Total synthesis of the immunosupressant (-)-pironetin (PA48153C).

Article abstract of DOI:10.1021/ol015531m

[reaction: see text]. Total synthesis of the immunosuppresant pironetin has been achieved by a synthetic route in which the connections between starting materials and the desired structure are readily discerned. Key steps include a diastereoselective Lewi

Full text of DOI:10.1021/ol015531m

ORGANIC
LETTERS
2001
Vol. 3, No. 5
707-710
Total Synthesis of the
Immunosupressant (−)-Pironetin
(PA48153C)
Gary E. Keck,* Chad E. Knutson, and Sarah A. Wiles
Department of Chemistry, UniVersity of Utah, 315 South 1400 East, RM 2020,
Salt Lake City, Utah 84112-0850
keck@chemistry.utah.edu
Received January 8, 2001
ABSTRACT
Total synthesis of the immunosuppresant pironetin has been achieved by a synthetic route in which the connections between starting materials
and the desired structure are readily discerned. Key steps include a diastereoselective Lewis acid mediated crotylstannane aldehyde addition,
a highly selective Lewis acid promoted Mukaiyama aldol reaction, an anti-selective SmI2 reduction of a â-hydroxyketone, and finally a lactone
annulation reaction.
Natural products possessing immunosuppressant activity have
received a great deal of attention since the introduction of
cyclosporin A (CsA)1 and FK-5062 for the treatment of
autoimmune diseases, especially those associated with organ
transplantation.3 Recently Kawada and co-workers reported
the isolation of 1 (PA-48153C)4 from the fermentation broth
of Streptomyces prunicolor PA-48153 and found it to possess
potent immunosupressant activity similar to that exhibited
by CsA (2) and FK-506 (3). Perhaps more interesting than
its activity per se is its differing mode of action: PA-48153C
was found to inhibit the responses of both T and B
lymphocytes to mitogens, while CsA and FK-506 affect only
T-cell activation.5 Kobayashi and co-workers have also
reported the isolation of a plant-growth regulator, given the
name pironetin, from Streptomyces sp. NK10958 whose
structure has proven to be identical with that of 1 (Figure
1).6
A major limitation of pironetin as a therapeutic agent is
its high cytotoxicity; however, certain structurally modified
derivatives of pironetin have recently shown promise.5
Moreover, these relatively simple structures exhibit potency
comparable to that shown by CsA. Another recent report in
the patent literature suggests efficacy of certain formulations
of such derivatives in assays against uterine and ovarian
tumors, in addition to activity as inhibitors of tubulin
polymerization and thus the cell cycle.7
(1) For general reviews, see: (a) Borel, J. F. Pharmacol. ReV. 1990, 41,
259. (b) Schreiber, S. L. Science 1991, 251, 283. (c) Sigal, N. H.; Dumant,
F. J. Annu. ReV. Immunol. 1992, 10, 519. (d) Rosen, M. K.; Schreiber, S.
L. Angew. Chem., Int., Ed. Engl. 1992, 31, 384. (e) Armistead, D. M.;
Harding, M. W. Annu. Rep. Med. Chem. 1993, 28, 207. (f) Schreiber, S.
L.; Albers, M. W. Brown, E. J. Acc. Chem. Res. 1993, 26, 412.
(2) Peters, D. H.; Fitton, A.; Plosker, G. L.; Faulds, D. Drugs 1993, 46,
746.
The desire to synthesize derivatives in which activity might
be dissociated from toxicity and the need to unambiguously
(5) Yasui, K.; Tamura, Y.; Nakatani, T.; Kawada, K.; Ohtani, M. J. Org.
Chem. 1995, 60, 7567.
(3) Thomson, A. W.; Starzl, T. E. Immunol. ReV. 1993, 136, 71.
(4) (a) Yoshida, T.; Koizumi, K.; Kawamura, Y.; Matsumoto, K.; Itazaki,
H. Japanese Patent kokai 5-310726, 1993; (b) European Patent 60389 A1,
1993.
(6) (a) Kobayashi, S.; Tsochia, K.; Harada, T.; Nishide, M.; Kurokawa,
T.; Nakagawa, T.; Shimada, N.; Kobayashi, K. J. Antibiot. 1994, 47, 697.
(b) Kobayashi, S.; Tsochia, K.; Harada, T.; Nishide, M.; Kurokawa, T.;
Nakagawa, T.; Shimada, N.; Iitaka, T. J. Antibiot. 1994, 47, 703.
10.1021/ol015531m CCC: $20.00 © 2001 American Chemical Society
Published on Web 02/14/2001

