Total synthesis of the novel immunosuppressant sanglifehrin A

Article abstract of DOI:10.1021/ja994285v

The total synthesis of the novel immunosuppressant sanglifehrin A (SFA, 1) is described. The approach is flexible, convergent, and stereoselective. The use of Paterson's aldol methodology was pivotal for the preparation of the novel, highly substituted sp

Full text of DOI:10.1021/ja994285v

3830
J. Am. Chem. Soc. 2000, 122, 3830-3838
Total Synthesis of the Novel Immunosuppressant Sanglifehrin A
K. C. Nicolaou,* F. Murphy, S. Barluenga, T. Ohshima, H. Wei, J. Xu, D. L. F. Gray, and
O. Baudoin
Contribution from the Department of Chemistry and The Skaggs Institute for Chemical Biology,
The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, and
Department of Chemistry and Biochemistry, UniVersity of California, San Diego, 9500 Gilman DriVe,
La Jolla, California 92093
ReceiVed December 7, 1999
Abstract: The total synthesis of the novel immunosuppressant sanglifehrin A (SFA, 1) is described. The approach
is flexible, convergent, and stereoselective. The use of Paterson’s aldol methodology was pivotal for the
preparation of the novel, highly substituted spirolactam fragment of SFA. The 22-membered macrocyclic core
of the molecule and the coupling of this fragment to the spirolactam moiety were successfully achieved using
selective intra- and intermolecular Stille reactions, respectively. Carbodiimide-based protocols were employed
for the synthesis of the tripeptide backbone.
Introduction
The discovery of immunosuppressive agents such as cy-
closporin A (CsA) and FK506 has led to a significant increase
in the success of organ and bone marrow transplantation and
to a greater understanding of the molecular basis of signal
transduction pathways.1 These structurally distinct natural
products form two different drug-protein complexes that inhibit
the phosphatase activity of the intracellular signaling molecule
calcineurin.2 Such inhibition blocks T-cell activation and
prevents host rejection of transplants. Recently, an exciting new
immunosuppressive compound, sanglifehrin A (SFA, 1), was
discovered by scientists at Novartis3 during their screening for
compounds that would interfere with signaling molecules other
than calcineurin. Produced by Streptomyces sp A92-309110
found in a soil sample in Dembo-Bridge in Malawi, SFA
possesses impressive biological properties.4 These include strong
binding to cyclophilin, immunosuppressive activity, and inhibi-
tion of both T-cell and B-cell proliferation.3,4 Studies concerning
the mode of action of SFA and analogues thereof should advance
our understanding of the immune response at the molecular level
and thereby facilitate the design of immunosuppressants in the
future.
Figure 1. Structure and retrosynthetic analysis of sanglifehrin A.
The structure of 1 has been fully elucidated by spectroscopic
and X-ray crystallographic techniques5 and is shown in Figure
1. Key features include a novel, highly substituted [5,5]
spirolactam moiety and a 22-membered macrocycle possessing
a peptidic backbone characterized by unusual â-substituted
piperazic acid and m-hydroxyphenylalanine units. The macro-
cycle also contains four contiguous chiral centers at positions
14-17 as well as endo- and exocyclic E,E cumulated diene
units. Such challenging structural novelty combined with
exciting biological activity led us to tackle the total synthesis
of 1. In planning our approach, we hoped to develop a
convergent, flexible, and stereocontrolled route that would
minimize protecting group manipulations and provide a platform
from which entry to analogues of biological interest could
ultimately be realized. A detailed account of our successful
endeavors toward the first total synthesis of 1 is presented in
this paper.6
(1) (a) Bierer, B. E. Curr. Opin. Hematol. 1993, 1, 149-159. (b)
Schreiber, S. L. Science 1991, 283, 283-287.
(2) Liu, J.; Farmer, J. D.; Lane, W. S.; Friedman, J.; Weissman, I.;
Schreiber, S. L. Cell 1991, 66, 807-815.
Results and Discussion
(3) Fehr, T.; Oberer, L.; Quesniaux Ryffel, V.; Sanglier, J.-J.; Schuler,
W.; Sedrani, R., Sandoz Ltd., PCT Int. Appl. WO 9702285A/970123, 1997.
(4) Sanglier, J.-J.; Quesniaux, V.; Fehr, T.; Hofmann, H.; Mahnke, M.;
Memmert, K.; Schuler, W.; Zenke, G.; Gschwind, L.; Maurer, C.; Schilling,
W. J. Antibiot. 1999, 52, 466-473.
(5) Fehr, T.; Kallen, J.; Oberer, L.; Sanglier, J.-J.; Schilling, W. J.
Antibiot. 1999, 52, 474-479.
Retrosynthetic Analysis. Recognizing the importance of
developing a convergent synthesis of 1, we divided the molecule
(6) Preliminary communication: Nicolaou, K. C.; Xu, J. Y.; Murphy,
F.; Barluenga, S.; Baudoin, O.; Wei, H. Gray, D. L. F.; Ohshima, T. Angew.
Chem., Int. Ed. 1999, 38, 2447-2451.
10.1021/ja994285v CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/06/2000

Synthesis of Sanglifehrin A
J. Am. Chem. Soc., Vol. 122, No. 16, 2000 3831
Figure 3. Retrosynthetic analysis of spirolactam fragment 2 of
sanglifehrin A.
Figure 2. Retrosynthetic analysis of the macrocyclic core 3 of
sanglifehrin A.
would require an antialdol coupling reaction between the achiral
aldehyde 8 and ethyl ketone 9, followed by in situ carbonyl
reduction.
into two principal fragments resulting from the cleavage of the
C25-C26 σ bond of the exocyclic diene unit (Figure 1). This
disconnection revealed the E-vinylstannane 2 and the E-
trisubstituted alkenyl iodide 3, union of which was envisaged
via a palladium-mediated Stille coupling.7 Notably, intramo-
lecular protection of the C15,C17 diol unit and the C53 methyl
ketone in 1 via ketalization was proposed given their convenient
proximity.
