Divalent ligand for intramolecular complex formation to streptavidin

Article abstract of DOI:10.1039/b505700k

Monovalent ligand 4 and divalent ligand 8 have been synthesized, and their thermodynamic parameters of complexation to avidin and streptavidin have been analyzed in terms of multivalent binding. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b505700k

View Article Online / Journal Homepage / Table of Contents for this issue
C O M M U N I C A T I O N
Divalent ligand for intramolecular complex formation to
streptavidin†
Joan-Antoni Farrera,*a Pedro Hidalgo-Ferna´ndez,a Jurry M. Hannink,b Jurriaan Huskens,c
Alan E. Rowan,b Nico A. J. M. Sommerdijkd and Roeland J. M. Nolteb
a Departament de Quimica Organica, Universitat de Barcelona, Mart´ı i Franque´s 1, 08028,
Barcelona, Spain. E-mail: jfarrera@ub.edu
b Institute for Molecules and Materials, Radboud Universiteit Nijmegen, P.O. Box 9010, 6525,
ED Nijmegen, The Netherlands
c Laboratory of Supramolecular Chemistry and Technology, MESA+ Institute for
Nanotechnology, University of Twente, P.O. Box 217, 7500, AE Enschede, The Netherlands
d Laboratory for Macromolecular and Organic Chemistry, Eindhoven University of Technology,
P.O. Box 513, 5600, MB Eindhoven, The Netherlands
Received 25th April 2005, Accepted 17th May 2005
First published as an Advance Article on the web 24th May 2005
Monovalent ligand 4 and divalent ligand 8 have been
synthesized, and their thermodynamic parameters of com-
plexation to avidin and streptavidin have been analyzed in
terms of multivalent binding.
informative could be the use of divalent ligands having two units
of a monovalent ligand of moderate binding affinity for Av and
SAv. With these reversible ligands it should be possible to easily
measure the binding constants of the complexes with Av and
SAv, giving an indication of how appropriate a given spacer is
for each protein in order to form intramolecular complexes.8
For this purpose, monovalent ligand 4 and divalent ligand
8 have been designed and synthesized, and their binding
parameters to Av and SAv measured. Ligand 4 is based on
the dimethylpropanediurea bicyclic framework. Molecular clips
based on this compound have recently been described as host
molecules for neutral guests.9 Like BT, ligand 4 has a urea group
that might allow the formation of similar hydrogen bonding
interactions with the protein residues of the binding pockets of
Av and SAv. Ligand 4 has also a valerate side chain that, as
in BT, might contribute to the binding through hydrophobic
interactions of the methylene groups and hydrogen bonding of
the carboxylate group. But unlike BT, in 4 the valerate chain is
bound to the bridgehead position, there is a second ureido group
present in the molecule and the molecular framework is of the
bicyclo[3.3.1] type with an isopropylidene group as the shortest
bridge. Ligand 8 contains two units of the monovalent ligand
4 linked to glutamic acid through a diamine. The amino group
of 8 is expected to increase its aqueous solubility at neutral pH,
allowing also further functionalisation of the molecule.
Ligand 4 was prepared in 90% overall yield by reacting methyl
7,7-dimethyl-6,8-dioxooctanoate with two equivalents of urea in
toluene and TFA as a catalyst, followed by saponification. Lig-
and 8 was synthesized in 25% overall yield by condensation of Z-
protected, activated glutamic acid to N-Boc-1,6-hexanediamine,
followed by cleavage of the Boc group, condensation of the
resulting diamine to ligand 4 with DPPA and cleavage of the
Z group (see Scheme 1 and Supporting Information†).
The dye HABA was used to probe the binding of 4 and 8 to
Av and SAv. This dye, in its unbound form, has an absorption
maximum at 348 nm, which shifts to 500 nm when bound to
Av or SAv.10 Addition of 4 or 8 to an aqueous solution of
HABA containing either Av or SAv resulted in a decrease in
the absorption at 500 nm and a corresponding increase in the
absorption at 348 nm, proving that both 4 and 8 bind to the same
binding sites as HABA and BT. Since the binding constants of
HABA to Av and SAv were already known (Ka = 1.7 × 105 M−1
and 7.3 × 103 M−1, respectively),6–10 the binding constants of 4
and 8 to Av and SAv could be measured in spectrophotometric
competition experiments with HABA. The obtained binding
constants (Table 1) are lower than the binding constants of
BT, which is not surprising considering the good fit of the
natural ligand into the binding site in terms of van der Waals,
Avidin (Av) and streptavidin (SAv) are tetrameric proteins, well-
known for their high binding (Ka ≈ 1015 M−1 and 2.5 × 1013 M−1,
respectively) to biotin (BT), which has allowed to use them in
many biochemical applications.1 X-Ray diffraction and binding
studies of BT and analogues with Av and SAv have suggested
a major contribution to binding of the ureido group of BT,
with a lower contribution coming from the valerate side chain
and the thiolane ring of BT.2,3 The formation of the complex
SAv–BT (DG◦ = −76.4 kJ mol−1) is enthalpically driven (DH◦ =
−134 kJ mol−1), with an unfavorable contribution from entropy.3
Other ligands structurally not related to BT, e.g. azobenzene
dyes such as 2-(4ꢀ-hydroxyphenylazo)benzoic acid (HABA) have
been shown to bind to both Av and SAv at the same binding
site as BT. However, in the case of HABA, the binding to SAv
(DG◦ = −22.0 kJ mol−1) is not enthalpically driven (DH◦
=
+7.1 kJ mol−1).3
In most applications of the SAv–BT system,4 for instance
in the preparation of protein monolayers based on avidin–
polymer amphiphiles,5 it is desirable to have a unique type of
complex and a well-defined supramolecular architecture. This is
usually accomplished by working under saturation conditions,
where the four binding sites are occupied by derivatized BT
molecules. However, in some specific applications, it can be
difficult or even impossible to occupy the four binding sites
of SAv, leading to the formation of isomeric complexes as
well as complexes of different stoichiometry. In this case the
use of divalent ligands, i.e. those in which two units of a
monovalent ligand are covalently linked via a spacer arm, could
be advantageous owing to the smaller number of complexes they
form with SAv. In order to design these divalent ligands, the
optimal spacer length and polarity should be known. Previous
works were based on bisbiotin compounds, which were shown
to induce Av or SAv oligomerization when the length of the
spacer was not optimal for the formation of the intramolecular
complex.6,7 Unfortunately, the techniques used to detect the
extent of oligomerization (electron microscopy and size exclu-
sion chromatography) give no information about the stability
of the intramolecular complex, and only give an indication of
the length of the spacer that minimizes oligomerization. More
† Electronic supplementary information (ESI) available: Experimental
details and binding models. See http://www.rsc.org/suppdata/ob/
b5/b505700k/
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 3 9 3 – 2 3 9 5
2 3 9 3
©

