A two-directional strategy for the diversity-oriented synthesis of macrocyclic scaffolds

Article abstract of DOI:10.1039/c2ob26272j

Macrocyclic compounds represent a structural class with exceptional potential for biological activity; however, they have historically been underrepresented in screening collections and synthetic libraries. In this article we report the development of a highly step-efficient strategy for the diversity-oriented synthesis of complex macrocyclic architectures, using a modular approach based on the two-directional synthesis of bifunctional linear precursors and their subsequent combination in a two-directional macrocyclisation process. In this proof of principle study, the synthesis of 14 such compounds was achieved. Cheminformatic analysis of the compounds produced suggests that they reside in biologically relevant regions of chemical space and the compounds were screened for activity against two cancer cell lines.

Full text of DOI:10.1039/c2ob26272j

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 7545
www.rsc.org/obc
PAPER
A two-directional strategy for the diversity-oriented synthesis of macrocyclic
scaffolds†
Kieron M. G. O’Connell,a Henning S. G. Beckmann,a Luca Laraia,a,b Helen T. Horsley,c Andreas Bender,a
Ashok R. Venkitaramanb and David R. Spring*a
Received 4th July 2012, Accepted 3rd August 2012
DOI: 10.1039/c2ob26272j
Macrocyclic compounds represent a structural class with exceptional potential for biological activity;
however, they have historically been underrepresented in screening collections and synthetic libraries. In
this article we report the development of a highly step-efficient strategy for the diversity-oriented
synthesis of complex macrocyclic architectures, using a modular approach based on the two-directional
synthesis of bifunctional linear precursors and their subsequent combination in a two-directional
macrocyclisation process. In this proof of principle study, the synthesis of 14 such compounds was
achieved. Cheminformatic analysis of the compounds produced suggests that they reside in biologically
relevant regions of chemical space and the compounds were screened for activity against two cancer cell
lines.
therefore a high degree of structural diversity within a library
leads to an increased likelihood of achieving broad ranging bio-
Introduction
Diversity-oriented synthesis (DOS) aims to produce compound
collections that incorporate high degrees of structural diversity.1
These synthetic campaigns can provide a source of compounds
for use in screening experiments that attempt to discover novel
biologically active entities.2 The libraries produced by DOS find
particular utility in unbiased screening experiments, such as
phenotypic assays, where the precise nature of the biological
target is undefined. Experiments of this type give the potential for
the discovery of novel biological targets or modes of action, as
well as novel agents for targets that are already well understood.3
The potential for a given molecule to interact with biological
systems, and so exhibit biological activity, is intrinsically related
to the structure of that molecule.4 The macromolecules that make
up living systems represent 3D arrangements of functional
groups and potential binding regions that can interplay with
complementary regions and functionality displayed around small
molecules. The positions and orientations of these groups around
a small molecule determine the nature of the binding interactions
that molecule is capable of achieving. Therefore, the molecular
structure, and consequentially the molecular shape, of a small
molecule are fundamental to determining the biological activity
of that molecule.5 The presence of multiple structural classes and
logical activity across that library.
There are a number of ways that structural diversity can be
incorporated into a compound collection; however, variation of
the molecular scaffold is widely thought to be the most impor-
tant. The scaffold (or skeleton) of a molecule is usually con-
sidered to be the combination of rigidifying elements, such as
ring architectures, that comprise the core structure of the mo-
lecule.6 As such, varying the molecular scaffold has a far greater
influence on the overall shape of molecules than altering more
peripheral structural characteristics such as the attached func-
tional groups. Incorporating a high degree of scaffold diversity
into a library allows that library to span a comparatively large
area of chemical space.
