Evidence for separate binding and scaffolding domains in the immunosuppressive and antitumor marine natural product, pateamine A: Design, synthesis, and activity studies leading to a potent simplified derivative

Article abstract of DOI:10.1021/ja040065s

Pateamine A (PatA), a marine metabolite from Mycale sp., is a potent inhibitor of the intracellular signal transduction pathway emanating from the T-cell receptor leading to the transcription of cytokines such as interleukin-2 (IL-2). On the basis of the

Full text of DOI:10.1021/ja040065s

Published on Web 08/06/2004
Evidence for Separate Binding and Scaffolding Domains in
the Immunosuppressive and Antitumor Marine Natural
Product, Pateamine A: Design, Synthesis, and Activity
Studies Leading to a Potent Simplified Derivative
Daniel Romo,*,† Nam Song Choi,†,§ Shukun Li,† Ingrid Buchler,†,| Zonggao Shi,‡ and
Jun O. Liu*,‡
Contribution from the Department of Chemistry, Texas A&M UniVersity, P.O. Box 30012,
College Station, Texas 77842-3012 and Department of Pharmacology and Department of
Neuroscience, John Hopkins UniVersity, School of Medicine, 725 North Wolfe St.,
Baltimore, Maryland 21205
Received January 30, 2004; E-mail: romo@mail.chem.tamu.edu
Abstract: Pateamine A (PatA), a marine metabolite from Mycale sp., is a potent inhibitor of the intracellular
signal transduction pathway emanating from the T-cell receptor leading to the transcription of cytokines
such as interleukin-2 (IL-2). On the basis of the structure of PatA and initial biological results, a hypothesis
was developed regarding the presence of distinct binding and scaffolding domains in the PatA structure
with respect to interactions with its putative cellular receptor(s). Employing a highly convergent approach
involving a Hantzsch coupling strategy, we probed this hypothesis by preparing a simplified PatA derivative
(desmethyl, desamino PatA, DMDAPatA, 3). This derivative was prepared in 10 fewer synthetic steps relative
to PatA and was indeed found to exhibit equal to greater potency (IC50 0.81 ( 0.27 nM) in inhibition of IL-2
production relative to PatA (IC50 4.01 ( 0.94 nM) thus providing support for the binding/scaffolding domain
hypothesis. In addition, as a means to find more stable derivatives and gain further insights into structure-
activity relationships, several PatA derivatives were synthesized and assayed in the IL-2 reporter gene
assay. Several of these derivatives displayed lower potency but marked stability relative to the natural
product and provide further insights into the nature of the binding domain required for activity.
Introduction
interesting biological mechanisms of action. An important recent
discovery is the marine agent pateamine A (PatA (1), Figure
Natural products have proven to be extremely useful as probes
of biological processes.1 Examples include the immunosuppres-
sive, microbial secondary metabolites, cyclosporin A, FK506,
and rapamycin.2 Marine organisms are a rich source of bioactive
compounds, and many of these complex metabolites are prov-
ing useful as drug leads and biological probes.3 For example,
bryostatin,4 epothilone,5 discodermolide,6 and ecteinascidin7
show great potential as anticancer agents and have revealed
1), isolated from the Mycale sp.,8 which is showing great
promise as a biological probe of the complex intracellular
signaling pathways involved in T-cell activation.9 Preliminary
studies by a group at PharmaMar showed potent activity of PatA
in the mixed lymphocyte reaction and also in the mouse skin
graft rejection assay.10 More recent studies from our laboratories
suggest that PatA inhibits a specific intracellular signaling
pathway involved in T-cell receptor-mediated IL-2 production.11
In addition to its effect on TCR signaling pathway, PatA was
also found to induce apoptosis in certain mammalian cell lines,
especially those that are transformed with the oncogene Ras.12
† Department of Chemistry, Texas A&M University.
‡ Department of Pharmacology and Department of Neuroscience, John
Hopkins University, School of Medicine.
§ Work performed while on sabbatical from Chong Kun Dang Research
Institute (S. Korea).
| Current address: Pfizer, St. Louis, MO.
(8) Northcote, P. T.; Blunt, J. W.; Munro, M. H. G. Tetrahedron Lett. 1991,
32, 6411. This initial communication of pateamine A isolation and
2D structure elucidation described selective cytotoxic and antifungal
activity.
(9) Alexander Akhiezer, Ph.D. Thesis, Massachusetts Institute of Technology,
1999.
(10) Immunosuppressive activity was first discovered by Dr. Glynn Faircloth,
PharmaMar Inc., Cambridge, MA (private communication). PatA showed
potent activity in the mixed lymphocycte reaction, (IC50 2.6 nM) and in
the mouse skin graft rejection assay, PatA was found to be more potent
than cyclosporin A with only low toxicity at high doses but all doses were
active.
(11) Romo, D.; Rzasa, R. M.; Shea, H. A.; Park, K.; Langenhan, J. M.; Sun,
L.; Akhiezer, A.; Liu, J. O. J. Am. Chem. Soc. 1998, 120, 12 237-12 254.
(12) Hood, K. A.; West, L. M.; Northcote, P. T.; Berridge, M. V.; Miller, J. H.
Apoptosis 2001, 6, 207-219.
(1) Schreiber, S. L.; Hung, D. T.; Jamison, T. F. Chem. Biol. 1996, 3, 623-
639.
(2) (a) For a review on immunosuppressive natural products, see: (b) Hung,
D. T.; Jamison, T. F.; Schreiber, S. L. Chem. Biol. 1996, 3, 623-639.
(3) (a) Faulkner, D. J. Nat. Prod. Rep. 2002, 19, 1-48. (b) Newmann, D. J.;
Cragg, G. M.; Snader, K. M. Nat. Prod. Rep. 2000, 17, 215-234. (c)
Fenical, W. Chemical Studies of Marine Bacteria: Developing a New
Resource, Chem. ReV. 1993, 93, 1673-1683.
