Dynamic combinatorial chemistry with hydrazones: Libraries incorporating heterocyclic and steroidal motifs

Article abstract of DOI:10.1039/b917146k

We expand the possibilities in hydrazone based dynamic combinatorial chemistry with a series of new building blocks incorporating heterocyclic motifs. The synthetic procedure allows efficient access to building blocks with the general structure (MeO)₂CH-Heterocycle-C(O)NHNH₂, originating from heterocycles with an amine and an ester functionality. The equilibrium distribution of macrocyclic N-acyl hydrazones formed upon deprotection of the building blocks with TFA in organic solvents is reported. The mixing behaviour of these heterocycle-based building blocks with our cholate-based building blocks is described, particularly the observation of kinetic intermediates that disappear following 'proof-reading'. The Royal Society of Chemistry.

Full text of DOI:10.1039/b917146k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Dynamic combinatorial chemistry with hydrazones: libraries incorporating
heterocyclic and steroidal motifs†
Mark G. Simpson,a Michael Pittelkow,*‡a Stephen P. Watsonb and Jeremy K. M. Sanders*a
Received 19th August 2009, Accepted 20th November 2009
First published as an Advance Article on the web 11th January 2010
DOI: 10.1039/b917146k
We expand the possibilities in hydrazone based dynamic combinatorial chemistry with a series of new
building blocks incorporating heterocyclic motifs. The synthetic procedure allows efficient access to
building blocks with the general structure (MeO)2CH-Heterocycle-C(O)NHNH2, originating from
heterocycles with an amine and an ester functionality. The equilibrium distribution of macrocyclic
N-acyl hydrazones formed upon deprotection of the building blocks with TFA in organic solvents is
reported. The mixing behaviour of these heterocycle-based building blocks with our cholate-based
building blocks is described, particularly the observation of kinetic intermediates that disappear
following ‘proof-reading’.
of this work we have uncovered macrocyclic kinetic intermediates
and remarkable proof-reading effects in the DCLs.
Introduction
Dynamic combinatorial chemistry has emerged as an important
strategy in the search for selective hosts in supramolecular
Results and discussion
chemistry.1,2 Several reversible chemistries have been explored,
one of the most successful being exchange of hydrazones. This
has especially been explored with amino acid and steroid based
building blocks.3–10 Herein we describe the synthesis and properties
of a range of new hydrazone building blocks for dynamic combina-
torial libraries (DCLs) based on heterocyclic scaffolds. The build-
ing blocks have the general structure (MeO)2CH-Heterocycle-
C(O)NHNH2 with varying recognition abilities (Fig. 1).
A series of heterocycle based building blocks (15, 16, 21, 22, 25,
28, 31 and 38) was assembled in a slightly different fashion to their
steroidal and amino acid counterparts, to enable introduction of
a hydrazide under forcing conditions.8 The introduction of the
dimethyl acetal was achieved by means of N-alkylation of the
amine of the methyl or ethyl ester derivative of the heterocycle with
a meta- or para-substituted bromomethylbenzyl dimethyl acetal.
The hydrazide was then introduced by hydrazinolysis of the ester
under forcing conditions (Scheme 1).
Fig. 1 General structure of heterocyclic building blocks for hydrazone
based dynamic combinatorial chemistry.
Knowledge about the effects of changes in the heterocyclic
building block geometry on the distribution of the macrocyclic
N-acyl hydrazones formed and how this compares with cholate
building blocks was sought.8 The main goal, however, was to
determine the outcome of mixing heterocyclic with cholate based
building and to isolate and characterise the products. In the course
aUniversity Chemical Laboratory, University of Cambridge, Lensfield
Road, Cambridge, CB2 1EW, United Kingdom. E-mail: jkms@cam.ac.uk;
Fax: +44 1223 336017; Tel: +44 1223 336411
Scheme 1 General synthetic strategy for heterocyclic based building
blocks.
