Total synthesis of the novel, immunosuppressive agent (-)-pateamine A from Mycale sp. employing a β-lactam-based macrocyclization

Full text of DOI:10.1021/ja973549f

J. Am. Chem. Soc. 1998, 120, 591-592
Total Synthesis of the Novel, Immunosuppressive
591
Agent (-)-Pateamine A from Mycale sp. Employing a
â-Lactam-Based Macrocyclization
Robert M. Rzasa, Helene A. Shea, and Daniel Romo*
Department of Chemistry, Texas A&M UniVersity
College Station, Texas 77843-3255
Figure 1. Retrosynthetic analysis of pateamine A and structures of Boc-
and TCBoc-pateamine A.
ReceiVed October 10, 1997
Immunosuppressive natural products such as cyclosporin A,
FK506, rapamycin, deoxyspergualin, and didemnin B have proven
to be useful tools for dissection of cell signaling pathways.1
Moreover, these studies have recently culminated in the synthesis
of agents that exert cell and target specific responses.2 Pateamine
A (1) is a structurally novel and potent immunosuppressive natural
product isolated from Mycale sp. by Munro and Blunt off the
coasts of New Zealand.3 It uniquely combines a thiazole ring
and an E,Z-dienoate within a bis-lactone macrocycle bearing an
all-E trienylamine side chain. This novel array of functionality
makes pateamine A structurally dissimilar to other known
immunosuppressants. As a first step toward elucidating the
origins of its cellular effects, we now report a convergent,
stereochemically flexible route to pateamine A,4 wherein a
â-lactam ring was strategically implemented to introduce the C3
amino group and to serve as an activated acyl group for
macrocyclization.5 We have previously described the synthesis
of the C18-C24 fragment of pateamine A which aided in the
determination of the C24 absolute stereochemistry in collaborative
studies with Munro and co-workers.6 Importantly, this assignment
in conjunction with molecular modeling, extensive NMR studies,
and further chemical derivitization by the New Zealand group
provided a tentative stereochemical assignment of (-)-pateamine
A as 3R, 5S, 10S, 24S.7 With this information in hand, we
embarked on a total synthesis that has now verified the stereo-
chemical assignment and has enabled the synthesis of derivatives
for further biological studies.
Scheme 1
the stereochemistry at C10 could be readily retained (acylation)
or inverted (Mitsunobu8) during the joining of â-lactam 4 and
enyne acid 5. The end-game strategy would rely upon a â-lactam-
based macrocyclization to esterify the C24 alcohol and a Stille
coupling to append the trienylamine side chain. Concise and
efficient syntheses of â-lactam 4, enyne acid 5,9 and the
dienylamino stannane 6 were required at the outset.
The dienylamino stannane 6 was readily prepared in two steps
from enyne alcohol 7 (Scheme 1).10 A one-pot tosylation and
displacement with dimethylamine provided the enyne amine 8.11
Stannylcupration of this alkyne by the method of Oehlschlager
gave the desired stannane 6 as a mixture of regioisomers which
could be enriched in the desired isomer (9:1).12
The synthesis of â-lactam 4 commenced with a Nagao acetate
aldol reaction13 between aldehyde 1014 and the (S)-valinol-derived
thiazolidinone 9 (diastereomeric ratio (dr) > 19:1; Scheme 2).
Following alcohol protection, aminolysis gave the amide 11 which
was then converted to the corresponding thioamide with the
Belleau reagent.15 A modified Hantzsch thiazole synthesis16
delivered thiazole ester 12. Half-reduction of this ester followed
by a Wadsworth-Emmons reaction simultaneously homologated
and introduced the (S)-phenylglycine-derived auxiliary required
for an asymmetric conjugate addition.17 Introduction of the C5
methyl group by the method of Hruby18 proceeded smoothly to
give a mixture of methyl adducts (dr 6.4:1) from which the major
diastereomer could be isolated in 77% yield. Transamidation
delivered the Weinreb amide 15 with the desired stereochemistries
at C5 and C10.19 What remained was installation of the â-lactam,
and to accomplish this task, we were attracted to the method of
Miller involving an intramolecular Mitsunobu reaction of an aldol
product.20 In the event, half-reduction of Weinreb amide 15 to
aldehyde 16 followed by a Nagao acetate aldol reaction gave the
Several issues guided our retrosynthetic plan (Figure 1). These
included the known lability of the C3 amino group,3 the
isomerization-prone E,Z-dienoate, the desire to incorporate and
liberate the polar amino groups at a late stage in the synthesis,
the flexibility of attaching various side chains to the macrocyclic
core structure, and finally the uncertainty of the stereochemical
assignment. With this last consideration in mind and due to the
distance between stereocenters, we resolved to introduce the four
stereocenters by reagent control. In addition, we realized that
some inherent flexibility was built into the synthetic plan since
(1) Hung, D. T.; Jamison, T. F.; Schreiber, S. L. Chem. Biol. 1996, 3, 623-
639.