(4) Amador, M. L.; Jimeno, J.; Paz-Ares, L.; Cortes-Funes, H.; Hidalgo, M.
Ann. Oncol. 2003, 12, 1607-1615.
(5) Galmarini, C. M.; Dumontet, C. Idrugs 2003, 6, 1182-1187.
(6) Hung, D. T.; Nerenberg, J. B.; Schreiber, S. L. J. Am. Chem. Soc. 1996,
118, 11 054-11 080.
(7) For a lead reference describing Phase II clinical studies, see: Cvetkovic,
R. S.; Figgitt, D. P.; Plosker, G. L. Drugs 2002, 62, 1185-1192.
9
10582
J. AM. CHEM. SOC. 2004, 126, 10582-10588
10.1021/ja040065s CCC: $27.50 © 2004 American Chemical Society

Binding and Scaffolding Domains in Pateamine A
A R T I C L E S
Figure 1. Structures of pateamine A (1), Boc-pateamine A (2), and des-methyl, des-amino pateamine A (DMDAPatA, 3). Also shown are the putative
binding (rigid regions) and scaffolding (flexible) domains.
Scheme 1
Analysis of the PatA structure reveals a rigid eastern half
(C6-C24) including the thiazole, dienoate, and the triene side
chain, due to extended conjugation, and a more flexible western
half (C1-C5). Furthermore, C3-Boc-PatA was found to have
only 3-4-fold lower activity than PatA.11 These facts in
conjunction with simple molecular modeling studies led us to
hypothesize that PatA consists of separate binding (C6-C24)
and scaffolding (C1-C5) domains (Figure 1). Herein, we
describe support for this hypothesis garnered from the total
synthesis and biological analysis of a simplified derivative,
devoid of the C3-amino and C5-methyl groups (3, DMDAPatA)
that builds on our previously described total synthesis but
importantly includes improvements to several reaction steps
leading to a more practical synthesis.11 For example, the
synthesis of DMDAPatA incorporates a practical, convergency-
building strategy involving a Hantzsch thiazole coupling. We
also describe the synthesis and biological testing of PatA
derivatives that were designed to impart stability to the labile
PatA structure. In particular, the goal was to stabilize the acid
sensitive triallylic acetate moiety (i.e., C10 position).
C3-amino and C5-methyl groups to provide evidence for this
hypothesis and potentially arrive at a more readily prepared PatA
derivative useful as a potential immunosuppressive or anticancer
agent and also as a probe for studying the T-cell receptor
mediated signaling pathway.
Synthesis of PatA Derivatives. For the synthesis of
DMDAPatA 3, we returned to a strategy that we had considered
and attempted for the total synthesis of PatA but were unable
to bring to fruition.14 This entailed a more convergent strategy
to the C1-12 thiazole-containing fragment employing a Hantz-
sch coupling reaction (Scheme 2, 10 f 11 f 12). Pattenden
made use of a related-strategy in his synthesis of PatA.15
Since our initially reported total synthesis of PatA, we have
found a more practical approach to the acetate aldol adduct 6
previously described, employing the modified Nagao method
of Vilarrasa and co-workers.16 Use of the same substrates, but
employing TiCl4 as Lewis acid, led to a more practical and
scaleable procedure which could be readily performed on
multigram scale to give the â-hydroxy amide 6 (Scheme 1).
Subsequent conversion to the thioamide 7 was performed as
previously described.11
The requisite R-bromoketone 10 required for Hantzch cou-
pling was obtained by esterification and bromination of 6-oxo-
heptanoic acid (8). Hantzsch thiazole coupling between bromo-
ketone 10 and thioamide 711 using modified Meyers’ condi-
tions17 provided thiazole 12 in good overall yield (62%). A
critical prerequisite for optimal yields in this coupling was
purification of the intermediate thiazoline 11 prior to the de-
hydration step in contrast to our previous applications of this
reaction, in which this process could be performed in a single
pot11 and as previously described by Meyers.17 Deprotection
of the TIPS ether followed by a Mitsunobu coupling with the
more robust TIPS protected version of the previously described
enyne acid 1411 gave the macrocyclic precursor 15. Deprotection
Results and Discussion
Structural Analysis of PatA and DMDAPatA. Inspection
of the PatA structure reveals a highly unsaturated C6-C24
region consisting of a thiazole, a hydroxy dienoate, and a triene.
In addition, a more flexible region (C1-C5) bearing amino and
methyl substituted stereogenic carbons is found in the mac-
rocycle. This simple and cursory structural analysis and sim-
ple molecular modeling studies in combination with the pre-
viously reported potent biological activity of an acylated
C3-amine derivative (e.g., 2, R ) Boc)11 led us to a simple
hypothesis. Does PatA possess separate binding (C6-C24) and
scaffolding (C1-C5) domains? This notion was bolstered by
consideration of the immunosuppressive agents CsA, FK506,
and rapamycin, each of which has been functionally dissected
into two structural domains responsible for interacting with
distinct protein targets.2 Many protein ligands are known to
change conformations on binding to their receptors or alterna-
tively, the binding event leads to a more defined conformation.13
However, considering the preliminary biological data and simple
structural analysis described above, we set out to synthesize
the simplified PatA derivative 3 (DMDAPatA) devoid of the
(14) Unpublished results of Drs. Robert Rzasa and William D. Schmitz (TAMU).
(15) Remuinan, M. J.; Pattenden, G. Tetrahedron Lett. 2000, 41, 7367-7371.
