 with M=SiMe3)
9
Scheme 1. Attempted TADA approaches.
olefination provided 14 in a yield of 63%, which was
de-C-silylated with Bu4NF, providing 15. As a model
experiment, we explored the macrolactone formation
using 15 via an intramolecular sp–sp2 cross-coupling
strategy. Thus, the Pd-catalyzed intramolecular Sono-
gashira coupling of 15 was explored using pyrrolidine

8444
T. Anzo et al. / Tetrahedron Letters 48 (2007) 8442–8448
1) I2, CH2Cl2, 0 ºC
HO2C
P(O)(OEt)2
2) 12, EDCI•HCl, Et3N,
HOBt, CH2Cl2, reflux
Me
7
8
12
(98% for 2 steps)
CO2Me
MOMO
Me
Me
CHO
TMS
O
O
I
1) LiCl, DBU,
Me
R
CH3CN
+
7
Me
2) Bu4NF,
THF, 0 ºC
O
O
Me
O
O
Ph
13 see, ref. 3a
Ph
14 R=TMS (63%)
15 R=H (quant.)
CO2Me
CO2Me
MOMO
MOMO
Me
Me
O
Me
Me
+
O
Me
O
O
H
O
H
Pd(PPh3)4, CuI,
pyrrolidine, rt, 1 h
Me
15
Me
Me
H
H
O
O
O
16-exo (9%)
Ph
16-endo (10%)
Ph
Scheme 2. The Sonogashira coupling/TADA reaction of 15.
as a base at room temperature. After 1 h, this coupling
produced two products. Interestingly, these products
were identified to be an endo-TADA adduct 16-endo
and an exo-TADA one 16-exo, which were isolated in
10% and 9% yields, respectively. Neither the Sonogash-
ira coupling product nor other by-products were iso-
lated from the reaction mixture. We consider that the
low yields of 16-endo and 16-exo are mainly due to the
difficulty in the formation of 24-membered intermediate
in the intramolecular cyclization step. On the other
hand, we conclude that the TADA reaction occurred
instantly after the Sonogashira coupling product was
formed. It should be emphasized that the IMDA reac-
tion of 15 was not the first event. This was evidenced
by the fact that the IMDA reaction of 14 started by
heating 14 at 80 °C and completed after 4 days, provid-
ing an approximately 1:2 mixture (1H NMR analysis) of
endo- and exo-adducts (not shown) in a combined yield
of 93%. We then focused our efforts on the synthesis of
substrates 6 (R = SnBu3 or B(OH)2) via the functional-
ization of the acetylene moiety in 15. Despite extensive
efforts,7 however, we could not find efficient conditions
for the synthesis of these terminally metalated olefins
as the substrates for the attempted intramolecular Stille
or Suzuki–Miyaura coupling reaction.
Next, we explored the synthesis of vinylsilane 6
(M = SiMe3) as the substrate for intramolecular Hiyama
cross-coupling.8 The synthesis of 6 was expected to be
achieved by the HWE reaction of 7 and aldehyde 9
with an (E,E,E)-trienyl trimethylsilane moiety. For the
synthesis of 9, we explored the synthesis of pentadienyl
phosphonate 11 as the HWE reaction partner of the pre-
viously reported aldehyde 10.3a The synthesis of 11 was
achieved as shown in Scheme 3. Thus, the LiAlH4 reduc-
tion of commercially available C-silylated propargylic
alcohol 17 provided trans-allylic alcohol 18,9 which
was converted into unsaturated aldehyde 19 with
MnO2.10 The Wittig reaction of 19 with Ph3P = C(Me)-
CO2Et provided the a,b:c,d-unsaturated ester 20. The
diisobutylaluminum hydride (DIBAL-H) reduction of
20 provided pentadienyl alcohol 21 (E:Z = >20:1).
Bromination of 21 with a mixture of CBr4 and Ph3P,
followed by a thermal Arbusov rearrangement of the
resulting bromide 22 with triethyl phosphite, eventually
produced 11.