View Article Online
Scheme 1 Reagents and conditions: (i) N-(2-methyl-1-propenyl)morpholine; (ii) H2O; (iii) H2NCONH2, TFA, toluene, 10 h reflux; (iv) 2 M NaOH,
reflux; (v) H2N(CH2)6NHBoc, AcOEt, rt, 48 h; (vi) TFA, rt, 1 h; (vii) DPPA, 4, Et3N, DMSO; (viii) H2, Pd/C, MeOH.
Table 1 Binding constant and thermodynamic parameters of SAv and
consequence of a less favourable orientation of the ligand inside
the binding pocket due to steric repulsive interactions between
the isopropylidene bridge of the ligand and protein residues at
the binding site. London dispersion interactions between water
molecules and apolar parts of the ligand, which will be lost in
the complex with the protein, might also be partly responsible
for the small enthalpic change of the binding.
Av complexes in watera
Complex Ka/M−1
DG/kJ mol−1 DH/kJ mol−1 TDS/kJ mol−1
b
SAv-4
SAv-8
Av-4
2.8 × 103
−19.7b
−8.4c
+11.3
1.1 × 106b −34.5b
b
5.3 × 105
8.0 × 105
6.0 × 107
1 × 107
−32.7b
−33.7c
−44.4b
−39.9c
c
b
c
−11.7c
+29.3c
+22.0
+69.2
It has been established that the minimal distance a fully
extended spacer should span between two BT carboxylate
Av-8
carbonyls, in order to form an intramolecular complex to Av, is
6
˚
25 A. In the case of SAv the minimal distance for intramolecular
a 298 K, pH = 7.3. b Competition experiment with HABA as measured
by spectrophotometry (estimated error on log Ka of 5%). c Isothermal
titration calorimetry.
13
˚
complex formation was determined to be 20 A. The spacer
˚
in ligand 8, when fully extended, has a length of circa 27 A
(measured between the amide carbonyls), which should be
enough to allow the formation of an intramolecular complex
with both Av and SAv. Size exclusion chromatography and light
scattering measurements of a mixture of ligand 8 and SAv (or
Av) did not show any indication of oligomer formation.
hydrogen-bonding and hydrophobic interactions. Nevertheless,
they are of the same order of magnitude as other ligands such
as HABA.
The spectrophotometric titration curve of SAv with 8, in
competition with HABA, was fitted to a 1 : 1 binding
model corresponding to the formation of the divalent (in-
tramolecular) complex assuming independent binding. The
binding constant for the simultaneous binding of both units
of dimethylpropanediurea to half SAv (two proximal binding
sites) amounted to 1.1 × 106 M−1. From this value and the
intrinsic binding constant (Ki) of 2.8 × 103 M−1 an effective
In order to get further insight into the origin of the binding
process, microcalorimetric titrations were performed on Av and
SAv, and ligand 4. The thermodynamic parameters (Table 1)
show that the formation of both complexes (Av-4 and SAv-4)
has a large entropic contribution (65% and 57%, respectively), an
intermediate situation between the enthalpically driven binding
of BT and the entropically driven binding of HABA. A large
entropic contribution is usually an indication of the occurrence
of hydrophobic interactions, and could be interpreted by the
release of water molecules from the hydration shell of the ligand,
as well as from the binding pocket.11 Reduction of rotational
freedom of the valerate side chain of 4 upon binding to the
protein should lead to a negative contribution to DS◦. Inter-
estingly, however, electrospray mass spectrometry in aqueous
solution showed the presence of dimers and higher aggregates
of 4. In these self-associated structures both the urea and the
carboxylic acid groups are expected to be involved, as it is
observed in the crystal structure of similar compounds.12 This
self-association might also restrict the rotational freedom of the
valerate side chain of the free ligand (4), leading to an almost
neutral contribution to the DS◦ of binding.
˚
molarity (EM) of 70 mM is calculated. Based on the 20 A
distance between proximal binding sites in SAv, a divalent
ligand of optimal length is estimated to generate an effective
concentration8–14 for the second binding (Ceff) of approximately
99 mM. This value, together with the corresponding Ki, leads
to a higher limit of 1.6 × 106 M−1 for the expected binding
constant of SAv to an optimal divalent ligand based on 4,
assuming independent binding. The fact that the experimental
binding constant of 8 to SAv is so close to this higher limit
suggests that the length of the spacer in 8 is appropriate to
induce intramolecular complexation to SAv.
On the other hand, the spectrophotometric titration curve
of Av with 8 in competition with HABA, fitted to the same 1
: 1 binding model (Fig. 1), gave a binding constant of 6.0 ×
107 M−1. From this value and the Ki of 5.3 × 105 M−1, an
EM of 0.11 mM is calculated. In the case of Av, considering
On the other hand, the small value for DH◦ associated to
the complexation of ligand 4, compared to BT, points to the
formation of fewer and/or weaker hydrogen-bonds involving
the ureido and the carboxylate groups of 4. This is probably the
˚
the distance of approximately 25 A between proximal binding
2 3 9 4
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 3 9 3 – 2 3 9 5

