Total Synthesis of Prostaglandin 15d-PGJ2 and Investigation of its Effect on the Secretion of IL-6 and IL-12

Article abstract of DOI:10.1021/acs.orglett.5b02181

An efficient synthesis of 15-deoxy-Δ12,14-prostaglandin J₂ (15d-PGJ₂, 1) is reported. The route described allows for diversification of the parent structure to prepare seven analogues of 1 in which the positioning of electrophilic sites is varied. These analogues were tested in SAR studies for their ability to reduce the secretion of proinflammatory cytokines. It was shown that the endocyclic enone is crucial for the bioactivity investigated and that the conjugated ω-side chain serves in a reinforcing manner.

Full text of DOI:10.1021/acs.orglett.5b02181

Letter
pubs.acs.org/OrgLett
Total Synthesis of Prostaglandin 15d-PGJ2 and Investigation of its
Effect on the Secretion of IL‑6 and IL-12
Julian Egger,‡ Stefan Fischer,‡ Peter Bretscher,† Stefan Freigang,† Manfred Kopf,*,†
and Erick M. Carreira*,‡
‡Laboratory of Organic Chemistry, ETH Zurich, HCI H335, Wolfgang-Pauli-Strasse 10, 8093 Zurich, Switzerland
†Institute for Integrative Biology, ETH Zurich, HPL-G 12, Schafmattstrasse 27, 8093 Zurich, Switzerland
S
* Supporting Information
ABSTRACT: An efficient synthesis of 15-deoxy-Δ12,14-
prostaglandin J2 (15d-PGJ2, 1) is reported. The route
described allows for diversification of the parent structure to
prepare seven analogues of 1 in which the positioning of
electrophilic sites is varied. These analogues were tested in
SAR studies for their ability to reduce the secretion of
proinflammatory cytokines. It was shown that the endocyclic
enone is crucial for the bioactivity investigated and that the
conjugated ω-side chain serves in a reinforcing manner.
15-Deoxy-Δ12,14-prostaglandin J2 (15d-PGJ2, 1) is a recently
described prostanoid eicosanoid receiving considerable atten-
tion as a consequence of its extraordinary biological activity in
the modulation of inflammatory and apoptotic processes.1
While the majority of the other prostaglandins act in a
proinflammatory fashion, 15d-PGJ2 (1) was recognized for its
contrary effects and is therefore referred to as “the anti-
inflammatory prostaglandin”.1a Cyclopentenone prostaglandins
such as 15d-PGJ2 (1) differ from the other prostaglandins not
only in their biological mode of action, but also in their
structural properties.2 Prostaglandin 1, for instance, contains a
cross-conjugated trienone and therefore has become an
attractive synthetic target that has been the subject of four
total syntheses prior to this work.3 Recently, we have
established an efficient synthetic route to the structurally
similar cyclopentenone epoxyisoprostanes EC (2) and EI (3)
and have investigated their role in inhibiting secretion of
proinflammatory cytokines IL-6 and IL-12.4 Compounds 2 and
3 differ from 15d-PGJ2 (1) by the presence of an allylic epoxide
and the disposition of the carboxylic acid (Figure 1). Herein,
Figure 2. 15d-PGJ2 (1) and analogues 22−28 with varying double-
bond pattern. The 1,6-dienone in 1 and 1,4-enone in 26 are
highlighted in red.
deactivating compared to the analogue 1,4-enone that was
identified as the more potent compound (Figure 2, 1 vs 26).
The two distinct enone systems are highlighted in red in
compounds 1 and 26 in Figure 2.
In previous work, we have shown that the potency of the
epoxyisoprostanes 2 and 3 is higher than that of 15d-PGJ2 (1)
and that the lactone resulting from epoxide opening by the
carboxylic acid in EC (2) affords a highly potent anti-
inflammatory agent.5 Furthermore, we demonstrated that the
endocyclic enone (indicated by
in Figure 1) is crucial for
bioactivity and that 3 is most probably a precursor of 2.5
Prostaglandin 1 also incorporates an endocyclic enone but
shows much lower potency. The difference in the positioning of
the carboxylic acid in 1 versus 2 and 3 suggests that the inability
to form a γ-lactone as observed for the latter two could be
Figure 1. Structures of 15d-PGJ2 (1), EC (2), and EI (3).
we implement a modification of our approach to EC (2)
toward the synthesis of 1, enabling the preparation of a
collection of seven enone analogues (Figure 2). Biological
testing of the analogue structures disclosed that for 15d-PGJ2
(1) the endocyclic enone is of critical importance. Furthermore,
we could show that the 1,6-dienone in the ω-side chain acts
Received: July 28, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.5b02181
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
responsible for its distinct activity. Additionally, we wondered
whether the variations in biological activity could be derived
from differences in the nature of the electrophilic sites
secretion, we synthesized all seven possible electrophilic site
analogues 22−28 (Figure 2).
