Synthesis of natural-product-like molecules with over eighty distinct scaffolds

Article abstract of DOI:10.1002/anie.200804486

(Chemical Equation Presented) Seeking scaffold diversity: A synthetic approach for the combinatorial variation of the scaffolds of small molecules is described. Using just six basic reaction types, compounds with 84 distinct scaffolds were prepared. The c

Full text of DOI:10.1002/anie.200804486

Communications
DOI: 10.1002/anie.200804486
Scaffold Diversity
Synthesis of Natural-Product-Like Molecules with Over
Eighty Distinct Scaffolds**
Daniel Morton, Stuart Leach, Christopher Cordier, Stuart Warriner, and
Adam Nelson*
Angewandte
Chemie
104
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 104 –109

Angewandte
Chemie
A key challenge in chemical biology is the design and
synthesis of compound libraries spanning large tracts of
biologically relevant chemical space.[1–3] Libraries with high
molecular shape diversity, a prerequisite for broad biological
activity,[2] are particularly valuable in phenotypic screens.
Despite this, organic chemistry is dominated by a remarkably
small number of molecular scaffolds: in a recent study[4] of
known cyclic organic molecules, 0.25% of the molecular
frameworks were found in 50% of the known compounds!
The uneven distribution of frameworks may reflect the way
that chemical space is generally explored, with chemists
tending to prepare compounds based on frameworks that are
already known.[4]
Diversity-oriented synthesis (DOS) involves the prepara-
tion of compound libraries with high substitutional, stereo-
chemical, and/or scaffold diversity.[5] Varying molecular scaf-
folds is challenging though ingenious innovations have
allowed the parallel synthesis of libraries based on multiple
(ca. 6–30) scaffolds.[6] Developing approaches in which the
ethos of DOS is retained—that is, that the synthesis is
deliberate and simultaneous—is, nonetheless, extremely
demanding. High-throughput screens of DOS libraries have,
however, yielded useful small molecule tools for chemical
genetic studies of cellular protein functions.[7]
Our approach to the combinatorial variation of molecular
scaffolds involved the attachment of pairs of unsaturated
building blocks to a fluorous-tagged linker (Scheme 1).[8] The
fluorous tag allowed the removal of excess reagents at each
stage by fluorous-solid phase extraction (F-SPE) alone.[9] We
used two general types of building blocks: “propagating” and
“capping” building blocks (Scheme 2). To increase the
structural diversity of the final compounds, we attached the
building blocks using combinations of temporary silaketal
tethers[10] and bonds that would remain as a vestige in the final
compounds. For example, the cyclopentene building block 9
could be attached to the linker (1 or 2) using either a
Fukuyama–Mitsunobu reaction[11] (!3) or a diisopropylsily-
lene tether (!4). Following deacetylation, a “capping”
building block, such as 15(Sil,All)!5 or 25!6, was attached
to yield metathesis substrates. Metathesis cascades were used
to “reprogram” the scaffolds of the metathesis substrates to
yield those of the final products (for example 5!7 or 6!8).
Each metathesis cascade was expected to initiate at the
terminal alkene[12] of the “capping” building block, leading to
the release of only cyclized products from the fluorous-tagged
linker. It was expected that the ability to vary the pairs of
building blocks used, together with the nature of the attach-
Scheme 1. Outline of our approach to the combinatorial variation of
the scaffolds of small molecules. The labels in parentheses define the
substituents: Ac=acetyl; Sil=diisopropylsilyl; All=allyl.
ment reactions, would yield a small molecule library with
extremely high scaffold diversity.
The building blocks were prepared on a multi-gram scale.
Chiral building blocks were generally prepared in enantio-
merically-enriched form, often using enzymatic desymmet-
risation[13] to induce asymmetry. We started by assessing the
proficiency of some propagating building blocks in simple
metathesis cascades (see Supporting Information). Diallyl-
ated derivatives of the building blocks 9, 12, 13, 14, 17, and 18
were metathesized, and a range of products with alternative
scaffolds were obtained. The most promising results were fed
into the design of our synthetic scheme.
The propagating building blocks were first attached to the
fluorous-tagged linker, with, in general, removal of excess
reagents by F-SPE alone (Table 1). Thus, the building blocks
9(H,Ac), 9’(H,Ac), (ꢀ )-10(H,Ac), 11(H,Ac), 12(H,Ac), 13-
(H,Ac), 15(H,Ac), and 16(H,Ac) were attached in good to
excellent yield to the fluorous-tagged sulfonamide 1 using a
Fukuyama–Mitsunobu reaction.[8] In addition, the building
blocks 9(Sil,Ac), 12(Sil,Ac), 13(Sil,Ac), 14(Sil,Ac), and 15-
(Sil,Ac) were activated and attached[14] via a diisopropylsilyl-
ene tether to the fluorous-tagged alcohol 2. Some of the
[*] Dr. D. Morton, Dr. S. Leach, Dr. C. Cordier, Dr. S. Warriner,
Prof. A. Nelson
School of Chemistry and Astbury Centre for Structural Molecular
Biology, University of Leeds, Leeds, LS2 9JT (UK)
Fax: (+44)113-343-6565
E-mail: a.s.nelson@leeds.ac.uk
Homepage: http://www.asn.leeds.ac.uk
[**] We thank EPSRC and the Wellcome Trust for funding.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200804486.
Angew. Chem. Int. Ed. 2009, 48, 104 –109
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
105