(16) Gonza´lez, A.; Aiguade´, J.; Urp´ı, F.; Vilarrasa, J. Tetrahedron Lett. 1996,
37, 8949-8952.
(13) Rosen, M. K.; Belshaw, P. J.; Alberg, D. G.; Schreiber, S. L. Bio. Med.
Chem. Lett. 1992, 2, 747-753.
(17) Aguilar, E.; Meyers, A. I. Tetrahedron Lett. 1994, 35, 2473-2476.
9
J. AM. CHEM. SOC. VOL. 126, NO. 34, 2004 10583

A R T I C L E S
Romo et al.
Scheme 2
(a) HOCH2CCl3, SOCl2, benzene, reflux, 5 h, 61%; (b) (i) TMSOTf, i-Pr2NEt, 0 °C,1.5 h (ii) NBS, NaHCO3, -78 °C, 2 h (79%, 2 steps); (c) 7, 2,6-lut.,
CH2Cl2, 25 °C, 18 h; (d) TFAA, py., Hu¨nig’s base, 0f25 °C, 3 h, 62% (2 steps); (e) TBAF, 20 mol % AcOH, THF, -20 °C, 3 h, 97%; (f) PPh3, DIAD,
THF, -20 °C, 1 h, 91%; (g) TBAF, 50 mol% AcOH, THF, 25 °C, 8 h, 80%; (h) 10% Cd/Pb, THF/1M NH4OAc, 25 °C, 3 h, 89%; (i) 2,4,6-trichlorobenzoyl
Chloride, Et3N, DMAP, toluene, THF (0.001 M), 25 °C, 2 h, 84%; (j) Pd(CaCO3)/Pb, H2, MeOH, 25 °C, 3 h, 56%; (k) 10 mol % [Pd2dba3‚CHCl3:
AsPPh3)1:8], 20, THF, 25 °C, 14 h, 76%.
of the TIPS ether and trichloroethyl ester18 of diester 15 followed
by Yamaguchi macrocyclization19 gave bis-macrolactone 18.
Lindlar reduction gave E,Z-diene 19, and following a Stille
coupling20 with the previously described dienyl stannane 20,11
DMDAPatA (3) was obtained in 14 steps (longest linear
sequence) from aldehyde 5.
effects on biological activity of a more rigid macrocycle (i.e.,
enyne vs dienoate) for BocPatA (i.e., enyne 32) in conjunction
with modified terminal functional groups on the diene were
also investigated (i.e., enynes 29-31). These derivatives were
readily prepared by omission of the Lindlar reduction step
(Scheme 3).
Due to the aforementioned activity of C3-Boc PatA,11 two
additional C3-acyl amino derivatives were prepared with the
expectation that they should have similar potency. In this regard,
the C3-phenyl carbamate 35 and the C3-trifluoroacetamide 36
were synthesized by deprotection of Boc-macrocycle 24 fol-
lowed by acylation to give macrocycles 33 and 34 (Scheme 3).
Subsequent Lindlar reduction and Stille coupling gave the
C3-acylated derivatives 35 and 36.
The ability of these derivatives to inhibit T-cell receptor-
mediated IL-2 production was analyzed using an IL-2 reporter
gene assay. In this assay, a plasmid encoding a reporter gene
(luciferase) under the control of the IL-2 promoter was first
introduced into Jurkat T cells by transfection. The transfected
Jurkat T cells are then stimulated with two pharmacological
agents, phorbol myristyl acetate (PMA), which activates protein
kinase C, and ionomycin, which allows calcium to enter T cells
leading to activation of calmodulin and calcineurin. Together,
PMA and ionomycin recapitulate T-cell receptor signaling,
leading to the activation of the luciferase reporter gene by
activating the IL-2 promoter. Concurrent with the IL-2 lucifer-
ease reporter gene, we also transfected the same population of
Jurkat T cells with another plasmid encoding a constitutively
expressed â-galactosidase as an internal control. The ability of
PatA and its analogues to block T cell receptor-mediated IL-2
expression was determined after normalization of the IL-2
luciferase activity against the â-galactosidase activity from the
same sample. As â-galactosidase gene expression is constitutive
and not affected by stimulation through the T cell receptor, the
normalization against â-galactosidase activity excludes the
The synthesis of additional PatA derivatives with only minor
structural variations began with the previously described â-lac-
tam 21 (Scheme 3). The synthesis of all derivatives in this series
followed a process similar to that previously employed in the
total synthesis of PatA.11 We found it best to introduce the side
chain via a Stille coupling reaction as the final step in the
synthesis due partly to the polarity introduced by the tertiary
amine but primarily due to the instability associated with the
triallylic ester moiety (i.e., C10, PatA numbering). Derivatives
26-31 were prepared to determine the structural tolerance of
the terminal functional group on the side chain of PatA and
also to potentially improve the stability of the acid labile triallylic
acetate moiety by removal of one unsaturation. For this purpose,
the required stannanes 37-39 for side chain-modified PatA
derivatives were prepared by standard conditions and stannyl-
cupration was performed as described previously for vinyl
stannane 20 (Scheme 4).11 The C3-Boc protecting group had
previously been found to have a minor effect on activity
(3-4-fold decrease in activity)11 and improve stability and
therefore, for ease of handling, this group was retained in all
derivatives. PatA derivatives with a diene rather than a triene
side chain bearing a hydrogen bond donor (i.e., dienyl alcohol
26), a less basic hydrogen bond acceptor (i.e., methyl ether 27),
and the parent dimethyl amino group (i.e., dienyl dimethylamine
28) were also synthesized. In addition, derivative 25 bearing
the identical side-chain found in PatA with the exception of
one unsaturation and the C16 methyl group was prepared. The
(18) Dong, Q.; Anderson, C. E.; Ciufolini, M. A. Tetrahedron Lett. 1995, 36,
5681-5682.
