Two-phase dynamic combinatorial discovery of a spermine transporter

Article abstract of DOI:10.1039/b902842k

The discovery, in a two-phase dynamic combinatorial library, of an unexpected linear receptor and transporter for spermine is described.

Full text of DOI:10.1039/b902842k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Two-phase dynamic combinatorial discovery of a spermine transporterw
´
Ruth Perez-Fernandez, Michael Pittelkow, Ana M. Belenguer, Laura A. Lane,
Carol V. Robinson and Jeremy K. M. Sanders*
´
Received (in Cambridge, UK) 10th February 2009, Accepted 23rd March 2009
First published as an Advance Article on the web 28th May 2009
DOI: 10.1039/b902842k
The discovery, in a two-phase dynamic combinatorial library, of
an unexpected linear receptor and transporter for spermine is
described.
When a mixture of building blocks 1, 2 (dissolved in
aqueous buffer solution) and 3 (dissolved in chloroform) was
allowed to oxidise and equilibrate after addition of N-methyl
morpholine (NMM) as a base and Bu3N as a phase-transfer
reagent, a DCL was formed (Fig. 2). Analysis of the composition
of the library was by LC-MS, injecting each phase separately.
The presence in the organic phase of building blocks that are
usually only soluble in aqueous phases was encouraging. The
experimental conditions established in previous work were
used, ensuring equilibration under thermodynamic control,
where it was shown that more of the water soluble materials
could be forced into the organic phase using Bu3N as a
phase-transfer reagent.5a None of the building block 3 was
observed in the aqueous phase in any of the experiments
performed, nor was any hydrolysis of the methyl ester 3
observed. Likewise, none of the dicarboxylic acid building
block (2) was observed in the organic phase.
Dynamic combinatorial chemistry has emerged as a powerful
tool for the discovery of new molecular receptors. Dynamic
combinatorial libraries (DCLs) rely on reversible chemical
reactions that interchange under thermodynamic control.1,2
Introducing a guest that interacts with members of the library
will shift the equilibrium composition, amplifying the strong
binders at the expense of other library members.3 The
amplified host can then be isolated and the host–guest system
characterised. Most of the molecular receptors discovered
in DCLs have drawn inspiration from previously known
synthetic receptors,4 and the discovery of structurally
unexpected receptors using DCLs is still relatively rare.5
We present here the discovery of an unexpected receptor
for the polyamine spermine using our recently-described
two-phase DCL approach and we also show that the receptor
acts to carry spermine across a bulk liquid membrane.5 In the
course of this work we developed a sensitive analytical method
to selectively quantify small amounts (mM) of spermine in
solution. Spermine is an attractive target for recognition as it
has many biological roles in normal cell division and growth,6
it forms strong interactions with DNA and RNA,7 and it has
an undefined geometry in solution.8
A DCL was set up with building blocks 1,9 2,10 and 3
(Fig. 1). Recently we described the idea of two-phase dynamic
combinatorial chemistry in which building blocks in one phase
combine with building blocks in the other phase, increasing
the available structural diversity.5 In particular we addressed
the possibility of achieving molecular recognition of polar
templates by non-polar receptors and vice versa. The discovery
of receptors soluble in one solvent for compounds that are
soluble in a different, immiscible, solvent is a challenge for
design, synthesis and characterisation. Spermine appeared to
be an ideal target for this approach.
The top chromatogram in Fig. 2 shows that the aqueous
phase exclusively contains the linear dimer of 2 (22), both in
the templated and the untemplated library. In the organic
layer, the untemplated library (Fig. 2, middle) contains the
linear dimer of 3 (32), the cyclic trimer of 1 (13), the linear
trimer of 1 and 3 (3ꢀ1ꢀ3), and small amounts of the linear
tetramer (3ꢀ1ꢀ1ꢀ3). The transfer of the water soluble building
block 1 and the library members containing 1 is enhanced by
the Bu3N. In the spermine templated library (Fig. 2, bottom)
an amplification of the linear trimer (3ꢀ1ꢀ3) by 50% is observed
by integration of the peaks in the HPLC chromatogram at the
expense of the other library members present in the organic
phase. This type of amplification in a DCL is an indication of
a favourable interaction between the species and the template.3
The change in composition of the linear tetramer (3ꢀ1ꢀ1ꢀ3) is
less than 2%.