Communications
Table 1: Attachment of propagating building blocks and deacetylation.
Attachment of propagating building block Deacetylation
Linker Building
Mass recovery[b] Product[d] Mass recovery[b]
block[a]
(purity[c])
(purity[c])
1
2
1
1
1
1
2
1
2
2
1
2
1
9(H,Ac)
9(Sil,Ac)
9’(H,Ac)
(ꢀ)-10(H,Ac) >98% (90%)
>98% (98%)
85%[e] (>95%)
>98% (98%)
3
4
3’
>98% (95%)
>98% (93%)
>98% (95%)
(ꢀ)-31 >98% (90%)
11(H,Ac)
12(H,Ac)
12(Sil,Ac)
13(H,Ac)
13(Sil,Ac)
14(Sil,Ac)
15(H,Ac)
15(Sil,Ac)
16(H,Ac)
>98% (90%)
>98% (90%)
88%[e] (>95%)
>98% (89%)
61% (>95%)
58%[e] (>95%)
>98% (>95%)
95%[e] (90%)
96% (>95%)
32
33
34
35
36
37
38
39
40
>98% (90%)
>98% (90%)
>98% (95%)
>98% (90%)
>98% (86%)
>98% (94%)
>98% (>95%)
>98% (90%)
>98% (90%)
[a] Method for attachment to 1: 1) building block (4 equiv), DEAD
(4 equiv), PPh3 (4 equiv), THF, 08C, 1 h; 2) F-SPE; Method for attach-
ment to 2: 1) building block (5.5 equiv), NBS (5 equiv), CH2Cl2, 08C!
RT, 15 min; 2) inverse addition of fluorous-tagged linker, DMAP (50
mol%), Et3N (15 equiv), 08C!RT; 3) F-SPE; [b] Purification by F-SPE
only unless otherwise indicated. [c] Determined by analytical HPLC or
1
500 MHz H NMR spectroscopy. [d] Method: saturated NH3 in MeOH.
[e] Purified, in addition, by filtration through a pad of florisil.
required. The phosphine, P(CH2OH)3,[16] was used to
remove the catalyst G I. F-SPE was used to separate the
released metathesis products from possible fluorous-tagged
contaminants: remaining substrate, the remnant of the
linker,[8] unreleased by-products, and, for reactions catalysed
by f-HG II, the catalyst. The products derived from silaketal
substrates were treated with HF·pyridine and quenched with
Me3SiOMe. Finally, the released products were purified.
The linkers 1 and 2 were instrumental to the success of our
approach.[8] First, the fluorous tag allowed the assembly of the
metathesis substrates without column chromatography.
Second, linker design ensured that only metathesized prod-
ucts were released. Thus although yields varied widely, F-SPE
enabled the very easy isolation of just the released metathesis
products which were subsequently purified. The average yield
of the released products, based on the purity of the metathesis
substrates, was 46%.
Our synthetic approach yielded products with a diverse
range of scaffolds. In most cases, all of the unsaturated groups
in the substrate were involved in the dominant metathesis
pathway (for example, 41!42; Scheme 5). However, com-
petition between the formation of different ring sizes
extended the structural diversity possible. For example, the
alkene in the propagating building block of 43 did not
participate in the subsequent cascade, and thus a bridged
macrocycle (44) was formed. In four cases, the competition
between alternative cyclizations was sufficiently close that
products with two distinct scaffolds were isolated from the
same reaction. The metathesis substrate 45 has two more
methylene groups than 41, which had a key effect on its fate:
competition between seven- and thirteen-membered ring
formation was close, and, after desilylation, two products (46
and 47) were obtained.
Scheme 2. Structures of the building blocks used in our diversity-
oriented approach. Enantiomeric structures are distinguished thus: 9
and 9’.
silaketal formations did not proceed to completion: filtration
through a short pad of florisil allowed the unreacted fluorous-
tagged alcohol 2 to be removed. Each intermediate was
deacetylated in excellent yield by treatment with saturated
ammonia in methanol and the products were used without
purification (Table 1).
Next, “capping” building blocks were attached through
silaketal formation, esterification or a variant of the Mitsu-
nobu reaction. In general, F-SPE alone was used to purify
each of the resulting metathesis substrates. In total, 86
metathesis substrates were prepared; a wide range of selected
examples are shown in Scheme 3.
The metathesis substrates were each treated with a
metathesis catalyst in refluxing dichloromethane (see
Scheme 3 for selected examples, and Supporting Informa-
tion). In general, G I (Scheme 4) was used with substrates in
which one or both of the building blocks had been attached
using a silaketal tether;[14] with other substrates, the fluorous-
tagged catalyst[15] f-HG II was usually used. The reactions
were monitored, and additional catalyst was added as
106
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 104 –109