Compounds containing macrocyclic ring structures (a ring
size of 12 atoms or above) are capable of extremely potent bio-
logical activity and specificity. This can be clearly illustrated by
the exquisite biological activity of many macrocyclic natural pro-
ducts that has been harnessed by medicinal chemists to provide
chemotherapy for a broad range of conditions. Currently macro-
cyclic structures account for over 100 approved drugs covering
examples of antibiotic, immunosuppressant and anticancer
chemotherapeutics.7
This biological activity has been attributed to the ability of
macrocyclic compounds to represent a compromise between
structural pre-organization and conformational flexibility.8 Their
cyclic structure means they have less conformational freedom
than an equivalent acyclic compound and so suffer a corres-
pondingly smaller entropic penalty upon binding to a receptor.
However, unlike smaller cyclic systems, they are not completely
aDepartment of Chemistry, University of Cambridge, Cambridge, UK
CB2 1EW. E-mail: spring@ch.cam.ac.uk
bUniversity of Cambridge, Medical Research Council Cancer Cell Unit,
Hutchison/MRC Research Unit, Hills Road, Cambridge, UK CB2 0XZ
cUCB Celltech, 208 Bath Rd., Slough, UK SL1 3WE
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c2ob26272j
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7545–7551 | 7545

View Online
Fig. 1 Two-directional strategy for macrocyclisation used in the DOS.
rigid and so potentially have the ability to mould to a surface to
achieve optimal binding. In addition to this, the comparatively
large size of macrocyclic compounds gives the possibility for
binding across extended surfaces that are difficult to access with
more traditional small molecules.9 For these reasons macrocycles
can be thought to occupy something of a middle ground between
(rule of 5 compliant) small molecules and biological treatments.
They can be capable of the desirable pharmacokinetic properties
associated with small molecules in terms of oral bioavailability
and cell permeability, but also allow for the possibility of modu-
lating difficult targets usually attenuated by biological treatments.
However, despite this unique activity, they are widely thought to
be underrepresented in screening collections.8,10 This apparent
deficiency is normally attributed to the synthetic difficulties
associated with the synthesis of macrocycles: the necessity for
high dilution conditions, and the substrate dependant nature of
the macrocyclisation step, being poorly compatible with library
synthesis. However, this is beginning to change and recent years
have seen several DOS campaigns targeted at macrocyclic
structures.11
Herein, we report a proof of principle study on the develop-
ment of a novel strategy for the DOS of complex macrocyclic
compounds. This strategy involves the initial two-directional
synthesis of symmetrical ‘linear’ precursors bearing reactive
functionality at two positions.12 Linear precursors containing
mutually complementary functionality are then able to react
together in a two-directional macrocyclisation process to give the
desired molecular architectures.13 This strategy allows a com-
plexity generating reaction to be performed at two positions on a
linear species while concurrently creating a macrocyclic scaffold,
allowing an extremely broad scope for the degree of complexity
that can be produced in a single synthetic step. The nature of the
strategy also means that the linker units between the reacting
groups can be regarded as ‘scaffold elements’ with regards to the
eventual macrocyclic structure. Combining appropriately functio-
nalised scaffold elements is then able to lead directly to distinct
macrocyclic ring sizes and scaffolds. This strategy has the poten-
tial to provide extremely efficient access to a range of
macrocyclic scaffolds because the combinatorial variation of
scaffold elements can lead to the generation of a unique scaffold
from each macrocyclisation process. Fig. 1 shows an overview
of the two-directional strategy.
Results and discussion
The two reactions utilized to accomplish the two-directional
macrocyclisation process were: the Diels–Alder reaction and the
copper-catalysed azide–alkyne cycloaddition (CuAAC).14 The
Diels–Alder reaction is
a complexity-generating reaction,
capable of creating a cyclohexene ring and up to four stereo-
centres. The CuAAC was chosen for its broad substrate tolerance
and because the resulting triazoles are regarded as biologically
relevant species with the ability to act as peptidomimetic struc-
tural elements.15
In order to create a convergent strategy for the macrocyclisa-
tions, one of the linear precursors was based on bis-enyne
amides (1), which are easily accessible from the corresponding,
commercially available, dicarboxylic acids (scaffold element 1).