Synthesis of the C13-C19 Vinylstannane Ketal Carboxy-
lic Acid 5. The sequence leading to the required 5 is shown in
Scheme 1. The known R,â-unsaturated ester 1011 was initially
converted to the triisopropylsilyloxy (TIPS) ether 11 (TIPSCl,
imidazole, 95%) which was then reduced with DIBAL (87%)
to the corresponding allylic alcohol 12. mCPBA-mediated
epoxidation of allylic alcohol 12 provided epoxide 13 in
quantitative yield and in good selectivity (â:R epoxide ratio ∼6:
1). Regiospecific ring opening12 of epoxide 13 with 3-butenyl-
magnesium bromide13 in the presence of CuI led to olefinic diol
14 (79% yield), which was chemoselectively converted to the
primary pivaloate ester 15 (95%). After removal of the TIPS
protecting group with tetra-n-butylammonium fluoride (TBAF,
81%), Wacker oxidation14 of diol pivaloate ester 16 was
followed by the proposed acid-induced ketalization (vide supra)
of the intermediate methyl ketone providing ketal 17 in excellent
yield (88% for two steps). Unmasking of the C18 primary
hydroxyl function was achieved by debenzylation of ketal 17
(H2, 10% Pd/C) providing hydroxy ketal 18 which was oxidized
to aldehyde ketal 19 (83% for two steps) using TPAP/NMO.15
A one-carbon homologation of 19 to acetylenic ketal 20 was
next achieved in 98% yield using the Ohira-Bestmann reagent16
in the presence of K2CO3. Conveniently, the pivaloate group
was also cleanly removed during this operation once an excess
of K2CO3 was added. A key requirement for the success of this
reaction was that an equimolar amount of the diazophosphonate
and base be premixed prior to the addition of aldehyde ketal
19; otherwise the reaction is particularly sensitive to epimer-
ization at C17. Confirmation of the expected stereochemical
outcome of the epoxidation and the ring-opening reactions was
provided by X-ray crystallographic analysis of acetylenic ketal
20 (see Figure 4).17 A highly regio- and stereoselective
Our approach to the macrocyclic core of 1 involves a novel
Stille macrocyclization8 and a projected racemization-free amide
bond formation (Figure 2). Thus, sequential disconnection of
the C19-C20 and N12-C13 σ bonds revealed the bis(vinyl)-
iodoamine 4 and the vinylstannane ketal carboxylic acid 5 as
potential key intermediates. This approach was expected to be
challenging principally because formation of the desired 22-
membered macrocycle would require the chemoselective par-
ticipation of the C20 vinyl iodide. To our knowledge, when we
began this project there were no reports in the literature of such
an intramolecular chemoselective Stille cyclization. Key pro-
jected steps for the synthesis of 4 and 5 include stereoselective
epoxidation and regioselective epoxide opening for the introduc-
tion of the C15 and C14 stereocenters, respectively, asymmetric
allylboration for the incorporation of the C23 stereogenic center,
and carbodiimide-based coupling protocols for the generation
of the tripeptide backbone.
Figure 3 outlines the retrosynthetic analysis of the spirolactam
fragment of 1. We envisioned the introduction of the desired
C30 and C31 stereogenic centers via asymmetric crotylboration
of the spirolactam aldehyde 6. Key steps proposed for the
conversion of stereopentad 7 to 6 include a stereoselective
Ireland-Claisen rearrangement,9 a substrate-controlled hydro-
boration, and a thermodynamically controlled spirolactamization
for the incorporation of the C40, C38, and C37 stereocenters.
We planned to synthesize the C31-C39 fragment 7 using
methodology developed by Paterson and co-workers.10 This
(11) Miyashita, M.; Hoshino, M.; Yoshikoshi, A. J. Org. Chem. 1991,
56, 6483-6485.
(12) Tius, M. A.; Fauq, A. H. J. Org. Chem. 1983, 48, 4131-4132.
(13) Patel, D. J.; Hamilton, C. L.; Roberts, J. D. J. Am. Chem. Soc. 1965,
87, 5144-5157.
(7) Stille, J. K. Angew. Chem., Int. Ed. Engl. 1986, 25, 508-524.
(8) Duncton, M. A. J.; Pattenden, G. J. Chem. Soc., Perkin Trans. 1
1999, 1235-1247.
(9) (a) Ireland, R. E.; Mueller, R. H.; Willard, A. K. J. Am. Chem. Soc.
1976, 98, 2868-2877. (b) Ireland, R. E.; Wipf, P.; Armstrong, J. D. J.
Org. Chem. 1991, 56, 650-657. For a recent review see: (c) Periera, S.;
Srebnik, M. Aldrichim. Acta 1993, 26, 17-27.
(10) (a) Vulpetti, A.; Bernardi, A.; Gennari, C.; Goodman, J. M.;
Paterson, I. Tetrahedron 1993, 49, 685-696. (b) Paterson, I.; Goodman, J.
M.; Lister, M. A.; Schumann, R. C.; McClure, C. K.; Norcross, R. D.
Tetrahedron, 1990, 46, 4663-4684. (c) Paterson, I.; Perkins, M. V.
Tetrahedron Lett. 1992, 33, 801-804.
(14) Tsuji J. in ComprehensiVe Organic Synthesis; Trost, B. M., Fleming,
I., Eds.; Pergamon Press: New York, 1991; Vol. 7, p 449.
(15) (a) Griffith, W. P.; Ley, S. V. Aldrichim. Acta 1990, 23, 13-19.
(b) Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P. Synthesis 1994,
639-666.
(16) (a) Ohira, S. Synth. Commun. 1989, 19, 561-564. (b) Muller, S.;
Liepold, B.; Roth, G. J.; Bestmann, H. J. Synlett 1996, 521-522.
(17) The crystal data, atomic coordinates, etc., have been deposited with
the Cambridge Crystallographic Data Centre.

3832 J. Am. Chem. Soc., Vol. 122, No. 16, 2000
Scheme 1. Synthesis of Carboxylic Acid 5a
Nicolaou et al.
Figure 4. X-ray crystal structure of hydroxy alkyne 20.