Angewandte
Chemie
Scheme 3. Selected examples of metathesis cascades leading to skeletally diverse products. Methods: Attachment of 21–26: 1) building block
(4 equiv), DEAD (4 equiv), PPh3 (4 equiv), THF, 08C, 1 h; 2) F-SPE. Attachment of 27: 1) building block (4 equiv), DEAD (4 equiv), PPh3 (4 equiv),
THF, 08C ! reflux; 2) F-SPE. Silaketal formation: 1) building block (5.5 equiv), NBS (5 equiv), CH2Cl2, 08C ! RT, 15 min; 2) inverse addition of
substrate (1 equiv, 0.2m), DMAP (50 mol%), Et3N (15 equiv), 08C ! RT; 3) F-SPE. Attachment of 28–30: 1) building block (2 equiv), EDC
(2 equiv), DMAP (10 mol%), CH2Cl2, RT; 2) F-SPE. G-I: 1) catalyst G I, CH2Cl2, 458C; 2) Et3N (86 equiv), P(CH2OH)3 (86 equiv), then silica, then
filter through Celite; 3) F-SPE; 4) HF·pyridine, THF, then Me3SiOMe. HG-II: 1) catalyst HG II, CH2Cl2, 458C; 2) Et3N (86 equiv), P(CH2OH)3
(86 equiv), then silica, then filter through Celite; 3) F-SPE. f-HG-II: 1) catalyst f-HG II, CH2Cl2, 458C; 2) F-SPE. DEAD=diethyl azodicarboxylate;
NBS=N-bromosuccinimide; DMAP=4-dimethylaminopyridine; EDC=N’-(3-dimethylaminopropyl)-N-ethylcarbodiimide.
The diversity of organic compounds has often been
analysed in a chemically intuitive manner in terms of the
ring systems in structures.[2,14,18] We have assessed the skeletal
diversity of our library in terms of an hierarchical approach
that formalises the relationship between molecular scaffolds
(Figure 1).[17] In this approach, terminal side chains are
pruned to give a scaffold which is defined by the rings and
unsaturated groups that rigidify the molecule. Rings are then
iteratively removed according to intuitive prioritisation rules
Scheme 4. Some catalysts for metathesis reactions.
Angew. Chem. Int. Ed. 2009, 48, 104 –109
to reveal the last remaining ring—the “parental” scaffold. The
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
107