(8) Mitsunobu, O. Synthesis 1981, 1-28.
(2) Diver, S. T.; Schreiber, S. L. J. Am. Chem. Soc. 1997, 119, 5106-
(9) We have described the synthesis of enyne acid 5,6 although the acid
used in the present study originated from a Noyori reduction of ethyl aceto-
acetate and was determined to be 94% enantiomeric excess by chiral GC.
(10) Alcohol 7 was generously provided by Dr. P. Weber (F. Hoffman-La
Roche Ltd., Switzerland).
5109.
(3) Northcote, P. T.; Blunt, J. W.; Munro, M. H. G. Tetrahedron Lett. 1991,
32, 6411-6414. This initial paper on pateamine A reported antifungal and
selective cytotoxic activity. Immunosuppressive activity was found by Dr.
Glynn Faircloth, PharmaMar Inc., Cambridge, MA (private communication):
MLR (mixed lymphocyte reaction) IC50 ) 2.6 nM; LCV (lymphocyte viability
assay)/MLR ratio >1000.
(11) Pisano, J. M.; Firestone, R. A. Synth. Commun. 1981, 11, 375-378.
(12) Aksela, R.; Oehlschlager, A. C. Tetrahedron 1991, 47, 1163-1167.
(13) Nagao, Y.; Hagiwara, Y.; Kumagai, T.; Ochiai, M.; Inoue, T.;
Hashimoto, K.; Fujita, E. J. Org. Chem. 1986, 51, 2391-2393.
(14) Prepared by MnO2 oxidation of the corresponding alcohol, see: Corey,
E. J.; Bock, M. G.; Kozikowski, A. P.; Rao, A. V. R.; Floyd, D.; Lipshutz, B.
Tetrahedron Lett. 1978, 19, 1051-1054.
(4) For other synthetic studies of pateamine A, see: Critcher, D. J.;
Pattenden, G. Tetrahedron Lett. 1996, 37, 9107-9110.
(5) For the use of â-lactams in intermolecular acylations, see: (a) Ojima,
I. In The Organic Chemistry of â-Lactams; Georg, G. I., Ed.; VCH
Publishers: New York, 1992; Chapter 4, pp 197-255. (b) Ojima, I. Acc. Chem.
Res 1995, 28, 383-389. For an intramolecular acylhydrazine addition to a
â-lactam forming a 10-membered ring, see: (c) Gardner, B.; Nakanishi, H.;
Kahn, M. Tetrahedron 1993, 49, 3433-3448. For transamidations of â-lactams
to medium-sized rings, see: (d) Hesse, M. Ring Enlargement in Organic
Chemistry; VCH: New York, 1991; Chapter VI.
(15) Lajoie, G.; Lepine, F.; Maziak, L.; Belleau, B. Tetrahedron Lett. 1983,
24, 3815-3818.
(16) Aguilar, E.; Meyers, A. I. Tetrahedron Lett. 1994, 35, 2473-2476.
(17) This type of homologation has been previously reported with the (S)-
phenylalanine-derived auxiliary, see: Broka, C. A.; Ehrler, J. Tetrahedron
Lett. 1991, 32, 5907-5910.
(6) Rzasa, R.; Romo, D.; Stirling, D. J.; Blunt, J. W.; Munro, H. M. G.
Tetrahedron Lett. 1995, 36, 5307-5310.
(7) (a) Blincoe, S. N. M.Sc. Thesis, University of Canterbury, New Zealand,
1994. (b) Stirling, D. J. Ph.D. Thesis, University of Canterbury, New Zealand,
1996.
(18) Li, G.; Patel, D.; Hruby, V. Tetrahedron: Asymmetry 1993, 4, 2315-
2318.
(19) X-ray analysis performed on an imide obtained as a byproduct during
the transamidation reaction verified the C5 and C10 stereochemistry. See the
Supporting Information for details.