Scheme 2. Alternative Synthesis of Hydroxy Alkyne 20a
a Reagents and conditions: (a) 2.0 equiv of TIPSCl, 3.0 equiv of
imidazole, DMF, 60 °C, 24 h, 95%; (b) 3.0 equiv of DIBAL, CH2Cl2,
-78 °C, 2 h, 87%; (c) 1.5 equiv of mCPBA, CH2Cl2, -25 °C, 100%,
â:R epoxide ratio ∼6:1; (d) 5.0 equiv of H2CdCHCH2CH2MgBr, 1.0
equiv of CuI, Et2O/THF (1:1), -40 f -20 °C, 18 h, 79%; (e) 25
equiv of PivCl, 50 equiv of pyridine, 25 °C, 24 h, 95%; (f) 2.0 equiv
of TBAF, THF, 25 °C, 1 h, 81%; (g) 0.1 equiv of PdCl2, 1.5 equiv of
benzoquinone, DMF/H2O (7:1), 25 °C, 3 h; (h) 0.05 equiv of
TsOH‚H2O, benzene, reflux, 88% for two steps; (i) H2, 0.1 equiv of
10% Pd/C, EtOH, 25 °C, 1 h, 100%; (j) 0.05 equiv of TPAP, 3.0 equiv
of NMO, 4 Å MS, CH2Cl2, 25 °C, 20 min; (k) 5.0 equiv of
MeC(O)C(dN2)PO(OMe)2, 5.0 equiv of K2CO3, MeOH, 0 f 25 °C,
13 h, then 5.0 equiv of K2CO3, 25 °C, 24 h, 73% for two steps; (l) 4.0
equiv of nBu3SnH, 0.3 equiv of PdCl2(PhCN)2, 0.6 equiv of P(o-tol)3,
4.0 equiv of iPr2NEt, CH2Cl2, -20 °C, 1 h, 80%; (m) 0.05 equiv of
TPAP, 3.0 equiv of NMO, 4 Å MS, CH2Cl2, 25 °C, 15 min; (n) 6.0
equiv of NaClO2, 2.0 equiv of NaH2PO4, 10 equiv of 2-methyl-2-butene
(2 M in THF), tBuOH:H2O (5:1), 25 °C, 15 min (good yield, see
Scheme 6). mCPBA m-chloroperbenzoic acid, TBAF ) tetra-n-butyl-
ammonium fluoride, TsOH ) p-toluensulfonic acid, TPAP ) tetra-n-
propylammonium perruthenate, and NMO ) N-methylmorpholine
N-oxide.
a Reagents and conditions: (a) 2.5 equiv of (E)-crotyldiisopinocam-
pheylborane, THF, -78 °C; then 15 equiv of NaBO3‚4H2O, THF/H2O
(1:1), 25 °C, 12 h, 67%; (b) 1.5 equiv of TBSOTf, 2.0 equiv of 2,6
lutidine, CH2Cl2, 0 f 25 °C, 12 h, 97%; (c) O3, 0.2 equiv of Sudan
7B, CH2Cl2, -78 °C; then 1.5 equiv of PPh3, -78 f 25 °C, 12 h; (d)
3.0 equiv of (EtO)2P(dO)CH2CO2Et, 3.0 equiv of NaH, THF, -78 f
25 °C, 2 h, 82% for two steps; (e) 2.5 equiv of DIBAL, -78 °C, 0.5
h, 92%; f. 2.0 equiv of mCPBA, -30 °C, 17 h, 92%, â:R epoxide ratio
∼79:21; (g) 6.0 equiv of 29, -40 °C, 3 h; (h) 1.3 equiv of PivCl,
pyridine, 0 f 25 °C, 78% for two steps; (i) HF:CH3CN:H2O, 1:10:1,
25 °C, 2 h, 76%; (j) 5.0 equiv of K2CO3, MeOH, 25 °C, 48 h, 92%.
mCPBA ) m-chloroperbenzoic acid.
Silylation of the resulting acetylenic alcohol 23 with TBSOTf
provided acetylenic silyl ether 24 (97%), which was transformed
in a two-step sequence (ozonolysis, followed by phosphonate
anion olefination) to the acetylenic R,â-unsaturated ester 26 via
acetylenic aldehyde 25 (82% for two steps). Reduction of 26
with DIBAL then furnished acetylenic allylic alcohol 27 in 92%
yield. The remaining two stereocenters at C14 and C15 were
introduced by a similar sequence of epoxide formation (27 f
28, 92% yield, â:R epoxide ratio ∼79:21) and ring-opening
reactions as described for allylic alcohol 12 (Scheme 1). This
alternative route, however, results in a more expedient arrival
at acetylenic ketal 20 by directly incorporating oxygenation at
C53 through the use of a nucleophilic 3-ketobutyl equivalent
palladium(0)-mediated hydrostannylation18 of 20 provided vi-
nylstannane ketal 21 (79%) which was oxidized in two steps to
the desired 5 in good overall yield.
An alternative approach to 20 that offers the advantage of
both a shorter synthetic sequence and higher overall yield is
presented in Scheme 2. This sequence is primed by the
asymmetric crotylboration (67%) of propargylic aldehyde 2219
with Brown’s trans-crotyl (+) diisopinocampheylborane.20
(18) Zhang, H. X.; Guibe´, F.; Balavoine, G. J. Org. Chem. 1990, 55,
1857-1867.
(19) Danheiser, R. L.; Carini, D. J.; Fink, D. M.; Basak, A. Tetrahedron
1983, 39, 935-937.
(20) Brown, H. C.; Bhat, K. S. J. Am. Chem. Soc. 1986, 108, 293-294.

Synthesis of Sanglifehrin A
J. Am. Chem. Soc., Vol. 122, No. 16, 2000 3833
Scheme 3. Synthesis of Fragment 38a
Scheme 4. Synthesis of Model Fragment 44 (C2-epi as
Compared to Sanglifehrin A)a
a Reagents and conditions: (a) 2.0 equiv of DBU, CH2Cl2, 25 °C, 2
h, 90%; (b) 0.7 mol % of [(S,S)-Et-DuP-Rh)+TfO-, 60 psi, 96 h, 98%
ee, 90%; (c) H2, 10% Pd/C, MeOH, 25 °C, 12 h, 96%; (d) 3.0 equiv
of EDC, 3.0 equiv of HOAt, CH2Cl2, 0 f 25 °C, 3 h, 78%; (e) 2.0
equiv of LiOH, THF/H2O (3:1), 0 f 25 °C, 1.5 h, 89%. DBU ) 1,8-
diazobicyclo[5.4.0]undec-7-ene, [(S,S)-Et-DuP-Rh]+TfO- ) (+)-1,2-
bis[(2S,5S)-2,5-diethylphospholano]benzene (1,5-cyclooctadiene)rhod-
ium(I) trifluoromethanesulfonate, Boc-Val-OH ) N-(tert-butoxycarbonyl)-
L-valine), HOAt ) 1-hydroxy-7-azabenzotriazole, and EDC ) 1-(3-
(dimethylamino)propyl)-3-ethylcarbodimide hydrochloride.
a Reagents and conditions: (a) 2.0 equiv of EDC, 0.1 equiv of PPy,
1.0 equiv of iPr2NEt, CH2Cl2, 0 f 25 °C, 80%; (b) TFA:CH2Cl2 (1:1),
0 f 25 °C, 2 h; (c) 1.0 equiv of HOAt, 3.0 equiv of iPr2NEt, 1.2
equiv of EDC, CH2Cl2, 0 f 25 °C, 3.5 h, 66% for two steps; (d) TFA:
CH2Cl2 (1:10), 0 f 25 °C, 2 h (good yield). PPy ) 4-pyrrolidinopy-
ridine, TFA ) trifluoroacetic acid.