bGlaxo Wellcome Medicines Research Centre, Gunnels Wood Road,
Stevenage, Hertfordshire, SG1 2NY, United Kingdom
† Electronic supplementary information (ESI) available: Experimental and
analytical details for all compounds. See DOI: 10.1039/b917146k
Synthesis
‡ Present address: Department of Chemistry, University of Copenhagen,
Universitetsparken 5, DK-2100, Copenhagen Ø, Denmark; E-mail:
pittel@kiku.dk
The dimethyl acetals 5 and 6 were synthesised in quantitative yield
by treating the aldehydes 3 and 4 with a catalytic amount of pTsOH
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1181–1187 | 1181
©

View Article Online
in MeOH at room temperature.11 The bromomethyl benzaldehydes
(3 and 4) were synthesised from the respective a-bromo tolunitriles
(1 and 2) by means of a modification of the procedure of Schlenoff
et al. (Scheme 2).12 Alternatively, 3 and 4 were synthesised on a
larger scale from the symmetrical dialdehydes (7 and 8) by mono
reduction using NaBH4 in MeOH to give the benzylic alcohols (9
and 10), followed by bromination with aqueous HBr.13,14
An identical methodology was used to synthesise the (R)-
stereoisomers of the THIQ building blocks (21 and 22) via
the methyl esters (19 and 20) (Fig. 2.). The intermediates 19
and 20 were prepared from (R)-1,2,3,4-tetrahydroisoquinoline-3-
carboxylic acid (17) and its methyl ester (18).
Fig. 2 (R)-Stereoisomer of tetrahydroisoquinoline (THIQ) building
blocks 21 and 22.
For further studies on the effect of the heterocycle struc-
ture on product distributions in DCLs, ethyl-4-methyl-5-
imidazolecarboxylate (23) was alkylated with 5 utilising K2CO3
in DMF to afford 24 (only one isomer isolated). Hydrazinolysis of
24 with a large excess of hydrazine monohydrate in boiling MeOH
afforded the hydrazide (25, Scheme 4).
Scheme 2 Syntheses of acetal aromatic linkers. Reaction conditions:
a) DIBAL, Toluene (72–80%), H+. b) NaBH4, MeOH (65–68%). c) HBr
(aq), toluene (82–85%). d) pTsOH, MeOH (quant.).
The ester 12 was synthesised from (S)-1,2,3,4-
tetrahydroisoquinoline-3-carboxylic acid (11) by the procedure of
Nakamura et al.15 and then alkylated with 5 or 6 utilising K2CO3
in DMF to form 13 and 14. Hydrazinolysis was accomplished
with 10 equivalents of hydrazine monohydrate in boiling MeOH
(Scheme 3) to give the hydrazides 15 and 16.
Scheme 4 Synthesis of imidazole (Imid) based building block 25.
Reaction conditions: a) 5, K2CO3, DMF, heat (43%). b) NH2NH2¥H2O,
MeOH, reflux (63%).
The building blocks 28 and 31 were synthesised from
ethyl piperidine-4-carboxylate (26, via the acetal 27) and ethyl
piperidine-2-carboxylate (29, via the acetal 30) respectively
(Schemes 5 and 6).
Scheme 5 Synthesis of a piperidine (Pip4) based building block (28).
Reaction conditions: a) 5, K2CO3, DMF, heat. b) NH2NH2¥H2O, MeOH,
reflux (50%, 2 steps).
Scheme 3 Synthesis of tetrahydroisoquinoline (THIQ) building blocks 15
and 16 illustrated by the (S)-stereoisomer. Reaction conditions: a) SOCl2,
MeOH (82%). b) 5 or 6, K2CO3, DMF, 100 ◦C (90%). c) NH2NH2¥H2O,
MeOH, reflux (90–92%).
The indole-based building block 38 was synthesised from
indole-3-carboxylic acid (34, Scheme 8). In this case the dimethyl
1182 | Org. Biomol. Chem., 2010, 8, 1181–1187
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 6 Synthesis of a piperidine (Pip2) based building block (31).
Reaction conditions: a) 5, K2CO3, DMF, heat. b) NH2NH2¥H2O, MeOH,
reflux (53%, 2 steps).
acetal was introduced via EDC mediated amide coupling of the
amino dimethyl acetal 33.
The p-methylamino-benzaldehydedimethylacetal (33) was syn-
thesised from 4-cyanobenzaldehyde (32). The aldehyde was pro-
tected as its dimethyl acetal (32) by means of NH4Cl in MeOH,16
and the nitrile was reduced with LiAlH4 in THF (Scheme 7).17
Scheme 8 Synthesis of an indole (Ind) based building block (38).
