Soluble-polymer supported synthesis of a prostanoid library: identification of antiviral activity.

Article abstract of DOI:10.1021/ol991130j

[formula: see text] The prostaglandins are potent natural products taking part in many biological processes. The "convergent generation of diversity" from a "toolbox" of prostanoid components, augmented with additional polymer-supported transformations, c

Full text of DOI:10.1021/ol991130j

ORGANIC
LETTERS
1999
Vol. 1, No. 11
1859-1862
Soluble-Polymer Supported Synthesis of
a Prostanoid Library: Identification of
Antiviral Activity
,‡
,†
Kyung Joo Lee,† Ana Angulo,‡ Peter Ghazal,* and Kim D. Janda*
Departments of Chemistry and Immunology, The Scripps Research Institute,
10666 North Torrey Pines Road, La Jolla, California 92037
kdjanda@scripps.edu
Received October 7, 1999
ABSTRACT
The prostaglandins are potent natural products taking part in many biological processes. The “convergent generation of diversity” from a
“toolbox” of prostanoid components, augmented with additional polymer-supported transformations, can enable construction of valuable
libraries. A parallel-pool strategy was used to assemble a small library of prostanoids. The inhibition of a herpes-family virus demonstrated
the potential for new drug discovery.
The synthesis and screening of chemical libraries represents
the current forefront in the search for compounds possessing
biological activity. Of primary importance in this endeavor
is the construction of complex molecules on polymer
supports through adaptation of sophisticated organic reac-
tions. A number of natural product analogues have recently
highlighted the advancements in this area.1 Our contributions
have included the syntheses of the prostaglandin E2 (PGE2)
methyl ester and the related PGF2R on soluble polystyrene.2
The prostaglandin family of natural products constitutes
perhaps the most physiologically potent nonprotein molecules
found in mammals. These relatively low molecular weight
and delicate structures play a vital role in the processes of
inflammation and tissue repair and in the immune response.3
Given their enormous potential therapeutic benefits, extensive
efforts have been directed at the design and synthesis of
pharmacologically useful analogues.4 Herein, we report the
development of our previous soluble-polymer technology for
the preparation of a prostanoid library. Moreover, evaluation
of the library for inhibition of cytomegalovirus (CMV), a
herpes-family virus, revealed a prostanoid with potent
antiviral activity.
The “convergent generation of diversity” from a “toolbox”
of prostanoid components (enones, R-chains, ω-chains),
augmented with additional polymer-supported transforma-
tions, can enable the assembly of arrays of valuable
compounds. Central to this approach is a “parallel-pool”
library strategy that incorporates the well-known techniques
of split-mix and parallel combinatorial synthesis.5 In this way,
small pools of compounds (4-16 members) are manipulated
during library construction through a desired number of
† Department of Chemistry.
‡ Department of Immunology.
(1) Watson, C. Angew. Chem. Int. Ed. 1999, 38, 1903-1908.
(2) (a) Chen, S.; Janda, K. D. Tetrahedron Lett. 1998, 39, 3943-3946.
(b) Chen, S.; Janda, K. D. J. Am. Chem. Soc. 1997, 119, 8724-8725. For
prostaglandin syntheses on a solid support, see: (c) Thompson, L. A.;
Moore, F. L.; Moon, Y.-C.; Ellman, J. A. J. Org. Chem. 1998, 63, 2066-
2067.
(4) Collins, P. W.; Djuric, S. W. Chem. ReV. 1993, 93, 1533-1564.
(5) Reviews: (a) Molecular DiVersity and Combinatorial Chemistry:
Libraries and Drug DiscoVery; Chaiken, I. M., Janda, K. D., Eds.; American
Chemical Society, 1996. (b) Balkenhohl, F.; Bussche-Hunnefeld, C. v. d.;
Lansky, A.; Zechel, C. Angew. Chem., Int. Ed. Engl. 1996, 35, 2288-
2337.
(3) AdVances in Prostaglandin, Thromboxane, and Leukotriene Research;
Raven: New York; Vols. 11-19, 1983-1989.
10.1021/ol991130j CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/02/1999