At this stage, we decided to initiate a model study to test the
efficacy of the proposed Stille coupling for the preparation of
the 22-membered macrocycle.25 To this end, we chose to replace
the trisubstituted vinyl iodide at C23 in tripeptide derivative 3
(Figure 2) with a hydrogen atom. As shown in Scheme 4, the
synthesis of model tripeptide ester derivative 44 (note the C2-
epi stereochemistry of this model system as compared to
sanglifehrin A) was achieved by a short four-step sequence.
Thus, coupling of alcohol 3926 with the Boc-protected piperazic
acid derivative 4027 (C2-epi as compared to sanglifehrin A) in
the presence of EDC provided 2(R)-di-Boc-piperazic acid ester
41 in good yield (80%). Treatment of 41 with trifluoroacetic
acid (TFA) removed the Boc protecting groups from the nitrogen
atoms providing piperazic acid ester 42 which was regioselec-
tively acylated with dipeptide carboxylic acid derivative 38 at
the less hindered â-position in the presence of EDC/HOAt (66%
for two steps) providing tripeptide ester N-Boc derivative 43.
Removal of the Boc protecting group from 43 with TFA then
furnished tripeptide ester derivative 44 in good yield.
Given the success of this approach, we next set about
preparing the fully functionalized tripeptide fragment of 1 in
an identical manner (Scheme 5). The requisite hydroxybis(vinyl
iodide) 47 was readily prepared from the known iodo aldehyde
4528 in two steps. Thus, the chromium(II)-mediated Takai
reaction29 was used for the stereoselective introduction of the
C19-C20 (E)-vinyl iodide (57% yield), and the C17 hydroxyl
group of the resulting product was unmasked by desilylation
with TBAF (88% yield) to afford 47 via bis(vinyl iodide) silyl
for the ring-opening reaction and by simplifying the deprotecting
strategy. Thus, regioselective ring opening of acetylenic epoxide
28 with ketal-magnesium bromide 2921 followed by chemose-
lective pivaloate formation provided acetylenic hydroxy ketal
pivaloate 31 via acetylenic hydroxy ketal 30 (66% for two steps).
Removal of the tert-butylsilyloxy group (HF/CH3CN/H2O, 1:10:
1, 76% yield) was accompanied by transketalization furnishing
acetylenic internal ketal pivaloate 32. Concomitant cleavage of
the pivaloate and triethylsilyl groups in acetylenic internal ketal
pivaloate 32 in the presence of K2CO3 then completed the
sequence leading to acetylenic internal ketal alcohol 20 in 92%
yield.
Synthesis of the Tripeptide Fragment of SFA. The syn-
thesis of the macrocyclic core of 1, requiring an efficient
preparation of the tripeptide fragment 4 (Figure 2), began with
the stereoselective assembly of the intermediate dipeptide
carboxylic acid derivative 38 as outlined in Scheme 3. The key
step involved the enantioselective hydrogenation of the R,â-
didehydro amino acid derivative 34. Thus, the condensation of
N-benzyloxycarbonylglycine phosphonate 33 with m-hydroxy-
benzaldehyde in the presence of DBU furnished the diastereo-
merically pure (E)-34 in 90% yield.22 Asymmetric hydrogena-
tion23 of 34 using catalytic [(S,S)-Et-DuP-Rh]+TfO- provided
the amino acid derivative 35 in excellent ee (98%) and yield
(90%). Hydrogenolysis of the Cbz group from 35 (96%)
followed by carbodiimide-mediated coupling of the resulting
amino acid methyl ester 36 in the presence of HOAt24 with Boc-
protected valine then led to dipeptide derivative 37 (78%), which
underwent smooth hydrolysis upon treatment with LiOH
providing the dipeptide carboxylic acid derivative 38 (89%).
With the requisite C7-N12 fragment in hand, the tripeptide
fragment was ready to be assembled.
(25) Nicolaou, K. C.; Ohshima, T.; Murphy, F.; Barluenga, S.; Xu, J.
Y.; Winssinger, N. Chem. Commun. 1999, 809-810.
(26) Nicolaou, K. C.; Stylianides, N. A.; Ramphal, J. Y. J. Chem. Soc.,
Perkin Trans 1 1989, 11, 2131-2132.
(27) Hale, K. J.; Cai, J.; Delisser, V.; Manaviazar, S.; Peak, S. A.; Bhatia,
G. S.; Collins, T. C.; Jogiya, N. Tetrahedron 1996, 52, 1047-1068.
(28) Nicolaou, K. C.; King, N. P.; Finlay, M. R. V.; He, Y.; Roschangar,
F.; Vourloumis, D.; Vallberg, H.; Sarabia, F.; Ninkovic, S.; Hepworth, D.
Bioorg. Med. Chem. 1999, 7, 665-697.
(29) (a) Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986, 108,
7408-7401. (b) Evans, D. A.; Black, W. C. J. Am. Chem. Soc. 1993, 115,
4497-4513.
(21) Hsung, R. P. Synth. Commun. 1990, 20, 1175-1179.
(22) Schmidt, U.; Griesser, H.; Leitenberger, V.; Lieberknecht, A.;
Mangold, R.; Meyer, R.; Reidl, B. Synthesis 1992, 487-490.
(23) Burk, M. J.; Feaster, J. E.; Nugent, W. A.; Harlow, R. L. J. Am.
Chem. Soc. 1993, 115, 10125-10318.
(24) Carpino, L. A. J. Am. Chem. Soc. 1993, 115, 4397-4398.

3834 J. Am. Chem. Soc., Vol. 122, No. 16, 2000
Scheme 5. Synthesis of Fragment 4a
Nicolaou et al.