(19) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Chem. Soc.
Jpn. 1979, 52, 1989-1993.
(20) Farina, V.; Krishnan, B. J. Am. Chem. Soc. 1991, 113, 9585-9595.
9
10584 J. AM. CHEM. SOC. VOL. 126, NO. 34, 2004

Binding and Scaffolding Domains in Pateamine A
A R T I C L E S
Scheme 3
(a) PPh3, DIAD, THF, -20 °C, 2H, 71%; (b) (i) HF‚py. THF, 25 °C, 24 h, 86%, (ii) Et4NCN, CH2Cl2, 25 °C, 5 h, 65%; (c) Pd(CaCO3)/Pb, H2, MeOH,
12 h, 99% (d) 20 or 37-39, 10 mol % [Pd2dba3‚CHCl3: AsPPh3 ) 1:8], THF, 25 °C, 10-18 h, 11-76%; (e) (i) 20% TFA, CH2Cl2, 0 °C, 15 h, 95%, (ii)
PhCOCl or (CF3CO)2O, DMAP, py., CH2Cl2, 25 °C, 5 h, 99%.
Scheme 4
rather than nitrogen at the terminus of the side chain (i.e., 29
and 30) were found to have activities in the IL-2 reporter gene
assay ranging between 55 and 335 nM, respectively. However,
enyne derivatives 31 and 32 with side chains more closely
resembling the natural product (i.e., amino end groups) had no
activity. Conversely, the dienoate derivatives (i.e., 26 and 27)
that should have macrocyclic conformations similar to the
(a) Ag2O, MeI, CH3CN, reflux, 9 h, 25%; (b) (i) TsCl, py., CH2Cl2 0
natural product but bearing oxygen rather than nitrogen at the
°C, 8 h, 65% (ii) dimethyl amine (g), THF, -78 °C, 6 h, 85%.
terminus of the side chain had very low activity. However, once
nitrogen is introduced into the side chain, as in derivative 28,
activity is restored (328 nM) to some degree. Thus, it would
nonspecific cytotoxic effects to cells.21
possibility that inhibition of IL-2 reporter gene is due to
appear that an oxygenated side-chain, regardless of hydrogen
As seen in Table 1, most derivatives were in general less
bonding capabilities, compensates to some extent for the change
potent than PatA (1). As expected, the C3-phenyl carbamate
in macrocycle conformation that occurs upon introduction of
derivative 35 possessed an activity that is approximately 4-fold
an enyne. However, oxygen rather than nitrogen on the side
lower than PatA (4 nM) in analogy to BocPatA (3). What is
chain leads to greatly diminished activity when coupled to the
not readily explained is the reduced activity of the trifluoro-
natural dienoate-containing macrocycle. Further analysis and
acetamide 36 (∼303 nM). While we cannot offer an explanation
understanding of these results in regard to their relevance for
at this time, we note that introduction of fluorine atoms into
protein ligand interactions must await structural characterization
protein ligands often lead to results that are not readily
of the interactions of the putative cellular protein receptor(s)
rationalized.22 A possible trend is observed upon com-
with these PatA derivatives.
parison of dienoate macrocycles 26-28 and enynoate macro-
cycles 29-32. Enyne derivatives having an expectedly more
rigid macrocycle than the natural product and bearing oxygen
The most intriguing derivative and one that provides direct
support for the binding/scaffolding hypothesis was DMDAPatA
3. This derivative displayed similar to greater potency (IC50 0.8
( 0.3 nM) relative to PatA (IC50 4.0 ( 0.9 nM) in its ability to
(21) Su, B.; Jacinto, E.; Hibi, M.; Kallunki, T.; Karin, M.; Ben-Neriah, Y. Cell
inhibit expression of the IL-2 reporter gene in stimulated Jurkat
T cells (Table 1). To ensure that the chemical modifications in
DMDAPatA did not change its mechanism of action, we
1994, 77, 727-736.
(22) Filler, R., Kobayashi, Y., Yagupolskii, L. M., Eds. Organofluorine
Compounds in Medicinal Chemistry and Biomedical Applications; Elsevi-
er: Amsterdam, 1993.
9
J. AM. CHEM. SOC. VOL. 126, NO. 34, 2004 10585

A R T I C L E S
Romo et al.
Table 1. IL-2 Reporter Gene Assay (transfected Jurkat cells) Activity of Pateamine A and Derivatives
a Not active. b Inhibition activity was observed, but it did not reach 50% even with the highest concentration tested. c It should be noted that the IC50
value for PatA in this particular assay is 10 fold higher than that previously reported (ref 11). It seems that Jurkat cells appear to vary in their sensitivity to
PatA, depending in part on the number of passages they have undergone. All IC50 values listed in this table were determined using the same population of
Jurkat T cells.
of signaling leading to IL-2 promoter activation. Together, the
results with DMDAPatA support the hypothesis that the
C1-C5 segment of PatA does not interact directly with its
putative cellular receptor leading to inhibition of IL-2 transcrip-
tion but that this sector of the molecule may simply serve as a
scaffold to define and maintain the macrocyclic conformation.
Figure 2. Reverse transcription-polymerase chain reaction assay to
Importantly, this derivative was also more stable than PatA
determine the level of mRNA for both IL-2 and PGK genes in the presence
(stable in CDCl3 for 3-4 weeks at 25 °C; PatA decomposes in
and absence of PatA and DMDAPatA. Lane 1, PMA + ionomycin; 2, PMA
CDCl3 at 25 °C in <10 min).
+ ionomycin + 0.1 nM PatA; 3, PMA + ionomycin + 5 nM PatA; 4,
PMA + ionomycin + 0.1 nM DMDAPatA; 5, PMA + ionomycin + 5 nM
DMDAPatA; 6. DMSO (carrier solvent control).