These bis-enyne amides are suitable for use immediately in
the CuAAC reaction and can be transformed into the requisite
conjugated 1,3-dienes (2) for the Diels–Alder reactions by ring-
closing enyne metathesis (RCEYM). The complementary react-
ing partners: bis-maleimides (3) and bis-azides (4) were then
obtained from commercially available diamines (scaffold
element 2). Maleimides were chosen as dienophiles due to their
high reactivity and for the plane of symmetry through the alkene,
which circumvents this regioselectivity issue. The Diels–Alder
reaction between bis-dienes 2 and bis-maleimides 3 then gave
macrocycles of type 5, all of which were obtained as racemates.
The bis-azide species used were bis-amides, synthesised from
the linear diamines and azido-acids (6). The direct synthesis of
bis-azides from the corresponding diamines was avoided as these
species would have unacceptably high nitrogen to carbon
ratios.16 The use of azido-acids to ‘cap’ the diamines and
provide the desired azide functionality, also introduced a third
7546 | Org. Biomol. Chem., 2012, 10, 7545–7551
This journal is © The Royal Society of Chemistry 2012

View Online
scaffold element (scaffold element 3) into the linear precursors.
The reaction between bis-enyne amides 1 and bis-azides 4, gave
macrocycles of type 7. It would also have been possible to add a
fourth scaffold element to the strategy by varying the choice of
amine used to synthesise 1; however, for this proof of principle
study, work focused solely on the use of allylpropargylamine.
Both of the two-directional macrocyclisations chosen left alkene
groups in the final macrocycles, which could potentially be used
as handles for further reactivity.
The linear precursors were all obtained in one or two steps
from commercially available materials; three bis-enyne amides
(1a–c) were synthesised from the corresponding dicarboxylic
acids by coupling with allylpropargylamine using EDC·HCl and
DMAP mediated conditions (Scheme 1). Treatment of 1a–c with
10 mol% Grubbs’ first generation catalyst under an ethylene
atmosphere then furnished the desired bis-dienes (2a–c).17 Three
bis-maleimides (3a–c) were obtained by treatment of the linear
diamines with N-methoxycarbonyl maleimide.18 Azido-acids 6a
and 6b were synthesised using the diazo-transfer agent developed
by Stick et al.,19 then coupled to two diamines: 1,5-diamino-
pentane and p-xylenediamine (scaffold element 2a and b) to give
four bis-azides. When 1,5-diaminopentane was used, coupling
was achieved using HATU to facilitate the reaction, giving 4ai
and 4aii. When p-xylenediamine was used, all peptide-coupling
reagents proved ineffective, and it was necessary to first convert
the azido-acids to the corresponding acid chlorides with oxalyl
chloride to achieve the synthesis of 4bi and 4bii.
dichloroethane, which was chosen as the solvent because it pro-
vided good solubility properties for both the bis-dienes and bis-
maleimides. It was found to be necessary to heat the reactions to
120 °C to achieve macrocyclisation, and the reactions were
carried out in sealed tubes with heating for 18 hours, at concen-
trations of around 10 mM. Under these conditions six macro-
cyclisations were achieved. For four examples (compounds
8–11), a single diastereomer of the macrocyclic product was iso-
lated; for the other two examples (12 and 13), two diastereomers
were isolated (Scheme 2).20 The products were obtained in un-
optimised isolated yields (11–33%) after flash chromatography
and HPLC purification to remove starting materials and non-
cyclised materials.
In the cases where a single diastereomer was obtained, it was
always found to be the result of the Diels–Alder reaction occur-
ring in the endo orientation at both reaction sites. This was deter-
mined by NOESY spectroscopy. For the two examples where
two diastereomeric products were obtained, the additional
product was the result of one Diels–Alder process proceeding
via an endo-transition state and the other via an exo-transition
state.