S0002-7863(97)03549-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 01/13/1998

592 J. Am. Chem. Soc., Vol. 120, No. 3, 1998
Communications to the Editor
Scheme 2a
a Key: (a) N-ethylpiperidine, CH2Cl2, -40 °C (dr >19:1, 84%); (b) CH2Cl2, 0 °C (99%); (c) CH2Cl2, 0 °C (86%); (d) THF, 25 °C (82%); (e) DME,
-20 f 0 °C (94%); (f) CH2Cl2, -90 °C (93%); (g) THF, 25 °C (90%); (h) THF/DMS (3:2), -20 °C (dr 6.4:1, 77% isolated, major diastereomer); (i)
Me3Al, CH2Cl2, 0 °C (82%); (j) CH2Cl2, -90 °C (95%); (k) N-ethylpiperidine, CH2Cl2, -40 °C (dr, >19:1, 90%); (l) Me3Al, CH2Cl2, 0 °C (90%); (m)
THF, 25 °C (92%); (n) THF, 0 °C (96%); (o) CH2Cl2 (95%); (p) THF, 0 °C (95%); (q) THF, -20 °C (86%); (r) py, THF, 25 °C (86%); (s) CH2Cl2, 0
°C; (t) py, 25 °C; (u) MeOH, 25 °C (3 steps, 50%); (v) 1:4 Pd(15 mol %)/ligand, THF, 25 °C (27%, 57% based on recovered 20; (w) THF, 25 °C (80%).
desired aldol adduct in 90% yield. Conversion to the N-
(benzyloxy)amide and intramolecular Mitsunobu reaction gave
the â-lactam 17 in excellent overall yield. In preparation for the
â-lactam-based macrocyclization, reduction of the N-O bond was
involving acid deprotection of the Boc-macrocycle 19, reprotec-
tion with TCBoc-Cl, and Lindlar reduction provided the TCBoc-
macrocycle 20. Stille coupling of this vinyl bromide and stannane
6 employing the conditions of Farina30 gave TCBoc-pateamine 3
in 27% yield (57% based on recovered 20), and deprotection using
Cd/Pb couple gave (-)-pateamine A (1) in 80% yield. The
synthetic material displayed physical and spectral properties
21
cleanly effected with SmI2 in 96% yield and the resulting
â-lactam was protected as the tert-butyl carbamate.22 Deprotec-
tion of the triisopropylsilyl ether afforded the final key intermedi-
ate, â-lactam 4.
1
including H and 13C NMR, CD, IR, and UV identical to the
natural product.
With the required fragments in hand, their coupling to provide
pateamine A began by a Mitsunobu coupling between the alcohol
4 and the enyne acid 5 providing the required C10 (S)-
stereochemistry.23 Deprotection of the tert-butyldimethylsilyl
ether gave alcohol 18 and set the stage for the key â-lactam-
based macrocyclization. Building on conditions developed by
Palomo for the intermolecular alcoholysis of â-lactams employing
KCN in DMF, macrocyclization conditions were developed to
avoid the use of large volumes of DMF required for high
dilution.24 Use of 9.0 equiv of CH2Cl2-soluble Et4NCN cleanly
promoted the â-lactam-based macrocyclization in 59-68% yield
at ambient temperature. With the Boc-protected macrocycle 19
in hand, Lindlar reduction followed by Stille coupling (not shown)
provided Boc-pateamine A (2). Although we have been unable
to deprotect this compound without decomposition of the acid-
sensitive pateamine A molecule,25 this derivative allowed us to
obtain tentative confirmation of the relative stereochemistry26 and
to obtain some interesting and important, preliminary biological
results.27 Pateamine A was ultimately obtained employing the
trichloro-tert-butoxycarbamate (TCBoc) protecting group by
reasoning that the mild reductive conditions recently reported by
Ciufolini for the trichloroethoxy carbamate28 would be compatible
with both the triene and dienoate.29 A three-step sequence
In summary, a convergent synthesis has been developed for
(-)-pateamine A. During the course of the synthesis, we found
that N-O cleavage of a N-(benzyloxy)-â-lactam can be cleanly
effected with SmI2 and that Stille coupling of a vinyl bromide in
the presence of an allylic or triallylic acetate can be competitive
with π-allyl formation. In addition, a novel â-lactam-based
macrocyclization was employed to construct the pateamine A
dilactone. This synthesis confirms the stereochemistry as 3R, 5S,
10S, 24S and sets the stage for delineation of its mechanism of
action by providing access to further quantities of pateamine A
and designed structural derivatives.