annulation procedure recently developed in our laboratories
to install the lactone moiety by reaction of the lithium enolate
of methyl acetate with â-acetoxy aldehyde 4.11 This material
was in turn envisioned to arise by the Lewis acid promoted
Mukaiyama aldol union of silyl enol ether 5 and aldehyde
6. Aldehyde 6 should be accessible from â-benzyloxy
aldehyde 7 by chain extension using procedures previously
developed in our laboratories.12
The preparation of aldehyde 6 commenced with the
chelation-controlled addition of (Z)-crotyltri-n-butylstannane
to the â-benzyloxy aldehyde 7, using TiCl4 as previously
described,12 to give the desired anti,syn homoallylic alcohol
8 in 89% yield after purification by chromatography. After
conversion of the free hydroxyl to the corresponding methyl
ether 9 by reaction with KH/MeI (92% yield), a hydro-
boration-oxidation of the terminal olefin was employed to
secure aldehyde 11 (92%). In this instance, direct oxidation
of the organoborane formed from reaction of 9 with 9-BBN
proved to be low yielding. Better results were obtained by
isolation of the intermediate alcohol 10, followed by oxida-
tion using the method of Ley (TPAP/NMO).13 Installation
of the requisite E alkene was initiated by treatment of this
aldehyde with CBr4/P(Ph)3 according to the Corey-Fuchs
protocol14 to afford the homologated dibromoalkene 12 in
68% yield for two steps from alcohol 10. Reaction of this
material with n-BuLi and quenching with methyl iodide
furnished alkyne 13 (99%). Reduction of alkyne 12 with
lithium/ammonia then effected conversion of the alkyne to
the corresponding E alkene and also effected removal of the
benzyl protecting group to give alcohol 14 (84%); aldehyde
6 was prepared from this material by Ley oxidation and used
immediately in the subsequent Mukaiyama aldol reaction
(Scheme 1).
Figure 1. Pironetin, CsA, and FK506.
verify the absolute stereochemistry of 1 have prompted
several investigations which have culminated in successful
total syntheses. Kawada and co-workers reported the first
synthesis of 1, which utilized a carbohydrate approach to
the 2-pyranone ring system.8 Gurjar has disclosed two syn-
theses9 in which the stereochemistry was set via combinations
of epoxide ring openings and Evans aldol reactions. Kitahara
and Chida have also reported syntheses in which nucleophilic
additions to epoxides play prominent strategic roles.10
The route chosen for experimental investigation is outlined
in Figure 2. We envisioned application of the lactone
The preparation of the methyl ketone required for the aldol
coupling proved somewhat more difficult than originally
envisioned, since it was found that a PMB protecting group
for the â-oxygen function was necessary15 to preclude
adventitious acetate migration prior to the installation of the
lactone moiety (vide infra). Although the corresponding
benzyl ether was easily synthesized, it proved more trouble-
(7) Yasui, K.; Tamura, Y.; Nakatani, T.; Horibe, I.; Kawada, K.; Koizumi,
K.; Suzuki, R.; Ohtani, M. J. Antibiot. 1996, 49, 173.
(8) (a) Gurjar, M. K.; Henri, J. T. Jr.; Bose, D. S.; Rau, A. V. R.
Tetrahedron Lett. 1996, 37, 6615. (b) Gurjar, M. K.; Chakrabarti, A.; Rao,
A. V. R. Heterocycles 1997, 45, 7.
(9) Watanabe, H.; Watanabe, H.; Kitahara, T. Tetrahedron Lett. 1998,
39, 8313.
(10) Chida, N.; Yoshinaga, M.; Tobe, T.; Ogawa, S. Chem. Commun.
1997, 1043.
(11) Keck, G. E.; Li, X.-Y.; Knutson, C. E. Org. Lett, 1999, 1, 411.
(12) (a) Keck, G. E.; Abbott, D. E. Tetrahedron Lett. 1984, 25, 1883.
(b) Keck, G. E.; Savin, K. A.; Cressman, E. N. K.; Abbott, D. E. J. Org.
Chem. 1994, 59, 7889.
(13) Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P. Synthesis
1994, 639.
(14) Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769.
(15) It was straightforward to prepare the corresponding benzyl ether
by using the Evans oxazolidinone alkylation procedure with chlorometh-
ylbenzyl ether as the alkylating agent, and the derived â-benzyloxy methyl
ketone intermediate was used sucessfully in condensation with aldehyde 6.
However, all attempts to remove the benzyl ether at the stage of intermediate
21 were either compromised by olefin reduction or by acetate migration.
Also, it proved impractical to prepare the p-methoxybenzyl ether 18 using
the simple sequence employed to prepare the corresponding benzyl ether.
Figure 2. Antithetic analysis
708
Org. Lett., Vol. 3, No. 5, 2001