NMR experiments were unable to determine whether the
Diels–Alder reactions took place on the same face or the oppo-
site face of the dienes, making assignment of relative stereo-
chemistry between the two 5–6–5 ring systems difficult.21
However, for compound 8 an X-ray crystal structure was
obtained (Scheme 3) which confirmed the stereochemistry at
each of the stereocentres and so showed that, in this case, the
reactions had occurred on opposite faces of the bis-diene.22 In
the absence of further information, the rest of the Diels–Alder
products (9–13) are drawn with analogous stereochemistry;
The two-directional Diels–Alder macrocyclisation proved suc-
cessful; creating extremely complex molecular architectures con-
taining six stereocentres in a single transformation from ‘flat’
achiral precursors. The reactions were carried out in 1,2-
Scheme 1 Preparation of linear precursors.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7545–7551 | 7547

View Online
Scheme 2 Two-directional Diels–Alder macrocyclisation products; diastereomeric pairs are outlined. Compounds 9–13 are drawn with analogous
relative stereochemistry to compound 8; however, for these compounds it is not certain which endo–endo or endo–exo isomer was isolated.
however, for these cases either of the endo–endo products could
be present.
ineffective, possibly due to the low solubility of the substrates in
the BuOH/H2O mixtures that the reactions are usually per-
t
Compound 8 was also used to illustrate the potential for the
further elaboration of these compounds. In order to provide
enough material for experimentation, the synthesis of this com-
pound was scaled up to 1.6 mmol scale to provide 270 mg of the
Diels–Alder product.23 The alkene groups resulting from the
Diels–Alder cyclisations give the potential for a multitude of
synthetic transformations to be performed after the macrocyclisa-
tion step; for example, hydrogenation and dihydroxylation
(Scheme 3). Hydrogenation with hydrogen gas and Pd/C pro-
ceeded smoothly to give fully saturated compound 14 in 50%
yield. The hydrogenation of alkenes within cyclic systems can
cause large changes in the shape of a molecule and so alter the
biological interactions it is capable of achieving.24 In this case, it
also appeared to affect the conformational flexibility of the
macrocycle with NMR suggesting that 14 exists as a number
of interconverting conformers at ambient temperature. The di-
hydroxylation of 8 proved more problematic; due, in the most
part, to difficulties in separating and isolating the products of the
reaction. After some optimisation, a 23% yield of 15 was
achieved using eight equivalents of NMO and HPLC purification
of the crude reaction mixture.
formed in. Using CuI and DIPEA in THF provided a viable
system for the macrocyclisations; however, the solubility of the
substrates again proved troublesome and so the reactions were
only performed with the most soluble bis-enyne, which was
found to be 1b. The reactions were carried out at 20–30 mM
concentration at 80 °C in sealed tubes and gave the desired pro-
ducts 16–19 in unoptimised 7–23% isolated yields (Scheme 4).25
Between the Diels–Alder and CuAAC macroyclisations 14
complex macrocyclic products were produced, providing a clear
indication of the validity of this approach to DOS. The com-
pounds were produced in three to five synthetic steps from com-
mercially available materials, and in most cases, a single step
from simple, easily accessible, linear precursors. This approach
therefore represents an effective strategy for the rapid generation
of molecular complexity. Nine macrocyclic ring sizes are repres-
ented within the compound collection: 21, 22, 23, 24, 27, 28,
30, 31 and 32; indicating coverage of a relatively large area of
macrocyclic chemical space.