The HWE olefination of 103a and 11 using potassium
hexamethyldisilazide (KHMDS) as the base provided
(E,E,E)-triene 23 (Scheme 4). Deprotection of the
TBDPS (t-BuPh2Si) group in 23, followed by

T. Anzo et al. / Tetrahedron Letters 48 (2007) 8442–8448
8445
TMS
LiAlH4, DME,
0 ºC
TMS
TMS
MnO2, CH2Cl2
OH
OH
17
18
TMS
CBr4, Ph3P,
CH2Cl2
1) Ph3P=C(Me)CO2Et
benzene
2) DIBAL-H, CH2Cl2,
-78 ºC
CHO
Me
R
19
20 R=CO2Et (69% for 3 steps)
21 R=CH OH (87%)
E:Z >20:1
2
TMS
TMS
P(OEt)3,
110 ºC
Br
P(O)(OEt)2
Me
Me
22
11 (56% for 2 steps)
Scheme 3. Synthesis of pentadienyl phosphonate 11.
OTBDPS
TMS
1) HF•pyridine,
KHMDS,
THF:pyridine (3:1)
THF, -78 ºC to rt
10 + 11
2) Dess-Martin
Me
periodinane,CH2Cl2
O
O
Ph
TMS
23 (48%)
R
LiCl, DBU,
Me
CH3CN, 0 ºC
O
O
6 (M=TMS)
7 + 9
Ph
(73%)
24 R=CH2OH
9 R=CHO (62% for 2 steps)
Scheme 4. Synthesis of pentadienyl trimethysilane 6.
Dess–Martin oxidation11 of the resulting primary alco-
hol 24, provided aldehyde 9. Then, the HWE coupling
of 7 and 9 smoothly provided 6 (M = TMS), the sub-
strate for the intramolecular Hiyama cross-coupling.
coupling product 5 nor other isolable products were
found in the reaction mixture. We consider that the
rather low yields of the TADA adducts were the result
from the instability of the poly-unsaturated nature of
substrate 6 under the transition-metal-catalyzed reaction
conditions. In fact, a number of high-polar materials
were found in the reaction mixture (TLC monitoring)
during the Hiyama cross-coupling/TADA event. We
could not isolate or identify these products. Compared
to the previous results obtained from the stereoselective
(endo- and p-facial) and high-yielding IMDA reactions
realized in the total synthesis of tubelactomicins,3,5 the
endo/exo-selectivity observed in the Hiyama coupling/
TADA reaction of 6 was low.12 As expected from the
result of the Sonogashira coupling/TADA reaction of
15, however, it should be emphasized that the sequential
intramolecular cyclization reactions of 6 proceeded with
As shown in Scheme 5, the Pd-catalyzed intramolecular
Hiyama cross-coupling of 6 was examined in the pres-
ence of semi-catalytic amount of tetrabutylammonium
fluoride (TBAF, four-times addition of each 3 mol %
amount) at 60 °C for 20 h. As a result, two TADA
adducts 25-endo and 25-exo were obtained. The struc-
tures of these two adducts were determined by extensive
1H NMR analysis. Furthermore, one of the adducts,
that is, 25-endo, was identified with a synthetic inter-
mediate in the previous total synthesis of 4.5 Although
the yields of 25-endo and 25-exo were not necessarily
remarkable, neither the intermediary Hiyama cross-

8446
T. Anzo et al. / Tetrahedron Letters 48 (2007) 8442–8448
CO2Me
CO2Me
MOMO
MOMO
Me
Me
O
Me
Me
O
6
O
H
O
O
H
Me
(M=TMS)
Me
+
[allylPdCl]2,
TBAF (3 mol% x 4),
degassed THF
60 ºC, 20 h
Me
Me
H
H
O
O
O
Ph
Ph
25-endo (10%)
25-exo (10%)
CO2Me
MOMO
Me
Me
O
4 M HCl/
O
H
Me
MeOH (1:3)
ref.5
25-endo
4
Me
H
OH OH
26 (99%)
CO2Me
MOMO
Me
Me
1) TsCl, Et3N,
O
O
H
ref.3b
DMAP, CH2Cl2
Me
1
26
2) NaBH4, DMSO,
100 ºC
Me
R
H
OH
27 R=CH2OTs (97%)
28 R=Me (62%)
Scheme 5. The Hiyama coupling/TADA reaction of 6 and formal syntheses of 1 and 4.