Conclusion
In summary, a hypothesis was developed regarding a potential
binding and scaffolding domain in the immunosuppressive and
anticancer marine natural product, pateamine A. This hypothesis
was premised on preliminary biological studies, structural
analysis, and simple molecular modeling studies of PatA and
derivatives. To lend support to this hypothesis, the simplified
derivative DMDAPatA 3 devoid of the C3-amino and C5-methyl
groups was designed, synthesized, and found to have similar to
greater potency than PatA in the IL-2 reporter gene assay. This
result provides convincing evidence for the hypothesis that this
sector of the molecule (C1-C5) merely serves as a scaffold
for the remaining conformationally rigid sectors (C6-C24) of
the molecule including the thiazole, the dienoate, and the triene
side chain. These findings are reminiscent of similar receptor
binding proposals put forth in early studies of cyclic peptide
and macrocyclic immunosuppressive natural products, namely
cyclosporin A, FK506, and rapamycin.2 However, in those cases,
the domains were renamed effector and binding domains due
to the fact that these natural products were found to bind two
cellular proteins acting as molecular “glue.” Importantly, the
determined the effects of PatA and DMDAPatA in an orthogonal
assay, namely transcription of the endogenous IL-2 gene and
also the housekeeping gene PGK in response to stimulation by
PMA and ionomycin using a reverse-transcription-polymerase
chain reaction (RT-PCR) assay. As shown in Figure 2, IL-2
mRNA is induced by treatment of Jurkat T cells with PMA
plus ionomycin, similar to the induction of the IL-2 luciferase
reporter gene (Lane 1). Treatment with sub-optimal doses of
either PatA or DMDAPatA had no effect on the transcription
of the IL-2 gene (Figure 2, Lanes 2 and 4). In the presence of
higher concentrations of both PatA and DMDAPatA, transcrip-
tion of endogenous IL-2 gene is completely inhibited (Figure
2, Lanes 3 and 5), with the level of IL-2 mRNA similar to cells
without stimulation by PMA and ionomycin (Figure 2, Lane
6). It is noteworthy that higher concentrations of PatA or
DMDAPatA did not affect the transcription of PGK which
would indicate general inhibition of cell proliferation or
induction of apoptosis, strongly suggesting that the effects of
both PatA and DMDAPatA on IL-2 is due to specific inhibition
9
10586 J. AM. CHEM. SOC. VOL. 126, NO. 34, 2004

Binding and Scaffolding Domains in Pateamine A
A R T I C L E S
23.2; IR (neat) 2934, 1756, 1710 cm-1; HRMS (ESI) Calcd for
C9H12Br Cl3O3 [M+H]: 376.8903 Found: 376.8901.
synthesis of this derivative (14 versus 24 steps from crotyl
alcohol) is greatly simplified relative to PatA (1) being devoid
of two stereocenters and having greater stability.
Thiazole 12. To a cooled (-5 °C), stirred solution of R-bromoketo
ester 10 (4.50 g, 12.7 mmol) in CH2Cl2 (250 mL) was added 2,6-lutidine
(2.80 mL, 24.1 mmol) and thioamide 711 (3.72 g, 9.79 mmol). The
solution was stirred at 25 °C for 18 h. The reaction mixture was washed
with brine, dried over MgSO4 and concentrated in vacuo. Rapid
purification of the residue by flash column chromatography on SiO2
eluting with EtOAc:hexanes (1:4) gave 4.06 g (64%) of the intermediate
thiazoline 11 as a colorless oil and as a mixture of diastereomers which
was used directly in the next step. Partial spectral data is provided: 1H
NMR (300 MHz, CDCl3) δ 5.91 (dt, J ) 1.8, 9 Hz, 1H), 4.82-4.77
(m, 1H), 4.77(s, 2H), 3.29 (s, 2H), 2.91-2.81 (m, 2H), 2.72 (ddd, J )
7.2, 8.7, 14.1, 1H), 2.53 (t, J ) 7.5 Hz, 2H), 2.29 (s, 2H), 1.89 (t, J )
9.3 Hz, 2H), 1.78 (t, J ) 7.2 Hz, 2H), 1.63-1.48 (m, 4H); 13C NMR
(75 MHz, CDCl3) δ 172.1, 134.9, 108.5, 74.1, 68.9, 68.6, 43.6, 43.5,
43.4, 43.2, 40.9, 34.1, 25.2, 24.7, 23.7, 18.3, 18.1, 12.5. To a cooled
(0 °C), stirred solution of thiazoline 11 (3.46 g, 5.29 mmol) in CH2Cl2
(70 mL) was added Hu¨nig’s base (9.63 mL, 55.4 mmol), pyridine (1.53
mL, 18.9 mmol) and TFAA (2.61 mL, 18.5 mmol) and the solution
was stirred at 25 °C for 3 h and then diluted with 50 mL of CH2Cl2.
The organic layer was washed with satd. aqueous NaHCO3, brine, dried
over MgSO4 and concentrated in vacuo. Purification of the residue by
flash column chromatography on SiO2 eluting with EtOAc:hexanes
(1:4) gave 3.26 g (97%) of thiazole 12 as a yellow oil: 1H NMR (300
MHz, CDCl3) δ 6.78 (s, 1H), 5.89 (dd, J ) 0.6, 8.7 Hz, 1H), 4.80 (dt,
J ) 6.0, 9.0 Hz, 1H), 4.76 (s, 2H), 3.25 (dd, J ) 6.3, 14.1 Hz, 1H),
3.13 (dd, J ) 6.3, 14.1 Hz, 1H), 2.80-2.75 (m, 2H), 2.54-2.50 (m,
2H), 2.11 (s, 3H), 1.85-1.72 (m, 4H), 1.04 (s, 21H); 13C NMR (75
MHz, CDCl3) δ 172.1, 165.4, 156.6, 135.1, 121.5, 113.4, 95.2, 74.1,
70.3, 42.2, 33.9, 31.3, 28.8, 24.5, 24.2, 18.2, 12.5; HRMS (ESI) Calcd
for C24H40BrCl3NO3SSi [M+H]: 634.0717 Found: 634.0748.