The synthesis of analogues 22−28 commenced with
cyclopentenone 11. The ω-side chain was installed either by
aldol condensation or alkylation reactions, depending on the
requisite double-bond pattern in the targeted molecule
(Scheme 3). Trienone 13 was prepared as described before
(indicated by
in Figure 1). To test this hypothesis, we
synthesized 15d-PGJ2 (1) and its analogues varying in the
positioning of the double bond.
The synthetic endeavors commenced with preparation of a
precursor enabling installation of the α-chain, namely aldehyde
6. It was prepared employing a four-step synthetic sequence, as
shown in Scheme 1. 1,5-Pentanediol 4 was protected as the
Scheme 3. Divergent Synthesis of the Precursors 13, and
31−36 from 11
Scheme 1. Synthesis of the α-Chain Aldehyde 6
corresponding PMB-ether and the remaining alcohol subjected
to oxidation employing Swern conditions to yield aldehyde 5.
Wittig olefination of the aldehyde with the phosphonium salt
generated from ethyl 4-bromobutanoate and PPh3 exlusively
gave a (Z)-olefin. Selective reduction of the ethyl ester to
aldehyde 6 was accomplished using 1 equiv of DIBAL-H in
toluene at −78 °C to furnish 6.
Treatment of 6 with ketene as described by Nelson afforded
7 in 87% yield and 94% ee (Scheme 2).6 β-Lactone 7 was
Scheme 2. Synthesis of 15d-PGJ2 (1)
for the synthesis of 15d-PGJ2 (1, Scheme 2). Dienone 31 was
accessible by an aldol condensation reaction employing octanal
(30) as reaction partner. In contrast to the preparation of 13,
the elimination step required the addition of neutral Al2O3. In
the presence of HMPA as a cosolvent, alkylation of enolized
cyclopentenone 11 with allylic iodide 29 could be achieved.7
Enones 13, 31, and 32 not only served as precursors for 1 and
the analogues 26 and 27 (Figure 2), but were also further
elaborated into precursors 33−36 by treatment with Stryker’s
reagent (Scheme 3).8 Alterations in the stoichiometry of this
reagent allowed for selective enone reductions.5 Direct
alkylation of cyclopentenone 11 with 1-iodooctane to give
the precursor for analogue 22 with the saturated ω-chain was
met with failure under a variety of conditions. Therefore, for
the preparation of precursor 39 (Scheme 4), a strategy was
Scheme 4. Preparation of Analogue Precursor 39
subjected to ring opening to give β-keto-δ-hydroxy ester 8,
which after diazotization and protection yielded 9. Cyclization
of 9 through a C−H insertion reaction (Rh2(esp)2) furnished
cyclopentanone 10 in 62% yield and high diastereoselectivity
(10:1) as determined by 1H NMR spectroscopy. Heating of 10
in the presence of LiOH in THF/water at reflux led to an α,β-
enone. Subsequent decarboxylation under Krapcho conditions
afforded cyclopentenone 11, an intermediate of a previous
synthesis of 15d-PGJ2 (1) by Kobayashi and Acharya.3c,d The
ω-side chain was installed through the implementation of a
stepwise aldol condensation reaction with aldehyde 12 to
obtain PMB ether 13. Deprotection of the PMB ether (13 →
14) and stepwise oxidation of the alcohol to the acid furnished
15d-PGJ2 (1). To assess the nature of the electrophilic sites in 1
crucial for inhibition of proinflammatory IL-6 and IL-12
chosen in which the endocyclic enone was temporarily
masked.5 Epoxidation of the more reactive endocyclic enone
afforded epoxide 37. Treatment with Stryker’s reagent
furnished epoxide 38, which upon reductive epoxide opening
and elimination of the β-hydroxy group regenerated the
endocyclic enone to yield precursor 39.9
With the precursors in hand, adjustment of the oxidation
state of the masked primary alcohol (Scheme 5) remained for
completion of the synthesis of the targeted 15d-PGJ2 analogue
structures 22−28. Cleavage of the PMB ethers and two-step
oxidation of the resulting primary alcohols to the carboxylic
acids provided 22−28.
B
DOI: 10.1021/acs.orglett.5b02181
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
the biological activity investigated in this context is completely
lost. Analogues that contain the endocyclic enone but lack
additional activation by conjugation to the ω-side chain (27
and 22) still maintain detectable bioactivity, although they
show a marked decrease.
Scheme 5. Preparation 15d-PGJ2-Analogues 22−28
In summary, we have demonstrated the application of a
synthetic route previously developed for the synthesis of
epoxyisoprostanes EC (2) and EI (3)4 to the preparation of
structurally related prostaglandin 15d-PGJ2 (1). Cyclopente-
none 11 served as a key intermediate that enabled the
preparation of seven analogue structures of 1 that vary in their
disposition of electrophilic sites. With these analogues in hand
we could identify the endocyclic enone as critical for the activity
of 15d-PGJ2 (1) as inhibitor for the secretion of proin-
flammatory IL-6 and IL-12. Finally, cross-conjugation in the ω-
side chain of the parent prostanoid enhances the electrophilic
properties of the endocyclic enone, and this effect is enhanced
for the structure encompassing a 1,4-enone in the ω-side chain
in lieu of a 1,6-dienone, a feature which renders analogue 26 a
more potent compound than parent 15d-PGJ2 (1).
a
Combined yields over three steps.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.5b02181.