Figure 1. The parallel-pool synthesis of a prostanoid library: (a) (i) 2a, 2b, 2c, or 2d (5 equiv), Cp2Zr(H)Cl (5 equiv), THF, rt, 30 min,
(ii) MeLi (10 equiv), -50 °C, 10 min, then CuCN (5 equiv), -50 °C, 15 min, (iii) MeLi (5 equiv), -50 °C, 15 min, (iv) 1, THF, -50 °C,
40 min, (v) TMSCl (25 equiv), -50 °C, 40 min, then NEt3 (50 equiv), -50 to 0 °C, 15 min; (b) (i) MeLi (4 equiv), THF, -23 °C, 40 min,
(ii) 4a, 4b, 4c, or 4d (18 equiv), THF, -78 °C, 10 min, then -23 °C, 40 min; (c) H2, 5% Pd-BaSO4, quinoline, benzene/cyclohexane (1/1),
45 °C, 48 h; (d) 48% aqueous HF, THF, 45 °C, 6 h.
mixing and splitting steps, in conjunction with identical or
different parallel chemical operations on the individual pools.
Significantly, although a library of unbound pools is obtained
at the conclusion of reaction sequences, all prior pools
following convergence of the three prostanoid components
can be considered as part of the total library upon release
from the polymer. Each pool can then be assessed in a
particular biological assay and, if positive, deconvoluted by
synthesis of subpools and unique members. We describe the
preparation and evaluation of a small prostanoid library that
provides a model for eventual extension to larger libraries.
The soluble-polystyrene polymer 1 was functionalized with
ω chains in four separate reactions that yielded 3a-d and
then pool ω when the polymers were mixed in equal amounts
(Figure 1). The pool ω was split into four batches and each
reacted with a different R-chain triflate 4a-d that produced
the first set of polymer-supported prostanoids. The pools
5A-D represented 16 potential members of our prostanoid
library. However, our objective at this time was to evaluate
prostaglandin analogues that contained the natural side-chain
unsaturation. Hence, parallel reductions of the alkynyl groups
were carried out that afforded pools 6A-D. A comparison
1860
Org. Lett., Vol. 1, No. 11, 1999

1
of the H NMR spectra of individual polymer-supported
the effect of the library prostanoids on the replication of
murine CMV in cultured cells. After synthesis, each final
pool was tested and pool 1 effected a significant inhibition
of viral titer (Table 1). Members 7aa-7ad of this pool were
pools before and after hydrogenation clearly showed the
successful outcome of the reaction. The integration of all
vinylic resonances was doubled after reduction relative to
the distinct acetal-methine hydrogen derived from the linker
1
and used as a standard. These data, together with H NMR
data of 3a-d and pools 5A-D, indicated an overall excellent
efficiency of the convergent operations.
Table 1. Inhibition of MCMV Growth in NIH 3T3 Cells by
Prostanoidsa
Treatment of pools 6A-D, in parallel, with fluoride
liberated the final 16-member library as pools 1-4 (26-
38% yields based on average molecular weights) in which
each contained four compounds. As an example, pool 1
contained prostanoids 7aa, 7ab, 7ac, and 7ad denoted with
two letters that indicated the origin from toolbox R and
toolbox ω chains, respectively (Figure 2). The 1H NMR and
prostanoid(s)
viral titer (PFU/mL)b
activity (%)c
PGE2
PGE2 Me ester
pool 1
pool 2
pool 3
pool 4
7a a
7a b
7a c
7a d
1.32 × 105
3.60 × 105
1.70 × 104
4.02 × 105
1.17 × 105
1.02 × 105
6.80 × 103
1.10 × 105
1.30 × 105
1.60 × 105
47
>100
6
>100
42
36
2
39
46
57
a MCMV ) murine CMV; see Supporting Information for assay
procedure. b PFU ) plaque forming units. c The viral titer that remained
as a percentage compared to the control titer (without exogenous prostanoid)
after addition of 20 µM prostanoid(s).
individually examined for antiviral activity. The results of
these experiments showed that 7aa was clearly the most
potent.
Work by others on human CMV and ganciclovir, the most
commonly used antiviral agent, suggested that our compound
compared favorably, having approximately an order of
magnitude less potency.8 The remainder of the library, the
alkynyl prostanoids obtained after cleavage of pools 5A-
D, will be studied in future experiments. Although previous
studies have established a link between prostaglandins and
CMV replication, the exact nature of this connection is not
well understood. Interestingly, such studies have generally
revealed that blocking prostaglandin synthesis inhibits viral
growth,9 an effect contrary to that observed here. Yet, some
prostaglandins up-regulate and others down-regulate immune
function. Several of the prostanoids present in our library,
such as the previously unknown 7aa, might indeed boost
cell-mediated factors that inhibit the virus. Our results are
significant, since there are few antiviral agents or drug
therapies clinically available for CMV infection and these
often lack desirable efficacy.10
Figure 2. The prostanoids contained in pool 1 of the library.
HPLC analyses of pools 1-4 showed characteristic proton
signals and four well-separated chromatographic peaks. In
light of subsequent biological data (vide infra), each member
of pool 1 was independently synthesized, which confirmed
structure and stereochemistry. Interestingly, unlike other
prostaglandin analogues, 7ad could not be synthesized in
solution and was therefore prepared via our polymer-
supported synthesis followed by an additional purification
step that yielded analytical material. While the reasons for
success only on the polymer were not clear, control experi-
ments ruled out a linker effect and one possible solvent-like
microenvironment effect (e.g., the solution reaction failed
not only in the standard THF solvent but also in toluene).
Thus, soluble-polymer synthesis might provide access to
some valuable prostanoids not otherwise obtainable by
conventional methods. For all pools only minor amounts of
other compounds, such as des-R-chain prostanoids from
incomplete alkylation of pool ω, were evident. After analysis,
the final pools were entered into a screening assay for
inhibition of a productive CMV infection.
(6) (a) Davis-Poynter, N. J.; Farrell, H. E. Trends Microbiol. 1998, 5,
190-197. (b) Davis-Poynter, N. J.; Farrell, H. E. Immunol. Cell Biol. 1996,
74, 512-522.
(7) (a) Meyers, J. D.; Ljungman, P.; Fisher, L. D. J. Infect. Dis. 1990,
162, 373-380. (b) Grundy, J. E.; Shanley, J. D.; Griffiths, P. D. Lancet
1987, 996-999. (c) Quinnan, G. V., Jr.; Masur, H.; Rook, A. H.; Armstrong,
G.; Frederick, W. R.; Epstein, J.; Manischewitz, J. F.; Macher, A. M.;
Jackson, L.; Ames, J.; Smith, H. A.; Parker, M.; Pearson, G. R.; Parrillo,
J.; Mitchell, C.; Straus, S. E. J. Am. Med. Assoc. 1984, 252, 72-76.
(8) (a) Bedard, J.; May, S.; Barbeau, D.; Yuen, L.; Rando, R. F.; Bowlin,
T. L. AntiViral Res. 1999, 41, 35-43. (b) Cheraghali, A. M.; Kumar, R.;
Wang, L.; Knaus, E. E.; Wiebe, L. I. Biochem. Pharmacol. 1994, 47, 1615-
1625.
CMV is prevalent in most healthy adults in a latent state,
although it rarely reactivates and causes disease in immuno-
competent individuals.6 However, in cases of immuno-
suppression, as in AIDS patients and transplant recipients,
there can be dire consequences of morbidity and mortality.7
Since prostaglandins are immunomodulators, we investigated
(9) (a) Kline, J. N.; Hunninghake, G. M.; He, B.; Monick, M. M.;
Hunninghake, G. W. Exp. Lung Res. 1998, 24, 3-14. (b) Khyatti, M.;
Menezes, J. AntiViral Res. 1990, 14, 161-172. (c) Tanaka, J. J.; Ogura,
T.; Iida, H.; Sato, H.; Hatano, M. Virol. 1988, 163, 205-208.
(10) King, S. M. AntiViral Res. 1999, 40, 115-137.
Org. Lett., Vol. 1, No. 11, 1999
1861