While, to some degree, the presence of this range of ring sizes
is, in itself, indicative of the diversity of the compounds, we
sought to further explore and quantify the diversity and chemical
space coverage achieved by these compounds using compu-
tational analysis. The diversity of the compounds synthesized
was evaluated by calculating their 2D molecular descriptors
A range of conditions were trialled for the CuAAC macro-
cyclisations, including both CuI and CuSO4/ascorbic acid based
protocols. The CuSO4/ascorbic acid based systems proved
7548 | Org. Biomol. Chem., 2012, 10, 7545–7551
This journal is © The Royal Society of Chemistry 2012

View Online
(as implemented in MOE),26 followed by principal component
analysis (PCA) of the values obtained.27 Also included in the
analysis was a collection of 656 drugs (red dots) randomly
selected from the Drugbank28 database and the 95 compounds in
Drugbank that contain a macrocyclic ring (and molecular weight
< 1000 Da) (blue triangles). The inclusion of these additional
compound sets in the analysis allows the comparison of the
chemical space covered by the collections and also gives some
context to the positions of the library compounds in chemical
space with respect to known biologically active regions. Fig. 2
shows a scatter plot of the three compound collections.29 The
first thing that is apparent from this scatter plot is that the library
compounds (black dots) cover a relatively large area of chemical
space given the small number of compounds synthesised,
suggesting the successful incorporation of a high degree of
molecular diversity into the synthesis.
This is particularly impressive given the highly step-economi-
cal nature of the synthesis. In terms of the positions of the
library compounds with respect to the other compound collec-
tions it appears that the library compounds occupy an area
removed from the majority of the randomly selected drug com-
pounds and from the two main areas of concentration of the
macrocyclic drug compounds. However, there is clearly still sig-
nificant overlap with both compound collections suggesting that
the compounds produced in this campaign do have potentially
biologically relevant physical properties.
This computational assessment is useful, in terms of providing
a visually accessible representation of the abstract concepts of
molecular diversity and chemical space; however, a more
Fig. 2 Computational assessment of the position of the macrocycles in
chemical space calculated using 2D molecular descriptors and principle
component analysis.
Scheme 3 X-ray crystal structure of 8 proving the relative stereo-
chemistry; post-macrocyclisation modifications of 8.
Scheme 4 Two-directional CuAAC macrocyclisation.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7545–7551 | 7549

View Online
A. Nelson and C. O’Leary-Steele, Nat. Prod. Rep., 2008, 25, 719–737;
(d) W. R. J. D. Galloway, A. Isidro-Llobet and D. R. Spring, Nat.
Commun., 2010, 1, 80; (e) S. Dandapani and L. A. Marcaurelle, Curr.
Opin. Chem. Biol., 2010, 14, 362–370; (f) C. J. O’ Connor,
H. S. G. Beckmann and D. R. Spring, Chem. Soc. Rev., 2012, 41, 4444–
4456.
stringent test of the usefulness, and functional diversity, of a com-
pound collection such as this is the biological evaluation of the
compounds produced. With this in mind, the library members were
tested for antiproliferative activity against two cancer cell lines.
All of the macrocycles synthesised were tested for their ability
to inhibit the proliferation of an osteosarcoma cell line (U2OS)
and a human lung adenocarcinoma epithelial cell line (A549)
using a sulforhodamine B (SRB) assay (see ESI†). Two of the
library compounds were then found to inhibit the growth of one
or both cell lines. Compound 9 was found to inhibit the prolifer-
ation of the U2OS cell line with an IC50 of 141 μM, and com-
pound 19 was found to inhibit the proliferation of both the cell
lines with IC50 values of 131 μM (U2OS) and 125 μM (A549).30
While the IC50 values obtained for compounds 9 and 19
represent only moderate potency, it should be stressed that these
compounds are ‘starting points’ for biologically active scaffolds
that can be optimised with various appendages. As such, the dis-
covery of two hits from a library of 14 compounds represents a
successful screening campaign. Additionally, both compounds
are based around novel macrocyclic scaffolds, implying the dis-
covery of new areas of bioactive chemical space. The bioactivity
of these compounds also provides further support for the conten-
tion that varying the molecular scaffolds across a library is
important in terms of achieving functional diversity. As a result
of the synthetic strategy employed, both compounds have essen-
tially all of their functional groups in common with several other
library members suggesting that it must be the molecular scaf-
fold that is leading to the activity observed for these compounds.