■
Analogues 22−28 were tested together with parent 15d-PGJ2
(1) for their ability to reduce IL-6 and IL-12 secretion in bone
marrow-derived dendritic cells (BMDCs, Figure 3). All
S
Experimental procedures and full spectroscopic data for
all new compounds (PDF)
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail: carreira@org.chem.ethz.ch.
*E-mail: manfred.kopf@ethz.ch
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We are grateful to the ETH Zurich for generous support
through Grant ETH-18 09-1.
■
REFERENCES
■
(1) (a) Scher, J. U.; Pillinger, M. H. Clin. Immunol. 2005, 114, 100.
(b) Scher, J. U.; Pillinger, M. H. J. Investig. Med. 2009, 57, 703.
(c) Shibata, T.; Kondo, M.; Osawa, T.; Shibata, N.; Kobayashi, M.;
Uchida, K. J. Biol. Chem. 2002, 277, 10459. (d) Hardy, K. D.; Cox, B.
E.; Milne, G. L.; Yin, H.; Roberts, L. J. J. Lipid Res. 2011, 52, 113.
(2) (a) Straus, D. S.; Glass, C. K. Med. Res. Rev. 2001, 21, 185.
(b) Garzon
Proteomics 2011, 74, 2243. (c) Oeste, C. L.; Per
Spectrom. Rev. 2014, 33, 110.
(3) (a) Bickley, J. F.; Jadhav, V.; Roberts, S. M.; Santoro, M. G.;
Steiner, A.; Sutton, P. W. Synlett 2003, 2003, 1170. (b) Brummond, K.
M.; Sill, P. C.; Chen, H. Org. Lett. 2004, 6, 149. (c) Acharya, H. P.;
Kobayashi, Y. Tetrahedron 2006, 62, 3329. (d) Acharya, H. P.;
Kobayashi, Y. Tetrahedron Lett. 2004, 45, 1199. (e) Kim, N.-J.; Moon,
H.; Park, T.; Yun, H.; Jung, J.-W.; Chang, D.-J.; Kim, D.-D.; Suh, Y.-G.
J. Org. Chem. 2010, 75, 7458.
́
, B.; Oeste, C. L.; Díez-Dacal, B.; Per
́
ez-Sala, D. J.
Figure 3. Biological testing of 15d-PGJ2 (1) and its analogues 22−28
for their ability to reduce the secretion of proinflammatory cytokines
IL-6 and IL-12. R = C3H6CO2H.
́
ez-Sala, D. Mass
analogues were less potent than the parent 15d-PGJ2 (1)
with the exception of compound 26. Interestingly, dienone 26
shows higher potency than the parent compound, which
indicates that the 1,6-dienone in the ω-side chain of 15d-PGJ2
(1) has a less activating effect on the endocyclic enone than the
corresponding 1,4-enone in analogue 26. In this series of
analogues, it becomes clear that the endocyclic enone is crucial
for bioactivity as the analogues 23, 24, 25, and 28, all lacking
the endocyclic enone, show no effects in reducing IL-6 and IL-
12 secretion. Therefore, when the endocyclic enone is absent,
(4) Egger, J.; Bretscher, P.; Freigang, S.; Kopf, M.; Carreira, E. M.
Angew. Chem., Int. Ed. 2013, 52, 5382.
(5) (a) Egger, J.; Bretscher, P.; Freigang, S.; Kopf, M.; Carreira, E. M.
J. Am. Chem. Soc. 2014, 136, 17382. (b) Bretscher, P.; Egger, J.;
Shamshiev, A.; Trotzmuller, M.; Kofeler, H.; Carreira, E. M.; Freigang,
̈
̈
S.; Kopf, M. EMBO Mol. Med. 2015, 7, 593.
C
DOI: 10.1021/acs.orglett.5b02181
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(6) Zhu, C.; Shen, X.; Nelson, S. G. J. Am. Chem. Soc. 2004, 126,
5352.
(7) (a) Kanai, T.; Irifune, S.; Ishii, Y.; Ogawa, M. Synthesis 1989,
1989, 283. (b) Toyota, M.; Sasaki, M.; Ihara, M. Org. Lett. 2003, 5,
1193.
(8) Mahoney, W. S.; Brestensky, D. M.; Stryker, J. M. J. Am. Chem.
Soc. 1988, 110, 291.
(9) (a) Szostak, M.; Spain, M.; Procter, D. J. J. Org. Chem. 2012, 77,
3049. (b) Molander, G. A.; Hahn, G. J. Org. Chem. 1986, 51, 2596.
D
DOI: 10.1021/acs.orglett.5b02181
Org. Lett. XXXX, XXX, XXX−XXX