Building blocks 1 and 2 are aromatic thiols equipped with
carboxylate groups and are therefore soluble in aqueous
solution at neutral pH, whereas the mono-thiol building block
3 bearing a methyl ester group is soluble in chloroform. The
thiol functionality allows the building blocks to be oxidised to
disulfides in the presence of air; thiolate anions then induce
disulfide exchange under thermodynamic control.11
Department of Chemistry, University of Cambridge, Lensfield Road,
Cambridge, UK CB2 1EW. E-mail: jkms@cam.ac.uk
w Electronic supplementary information (ESI) available: Calibration
curves, LC-MS data and a description of how spermine concentrations
were determined. See DOI: 10.1039/b902842k
Fig. 1 Structures of the mono-thiol and di-thiol building blocks
(1, 2 and 3) and the polyamine template spermine tetrahydrochloride.
ꢁc
This journal is The Royal Society of Chemistry 2009
3708 | Chem. Commun., 2009, 3708–3710

View Article Online
Fig. 2 Reverse phase HPLC chromatograms (260 nm) of libraries
containing building blocks (1, 2 and 3). Top: control library, aqueous
phase (no template). Middle: control library, organic phase
(no template). Bottom: organic phase after addition of spermine
tetrahydrochloride as a template.
Fig. 4 (a) Job plot between the receptor (3ꢀ1ꢀ3) and free-base
The amplified linear trimer (Fig. 3) was conveniently
separated by chromatography on silica and isolated as the
carboxylate salt; Bu3NH1 was identified as the counter ion by
NMR analysis in CDCl3.
spermine giving 2
: 1 stoichiometry. (b) Nano-electrospray mass
spectrum of a 1 : 1 mixture of receptor and spermine.
The 2 : 1 stoichiometry was further indicated by
As spermine tetrahydrochloride is insoluble in CDCl3, and
since a solution of the trimer in CDCl3 did not bring
the spermine salt into solution, complexation studies were
attempted using free-base spermine. Upon addition of a
solution of free-base spermine to a solution of the receptor
in CDCl3, free-base Bu3N appeared in the 1H NMR spectrum,
demonstrating that the Bu3N had been replaced by spermine.
Unfortunately, the linear bis-disulfide trimer was not stable for
extended periods in the presence of excess free-base spermine.z
Equilibration of the bis-disulfide generated a library of
macrocycles from 1 (13 and 14) which started to precipitate,
and the dimer of 3 (32) which started to appear in the NMR
spectrum. This prevented further two-dimensional NMR
experiments to elucidate the structure of the complex in
solution. Nevertheless, it was possible to construct a Job plot
using free-base spermine (Fig. 4a), and this showed the
stoichiometry of the complex to be two molecules of receptor
to one molecule of spermine.
nano-electrospray ionisation (n-ESI) mass spectrometry
(MS) (Fig. 4b). The most intense peak in the mass spectrum
of a freshly prepared mixture of free-base spermine and the
receptor is clearly the 2 : 1 complex. Additionally, aggregates
larger than the 2 : 1 complex were also observed in the mass
spectra. These species can be accounted for by the small size
of the receptor–spermine complex compared to the fission
droplets formed in the ESI process. The micromolar
concentration of receptor and spermine allows for multiple
occupancies of the complex and any free receptor or spermine
within a single fission droplet and can result in ‘non-specific’
interactions.12 Most crucially the absence of a complex with
1 : 1 stoichiometry allows us to conclude that the 2 : 1
complex is preferentially formed.