2 Examples of the discovery of novel biologically active species using
DOS include: (a) G. L. Thomas, R. J. Spandl, F. G. Glansdorp,
M. Welch, A. Bender, J. Cockfield, J. A. Lindsay, C. Bryant,
D. F. Brown, O. Loiseleur, H. Rudyk, M. Ladlow and D. R. Spring,
Angew. Chem., Int. Ed., 2008, 47, 2808–2812; (b) E. E. Wyatt,
W. R. J. D. Galloway, G. L. Thomas, M. Welch, O. Loiseleur,
A. T. Plowright and D. R. Spring, Chem. Commun., 2008, 4962–4964;
(c) B. Z. Stanton, L. F. Peng, N. Maloof, K. Nakai, X. Wang,
J. L. Duffner, K. M. Taveras, J. M. Hyman, S. W. Lee, A. N. Koehler,
J. K. Chen, J. L. Fox, A. Mandinova and S. L. Schreiber, Nat. Chem.
Biol., 2009, 5, 154–156; (d) D. M. Huryn, J. L. Brodsky,
K. M. Brummond, P. G. Chambers, B. Eyer, A. W. Ireland,
M. Kawasumi, M. G. LaPorte, K. Lloyd, B. Manteau, P. Nghiem,
B. Quade, S. P. Seguin and P. Wipf, Proc. Natl. Acad. Sci. U. S. A., 2011,
108, 6757–6762; (e) R. W. Heidebrecht, C. Mulrooney, C. P. Austin,
R. H. Barker, J. A. Beaudoin, K. C.-C. Cheng, E. Comer, S. Dandapani,
J. Dick, J. R. Duvall, E. H. Ekland, D. A. Fidock, M. E. Fitzgerald,
M. Foley, R. Guha, P. Hinkson, M. Kramer, A. K. Lukens, D. Masi,
L. A. Marcaurelle, X.-Z. Su, C. J. Thomas, M. Weïwer, R. C. Wiegand,
D. Wirth, M. Xia, J. Yuan, J. Zhao, M. Palmer, B. Munoz and
S. Schreiber, ACS Med. Chem. Lett., 2012, 3, 112–117.
3 S. L. Schreiber, Science, 2000, 287, 1964–1969.
4 (a) W. R. J. D. Galloway, A. Bender, M. Welch and D. R. Spring, Chem.
Commun., 2009, 2446–2462; (b) M. D. Burke, E. M. Berger and
S. L. Schreiber, Science, 2003, 302, 613–618.
5 W. H. B. Sauer and M. K. Schwarz, J. Chem. Inf. Comput. Sci., 2003, 43,
987–1003.
6 T. E. Neilsen and S. L. Schreiber, Angew. Chem., Int. Ed., 2008, 47,
48–56.
7 E. M. Driggers, S. P. Hale, J. Lee and N. K. Terrett, Nat. Rev. Drug Dis-
covery, 2008, 7, 608–624.
8 L. A. Wessjohann, E. Ruijter, D. Garcia-Rivera and W. Brandt, Mol.
Diversity, 2005, 9, 171–186.
9 (a) N. K. Terrett, Drug Discovery Today: Technol., 2010, 7, e97–e104;
(b) E. Marsault and M. L. Peterson, J. Med. Chem., 2011, 54,
1961–2004.
Conclusions
In summary, we have reported a strategy for the DOS of
complex macrocyclic compounds using a two-directional syn-
thesis approach involving the modular synthesis of linear precur-
sors and their subsequent combination to form macrocyclic
architectures. The approach was used to rapidly synthesise
14 macrocyclic compounds including examples of nine ring
sizes, in three to five synthetic steps from commercially available
materials. Cheminformatic analysis of the compounds indicates
that they occupy biologically relevant positions in chemical
space, and two of the 14 compounds were found to exhibit anti-
proliferative activity against cancer cell lines, further implying
potential biological utility. Studies are ongoing to test these com-
pounds against a wider range of biological targets, and to apply this
synthetic strategy to the synthesis of a larger library of compounds.