Reaction conditions: a) TFA anhydride, BnOH, CH2Cl2 (68%). b) Methyl
bromoacetate, K2CO3, DMF, heat (76%). c) Pd/C, H2, THF (70%).
d) 33, EDC, DMAP, CH2Cl2 (59%). e) NH2NH2¥H2O, MeOH, room
temperature (70%).
Table 1 Distribution of libraries for the eight new building blocks
described in this work. For (38) the experiment was carried out in DMSO,
all others in CH2Cl2. The monomer concentration was 5 mM and TFA
was used as the acid
Scheme 7 Reaction conditions: a) NH4Cl, MeOH, heat. b) LiAlH4, THF
Monomera
Initial distribution
Equilibrium distribution
(91%).
15 or 21
16 or 22
25
[THIQ]2–6
[THIQ]2–6
[Imid]2–3
[pip4]2
[pip2]2
[Ind]2–4
[THIQ]2
[THIQ]2
[Imid]2
[pip4]2
[pip2]2
[Ind]2–4
The carboxylic acid (34) was transformed into its benzyl
ester (35) by treatment with trifluoroacetic anhydride and benzyl
alcohol in CH2Cl2. N-Alkylation of the indole (35) with methyl
bromoacetate and K2CO3 in DMF gave 36. Hydrogenation using
10% Pd/C afforded the carboxylic acid, which was reacted with
the amine (33) by means of EDC and a catalytic amount of DMAP
in CH2Cl2 to afford the amide (37).
28
31
38
a [THIQ]: 15 or 21, [Imid]: 25, [pip4]: 28, [pip2]: 31, [Ind]: 38.
Hydrazinolysis of the ester (37) was accomplished with 10 equiv-
alents of hydrazine monohydrate in MeOH at room temperature
to produce the hydrazide (38, Scheme 8).
rise to a slightly different initial distribution of products (compared
to 15 and 21) in which the cyclic trimer was the dominant species.
Over a period of seven days, however, this too exchanged to form
cyclic dimer as the overwhelmingly favoured product. Increasing
the concentration of 15 and 21 to 10 mM led to the formation of
significant amounts of precipitate after the addition of TFA.
Dilute solutions (5 mM) of 25, 28 and 31 in CH2Cl2 generated
significant amounts of precipitate 10 min after the addition of
the TFA. Filtering the precipitate off and analysing the filtrate by
LC-MS indicated that the only cyclic products from 25 were the
cyclic dimer and trimer. Only cyclic dimer could be observed in
the filtrate obtained after adding TFA to 5 mM solutions of 28
and 31. Repeating the deprotection using DMSO as the solvent
avoided formation of precipitate on TFA addition but did not
lead to a wider distribution of products: in all cases the only
product at equilibrium was the cyclic N-acyl hydrazone dimer.
The deprotection of a 5 mM solution of 38 led to the formation of
cyclic dimer, trimer and tetramer, according to LC-MS, with the
dimer once again the most favoured macrocycle. The distribution
of products in this case did not collapse and drift towards the cyclic
dimer. It is important to note that the equilibration in DMSO is
much slower that in CH2Cl2.
Cyclisation of heterocyclic building blocks
Deprotection of the dimethyl acetal moieties of the building blocks
was carried out using the previously reported conditions:8,10 the
building block was dissolved to give a 5 mM solution, TFA
(5% v/v) was then added and the reaction mixture stirred at room
temperature. The acid has two functions i) it removes the dimethyl
acetal to form the aldehyde, and ii) it catalyses the formation of
macrocyclic N-acyl hydrazones. The solvents used were anhydrous
CH2Cl2 for monomers 15, 16, 21, 22, 25, 28 and 31, and DMSO
for 38, for solubility reasons. The results of deprotecting these
dimethyl acetal hydrazides are given in Table 1, as determined by
HPLC-MS. At a monomer concentration of 5 mM 15, 16, 21 and
22 initially gave rise to a series of CH2Cl2-soluble macrocyclic N-
acyl hydrazones, from dimer through to hexamer. There was no
difference in distribution between the enantiomers 15 and 21; the
cyclic dimer [THIQ]2 was the most favoured macrocycle initially
and grew in abundance with time. After seven days the cyclic dimer
was virtually the only product present. In contrast 16 and 22 gave
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1181–1187 | 1183
©

View Article Online
The consistency of the observations concerning product distri-
bution is perhaps not too surprising, given that the structures of
15, 21, 25 and 31 are superficially, rather similar. What is a little
surprising is the effect of the structure of the building block on the
solubility of the macrocyclic N-acyl hydrazone formed in CH2Cl2.