The fact that the receptor binds a guest template in a phase
where the template is insoluble complicates the study of
the solution phase behaviour of the system. However, one
attractive application for such receptors is the transport of
molecular species across bulk liquid membranes. One of the
applications of artificial membrane carriers is as vehicles for
drug delivery.13
Transport studies were performed in a U-tube cell (Fig. 5),14
consisting of a chloroform phase separating two neutral
aqueous sources and receiving phases. When the U-tube was
charged with a 5 mM solution of the receptor in CHCl3
and the source phase was charged with a 600 mM solution
of spermine tetrahydrochloride with NMM and Bu3N,
significant transport of spermine to the receiving phase
took place.
Fig. 3 Structure of amplified trimer (3ꢀ1ꢀ3).
ꢁc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 3708–3710 | 3709

View Article Online
chemistry is reported by the Kiel group in an accompanying
communication.16 In addition, this is the first example of the
amplification of a linear species in a disulfide based DCL
where there is a possibility of forming cyclic species.
We thank EPSRC and the Danish Natural Science Research
Council for financial support, the EU Marie Curie Research
Training Network MRTN-CT-2006-035614 for providing the
forum for the Cambridge and Kiel groups to exchange insights
and results, and V. Saggiomo and U. Luning for helpful
¨
discussions.
Fig. 5 (a) U-tube experiment showing transport of spermine across
an organic phase using the receptor found in the DCL. (b) Trans-
ported spermine (average of 5 determinations) determined at different
sample periods by FIA-MRM.
Notes and references
z Disulfide exchange is dependent on pH and other physical experi-
mental changes, thus the bis-disulfide may re-equilibrate. The excess of
spermine (template) and different solvent conditions can modify the
‘library’ composition as compared to the original two-phase condi-
tions, thus allowing different oligomers to dominate the composition.
This leads to the loss of the 3ꢀ1ꢀ3 trimer in this experiment.
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. Ladame, Org. Biomol. Chem., 2008, 6, 219–226.
2 (a) M. M. Rozenman, B. R. McNaughton and D. R. Liu, Curr.
Opin. Chem. Biol., 2007, 11, 259–268; (b) B. de Bruin, P. Hauwert
and J. N. H. Reek, Angew. Chem., Int. Ed., 2006, 45,
2660–2663.
The system was able to transport approximately 60 mM of a
600 mM solution of spermine using 5 mM solution of the
bis-disulfide receptor molecule. The transport takes place over
approximately 30 minutes, after which it halts, apparently due
to the instability of the bis-disulfide under the experimental
conditions: the disulfides start to equilibrate to give a DCL,
diminishing the amount of receptor in the organic phase and
shutting down transport.z Preliminary transport studies using
the organic phase from a pre-equilibrated DCL (as in Fig. 2)
gave similar transport results. Control experiments showed no
detectable transport of spermine into the aqueous receiving
phase when carrier 3ꢀ1ꢀ3 was omitted.
3 P. T. Corbett, J. K. M. Sanders and S. Otto, Chem.–Eur. J., 2008,
14, 2153–2166.
4 (a) M. J. Marsella, H. D. Maynard and R. H. Grubbs, Angew.
Chem., Int. Ed. Engl., 1997, 36, 1101–1103; (b) V. Saggiomo and
The amount of spermine transported was measured using
a FIA-MSMS method (flow injection analysis coupled to
tandem mass spectrometry).15 Mass spectral quantification
of spermine was performed by electrospray ionisation in
positive ion mode, monitoring the mass transitions m/z
203 - 129 by a multiple reaction monitoring (MRM)
experiment. In MRM, the spermine (m/z 203) is isolated in
the ion trap and then fragmented into two daughter ions; the
more intense daughter ion (m/z 129) is isolated in the ion trap
and measured with high sensitivity. Only molecules with a m/z
of 129 fragmented from m/z of 203 are observed. A calibration
curve is constructed by comparing the known concentrations
of spermine in the calibration standard solutions (range
0.5 to 5 mM) to the peak area of the daughter ion m/z 129
obtained with the FIA-MRM method. The concentration of
spermine in the unknown sample could be calculated by
comparing the peak area obtained with FIA-MRM method
for the m/z 129 ions with the calibration curve. The transport
samples were diluted with water before analysis to reach a
concentration of spermine within the calibration range.