Scheme 6. Synthesis of Sanglifehrin A Model System 53
(C2-epi)a
a Reagents and conditions: (a) 6.0 equiv of CrCl2, 2.0 equiv of CHI3,
0 f 25 °C, dioxane/THF (9:1), 12 h, 57%; (b) 1.2 equiv of TBAF,
THF, 0 f 25 °C, 15 min, 88%; (c) 2.0 equiv of EDC, 0.1 equiv of
PPy, 1.0 equiv of iPr2NEt, 2.0 equiv of 48, CH2Cl2, 0 f 25 °C, 64%;
(d) TFA/CH2Cl2 (1:1), 0 f 25 °C, 2 h; (e) 1.0 equiv of HOAt, 3.0
equiv of iPr2NEt, 1.2 equiv of EDC, CH2Cl2, 0 f 25 °C, 3.5 h, 66%
for two steps; (f) TFA:CH2Cl2 (1:10), 0 f 25 °C, 2 h.
a Reagents and conditions: (a) 1.0 equiv of HATU, 4.0 equiv of
iPr2NEt, DMF, 0 f 25 °C, 10 h, 51% for three steps from 21 (Scheme
1); (b) 0.15 equiv of Pd2(dba)3‚CHCl3, 0.6 equiv of AsPh3, 10 equiv of
iPr2NEt, DMF, 25 °C, 3 h, 40%. HATU, O-(7-azabenzotriazole-1-yl)-
N,N,N′,N′-tetramethyluronium hexafluorophosphate; Pd2(dba)3‚CHCl3,
tris(dibenzylideneacetone)dipalladium (0)-chloroform adduct.
ether 46. Coupling of 47 with Boc-protected amino acid 4827
in the presence of EDC/PPy provided ester 2(S)-N-Boc-piperazic
ester 49 (64% yield) from which both Boc groups were removed
upon treatment with TFA to afford 2(S)-piperazic ester 50.
Regioselective acylation of 50 with dipeptide carboxylic acid
derivative 38 (Scheme 3) (EDC/HOAt) then furnished fragment
N-Boc-tripeptide ester 51 (66% yield for two steps), which was
deprotected with TFA providing tripeptide ester 4 in good yield
(Scheme 5). Interestingly, attempts to directly couple the
preformed tripeptide unit with 47 failed to provide 51 in good
yield (<30%).
Scheme 7. Synthesis of Sanglifehrin Macrocycle 3a
Synthesis of the Macrocyclic Core of SFA. The synthesis
of the 2(R)-sanglifehrin A model system 53 (note that this model
has the C2-epi stereochemistry of sanglifehrin A) is outlined in
Scheme 6. Thus union of tripeptide ester derivative 44 with
vinylstannane ketal carboxylic acid 5 in the presence of EDC/
HOAt resulted in a disappointingly low yield of tripeptide ester
52. Fortunately, the coupling was more efficiently effected using
HATU,24 which furnished 52 in 51% overall yield from
vinylstannane ketal 21 (see Scheme 1) with no detectable
epimerization. With the cyclization precursor in hand, the crucial
macrocyclization was tried next. In light of the reported
superiority of AsPh3 over PPh3 ligands in the palladium(0)-
mediated Stille coupling reaction.30 we decided to attempt the
cyclization in the presence of Pd2(dba)3‚CHCl3/AsPh3 and iPr2-
NEt. To our delight, the ring closure of 52 proceeded under
these conditions, albeit in 40% yield, at room temperature and
in the presence of the free phenol and amine functionalities
(Scheme 6).
a Reagents and conditions: (a) 1.0 equiv of HATU, 4.0 equiv of
iPr2NEt, DMF, 0 f 25 °C, 10 h, 50% for three steps from 21 (Scheme
1); (b) 0.15 equiv of Pd2(dba)3‚CHCl3, 0.6 equiv of AsPh3, 10 equiv of
iPr2NEt, DMF, 25 °C, 36 h, 62%.
cycle of 1 possessing the C18-C21 diene unit. Encouraged by
the successful construction of the C2-epi sanglifehrin A model
system via the intramolecular Stille coupling reaction, we
proceeded to build sanglifehrin A itself following the chartered
strategy.
The synthesis of the fully functionalized macrocyclic core
of 1 is presented in Scheme 7. Thus, in a manner identical to
that described for the model system (Scheme 6), the tripeptide
4 was coupled with vinylstannane ketal carboxylic acid 5
providing the final precursor before the key macrocyclization,
tripeptide ester 54 (HATU, iPr2NEt, 50% yield from vinylstan-
nane ketal 21, Scheme 1). Pleasantly, treatment of 54 with Pd2-
(dba)3‚CHCl3/AsPh3 in DMF (1.0 mM) for 36 h led to the
exclusive formation of the desired sanglifehrin A cyclic
intermediate 3 in an isolated yield of 62%. More prolonged
reaction times led to significantly reduced yields of sanglifehrin
In a parallel study, attempts were made to access 53 via a
ring-closing metathesis reaction; in contrast to the Stille reaction,
however, none of the desired macrocycle 53 was detected in
the complex mixture of products formed upon exposure to
(cy3P)RuCl2CHPh catalyst.31 Interestingly, Wagner and co-
workers32 recently reported the successful use of a ring-closing
metathesis reaction to access a more simplified model macro-
(30) Farina, V.; Krishnan, B. J. Am. Chem. Soc. 1991, 115, 9585-9696.
(31) Schwab, P.; France, M. B.; Ziller, J. W.; Grubbs, R. H. Angew.
Chem., Int. Ed. Engl. 1995, 34, 2039-2041.
(32) Martin, L.; Cabrejas, M.; Rohrbach, S.; Wagner, D.; Kallen, J.;
Zenke, G.; Wagner, J. Angew. Chem., Int. Ed. 1999, 38, 2443-2446.

Synthesis of Sanglifehrin A
J. Am. Chem. Soc., Vol. 122, No. 16, 2000 3835
Scheme 8. Synthesis of Fragment 65a
was elected to introduce the C40 stereocenter. Thus, formation
of the (Z)-boron enolate from diethyl ketone 55 and (+)-
diisopinocamphenyl boron triflate in the presence of iPr2NEt33
followed by the addition of methacrolein10a provided, after
silylation with TESCl, the syn aldol derivative TES-protected
aldol 56 in 74% overall yield. The absolute stereochemistry of
56 was tentatively assigned as 3R, 4R based on analogy with
examples reported by Paterson10a,34 for which reaction stereo-
control was rationalized in terms of a chairlike transition state
in which steric interactions between the Ipc ligands and the
substituents in the transition-state core are minimized. Coupling
of the (E)-dicyclohexylboron enolate derived from ethyl ketone
TES-protected aldol 56 and cy2BCl-iPr2NEt with 3-benzyloxy-
propanal followed by in situ reduction with lithium boro-
hydride10c provided stereopentad diol 57 in 72% yield (Scheme
8). The rationale for the sense of stereochemical control of such
one-pot aldol-reduction reactions has been discussed by
Paterson and co-workers.35 Proceeding with the synthesis, the
requisite Ireland-Claisen precursor acetonide ester 59 was
readily prepared in two steps from diol 57. Thus, treatment of
diol 57 with 2,2-dimethoxypropane in the presence of cam-
phorsulfonic acid (CSA) led to the concomitant formation of
the C33-C35 syn acetonide and the unmasking of the C37
hydroxyl group furnishing hydroxyacetonide 58 in 95% yield.