complete p-facial selectivity to provide the expected two
adducts. The adduct 25-endo was converted into (+)-
tubelactomicin E (4) previously.5 On the other hand,
de-O-benzylidene product 26, prepared from 25-endo,
was selectively tosylated. Deoxygenation of the resulting
primary tosylate 27 with NaBH4 in hot DMSO provided
28,3b from which (+)-tubelactomicin A (1) was synthe-
sized previously.3b
TADA adducts 25-endo and 25-exo as an approximately
1:1 mixture in a combined yield of 13%. As the case of
the intramolecular Hiyama cross-coupling applied to
6, the initially formed RCM product 5 was not found
in the reaction mixture. The TADA reaction of 5, which
was produced by the RCM of 31, proceeded at a lower
temperature compared to the case of the Hiyama cou-
pling/TADA reaction of 6. Although the p-facial selec-
tivity of the TADA reaction was again complete, the
endo/exo-selectivity was almost the same as the result
obtained from the sequential cyclization achieved using
6. Furthermore, another RCM product corresponding
to 5, which possesses a newly introduced Z-olefin part,
was not found in the reaction mixture. We did not
explore further for the improvement of the yields and
selectivity of this RCM/TADA reaction.
Finally, we were interested in the 24-membered lactone
formation via a ring-closing olefin metathesis (RCM)
approach.13 As summarized in Scheme 6, we synthesized
linear heptaene 31 from the aforementioned vinyl iodide
14. Thus, the Stille coupling of 14 and vinyltributylst-
annane provided conjugated diene 29 in a moderate
yield of 59%. Protodesilylation of the TMS group in
29 provided 30. The hemi-hydrogenation of the triple
bond in 30 was achieved efficiently using a Lindler cata-
lyst to provide 31, the substrate for the attempted RCM
approach. After some experimentation, we found that
the second-generation Grubbs catalyst14 was solely
effective for the RCM of 31,15 which provided the
In conclusion, we have accomplished the formal synthe-
ses of tubelactomicins A and E via a TADA approach.
Substrate 5 for the TADA reaction, that is, a highly
functionalized 24-membered macrolactone equipped
with an all-E conjugated tetraene part, was synthesized

T. Anzo et al. / Tetrahedron Letters 48 (2007) 8442–8448
8447
CO2Me
MOMO
Me
1) Pd2(dba)3, PPh3,
Me
O
DMF, 60 ºC
H2, Lindler cat.,
THF
O
Me
R
SnBu3
14
2) TBAF, THF
Me
O
O
29 R=TMS (59%)
30 R=H (85%)
Ph
CO2Me
MOMO
Me
Me
O
2nd generation
Grubbs catalyst
(20 mol% x 2)
O
Me
O
25-endo
CH2Cl2, reflux, 33 h
+
25-exo
Me
31 (79%)
(13% combined yield;
ca. 1:1 mixture)
O
Ph
Scheme 6. The metathesis approach to the TADA adducts.
K. Org. Lett. 2005, 7, 2261–2264; (b) Motozaki, T.;
Sawamura, K.; Suzuki, A.; Yoshida, K.; Ueki, T.; Ohara,
A.; Munakata, R.; Takao, K.; Tadano, K. Org. Lett. 2005,
7, 2265–2267.
by a Hiyama cross-coupling or a ring-closing olefin
metathesis. Furthermore, these 24-membered macrolac-
tone forming reactions triggered off a spontaneous
TADA reaction.
4. Hosokawa, S.; Seki, M.; Fukuda, H.; Tatsuta, K. Tetra-
hedron Lett. 2006, 47, 2439–2442.
5. Sawamura, K.; Yoshida, K.; Suzuki, A.; Motozaki, T.;
Kozawa, I.; Hayamizu, T.; Munakata, R.; Takao, K.;
Tadano, K. J. Org. Chem. 2007, 72, 6143–6148.
6. We have accomplished the total synthesis of macquari-
micins using a TADA approach, see Munakata, R.;
Katakai, H.; Ueki, T.; Kurosaka, J.; Takao, K.; Tadano,
K. J. Am. Chem. Soc. 2004, 126, 11254–11267.
7. We examined the following conditions for 15: (1)
Bu3SnH/BuLi/CuCN/THF/À78 °C (complex mixture);
(2) catechol-borane/cat.9-BBN/THF/0 °C and then
H2O workup (partial decomposition of 15 and 43%
recovery of 15); (3) catecholborane/Cp2Zr(H)Cl/
benzene/rt and then H2O workup (partial decomposition
of 15 and 44% recovery of 15). For Lipshutz reagent in
condition (1), see Lipshutz, B. H.; Ellsworth, E. L.;
Dimock, S. H.; Reuter, D. C. Tetrahedron Lett. 1989, 30,
2065–2068.