Communications
Figure 1. Hierarchical classification which illustrates the relationship
between parental scaffolds (blue), daughter scaffolds (red), and
molecular scaffolds (black).
compounds with > 80 scaffolds were prepared using a small
number of optimised procedures.
Excitingly, the compounds prepared had many of the
broad structural features that are reminiscent of, and have
evolved to constrain, natural products: isolated, fused, and
spirocyclic ring systems, intramolecular hydrogen bonding
motifs, unsaturation, and dense substitution. In this sense,
many of the compounds prepared are natural product-like. As
with polyketide biosynthesis,[19] a wholly remarkable number
of scaffolds with these broad structural features was acces-
sible using, in combination, a relatively small repertoire of
fundamental reaction types.
The key to our approach was the extraordinary scope of
ring-closing metathesis,[20] which allowed the preparation of a
library of much higher skeletal diversity than previous DOS
approaches. Combinatorial variation of molecular scaffold
was possible through the assembly of alternative metathesis
substrates from pairs of simple building blocks. The range of
accessible scaffolds was further extended both by varying the
linkages between building blocks, and competition between
the formation of alternative rings. Many of the diverse
scaffolds prepared have scope for easy further diversification
which may allow the discovery of novel bioactive small
molecule tools.
Scheme 5. Fate of some metathesis cascade reactions.
“parental” scaffold is related to scaffolds at lower levels of
hierarchy until the molecular scaffold is ultimately reached.
The vast majority of the 96 products prepared have a
distinct molecular scaffold. However, the scaffold diversity of
the library is even more striking at higher levels of the
hierarchy. This scaffold tree, like that of natural products,[3] is
dominated by the three classes of cyclic compounds: aza-
cycles, oxacycles, and carbacycles. There are 25 parental
scaffolds in total, and all ring sizes from 5 to 15 (except 10) are
represented at this level. The parental scaffolds are related to
54 “daughter” scaffolds, which are ultimately related to the 84
molecular scaffolds in the library. At all levels of hierarchy,
the scaffolds represented had unprecedented diversity. Fur-
thermore, around 65% of the deprotected scaffolds are not
found in molecules that have been previously prepared.
Our synthetic approach allowed the systematic explora-
tion of chemical space defined by scores of different
molecular scaffolds. Each final compound was prepared
from the building blocks in either four or five steps. Column
chromatography was avoided at all intermediate stages,
facilitating the application of the approach in parallel
format. Remarkably, our approach used combinations of
just six reaction types: the Mitsunobu reaction, silaketal
formation, esterification, deacetylation, metathesis, and desil-
ylation. The ethos of DOS was firmly retained since
Experimental Section
Experimental procedures, characterisation data, and NMR spectra
for all novel compounds are provided in the Supporting Information.
Received: September 11, 2008
Published online: November 17, 2008
Keywords: cascade chemistry · diversity-oriented synthesis ·
.
metathesis · molecular scaffolds · natural products
108
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 104 –109