10 C. M. Madsen and M. H. Clausen, Eur. J. Org. Chem., 2011,
3107–3115.
11 For examples see: (a) L. A. Wessjohann, B. Voigt and D. G. Rivera,
Angew. Chem., Int. Ed., 2005, 44, 4785–4790; (b) L. A. Marcaurelle,
E. Comer, S. Dandapani, J. R. Duvall, B. Gerard, S. Kesavan, M. D. Lee,
H. Liu, J. T. Lowe, J.-C. Marie, C. A. Mulrooney, B. A. Pandya,
A. Rowley, T. D. Ryba, B.-C. Suh, J. Wei, D. W. Young, L. B. Akella,
N. T. Ross, Y.-L. Zhang, D. M. Fass, S. A. Reis, W.-N. Zhao,
S. J. Haggarty, M. Palmer and M. A. Foley, J. Am. Chem. Soc., 2010,
132, 16962–16976; (c) A. Isidro-Llobet, T. Murillo, P. Bello,
A. Cilibrizzi, J. T. Hodgkinson, W. R. J. D. Galloway, A. Bender,
M. Welch and D. R. Spring, Proc. Natl. Acad. Sci. U. S. A., 2011, 108,
6793–6798; (d) F. Kopp, C. F. Stratton, L. B. Akella and D. S. Tan, Nat.
Chem. Biol., 2012, 8, 358–365.
12 For
a review on two-directional synthesis see: C. S. Poss and
S. L. Schreiber, Acc. Chem. Res., 1994, 27, 9–17; recent examples of
two-directional synthesis include: (a) M. Rejzek, R. A. Stockman and
D. L. Hughes, Org. Biomol. Chem., 2005, 3, 73–83; (b) M.
S. Karatholuvhu, A. Sinclair, A. F. Newton, M.-L. Alcaraz,
R. A. Stockman and P. L. Fuchs, J. Am. Chem. Soc., 2006, 128, 12656–
12657; (c) M. Diaz-Gavilan, W. R. J. D. Galloway, K. M. G. O’Connell,
J. T. Hodkingson and D. R. Spring, Chem. Commun., 2010, 46, 776–778;
(d) D. Robbins, A. F. Newton, C. Gignoux, J.-C. Legeay, A. Sinclair,
M. Rejzek, C. A. Laxon, S. K. Yalamanchili, W. Lewis, M. A. O’Connell
and R. A. Stockman, Chem. Sci., 2011, 2, 2232–2235; (e) P. Aggarwal,
G. Procopiou, D. Robbins, G. Harbottle, W. Lewis and R. A. Stockman,
Synlett, 2012, 423–427.
Acknowledgements
The authors would like to thank UCB Celltech, the EU, EPSRC,
BBSRC, MRC, Cancer Research UK, Wellcome Trust and the
Frances and Augustus Newman foundation for funding. H.S.G.
B. acknowledges a fellowship within the Postdoc-Program of the
German Academic Exchange Service (DAAD).
13 D. G. Rivera, O. E. Vercillo and L. A. Wessjohann, Org. Biomol. Chem.,
2008, 6, 1787–1795.
14 (a) H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed.,
2001, 40, 2004–2021; (b) J. E. Moses and A. D. Moorhouse, Chem. Soc.
Rev., 2007, 36, 1249–1262.
Notes and references
1 For review articles on DOS see: (a) D. R. Spring, Org. Biomol. Chem.,
2003, 1, 3867–3870; (b) M. D. Burke and S. L. Schreiber, Angew. Chem.,
Int. Ed., 2004, 43, 46–58; (c) C. Cordier, D. Morton, S. Murrison,
15 D. S. Pedersen and A. Abell, Eur. J. Org. Chem., 2011, 2399–2411.
7550 | Org. Biomol. Chem., 2012, 10, 7545–7551
This journal is © The Royal Society of Chemistry 2012

View Online
16 S. Bräse, C. Gil, K. Knepper and V. Zimmermann, Angew. Chem., Int.
Ed., 2005, 44, 5188–5240.