A considerable body of evidence was available to suggest
that the desired stereochemical outcome should be highly
preferred in this case.17 The diastereoselectivity associated
with nucleophilic additions to aldehydes possessing both R
and â stereocenters has been termed “merged 1,2 and 1,3
asymmetric induction”.17 In these Lewis acid promoted
reactions, the effects due to the stereocenters at the R and â
carbons can be reinforcing or opposing; only for the case in
which the directing effects of both centers operate in the
same direction is a very high level of diastereoselectivity
expected.
Scheme 1. Preparation of Aldehyde 6
That this reinforcing scenario applies in the present
circumstance is most easily seen by inspection of Figure 3.
some to obtain the PMB ether. This was ultimately ac-
complished using the asymmetric alkylation protocol devel-
oped by Meyers.16 Unsaturated amide 15 was formed by
acylation of the chiral auxiliary (1S,2S)-(+)-psuedodphedrine
with â,â-dimethylacryloyl chloride. Asymmetric alkylation
of 15 with ethyl iodide gave 16 (93%), which was reduced
using lithium amidoborane as described by Meyers to give
the primary alcohol. Protection of the hydroxyl as the PMB
ether gave 17, and oxidative cleavage of the olefin then gave
the requisite ketone 18 (52% overall from 16). Conversion
of 18 to the trimethylsilyl enol ether 5 was accomplished in
the normal way using LiHMDS and TMSCl (Scheme 2).
Figure 3. A convenient description of “merged 1,2 and 1,3
asymmetric induction”.
In this attempted depiction of the preferred transition state
for the nucleophilic addition, the three-carbon backbone of
the â-alkoxy aldehyde, and the associated substituents, is
1
drawn as if it were /2 of a chair cyclohexane, using the
normal conventions for structural drawing of chair cyclo-
hexane. Orientation of the aldehyde carbonyl with oxygen
“up” as shown leads to the preferred Felkin-Ahn18 transition
state arrangement with the largest R substituent perpendicular
to the carbonyl group. In can be seen that in this preferred
Felkin arrangement the R-Me group is placed in the
“equatorial” position. The preferred trajectory of attack, at
approximately the tetrahedral angle, as indicated by the
dashed arrow, is also seen to be “equatorial”. At the â
position, the preferred arrangement of substituents which
minimizes steric interactions and dipole-dipole repulsions
can be seen to be that which places the large carbon
substituent “equatorial” and H “axial”.19
Scheme 2. Preparation of Enol Ether 5
(17) Evans, D. A.; Dart, M. J.; Duffy, J. L.; Yang, M. G. J. Am. Chem.
Soc. 1996, 118, 4322.
(18) Ahn, T. N.; Eisentein, O. NouV. J. Chim. 1977, 1, 61.
(19) One arrives at the correct predictions using this model simply by
using sterically based A values for conformational preferences in cyclo-
hexanes to predict the preferred arrangements of substituents at the R and
â carbons. However, dipole-dipole repulsions are also easily evaluated by
inspection. For example, it is easily seen that interchanging H and Râ will
give a 1,3-diaxial interaction between Râ and the aldehyde oxygen, while
interchanging H and the O substituent will give both a diaxial steric
interaction and also maximum dipole-dipole repulsion between the two
C-O bonds. Although there might be some ambiguity about which of these
alternative arrangements would be favored, it is a certainty that both would
be considerably higher in energy than the preferred arrangement depicted.
With the necessary components for the Mukaiyama aldol
in hand, we were now in a position to couple these fragments.
(16) Myers, A. G.; Yang, B. H.; Chen, H.; McKinstry, L.; Kopecky, D.
J.; Gleason, J. L. J. Am. Chem. Soc. 1997, 119, 6496.
Org. Lett., Vol. 3, No. 5, 2001
709