Thus, N-acyl hydrazones derived from 15 are freely soluble in
CH2Cl2 where those derived from 31 are only sparingly soluble.
The fact that changing the orientation of the aldehyde in 16 to
meta did not lead to a significantly different product distribution
at equilibrium was disappointing, given the success of this strategy
in our previous work with steroidal building blocks.8 Similarly the
lack of effect of changing the position of the dimethyl acetal with
respect to the hydrazide in 28. The marked stability of the cyclic
N-acyl hydrazone dimer seems to be a common factor in this.
Mixing steroidal and heterocyclic hydrazides and hydrazones
In an attempt to generate more diverse DCLs we investigated the
mixing of the new heterocyclic building blocks with the cholate
building blocks shown in Fig. 3.8
Fig. 4 Top: General structure of mixed dimeric macrocycles between 15
and the steroidal building blocks. Bottom: Structure of 47 showing some
nOe connections that define the overall conformation of the structure.
Under identical conditions of concentration and solvent es-
sentially the same outcome was observed at equilibrium by
deprotecting the building blocks 15 and 39 in one pot. The initial
distribution of products was quite different, however, in that the
dominant product was the heterotrimer previously observed by
LC-MS. Furthermore higher oligomers, such as mixed tetramers
and pentamers could also be observed. The trimers, tetramers
and pentamers were all ephemeral kinetic products that recycled
to form the heterodimer over a period of three days. This is an
example of proof-reading which can be observed with macrocyclic
N-acyl hydrazones (Fig. 5).8
Isolation of the heterodimer 47 allowed us to show, importantly,
that its formation was reversible. Thus, when a 2.5 mM solution of
47 in CH2Cl2 was treated with TFA (5% v/v), LC-MS showed the
formation of the homodimers derived from 15 and 39 after 24 h,
as well as a trace amount of the heterotrimer. The equilibrium
distribution of products was comparable to that obtained from
deprotecting the building blocks 15 and 39 in one pot and from
combining the separately preformed N-acyl hydrazones. This
confirms that the system is under thermodynamic control.
Since such mixed macrocyclic hydrazones had not been pre-
viously reported we studied their structures in some detail. The
Fig. 3 Structures of steroid based building blocks 39–46.8
All the mixing experiments were carried out at the same
concentration in CH2Cl2 to facilitate isolation of the macrocyclic
products by conventional preparative thin layer chromatography.
Combining a 5 mM CH2Cl2 solution of the library derived
from 15 with an equimolar 5 mM solution of the cyclic N-
acyl hydrazone dimer8 derived from 39 in CH2Cl2 slowly led to
the formation of the heterodimer 47 (Fig. 4), according to LC-
MS. Equilibrium was attained after 7 days and 47 was isolated
in 60% yield following neutralisation of the reaction mixture
with triethylamine. The remainder of the material consisted of
the homodimers derived from 15 and 39, together with a trace
amount of the heterotrimer derived from two heterocycle and one
steroid unit.
1184 | Org. Biomol. Chem., 2010, 8, 1181–1187
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
temperature was lowered to -60 ◦C. In contrast the cis N-acyl
hydrazone NH moved downfield from 8.82 ppm to 10.50 ppm on
lowering the temperature from ambient to -60 ◦C.
Furthermore when the pyridine-4-carboxylate substituent was
replaced with a benzoate group in heterodimer 49 the trans N-
acyl hydrazone NH was shifted upfield to 10.41 ppm. Notably,
no nOe connectivities could be observed between the trans N-acyl
hydrazone NH and the benzoate protons (49, Table 2).
When these mixing experiments were repeated with the enan-
tiomer 21 and 39 no significant effect on the yield of the
heterodimer 55 formed at equilibrium was found (Fig. 6). The
1H NMR spectrum of 55 was slightly different to that of 47 but
not sufficiently to merit further discussion. COSY and NOESY
indicated that the two N-acyl hydrazone geometries were again
present as was the significant downfield shift of the N-acyl
hydrazone NH resonance; this signal appears at 11.50 ppm.