The FIA-MRM method was optimised to obtain 5 replicate
injections of sample and standard solutions in 2.5 minutes
with good precision. The FIA-MRM method developed was
validated in terms of linearity, precision, accuracy and stability
in a solution of spermine (see ESIw for details).
U. Luning, Eur. J. Org. Chem., 2008, (25), 4329–4333;
¨
(c) Z. Rodriguez-Docampo, S. I. Pascu, S. Kubik and S. Otto,
J. Am. Chem. Soc., 2006, 128, 11206–11210.
5 (a) R. Perez-Fernandez, M. Pittelkow, A. M. Belenguer and J. K.
´ ´
M. Sanders, Chem. Commun., 2008, 1738–1740; (b) D. M. Epstein,
S. Choudhary, M. R. Churchill, K. M. Keil, A. V. Eliseev and
J. R. Morrow, Inorg. Chem., 2001, 40, 1591–1596.
6 (a) C. W. Tabor and H. Tabor, Annu. Rev. Biochem., 1984, 53,
749–790; (b) A. E. Pegg, Cancer Res., 1988, 48, 759–774.
7 V. A. Bloomfield, Biopolymers, 1977, 44, 269–282.
8 Some known spermine receptors, see: (a) H. Isobe, N. Tomita,
J. W. Lee, H.-J. Kim, K. Kim and E. Nakamura, Angew. Chem.,
Int. Ed., 2000, 39, 4257–4260; (b) L. Vial, R. F. Ludlow, J. Leclaire,
R. Pe
10253–10257.
9 B. M. R. Lienard, N. Selevsek, N. J. Oldham and C. J. Schofield,
rez-Fernandez and S. Otto, J. Am. Chem. Soc., 2006, 128,
´ ´
´
ChemMedChem, 2007, 2, 175–179.
10 T. Tsuboi, Y. Takaguchi and S. Tsuboi, Bull. Chem. Soc. Jpn.,
2008, 81, 361–368.
11 S. Otto, R. L. E. Furlan and J. K. M. Sanders, J. Am. Chem. Soc.,
2000, 122, 12063–12064.
12 L. A. Lane, B. T. Ruotolo, C. V. Robinson, G. Favrin and
J. L. P. Benesch, Int. J. Mass Spectrom., 2009, DOI:
10.1016/j.ijms.2009.03.006.
13 W. A. Ritchel, Handbook of Basic Pharmacokinetics, Drug
Intelligence, Hamilton, 4th edn, 1992.
14 (a) P. Breccia, M. van Gool, R. Perez-Fernandez, S. Martın-
´ ´ ´
Santamarıa, F. Gago, P. Prados and J. de Mendoza, J. Am. Chem.
´
Soc., 2003, 125, 8270–8284; (b) A. L. Sisson, J. P. Clare,
L. H. Taylor, J. P. H. Charmant and A. P. Davis, Chem. Commun.,
2003, 2246–2247; (c) C. Arnal-Herault, M. Michau and
´
M. Barboiu, J. Membr. Sci., 2008, 32, 94–99; (d) H. L. Rosano,
J. H. Schulman and J. B. Weisbuch, Ann. N. Y. Acad.Sci., 1961, 92,
457–469.
In conclusion, we have demonstrated for the first time that
unexpected receptors can be identified using two-phase
dynamic combinatorial chemistry. We have identified a receptor
for the biologically relevant polyamine spermine, showed that
the stoichiometry of the complex is 2 : 1 and that the receptor
works as a carrier of spermine across bulk liquid membranes.
A conceptually similar set of observations using different
15 (a) S. Millan, M. C. Sampedro, N. Unceta, M. A. Goicolea and
´
R. J. Barrio, Anal. Chim. Acta, 2007, 584, 145–152; (b) C. C. Sri,
S. K. Shukla and P. N. Sarma, J. Flow Injection Anal., 2008, 25,
20–23.
16 V. Saggiomo and U. Luning, Chem. Commun., 2009, DOI:
10.1039/b902847a.
¨
ꢁc
This journal is The Royal Society of Chemistry 2009
3710 | Chem. Commun., 2009, 3708–3710