We demonstrated a parallel-pool strategy as one approach
to the construction of prostanoid libraries. It is anticipated
that perhaps a 4-fold increase in overall diversity (i.e., up to
128 members) will be feasible with additional split-mix and
parallel operations. In this regard, mass spectrometry should
prove useful for examining the composition of larger
libraries.11 The advantages of the parallel-pool approach
versus individual compound syntheses will be fully appreci-
ated in the preparation of hundreds of compounds via several
libraries. Finally, from a lead compound as discovered herein,
a focused, second-generation library could provide even more
effective prostanoids as well as aid in the delineation of the
mechanism of action of these novel antiviral agents. Hence,
high-throughput screening of many new prostaglandin ana-
logues provides an excellent opportunity for drug discovery.
Acknowledgment. This work was supported by funding
from the Skaggs Institute for Chemical Biology and the
National Institutes of Health (K.D.J. (GM56154) and P.G.
(CA-66167 and AI-30627)). K.D.J. is an Arthur C. Cope
Scholar, P.G. is a Scholar of the Leukemia Society of
America, K.J.L. is a Fellow of the Korea Science and
Engineering Foundation (KOSEF), and A.A. is a Fellow of
the Universitywide AIDS Research Program. We thank Dr.
Peter Wirsching for collaboration and help in the preparation
of the manuscript.
Supporting Information Available: Experimental pro-
cedures, spectral data, and biological assay methods. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(11) (a) Walk, T. B.; Trautwein, A. W.; Richter, H.; Jung, G. Angew.
Chem. Int. Ed. 1999, 38, 1763-1765. (b) Schriemer, D. C.; Bundle, D. R.;
Li, L.; Hindsgaul, O. Angew. Chem. Int. Ed. 1998, 37, 3383-3387. (c)
Hughes, I. J. Med. Chem. 1998, 41, 3804-3811.
OL991130J
1862
Org. Lett., Vol. 1, No. 11, 1999