Fig. 5 Schematic representation of the proof-reading process.
1H NMR spectrum of 47 in CDCl3 has the simplicity one would
associate with a rigid macrocycle, existing in a single well-defined
conformation. Perhaps the most striking feature of the spectrum
is the N-acyl hydrazone NH resonance downfield shifted to
11.60 ppm. The COSY and NOESY spectra of 47 in CDCl3
allowed both cis E and trans E N-acyl hydrazone geometries to
be identified. These geometries have been identified in steroid-
based N-acyl hydrazone macrocycles previously but in separate
molecules. The cis E geometry was identified by virtue of an
nOe connectivity between the N-acyl hydrazone NH at 8.59 ppm
with the azamethine proton, Hm, at 7.58 ppm. As anticipated
the azamethine proton, Hm, has an nOe connectivity with Hl at
7.22 ppm.
The trans E geometry was assigned by virtue of the nOe
connectivity between the N-acyl hydrazone NH resonance at
11.60 ppm and the azamethine proton, Hq, at 9.00 ppm. Hq also
exhibits an nOe cross-peak with Hg , at 7.67 ppm. In addition nOe
connectivities could be observed between the N-acyl hydrazone
NH at 11.60 ppm, Hk, at 7.32 ppm and the a-pyridyl proton,
at 9.10 ppm. This last nOe connectivity strongly suggests that
the downfield shift of the N-acyl hydrazone NH at 11.60 ppm
is due to a hydrogen-bond between the trans N-acyl hydrazone
NH and the 12-pyridine substituent. In a variable tempera-
ture NMR experiment conducted with 47 in CDCl3 the trans
N-acyl hydrazone NH did not shift significantly downfield as the
Fig. 6 Structure of 55 and some nOe correlations confirming the
geometrical arrangement around the N-acyl hydrazone bond.
The yields, at equilibrium, of the heterodimers derived from 15
and the various steroids employed (Table 2) range from 10 to 90%,
indicating that the substituent at the 7 or 12-position of the steroid
controls the extent to which the heterodimer is formed instead of
a self-sorted mixture of homodimers.
The explanation for this difference in behaviour is not clear. It
seems most likely that the pyridine-3-carboxylate is situated such
that under acidic conditions it can be protonated and assist in
Table 2 Yields of heterodimers from 15 and various steroid building blocks. The structure of the steroids are shown in Fig. 4 (top)
Heterodimer
R1
R2
Yield of heterodimer (%)
47
48
49
50
51
52
53
54
H
H
H
H
pyridine-4-carboxylate
pyridine-3-carboxylate
benzoate
trifluoroacetate
H
H
H
H
62
10
55
40
48
88
90
45
H
pyridine-4-carboxylate
pyridine-3-carboxylate
OH
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1181–1187 | 1185
©

View Article Online
ring opening of the cis N-acyl hydrazone of the heterodimer. This
would then lead to kinetic instability of the heterodimer compared
to the parent homodimers. Thermodynamic arguments would,
however, seem to be more appropriate when accounting for the
difference in behaviour of the pyridine substituted deoxycholate
and chenodeoxycholate-based monomers.
The unexpected trifluoroacylation of the 12-OH, leading to
heterodimer 50, is a function of the position of the hydroxyl group
in the molecule. It is observed in the formation of 50 but not 54
so it would seem reasonable to link this with the closer proximity
of the cis N-acyl hydrazone in the former heterodimer. One can
envisage that a hydrazone could assist in the trifluoroacylation
of the 12-position either through general acid catalysis, whereby
the protonated imine nitrogen protonates TFA and so activates
it to nucleophilic attack by the 12-OH, or through nucleophilic
acid catalysis, whereby the imine nitrogen is acylated by self-
protonated TFA, which then undergoes intramolecular nucle-
ophilic attack by the 12-OH group to transfer the trifluoroacetyl
group.
Since a combination of a meta substituted cholate and a
para substituted THIQ favours formation of heterodimer, the
next step was to mix a meta substituted steroid with a meta
substituted THIQ (16). In the initial distribution of products
formed from deprotecting a 2.5 mM solution of 16 and 39 in
CH2Cl2 with the standard amount of TFA in one pot a whole
series of mixed macrocycles (heterodimer, trimers and tetramers)
could be observed by LC-MS. These were proof-read, however,
the result being that at equilibrium only a trace of the heterodimer
was present. No higher mixed macrocycles were present and the
homodimers derived from 16 and 39 were the only other products.