Angewandte
Chemie
[8] S. G. Leach, C. J. Cordier, D. Morton, G. J. McKiernan, S.
[1] M. A. Koch, L.-O. Wittenberg, S. Basu, D. A. Jeyaraj, E.
Gourzoulidou, K. Reinecke, A. Odermatt, H. Waldmann,
Proc. Natl. Acad. Sci. USA 2004, 101, 16721 – 16726.
[2] W. H. B. Sauer, M. K. Schwarz, J. Chem. Inf. Comput. Sci. 2003,
43, 987 – 1003.
[3] M. A. Koch, A. Schuffenhauer, M. Scheck, S. Wetzel, M.
Casaulta, A. Odermatt, P. Ertl, H. Waldmann, Proc. Natl.
Acad. Sci. USA 2005, 102, 17272 – 17277.
Warriner, A. Nelson, J. Org. Chem. 2008, 73, 2753 – 2759.
[9] a) D. P. Curran, Z. Luo, J. Am. Chem. Soc. 1999, 121, 9069 –
9072; b) J. A. Gladysz, D. P. Curran, I. T. Horvath, Handbook of
Fluorous Chemistry, Wiley, Weinheim, 2004.
[10] M. Bols, T. Skrydstrup, Chem. Rev. 1995, 95, 1253 – 1277.
[11] T. Fukuyama, C.-K. Jow, M. Cheung, Tetrahedron Lett. 1995, 36,
6373 – 6374.
[12] M. Ulman, R. H. Grubbs, Organometallics 1998, 17, 2484 – 2489.
[13] E. Garcia-Urdiales, I. Alfonso, V. Gotor, Chem. Rev. 2005, 105,
313 – 354.
[14] C. Cordier, D. Morton, S. Leach, T. Woodhall, C. OꢀLeary-
Steele, S. Warriner, A. Nelson, Org. Biomol. Chem. 2008, 6,
1734 – 1737.
[15] M. Matsugi, D. P. Curran, J. Org. Chem. 2005, 70, 1636 – 1642.
[16] H. D. Maynard, R. H. Grubbs, Tetrahedron Lett. 1999, 40, 4137 –
4140.
[17] A. Schuffenhauer, P. Ertl, S. Roggo, S. Wetzl, M. A. Koch, H.
Waldmann, J. Chem. Inf. Model. 2007, 47, 47 – 58.
[18] a) R. Nilakantan, N. Bauman, K. S. Haraki, J. Comput.-Aided
Mol. Des. 1997, 11, 447 – 452; b) M.-L. Lee, G. Schneider, J.
Comb. Chem. 2001, 3, 284 – 289; c) A. H. Lipkus, J. Chem. Inf.
Comput. Sci. 2001, 41, 430 – 438; d) X. Q. Lewell, A. C. Jones,
C. L. Bruce, G. Harper, M. M. Jones, I. M. McLay, J. Bradshaw, J.
Med. Chem. 2003, 46, 3257 – 3274; e) R. Kho, J. A. Hodges,
M. R. Hansen, H. O. Villar, J. Med. Chem. 2005, 48, 6671 – 6678;
f) E.-W. Lameijer, J. N. Kok, T. Bꢁck, A. P. Ijerman, J. Chem. Inf.
Model. 2006, 46, 553 – 562; g) P. Ertl, S. Jelfs, J. Mꢂhlbacher, A.
Schuffenhauer, P. Selzer, J. Med. Chem. 2006, 49, 4568 – 4573.
[19] J. Staunton, K. J. Weissman, Nat. Prod. Rep. 2001, 18, 380 – 416.
[20] A. Deiters, S. F. Martin, Chem. Rev. 2004, 104, 2199 – 2238.
[4] A. H. Lipkus, Q. Yuan, K. A. Lucas, S. A. Funk, W. F. Bartelt III,
R. J. Schenk, A. J. Trippe, J. Org. Chem. 2008, 73, 4443 – 4451.
[5] a) M. D. Burke, S. L. Schreiber, Angew. Chem. 2004, 116, 48 – 60;
Angew. Chem. Int. Ed. 2004, 43, 46 – 58; b) D. S. Tan, Nat. Chem.
Biol. 2005, 1, 74 – 84; c) C. Cordier, D. Morton, S. Murrison, A.
Nelson, C. OꢀLeary-Steele, Nat. Prod. Rep. 2008, 25, 719 – 737;
d) T. E. Nielsen, S. L. Schreiber, Angew. Chem. 2008, 120, 52 –
61; Angew. Chem. Int. Ed. 2008, 47, 48 – 56; e) R. J. Spandl, A.
Bender, D. R. Spring, Org. Biomol. Chem. 2008, 6, 1149 – 1158.
[6] a) M. D. Burke, E. M. Berger, S. L. Schreiber, J. Am. Chem. Soc.
2004, 126, 14095 – 14104; b) H. Oguri, S. L. Schreiber, Org. Lett.
2005, 7, 47 – 50; c) D. A. Spiegel, F. C. Schroeder, J. R. Duvall,
S. L. Schreiber, J. Am. Chem. Soc. 2006, 128, 14766 – 14767;
d) G. L. Thomas, R. J. Spandl, F. G. Glansdorp, M. Welch, A.
Bender, J. Cockfield, J. A. Lindsay, C. Bryant, D. F. J. Brown, O.
Loiseleur, H. Rudyk, M. Ladlow, D. R. Spring, Angew. Chem.
2008, 120, 2850 – 2854; Angew. Chem. Int. Ed. 2008, 47, 2808 –
2812; e) N. Kumagai, G. Muncipinto, S. L. Schreiber, Angew.
Chem. 2006, 118, 3717 – 3720; Angew. Chem. Int. Ed. 2006, 45,
3635 – 3638; f) E. E. Wyatt, S. Fergus, W. R. J. D. Galloway, A.
Bender, D. J. Fox, A. T. Plowright, A. S. Jessiman, M. Welch,
D. R. Spring, Chem. Commun. 2006, 3296 – 3298.
[7] D. P. Walsh, Y.-T. Chang, Chem. Rev. 2006, 106, 2476 – 2530.
Angew. Chem. Int. Ed. 2009, 48, 104 –109
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
109