essentially a single diastereomer (86% yield, diastereomeric
products not detected). It was now necessary to reduce the
â-hydroxy ketone to the anti-1,3-diol, a transformation which
has proven difficult in this system.9 However, application
of the recently described20 samarium diiode reduction
protocol afforded the desired 20 in good chemical yield
(91%) and with reasonable diastereoselectivity (4.6:1).21
Acetylation of the diol (Ac2O, NEt3; 97%), followed by
removal of the PMB protecting group, gave the desired
primary alcohol (22) in 92% yield. Key to the overall success
of this route is the finding that this deprotection can be
accomplished with no detectable 1,3-migration of the sec-
ondary acetate to the primary hydroxyl, a process which
precludes the use of many other common protecting groups.
Ley oxidation then furnished the aldehyde 4, now properly
functionalized to effect incorporation of the lactone ring in
a single step using the lactone annulation process.11 Lactone
annulation was effected by reaction of this material with the
lithium enolate of methyl acetate (initially at -78 °C for 15
min with warming to 0 °C and reaction at that temperaure
for 30 min) and afforded, after quenching and normal
extractive workup, the desired lactone 23 in 74% yield.
Finally, removal of the acetate by acid-catalyzed methanoly-
sis (MeOH, 3 N HCl, 65 °C, 8 h) gave (-)-pironetin in 86%
yield (Scheme 3).22
Scheme 3. Completion of the Route to Pironetin
Acknowledgment. Financial assistance provided by the
National Institutes of Health (through Grant GM-28961) and
by Pfizer, Inc., is gratefully acknowledged.
Supporting Information Available: Experimental pro-
cedures and spectral and analytical data. This material is
available free of charge via the Internet at http://pubs.acs.org.
Thus, the product structure arising from nucleophilic
addition is already drawn with the carbon backbone in the
normal extended zigzag arrangement and no cumbersome
manipulations are needed to ascertain the stereochemical
relationships in the product. This model is of course
consistent with both previous experimental results and the
model set forth by Evans;17 however, we find the use of the
conventions delineated above to be exceptionally convenient.
In the event, reaction of 5 with 6 using BF3‚Et2O as Lewis
acid proceeded in accord with expectation to yield 19 as
OL015531M
(20) Keck, G. E.; Wager, C. A.; Sell, T.; Wager, T. T. J. Org. Chem.
1999, 64, 2172.
(21) A similar level of stereoselectivity was obtained using tetramethyl-
ammonium triacetoxyborohydride reduction. In contrast, both reductions
were completely stereoselective for the anti diol when conducted on the
corresponding benzyl compound (19 with PMB replaced by benzyl).
(22) The spectral and analytical data for 1 were in excellent agreement
with that previously reported for the natural material and that from previous
synthetic efforts.
710
Org. Lett., Vol. 3, No. 5, 2001