Thiazole Enyne 15. To a cooled (-20 °C), stirred solution of
thiazole 12 (3.17 g, 4.99 mmol) in THF (50 mL) was added 15 mL of
1M TBAF (15.0 mmol) buffered with 20 mol % AcOH and the solution
was stirred at -20 °C for 3 h. The reaction mixture was diluted with
50 mL of CH2Cl2. The organic layer was washed with satd. aq.
NaHCO3, brine, dried over MgSO4 and concentrated in vacuo. Coarse
purification of the residue by flash column chromatography on SiO2
eluting with EtOAc:hexanes (1:4f1:1) gave 38 mg of alcohol 13 (2.30
g, 97%) as a light yellow oil which was used directly in the next step.
To a solution of DIAD (0.91 mL, 4.63 mmol) in THF (30 mL) was
added PPh3 (0.97 g, 3.70 mmol) as a solid and the solution was stirred
at ambient temperature for 30 min. The resulting heterogeneous mixture
was cooled (-20 °C) and the solution of acid 14 (0.60 g, 1.35 mmol)
in THF (0.2 mL) was added. After 20 min, a solution of alcohol 13
(0.39 g, 0.81 mmol) in THF (4.0 mL) was added and stirring was
continued for 1 h at -20 f -30 °C. The reaction was quenched by
addition of 10 mL pH 7 buffer followed by warming to 25 °C and
diluting with 100 mL of EtOAc. The layers were separated and the
aqueous layer was extracted with EtOAc. The combined organic layers
were washed with brine, dried over Na2SO4 and concentrated in vacuo.
Purification of the residue by flash column chromatography on SiO2
eluting with EtOAc:hexanes (1:15) gave 580 mg (91%) of thiazole
enyne 15 as a light yellow oil:1H NMR (300 MHz, CDCl3) δ 6.77 (s,
1H), 5.87 (dd, J ) 1.2, 9.6 Hz, 1H), 5.79-5.72 (m, 1H), 5.38 (d, J )
1.2 Hz, 1H), 4.71 (s, 2H), 4.13-4.07 (m, 1H), 3.36 (dd, J ) 6.9, 14.7
Hz, 1H), 3.24 (dd, J ) 6.9, 14.7 Hz, 1H), 2.74 (t, J ) 6.9 Hz, 1H),
2.48 (t, J ) 6.9 Hz, 2H), 2.39 (dd, J ) 5.4, 14.7 Hz, 1H), 2.23 (dd, J
) 5.4, 14.7 Hz, 1H), 2.26 (d, J ) 1.2 Hz, 1H), 1.99 (d, J ) 1.5 Hz,
3H), 1.77-1.69 (m, 4H), 1.24-1.22 (m, 2H), 1.11 (d, J ) 6.3 Hz,
3H), 1.03 (s, 21H); HRMS (ESI) Calcd for C33H50Br Cl3NO5SSi
[M+H]: 784.1428 Found: 784.1434.
As previously observed, C3-amino acylated derivatives of
PatA retain activity in the IL-2 reporter gene assay. In addition,
a subtle interplay between the dienoate sector (C18-C22) and
the triene side chain was revealed when dienoate versus
enynoate-containg macrocycles were compared. This appears
to suggest the importance of macrocycle conformation and side
chain functionality in binding of PatA to its putative cellular
receptor. Current studies are directed toward characterizing the
cellular protein receptor(s) of PatA and the results of these
studies will be reported in due course.
Experimental Section
Modified Nagao Acetate Aldol: â-Hydroxy Amide 6. To a cold
(-40 °C) stirred solution of thiazolidine thione 4 (6.59 g, 32.50 mmol)
in dry CH2Cl2 (115 mL) was added TiCl4 (35 mL, 1 M in CH2Cl2,
35.0 mmol). The resulting solution was stirred for 5 min at -40 °C,
then i-Pr2NEt (6.0 mL) was added. The resulting black slurry was
vigorously stirred for 2 h at -40 f 60 °C. Then a solution of aldehyde
5 (4.14 g, 27.89 mmol) in CH2Cl2 (5 mL) was added to the mixture at
-78 °C. The mixture was stirred at -78 °C for 15 min and then 20
mL of pH 7 buffer was added to quench the reaction. The aqueous
layer was extracted with EtOAc, the combined organic layer was
washed with brine, dried over MgSO4. Concentration and purification
by flash chromatography eluting with EtOAc:hexanes (1.5) gave aldol
adduct 6 (7.81 g, 80%) as a yellow oil with spectral and physical data
matching that previously reported and the remaining steps leading to
thioamide 7 were performed as previously described.11
Trichloroethyl Ester 9. To a stirred solution of 6-oxoheptanoic acid
(2.0 g, 12.58 mmol) in benzene (40 mL) was added trichloroethanol
(1.08 mL, 11.32 mmol) and SOCl2 (1.1 mL, 15.1 mmol). The solution
was refluxed for 8 h and then evaporated and diluted with 30 mL of
EtOAc. The layers were separated and the aqueous layer was extracted
with EtOAc. The combined organic layers were washed with brine,
dried over MgSO4 and concentrated in vacuo. Purification of the residue
by flash column chromatography on SiO2 eluting with EtOAc:hex-
anes (1:4) gave 2.1 g (61%) of ester 5 as a light yellow oil: 1H NMR
(300 MHz, CDCl3) δ 4.71(s, 2H), 2.46-2.42 (m, 4H), 2.11 (s, 3H),
1.65-1.62 (m, 4H); 13C NMR (75 MHz, CDCl3) δ 208.4, 171.8, 95.6,
74.1, 43.2, 33.8, 30.1, 24.3, 23.2; IR (neat) 2956, 1762, 1720 cm-1
.