1
Analysis of the H and 13C NMR spectra36 of the resulting 58
was in complete agreement with the expected stereochemical
outcome of the anti aldol bond construction and the in situ
reduction (vide supra).
Acylation of 58 with butyric anhydride in the presence of
Et3N provided acetonide ester 59 (98%), which underwent
enolization-silylation (LDA, TBSCl) to generate the required
acetonide keteneacetal 60. Thermolysis of 60 in toluene followed
by hydrolysis of the resulting silyl ester then provided acetonide
carboxylic acid 61 in 85% yield and as a single diastereoisomer.
The S stereochemistry of the ethyl-substituted chiral center and
the E geometry about the trisubstituted double bond were
tentatively assigned on the assumption that the Ireland-Claisen
rearrangement proceeded via the well-precedented chairlike
transition state9 (see structure 60, Scheme 8). This assumption
was subsequently confirmed unambiguously by X-ray analysis
of a more advanced intermediate (vide infra).
At this stage, the final relevant stereogenic center at C38 was
introduced by a substrate-controlled regio- and stereoselective
hydroboration of trisubstituted alkene acetonide carboxylic acid
61. Thus, treatment of 61 with borane at -25 °C followed by
oxidative workup provided a ∼5:1 mixture of diastereomeric
diols in favor of the desired acetonide diol 62. The stereochem-
istry of the newly formed stereocenters at C37 and C38 were
predicted by invoking Houk’s transition-state model37 in which
the largest group on the chiral center flanking the double bond
in 61 is anti to the attacking borane and allylic 1,3-strain is
minimized as illustrated in 63 (see Scheme 8). Attempts to
improve the selectivity were unsuccessful since the alkene was
unreactive in the presence of more bulky hydroborating reagents
(thexylborane and 9-BBN). Moreover, the employment of low
temperature (-25 °C) was critical to the success of the
a Reagents and conditions: (a) 1.3 equiv of (+)-Ipc2BOTf, 4.0 equiv
of iPr2NEt, THF, -78 °C, 2 h; then 5.0 equiv of methacrolein, -78 f
-10 °C, 10 h; H2O2 (30% aq)/CH3OH/H2O pH 7 buffer (1.3:5:1), 0
°C, 3 h; (b) 1.2 equiv of TESCl, 1.8 equiv of imidazole, CH2Cl2, 0 °C,
2 h, 74% for two steps; (c) 1.5 equiv of cy2BCl, 1.5 equiv of Et3N,
Et2O, 0 °C, 1.5 h; then 2.0 equiv of BnO(CH2)2CHO, -78 f -10 °C,
4 h; then 10 equiv of LiBH4, -78 f 25 °C, 12 h; 15 equiv of
NaBO3‚4H2O, THF/H2O (3:2), 15 °C, 12 h, 72%; (d) 0.1 equiv of CSA,
30 equiv of Me2C(OMe)2, acetone, 72 h, 95%; (e) 2.0 equiv of
(nPrCO)2O, 6.0 equiv of Et3N, CH2Cl2, 25 °C, 24 h, 98%; (f) 1.4 equiv
of LDA, 6.0 equiv of TBSCl, THF, -78 °C; then HMPA/THF (1:5),
-78 f 0 °C, 1 h; toluene, 70 °C, 2 h, 84%; (g) 5.0 equiv of BH3‚THF,
THF, -20 °C, 17 h; 15 equiv of NaBO3‚4H2O, THF/H2O (3:2), 15
°C, 12 h; 62/diastereoisomer ∼5:1 ratio; (h) 0.01 equiv of TPAP, 6.0
equiv of NMO, CH2Cl2, 25 °C, 8 h, 51% for two steps; (i) 20 equiv of
Me2Al-NH2, CH2Cl2, 25 °C, 2 h, 90%; TES ) triethylsilyl, cy2BCl )
dicyclohexylboron chloride, CSA ) camphorsulfonic acid, LDA )
lithium diisopropylamide, HMPA ) hexamethylphosphoramide, TPAP
) tetra-n-propylammonium perruthenate, NMO ) N-methylmorpho-
line-N-oxide, and DMP ) Dess-Martin periodinane.
A cyclic intermediate 3. Having successfully addressed the
synthesis of the macrocycle, the next task at hand was to develop
an approach to the spirolactam fragment of 1.
Synthesis of the Spirolactam Fragment of SFA. The
enantioselective synthesis of the spirolactam residue of 1 began
with the assembly of the key C31-N42 fragment acetonide
amide 65 (Scheme 8), which possesses six of the seven requisite
stereocenters. The short, nine-step sequence leading to 65 is
shown in Scheme 8. Asymmetric aldol methodology10 and
substrate-controlled hydroboration were used to install the array
of alternating oxygenated and methyl functionalities on the
C33-C38 backbone, while the Ireland-Claisen rearrangement9
(33) Brown, H. C.; Dhar, R. K.; Bakshi, R. K.; Pandiarajan, P. K.;
Singaran, B. J. Am. Chem. Soc. 1989, 111, 3441-3442.
(34) Paterson, I.; Hulme, A. N. J. Org. Chem. 1995, 60, 3288-3300.
(35) Bernardi, A.; Capelli, A. M.; Comotti, A.; Gennari, C.; Gardner,
M.; Goodman, J. M.; Paterson, I. Tetrahedron 1991, 47, 3471-3484.
(36) (a) Rychnovsky, S. D.; Skalitzky, D. J. Tetrahedron Lett. 1990, 31,
945-948. (b) Evans, D. A.; Rieger, D. L.; Gage, J. R. Tetrahedron Lett.
1990, 31, 7099-7100.
(37) Houk, K. N.; Rondan, N. G.; Wu, Y.-D.; Metz, J. T.; Paddon-Row,
M. N. Tetrahedron 1984, 40, 2257-2274.

3836 J. Am. Chem. Soc., Vol. 122, No. 16, 2000
Scheme 9. Synthesis of Spirolactam 6a
Nicolaou et al.
Figure 5. Computer modeling of natural and unnatural spirolactam
diols.