17 R. J. Spandl, H. Rudyk and D. R. Spring, Chem. Commun., 2008,
3001–3003.
18 (a) L.-X. Wang, J. Ni and S. Singh, Bioorg. Med. Chem., 2003, 11, 159–
166; (b) P. De, M. Li, S. R. Gondi and B. S. Sumerlin, J. Am. Chem.
Soc., 2008, 130, 11288–11289.
24 A. W. Hung, A. Ramek, Y. Wang, T. Kaya, J. A. Wilson, P. A. Clemons
and D. W. Young, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 6799–6804.
25 Attempts were then made to apply the alternative ruthenium-catalysed
azide–alkyne cycloaddition to this macrocylisation. However, the reaction
did not prove suitable with no productive reaction occurring in any case.
The reason for this intractability is not clear but it is possible that the
ruthenium can become complexed between the alkyne and alkene groups
of the enyne amides, leading to catalytically inactive species as observed
for 1,5-diynes in B. C. Boren, S. Narayan, L. K. Rasmussen, L. Zhang,
H. Zhao, Z. Lin, G. Lia and V. V. Fokin, J. Am. Chem. Soc., 2008, 130,
8923–2930.
19 E. D. Goddard-Borger and R. V. Stick, Org. Lett., 2007, 9, 3797–3800.
20 In all of the Diels–Alder cyclisations there was evidence for the formation
of additional cyclic and acyclic products by LCMS and HPLC but only
the compounds described were isolated in any significant quantity.
21 An additional difficulty in determining the structures of the compounds
by NMR was that a number of the products proved conformationally
dynamic on the NMR timescale resulting in complex spectra. A number
of VT experiments were run; however, they did little to improve the
spectra and so for these Diels–Alder reactions the structures were
assigned based on the complex room temperature spectra.
22 Crystal data for 8: Molecular formula C36H32N4O6·2CHCl3, M = 855.39,
monoclinic, a = 15.1710(2) Å, b = 17.9068(3) Å, c = 140740(2) Å, α =
90°, β = 90.417(1)°, γ = 90°, V = 3823.3(1) Å3, T = 180(2) K,
space group P2(1)/c, Z = 4, 38 105 reflections measured, 8725 indepen-
dent reflections (Rint = 0.0688). The final R1 values were 0.0559
(I > 2σ(I)). The final wR(F2) values were 0.1152 (I > 2σ(I)). The final R1
values were 0.0956 (all data). The final wR(F2) values were 0.1322 (all
data).
26 Molecular Operating Environment (MOE), 2011.10; Chemical Comput-
ing Group Inc., 1010 Sherbooke St. West, Suite #910, Montreal, QC,
Canada, H3A 2R7, 2011. Further details of this analysis are given in the
accompanying ESI.†
27 M. Ringner, Nat. Biotechnol., 2008, 26, 303–304.
28 D. S. Wishart, C. Knox, A. C. Guo, D. Cheng, S. Shrivastava, D. Tzur,
B. Gautam and M. Hassanali, Nucleic Acids Res., 2008, 36, D901–D906.
29 It should be noted that there are only 12 ‘black dots’ on the plot to rep-
resent the 14 library compounds, this is because the 2D molecular
descriptors chosen do not distinguish stereoisomers (or any other 3D
information) and so the two diastereomeric pairs (12a and 12b and 13a
and 13b) represent a single point each.
30 At present the mode of action of these active compounds is unknown.
The compounds were screened in an assay for their ability to arrest cells
in mitosis; however, this assay suggested that none of the compounds
have this ability and that 19 appears to induce apoptosis. See ESI† for
further information.
23 This scale represents 10 times the library scale where the reactions were
carried out on 0.156 mmol scale. The yield of this scaled up process
remained in line with the library reaction at 32%.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7545–7551 | 7551