Acknowledgments
We thank Dr. Masayuki Igarashi (the Institute of
Microbial Chemistry) for providing us the samples and
spectral copies (1H and 13C NMR, IR) of tubelacto-
micins. This work was supported by Grants-in-Aid for
Scientific Research on Priority Areas (A) ‘Creation of
Biologically Functional Molecules (18032069)’.
Supplementary data
The experimental procedures and 1H and 13C NMR
spectra for all new compounds. Supplementary data
associated with this article can be found, in the online
version, at doi:10.1016/j.tetlet.2007.10.002.
8. For a review on the Hiyama cross-coupling, see Hiyama,
T. In Metal-Catalyzed Cross-Coupling Reactions; Diede-
rich, F., Stang, P. J., Eds.; Wiley-VCH: Weinheim,
Germany, 1998; pp 421–453.
References and notes
9. Denmark, S. E.; Jones, T. K. J. Org. Chem. 1982, 47,
4595–4597.
1. (a) Igarashi, M.; Hayashi, C.; Homma, Y.; Hattori, S.;
Kinoshita, N.; Hamada, M.; Takeuchi, T. J. Antibiot.
2000, 53, 1096–1101; (b) Igarashi, M.; Nakamura, H.;
Naganawa, H.; Takeuchi, T. J. Antibiot. 2000, 53, 1102–
1107.
2. Takeuchi, T.; Igarashi, M.; Naganawa, H.; Hamada, M.
Japan Patent JP2001-55386A, 2001.
3. (a) Motozaki, T.; Sawamura, K.; Suzuki, A.; Yoshida, K.;
Ueki, T.; Ohara, A.; Munakata, R.; Takao, K.; Tadano,
10. Carter, M. J.; Fleming, I.; Percival, A. J. Chem. Soc.,
Perkin Trans. 1 1981, 2415–2434.
11. (a) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113,
7277–7287; (b) Ireland, R. E.; Liu, L. J. Org. Chem. 1993,
58, 2899; (c) Frigerio, M.; Santagostino, M.; Sputore, S. J.
Org. Chem. 1999, 64, 4537–4538.
12. Regarding the low endo/exo-selectivities observed in the
TADA reactions of 15, 6, and 31 (in Scheme 6), we do not
have reasonable explanation. On the other hand, we

8448
T. Anzo et al. / Tetrahedron Letters 48 (2007) 8442–8448
previously found the following experimental result during
2003; (b) Nicolaou, K. C.; Bulger, P. G.; Sarlah, D.
Angew. Chem., Int. Ed. 2005, 44, 4490–4527; For a recent
paper on the 12-membered macrolactone formation using
RCM, see Kanda, R. M.; Itoh, D.; Nagai, M.; Niijima, J.;
Asai, N.; Mizui, Y.; Abe, S.; Kotaka, Y. Angew. Chem.,
Int. Ed. 2007, 46, 4350–4355.
studies on the IMDA reactions for the stereoselective
construction of the octahydronaphthalene part. Thus,
when the bulkiness of the dienophile part increased, the
endo/exo-selectivity of the IMDA reaction reduced
remarkably. In the case of the substrate possessing a
CH@CHCOOR dienophile part in place of
a
14. Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett.
1999, 1, 953–956; For our synthetic achievement of the
mycoepoxydiene total synthesis using sequential ROM/
cross/RCM reaction, see Takao, K.; Yasui, H.; Yamam-
oto, S.; Sasaki, D.; Kawasaki, S.; Watanabe, G.; Tadano,
K. J. Org. Chem. 2004, 69, 8789–8795.
15. To maintain the reproducibility of the TADA reaction, it
is important to add the Grubbs catalyst (each 20 mol %)
for two times.
CH@CHCHO functionality, the IMDA reaction provided
a mixture of endo/exo-adducts in a 1:2 ratio. We speculate
that this lack of the endo/exo-selectivity may arise from
less effective secondary orbital interactions and/or increas-
ing steric hindrances of the dienophile part in the
transition-state of the IMDA reactions.
13. (a) For reviews on olefin metathesis, see Handbook of
Metathesis; Grubbs, R. H., Ed.; Wiley-VCH: Weinheim,