Scheme 10. Synthesis of Sanglifehrin A Spirolactam 2a
a Reagents and conditions: (a) PDC, CH2Cl2, 25 °C, 12 h, 76%; (b)
CH3CN/HF/H2O (20:1:1), 25 °C, 36 h, 95%; (c) H2, 20% Pd(OH)2/C
(cat.), EtOH, 25 °C, 12 h, 99%; (d) 0.08 equiv of RuCl2(PPh3)3, air,
C6H6, 25 °C, 3 h, 83%; (e) H2, 20% Pd(OH)2/C (cat.), EtOH, 25 °C,
12 h, 100% (f) 2.7 equiv of DMP, 4.0 equiv of pyridine, CH2Cl2, 25
°C, 30 min, 55%; (g) CH3CN/HF/H2O (20:1:1), 25 °C, 36 h, 95%.
DMP ) Dess-Martin periodinane.
hydroboration reaction since higher temperatures favored forma-
tion of the C33 isopropyl ether via intramolecular reduction of
the acetonide function.38 Oxidation of acetonide diol 62 in the
presence of its C37-C38 diastereoisomer with TPAP/NMO15
led to the direct formation of a mixture of diastereoisomeric
lactones from which the desired compound acetonide lactone
64 was isolated in 51% yield from 61 after chromatography.
The final step in the sequence leading to the C31-N42 fragment
required the conversion of 64 to acetonide amide 65. To this
end, treatment of 64 with dimethylaluminum amide39 furnished
65 in 90% yield. The next task at hand was the elaboration of
65 to the spirolactam core of 1.
a Reagents and conditions: (a) 5.0 equiv of (Z)-crotyldiisopinocam-
pheylborane, THF, -78 °C, 2 h, -78 f 25 °C, 1 h; NaBO3‚4H2O,
THF/H2O (3:2), 25 °C, 12 h, 67%, 71:â-isomer ∼7:3 ratio; (b) 2.4
equiv of TBSOTf, 3.6 equiv of 2,6-lutidine, CH2Cl2, -10 f 25 °C, 4
h, 92%; (c) O3, 100 equiv of Me2S, CH2Cl2, -78 f 25 °C, 45 h, 61%;
(d) 3.0 equiv of LDA, 3.0 equiv of TMSCH2CHdN-tBu, THF, -78
f 0 °C, 2.5 h, 68%; (e) H2, Lindlar catalyst, MeOH, 25 °C, 14 h,
92%; (f) 2.0 equiv of MeC(dO)C(dN2)P(dO)(OMe)2, 2.5 equiv of
K2CO3, CH3OH, 0 f 25 °C, 10 h, 98%; (g) 8.0 equiv of TBAF, THF,
45 °C, 48 h, 87%; (h) 1.2 equiv of NBS, 0.3 equiv of AgNO3, acetone,
25 °C, 30 min, 69%; (i) 0.1 equiv of Pd2(dba)3‚CHCl3, 0.8 equiv of
Ph3P, 2.2 equiv of nBu3SnH, 25 °C, 30 min, 70%.
Prior to presenting our synthesis of the spirolactam hydroxy-
aldehyde 6 (Scheme 9) it is relevant at this stage to comment
on computer modeling that was carried out on the natural and
unnatural spirolactams diol 66 and C37-epi-66, respectively,
shown in Figure 5. Molecular dynamics followed by minimiza-
tion with the CFF91 force field40 showed that the natural
spirolactam is more stable than the unnatural isomer by ∼5 kcal/
mol. Thus, under thermodymanic control, stabilizing anomeric
effects and destabilizing steric factors, considered together, are
expected to direct spirocyclization in favor of the natural
spirolactam. Moreover, the natural compound is probably further
stabilized by an intramolecular H-bond between HN-42 and HO-
C35. The confirmation of this assumption was provided by
X-ray crystallographic analysis of a key cyclized intermediate
(vide infra).
Two alternative routes to 6 are presented in Scheme 9. In
the first approach, 65 was transformed into acetonide ketoamide
67 on treatment with PDC. The crucial spirocyclization of 67
was then carried out in the presence of hydrofluoric acid in
aqueous acetonitrile. Under these conditions, acetonide hydroly-
sis and cyclization ensued and the spirolactam hydroxy benzyl
ether 68 was isolated as a single diastereoisomer in 95% yield.
Hydrogenolysis of benzyl ether 68 (H2, Pd(OH)2 catalyst 20
wt % on carbon) led to spirolactam diol 66, X-ray crystal-
lographic analysis of which was in complete accord with the
predicted stereochemistry at each of the seven stereogenic
(38) Coutts, L. D.; Cywin, C. L.; Kallmerten, J. Synlett 1993, 696-670.
(39) Basha, A.; Lipton, M.; Weinreb, S. M. Tetrahedron Lett. 1977,
4171-4174.
(40) Hwang, M. J.; Stockfisch, T. P.; Hagler, A. T. J. Am. Chem. Soc.
1994, 116, 2515-2525.

Synthesis of Sanglifehrin A
J. Am. Chem. Soc., Vol. 122, No. 16, 2000 3837
Figure 7. X-ray crystal structure of alkyne 77.
Scheme 11. Total Synthesis of Sanglifehrin A (1)a
Figure 6. X-ray crystal structure of spirolactam diol 66.
centers (see Figures 5 and 6). Chemoselective oxidation of the
primary hydroxyl function in 66 with oxygen in the presence
of tris(triphenylphosphine)ruthenium(II) dichloride catalyst41
then provided spirolactam hydroxyaldehyde 6 in 83% yield. In
a shorter sequence leading to 6, the primary hydroxyl group
was unmasked prior to the spirolactamization step. Thus,
debenzylation of 65 (H2, Pd(OH)2 catalyst 20 wt % on carbon)
was followed by bis-oxidation of the resulting acetonide
hydroxyamide 69 with Dess-Martin periodinane42 furnishing
keto aldehyde 70 in 55% yield. Complete dehydration of the
amide function in 69 resulted unless the oxidation was carried
out in the presence of excess pyridine. The use of PDC to effect
this oxidation led to significantly reduced yields of 70 (∼30%)
due to difficulties encountered during isolation. Acid-induced
spirocyclization of 70 then provided 6 stereoselectively and in
95% yield.