The outcome was the same when the preformed macrocyclic N-
acyl hydrazones derived from 16 and 39 were combined under the
same conditions.
Mixing 5 mM solutions of the other heterocycle-based
monomers 25, 28 and 31 in CH2Cl2 in the presence of steroid
39 led, uninformatively, to the formation of precipitates.
Mixing experiments were carried out between the para substi-
tuted steroid (56, Fig. 7) the meta substituted glycine extended
steroid (57) and the para substituted THIQs (15 and 21). In
the case of mixing experiments between 56 and 15 and 21, a
range of macrocycles were formed, including mixed trimers and
tetramers. The same was observed when 57 and 15 were mixed; the
heterodimer was not the major product and at equilibrium other
mixed macrocycles were also present as shown in Table 3.
This is an encouraging result, as diversity in parent libraries
might facilitate templating in DCLs.
Fig. 7 Structure of para-substituted building block (56) and extended
steroid building blocks (57).
Mixing isomers of heterocyclic hydrazides
The preference for the heterocyclic building blocks (15, 16, 21
and 22) to form cyclic dimers was further studied by making
libraries of all possible mixtures of regioisomers and stereoisomers.
This allowed assignment of the different species observed in
solution using LC-MS analysis. The LC-MS analysis revealed the
presence of six distinct homo- and hetero-dimers upon mixing
the four building blocks. In fact, ten different compounds are
possible but four of them are enantiomers of other species and
the actual number of expected peaks in the HPLC chromatogram
is therefore six. To confirm the reversible nature of the systems,
two cyclic dimers (58 and 59) were isolated (Fig. 8) and used
as starting material in library formation with the other isomers.
The libraries using these cyclic dimers gave library mixtures
with identical composition to the libraries formed directly from
the corresponding building blocks. This again confirms that the
libraries are equilibrating under thermodynamic control.
Conclusions
Table 3 Equilibrium distributions of mixed libraries and structures of
We have shown that building blocks for hydrazone based dynamic
combinatorial chemistry containing a hydrazide and a dimethy-
lacetal protected aldehyde can conveniently be prepared from the
corresponding amino acid methyl ester derivatives. A series of
steroidal and heterocyclic compounds have been prepared showing
the versatility of the synthetic protocol. Upon treatment of a
building block or of a mixture of building blocks in organic
solvents with TFA the dominant products were cyclic dimers at
thermodynamic equilibrium. Larger macrocycles were present as
kinetic intermediates in most experiments. The relative lack of
diversity in DCLs derived from single or mixed building blocks
extended steroid building blocks
Monomera
Equilibrium distribution
56 + (15 or 21)
[St]2, [THIQ]2, [St][THIQ], [St]3,
[THIQ][St]2, [THIQ]2[St],
[THIQ]2[St]2, [THIQ][St]3
[St]2, [THIQ]2, [St][THIQ],
[THIQ][St]2, [THIQ]2[St],
[THIQ]3[St]
57 + (15 or 21)
a [THIQ]: 15 or 21, [St]: 56.
1186 | Org. Biomol. Chem., 2010, 8, 1181–1187
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Int. Ed., 2002, 41, 898–952; (c) K. West and S. Otto, Curr. Opin. Drug
Dev., 2005, 5, 123–160; (d) S. Ladame, Org. Biomol. Chem., 2008, 6,
219–226.
2 (a) R. T. S. Lam, A. M. Belenguer, S. L. Roberts, C. Naumann, T.
Jarrosson, S. Otto and J. K. M. Sanders, Science, 2005, 308, 667–669;
(b) J.-L. Wietor, G. D. Pantos and J. K. M. Sanders, Angew. Chem., Int.
Ed., 2008, 47, 2689–2692; (c) A. Gonzalez-Alvarez, I. Alfonso and V.
Gotor, Chem. Commun., 2006, 2224–2226; (d) R. Pe´rez-Ferna´ndez, M.
Pittelkow, L. A. Lane, A. M. Belenguer, C. V. Robinson and J. K. M.
Sanders, Chem. Commun., 2009, 3708–3710.