R-Bromoketo Ester 10. To a cooled (0 °C) stirred solution of ester
9 (5.91 g, 21.5 mmol) in dry CH2Cl2 (350 mL) was added DIPEA
(18.5 mL, 0.106 mol) and TMSOTf (15.5 mL, 85.74 mmol). Stirring
was continued at 0 °C for 1.5 h. The reaction was quenched with
aqueous NaHCO3 (70 mL). The aqueous layer was extracted with
hexane, the combined organic layers were washed with brine, dried
over MgSO4 and concentrated in vacuo. The crude product was used
1
in the next step without purification. H NMR (300 MHz, CDCl3) δ
4.74 (s, 2H), 4.04 (s, 2H), 2.48 (t, 2H, J ) 8.2 Hz), 2.04 (t, 2H, J )
8.2 Hz), 1.78-1.64 (m, 2H), 1.59-1.50 (m, 2H).
To a cooled (-78 °C) stirred solution of the crude TMS enol ether
(8.1 g) in dry THF (350 mL) was added 2.2 g NaHCO3 and NBS (4.0
g, 22.5 mmol). The resulting mixture was stirred at -78 °C for 2 h
and then the reaction was quenched with aq. NaHCO3 (100 mL). The
aqueous layer was extracted with hexane, the combined organic layer
was washed with brine, dried over MgSO4 and concentrated in vacuo.
Purification of the residue was accomplished by directly loading onto
a flash column with SiO2 eluting with EtOAc:hexanes (25:1f10:1) to
give 6.04 g (79% for two steps) of bromoketo ester 10 as a pale yel-
low oil:1H NMR (300 MHz, CDCl3) δ 4.75(s, 2H), 3.88 (s, 2H),
2.74-2.68 (m, 2H), 2.56-2.44 (m, 2H), 1.74-1.66 (m, 4H); 13C NMR
(75 MHz, CDCl3) δ 201.7, 171.7, 95.1, 74.0, 39.4, 34.3, 33.7, 24.1,
Hydroxy Ester 16. To a stirred solution of silyl ether 15 (0.564 g,
0.72 mmol) in THF (10 mL) was added a 50% AcOH/TBAF solution
(3.60 mL, 3.60 mmol). The resulting solution was stirred at 25 °C for
9
J. AM. CHEM. SOC. VOL. 126, NO. 34, 2004 10587

A R T I C L E S
Romo et al.
8 h. The reaction mixture was diluted with 50 mL of CH2Cl2. The
organic layer was washed with NaHCO3 solution, brine, dried over
MgSO4 and concentrated in vacuo. Purification of the residue was
accomplished by directly loading onto a flash column chromatography
of SiO2 eluting with EtOAc:hexanes (1:1) gave 362 mg (80%) of alcohol
16 as a yellowish oil:1H NMR (500 MHz, CDCl3) δ 6.84 (s, 1H), 5.92
(dd, 1H, J ) 1.2, 6.9 Hz), 5.86-5.79 (m, 1H), 5.48 (s, 1H), 4.77 (s,
2H), 4.07-4.01 (m, 1H), 3.43 (dd, J ) 6.9, 14.7 Hz, 1H), 3.29 (dd, J
) 6.9, 14.7 Hz, 1H), 2.81 (t, 2H, J ) 6.6 Hz), 2.54 (t, 1H, J ) 6.9
Hz), 2.32 (d, 3H, J ) 1.5 Hz), 2.07 (d, 3H, J ) 1.2 Hz), 1.80-1.76
(m, 4H), 1.26 (d, 3H, J ) 6.3 Hz); 13C NMR (125 MHz, CDCl3) d
172.1, 171.4, 163.7, 158.5, 156.9, 153.2, 128.6, 128.1, 113.8, 105.2,
85.6, 83.7, 74.1, 71.6, 65.9, 48.9, 38.0, 33.9, 31.2, 28.7, 24.5, 23.6,
20.7, 14.4; HRMS (ESI) Calcd for C24H39Br Cl3NO5S [M+H]:
628.0094 Found: 628.0073
Macrocycle 18. To a stirred solution of alcohol 16 (342 mg, 0.54
mmol) in THF (7.5 mL) and 1M NH4OAc (7.5 mL) was added 10%
Cd/Pd couple (500 mg) The resulting solution was stirred at 25 °C for
8 h. The reaction mixture was diluted with 10 mL of EtOAc and then
filtered through a pad of diatomaceous earth. The organic layer was
washed with brine, dried over MgSO4 and concentrated in vacuo.
Purification by flash chromotagraphy on SiO2 eluting with EtOAc:
MeOH (10.1) gave hydroxy acid 17 as a colorless oil (240 mg, 89%).