The completion of the synthesis of the fully functionalized
left-hand fragment of 1 is shown in Scheme 10. Crotylboration
of spirolactam hydroxyaldehyde 6 with Brown’s cis-crotylborane
[(+)-Ipc2B(cis-crotyl)] provided a 7:3 mixture of diastereomeric
homoallylic alcohols (67% combined yield) that was readily
separated by chromatography. Literature precedents using
B-crotyl(Ipc)2 suggested that the major diastereoisomer was that
corresponding to the indicated stereochemistry at C33-C34 for
olefinic spirolactam diol 71. Unambiguous confirmation of this
assumption was established at a later stage in the sequence (vide
infra). Treatment of diol 71 with TBSOTf in the presence of
2,6-lutidine (92%) followed by ozonolysis and Me2S workup
provided spirolactam aldehyde 73 (61%) via olefinic spirolactam
diol bis(silyloxy ether) 72. A two-carbon homologation of
spirolactam aldehyde 73 to spirolactam aldehyde 76 was
achieved via the corresponding spirolactam R,â-unsaturated
aldehyde 74. Thus, reaction of 73 with the lithioderivative of
silyl aldimine 7543 followed by hydrogenation of the double
bond using Lindlar’s catalyst44 directly provided 76 in 64% yield
for two steps. Attempts to convert 71 directly to 74 via cross-
metathesis with acrolein45 failed. Next, 76 was transformed into
acetylenic spirolactam 77 in excellent yield with MeC(dO)C-
(dN2)P(dO)(OMe)2. X-ray crystallographic analysis of 77
confirmed its structure (see Figure 7), including the anticipated
stereochemical outcome of the crotylboration (6 f 71). At this
point the tert-butyldimethylsilyloxy protecting groups were
removed (77 f 78, TBAF, 87% yield) since the free hydroxyl
a Reagents and conditions: (a) 0.1 equiv of Pd2(dba)3‚CHCl3, 0.2
equiv of AsPh3, 10 equiv of iPr2NEt, DMF, 40 °C, 5 h, 45%; (b) 2.0
equiv of 2 N H2SO4, THF/H2O (4:1), 25 °C, 7 h, 50% conversion (by
HPLC), 33%.
groups were not expected to be detrimental to subsequent
transformations (Scheme 10). In light of the excellent re-
giospecificity and stereoselectivity observed during the pal-
ladium-catalyzed hydrostannylation of bromoalkynes,46 we
chose to access the targeted stannane 2 from acetylenic
spirolactam diol 78 via the intermediacy of bromoacetylenic
spirolactam 79. Thus, bromination of 78 with N-bromosuccin-
imide in the presence of a catalytic quantity of silver nitrate47
provided 79 (69%). This was converted to vinylstannane
spirolactam diol 2 on treatment with tri-n-butyltin hydride and
in situ-generated catalytic Pd(PPh3)4. With both key coupling
partners, 2 and sanglifehrin A cyclic intermediate 3 in hand,
the stage was set for the final coupling.
Final Stages of the Total Synthesis of SFA. The completion
of the total synthesis of 1 is presented in Scheme 11. Treatment
of a mixture of vinylstannane spirolactam 2 and sanglifehrin A
cyclic intermediate 3 with a catalytic amount of in situ-generated
palladium(0) tetrakistriphenylarsine30 in DMF at 40 °C led to
sanglifehrin A internal ketal 80 in 45% isolated yield. Finally,
1 was generated by unravelling ketal 80 through exposure to
aqueous sulfuric acid as described by the Novartis group.3
Synthetic 1 had physical properties identical to those of an
(41) Tomioka, H.; Takai, K.; Oshima, K.; Nozaki, H. Tetrahedron Lett.
1981, 22, 1605-1608.
(42) (a) Dess, D. B.; Martin, J. C. J. Org. Chem., 1983, 48, 4155-4156.
(b) Meyer, S. D.; Schreiber, S. L. J. Org. Chem. 1994, 59, 7549-7552.
(43) Corey, E. J.; Enders, D.; Bock, M. G. Tetrahedron Lett. 1976, 7-10.
(44) Righi, G.; Rossi, L. Synth. Commun. 1996, 26, 1321-1327.
(45) Blanco, O. M.; Castedo, L. Synlett 1999, 5, 557-558.
(46) (a) Zhang, H. X.; Guibe, F.; Balavoine, G. J. Org. Chem. 1990, 55,
1857-1867. (b) Boden, C. D. J.; Pattenden, G.; Ye, T. J. Chem. Soc., Perkin
Trans. 1 1996, 2417-2419.
(47) Hofmeister, H.; Annen, K.; Laurent, H.; Weichert, R. Angew. Chem.,
Int. Ed. Engl. 1984, 23, 720-721.

3838 J. Am. Chem. Soc., Vol. 122, No. 16, 2000
Nicolaou et al.
authentic sample kindly provided by Dr R. Metternich of
Novartis.
and possibly in the elucidation of the, as yet, unknown
mechanism of action of this exciting new immunosuppressant,
in particular.
Conclusion
Acknowledgment. We thank Drs R. Chadha, G. Siuzdak,
and D. H. Huang for X-ray crystallographic, mass spectrometric,
and NMR assistance, respectively. This work was financially
supported by The Skaggs Institute for Chemical Biology, the
National Institutes of Health, postdoctoral fellowships from
Ministerio de Educacion y Cultura, Spain (to S.B.) and Ligue
Nationale contre le Cancer, France (to O.B.), and grants from
Pfizer, Glaxo, Merck, Schering Plough, Hoffmann La Roche,
Dupont, Boehringer Ingelheim, Abbott Laboratories, and Bristol-
Myers Squibb.
In conclusion, a convergent, highly stereocontrolled total
synthesis of SFA has been achieved. The palladium-mediated
Stille coupling7 is the keystone of the strategy for the stereospe-
cific construction of the macrocycle and for the union of the
left and right-hand fragments of 1. Highlights of the synthesis
include the use of Paterson’s aldol methodology11 for the
synthesis of the spirolactam fragment and carbodiimide-based
protocols for the construction of the tripeptide backbone. The
flexibility of the described strategy allows it to be adapted for
the generation of a wide variety of sanglifehrin analogues.
Moreover, based on the developed chemistry, a solid-phase
version amenable to combinatorial synthesis is imaginable.48
The described research could ultimately facilitate chemical
biology studies in the field of immunosuppression in general
Supporting Information Available: Experimental methods
and characterization for 2, 3, 6, 11, 12, 14-18, 20, 21, 23-28,
30-32, 34-38, 41-43, 46, 47, 49-54, 56-59, 61, 64-74, 76-
80, and sanglifehrin A (1). This material is available free of
charge via Internet at http://pubs.acs.org. See any current
masthead page for ordering information and Web access
instructions.
(48) During the preparation of this paper, researchers from Novartis
reported a convenient procedure for the degradation of sanglifehrin A thereby
making fragments of this molecule readily available to be incorporated into
combinatorial libraries: Metternich, R.; Denni, D. Thai, B.; Sedrani, R. J.
Org. Chem. 1999, 64, 9632-9639.
JA994285V