Acknowledgment. These studies were made possible by generous
funding from the National Institutes of Health (GM 52964-01) and the
Robert A. Welch Foundation (A-1280). We thank Prof. M. H. G. Munro
and Prof. J. Blunt (University of Canterbury, NZ) for a very enjoyable
and information-rich collaboration. We thank Dr. Kaapjoo Park, Stephen
Cohn, and Ju Yang for expert assistance and the many skilled under-
graduate students who provided technical assistance for this project:
Joseph Langenhan (Allegheny College), Kristofor Voss, Erin Chavez,
W. Chad Shear, Dora Rios, Noel Crenshaw, and Chris Youngkin (Texas
A&M University). We thank Dr. Joe Reibenspies for performing the X-ray
analysis and Dr. Lloyd Sumner and Dr. Barbara Wolf of the Texas A&M
Center for Chemical Characterization for mass spectral analyses obtained
on instruments acquired by funding from the NSF (CHE-8705697) and
the TAMU Board of Regents Research Program.
(20) Guzzo, P. R.; Miller, M. J. J. Org. Chem. 1994, 59, 4862-4867.
(21) (a) Keck, G. E.; McHardy, S. F.; Wager, T. T. Tetrahedron Lett. 1995,
41, 7419-7422. (b) Chiara, J. L.; Destabel, C.; Gallego, P.; Marco-Contelles,
J. J. Org. Chem. 1996, 61, 359-360.
(22) Although we had found in model studies that a N-benzyloxy group
was sufficient for activation of a â-lactam toward intermolecular, nucleophilic
addition of an alkoxide, we reasoned that conversion to a carbamate protecting
group would enable the use of a deprotection method that would be compatible
with late-stage intermediates (unpublished results of Mr. Joe M. Langehan,
Texas A&M, NSF REU student). These findings will be described in detail
in a full account of this work.
Supporting Information Available: Physical and spectral data for
1
compounds 3-6, 11, 12, 15, and 17-20, H and 13C NMR spectra of
these intermediates, and selected procedures; spectral comparison of
natural and synthetic (-)-pateamine A; and X-ray structure (Chem 3D
representation) of transamidation byproduct (33 pages). See any current
masthead page for ordering and Internet access instructions.
JA973549F
(23) An inversion process was required at this stage as a result of an early
structural misassignment of the C10 stereocenter. (Private communication from
Prof. M. H. G. Munro).
(24) Palomo, C.; Aizpurua, H. M.; Cuevas, C.; Mielgo, A.; Galarza, R.
Tetrahedron Lett. 1995, 36, 9027-9030. This macrocyclization presumably
proceeds by way of an acyl cyanide, and studies are underway to detect this
intermediate in substrates devoid of an internal nucleophile.
(27) Prof. Jun Liu and co-workers (MIT) have found that synthetic Boc-
pateamine A (2) exhibits only 5-10 times less potency than pateamine A in
the human mixed lymphocyte reaction (IC50 of 10-20 nM). This result
suggests a viable site of attachment for linkers required for the synthesis of
hybrid molecules that will be used for putative, cellular target isolation.
(28) Dong, Q.; Anderson, C. E.; Ciufolini, M. A. Tetrahedron Lett. 1995,
36, 5681-5682. We thank Prof. Ciufolini for helpful discussions concerning
these deprotection conditions.
(25) While we have not assigned structures to the decomposition products,
it appears that the acid instability is due to the triallylic acetate portion of the
molecule since deprotection of compounds devoid of this functional array
(e.g., 19) proceeded without incident.
(29) The TCBoc group was chosen over the TROC (trichloroethoxy
carbamate) because we are, of course, interested in streamlining the synthesis
by introduction of the TCBoc group at the outset (i.e., â-lactam 4). The TCBoc
group will ensure the desired regioselective acylation during the â-lactam-
based macrocyclization.
(26) The 1H NMR of Boc-pateamine A derived from natural material was
kindly provided by Prof. Murray Munro and Prof. John Blunt (New Zealand).
No additional data were available for this derivative.
(30) Farina, V.; Krishnan, B. J. Am. Chem. Soc. 1991, 113, 9585-9595.