3 (a) G. R. L. Cousins, S.-A. Poulsen and J. K. M. Sanders, Chem.
Commun., 1999, 1575–1576; (b) R. L. E. Furlan, G. R. L. Cousins
and J. K. M. Sanders, Chem. Commun., 2000, 1761–1762; (c) R. L. E.
Furlan, Y. Ng, G. R. L. Cousins, J. E. Redman and J. K. M. Sanders,
Tetrahedron, 2002, 58, 771–778; (d) R. L. E. Furlan, Y. Ng, S. Otto and
J. K. M. Sanders, J. Am. Chem. Soc., 2001, 123, 8876–8877; (e) G. R. L.
Cousins, R. L. E. Furlan, Y. Ng, J. E. Redman and J. K. M. Sanders,
Angew. Chem., Int. Ed., 2001, 40, 423–428; (f) S. L. Roberts, R. L. E.
Furlan, S. Otto and J. K. M. Sanders, Org. Biomol. Chem., 2003, 1,
1625–1633; (g) J. Liu, K. R. West, C. R. Bondy and J. K. M. Sanders,
Org. Biomol. Chem., 2007, 5, 778–786.
4 S. M. Voshell, S. J. Lee and M. R. Gagne´, J. Am. Chem. Soc., 2006, 128,
12422–12423.
5 M.-K. Chung, C. M. Hebling, J. W. Jorgensen, K. Severin, S. J. Lee and
M. R. Gagne´, J. Am. Chem. Soc., 2008, 130, 11819–11827.
6 F. Bulos, S. L. Roberts, R. L. E. Furlan and J. K. M. Sanders, Chem.
Commun., 2007, 3092–3093.
7 (a) A. G. Orrillo, A. M. Escalante and R. L. E. Furlan, Chem. Commun.,
2008, 5298–5300; (b) Z. Rodriguez-Docampo and S. Otto, Chem.
Commun., 2008, 5301–5303.
8 M. G. Simpson, M. Pittelkow, S. P. Watson and J. K. M. Sanders, Org.
Biomol. Chem., DOI: 10.1039/b917145b, preceding paper.
9 N. H. Anderson, J. W. Neidigh, S. M. Harris, G. M. Lee, Z. Liu and H.
Tong, J. Am. Chem. Soc., 1997, 119, 8547–8561.
Fig. 8 Structure of homodimers of THIQs 58 and 59.
is, however, disappointing. Hydrazones derived from flexible
dipeptides turn out to generate more diversity and to be susceptible
to templating effects as described elsewhere.3
10 M. G. Simpson, S. P. Watson, N. Feeder, J. E. Davies and J. K. M.
Sanders, Org. Lett., 2000, 2, 1435–1438.
¨
11 H. Griesser, R. Ohrlein, W. Schwab, R. Ehrler and V. Ja¨ger, Org. Synth.,
2000, 77, 236.
Acknowledgements
12 L. Wen, M. Li and J. B. Schlenoff, J. Am. Chem. Soc., 1997, 119, 7726–
7733.
We thank BBSRC and The Danish Natural Science Research
Council for financial support and GlaxoWellcome for a CASE
award (MGS).
13 D. Tanner and O. Wennerstro¨m, Acta Chem. Scand., Ser. B, 1983, 37,
693–698.
14 Y. Endo, M. Ohno, M. Hirano, A. Itai and K. Shudo, J. Am. Chem.
Soc., 1996, 118, 1841–1855.
15 R. Nagata, N. Tanno, T. Kodo, N. Ae, H. Yamaguchi, T. Nishimura,
F. Antoku, T. Tatsuno, T. Kato, Y. Tanaka and M. Nakamura, J. Med.
Chem., 1994, 37, 3956–3968.
16 J. I. DeGraw, L. Goodman and B. R. Baker, J. Org. Chem., 1961, 26,
1156–1161.
References
1 (a) P. T. Corbett, J. Leclaire, L. Vial, K. R. West, J.-L. Wietor, J. K. M.
Sanders and S. Otto, Chem. Rev., 2006, 106, 3652–3711; (b) S. J. Rowan,
S. J. Cantrill, G. R. L. Cousins and J. K. M. Sanders, Angew. Chem.,
17 L. M. Soffer and M. Katz, J. Am. Chem. Soc., 1956, 78, 1705–1709.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1181–1187 | 1187
©