To a cooled (0 °C) stirred solution of hydroxy acid 17 (62.0 mg,
0.12 mmol) in THF (3.0 mL) was added Et3N (110 µL, 0.72 mmol)
and 2,4,6-trichlorobenzoyl chloride (98 µL, 0.63 mmol). The resulting
solution was stirred at 0 °C for 20 min and then added to a solution of
DMAP (152 mg, 1.26 mmol) in toluene (62 mL) at 25 °C and stirred
for 2 h. The reaction mixture was diluted with 50 mL of EtOAc. The
organic layer was washed with brine, dried over MgSO4 and concen-
trated in vacuo. Purification of the residue by flash column chroma-
tography on SiO2 eluting with EtOAc:hexanes (1:3) gave 50.0 mg (84%)
of macrocycle 18 as a pale yellow oil: 1H NMR (300 MHz, CDCl3) δ
6.82 (s, 1H), 6.06 (dd, J ) 1.2, 9.3 Hz, 1H), 5.93-5.86 (m, 1H), 5.35
(s, 1H), 5.33-5.26 (m, 1H), 3.34 (br d, J ) 7.5 Hz, 2H), 2.81-2.61
(m, 2H), 2.46-2.39 (m, 1H), 2.43 (d, J ) 1.5 Hz, 3H), 2.30 (d, J )
6.9 Hz, 2H), 1.96 (d, J ) 1.2 Hz, 3H), 1.90-1.68 (m, 3H), 1.57-1.45
(m, 3H), 1.28 (d, J ) 6.3 Hz, 3H); HRMS (ESI) Calcd for C22H27Br
Cl3NO4S [M+H]: 480.0844 Found: 480.0760.
δ 7.02 (d, J ) 11.7 Hz, 1H), 6.72 (s, 1H), 6.71 (dd, J ) 11.7 Hz, 1H),
6.08 (dt, J ) 4.5, 16.5 Hz, 1H), 5.97 (dq, J ) 1.2, 9.6 Hz, 1H), 5.36
(d, J ) 11.7 Hz, 1H), 5.23-5.12 (m, 1H), 3.27-3.14 (m, 2H), 2.93-
2.83 (m, 1H), 2.60 (ddd, J ) 4.5, 10.5, 1.4 Hz, 1H), 2.53-2.47 (m,
1H), 2.51 (s, 3H), 2.41-2.13 (m, 4H), 1.86 (s, 3H), 1.77-1.61 (m,
2H), 1.44-1.29 (m, 2H), 1.27 (d, J ) 6.6 Hz, 3H). This material was
directly used in the next reaction without further purification. To a
flask charged with Pd2dba3‚CHCl3 (17 mg, 0.016 mmol) and triphenyl
arsine (41 mg, 0.13 mmol) was added 1.0 mL of degassed THF prepared
by several freeze/thaw cycles. The final concentration of this palladium
catalyst stock solution was ∼0.031 M. To a solution of macrocycle 19
(66 mg, 0.14 mmol) and stannane 2011 (113 mg, 0.27 mmol) in 4.0
mL of THF was added 450 µL (0.014 mmol, 10 mol %) of palladium
catalyst stock solution. After stirring the resulting homogeneous solution
at 25 °C for 4 h, another 300 µL of palladium catalyst stock solution
was added and stirring was continued for an additional 12 h.
Concentration in vacuo and purification of the residue by flash column
chromatography on SiO2 eluting with EtOAc:Et3N (20:1) gave 54 mg
(76%) of DMDAPatA (3) as a pale yellow oil: 1H NMR (500 MHz,
C6D6) δ 7.47 (d, J ) 12.0 Hz, 1H), 6.71 (app dt, J ) 5.0, 9.0 Hz, 1H),
6.45 (app t, J ) 11.5 Hz, 1H), 6.32 (d, J ) 16.0 Hz, 1H), 6.21 (d, J )
16.0 Hz, 1H), 6.17 (s, 1H), 5.69 (t, J ) 7 Hz, 1H), 5.55 (d, J ) 11.5
Hz, 1H), 5.49 (d, J ) 9.0 Hz, 1H), 5.19-5.11 (m, 1H), 3.08-3.01
(m, 2H), 2.91 (d, J ) 6.5 Hz, 2H), 2.78 (dt, J ) 4.5, 14.0 Hz, 1H),
2.48-2.42 (m, 1H), 2.33 (ddd, J ) 4.0, 10.0, 14.5 Hz, 1H), 2.11 (s,
6H), 2.16-2.03 (m, 2H), 1.90 (d, J ) 1.0 Hz, 3H), 1.71 (2, 3H),
1.64-1.61 (m, 1H), 1.56-1.43 (m, 2H), 1.54 (s, 3H), 0.96-0.81 (m,
2H), 0.95 (d, J ) 6.5 Hz, 3H); HRMS (ESI) Calcd for C30H43N2O4S
[M+H]: 527.2944 Found: 527.2927.
Acknowledgment. We thank the NIH (NIGMS 052964-06
to D.R., NIGMS 055783-06 and CA10021-01 to J.O.L.) and
the Welch Foundation (A-1280) for support of this work. We
also thank Chong Kun Dang Research Institute (S. Korea) for
partial support of N.S.C. as a visiting scientist. We thank
Mr. Nicholas T. Burke, a Summer 2001 NSF-REU participant
(CHE-0071889) for technical assistance.
Supporting Information Available: General procedures for
synthesis of PatA derivatives 26-32 and 35-36 and 1H
NMR spectra for compounds 3, 9-10, 11-13, 15-16, 18-19,
26-32, and 35-36. This material is available free of charge
via the Internet at http://www.acs.pubs.org.
DMDA PatA (3). A slurry of Pd/CaCO3 poisoned with Pb (52 mg)
and macrocycle 18 (117 mg, 0.24 mmol) in 10 mL of MeOH was
evacuated under water aspirator pressure and purged with H2. After
stirring at 25 °C for 3 h under 1 atm of H2, the reaction was filtered
through diatomaceous earth and concentrated in vacuo. Passage through
a plug of SiO2 eluting with EtOAc:hexanes (50:50) gave 66 mg (56%)
of E,Z-macrocycle 19 as a colorless oil: 1H NMR (300 MHz, CDCl3)
JA040065S
9
10588 J. AM. CHEM. SOC. VOL. 126, NO. 34, 2004