View Article Online
Acknowledgements
This research was supported by Generalitat de Catalunya
(2001SGR00045 and 2002BEAI400180), by Ministerio de Cien-
cia y Tecnologia (BQU2002-02424) of Spain, and by the
Netherlands Science Foundation (NWO-CW).
References
1 M. Wilchek and E. A. Bayer, Methods Enzymol., 1990, 184, 5.
2 N. M. Green, Adv. Protein Chem., 1975, 29, 85.
3 (a) P. C. Weber, D. H. Ohlendorf, J. J. Wendoloski and F. R. Salemme,
Science., 1989, 243, 85; (b) P. C. Weber, J. J. Wendoloski, M. W.
Pantoliano and F. R. Salemme, J. Am. Chem. Soc., 1992, 114, 3197;
(c) P. C. Weber, M. W. Pantoliano, D. M. Simons and F. R. Salemme,
J. Am. Chem. Soc., 1994, 116, 2717.
Fig. 1 Absorption change at 500 nm in the titration of Av 8.7 × 10−6
M
and HABA 8.1 × 10−5 M with ligand 8. Phosphate buffer 0.1 M, pH =
7.3. The solid line represents the fit of the data to the (1 : 1) binding
model.
4 (a) H. Ebato, J. N. Herron, W. Mu¨ller, Y. Okahata, H. Ringsdorf and
P. Suci, Angew. Chem., Int. Ed. Engl., 1992, 31, 1087; (b) W. Mu¨ller,
H. Ringsdorf, E. Rump, G. Wildburg, X. Zhang, L. Angermaier, W.
Knoll, M. Liley and J. Spinke, Science., 1993, 262, 1706.
5 J. M. Hannink, J. J. L. M. Cornelissen, J. A. Farrera, P. Foubert, F. C.
De Schryver, N. A. J. M. Sommerdijk and R. J. M. Nolte, Angew.
Chem., Int. Ed., 2001, 40, 4732.
sites, a divalent ligand of optimal length is estimated to generate
a Ceff of about 51 mM. This Ceff value, together with the
corresponding Ki, leads to a higher limit of about 2.9 × 1010 M−1
for the expected binding constant of Av to an optimal divalent
ligand based on 4, assuming independent binding. The fact that
the experimental binding constant is about 400 times lower
than this higher limit suggest that the spacer of the divalent
ligand could still be improved in order to induce intramolecular
complexation to Av. The thermodynamic parameters (Table 1),
obtained in a microcalorimetric titration experiment, show that
the complexation of 8 to Av is strongly entropy-driven, with
an endothermic change in enthalpy of 29 kJ mol−1. This severe
enthalpy loss of the divalent ligand complexation compared to
the monovalent ligand complexation suggests that the fit of the
dimethylpropanediurea units of 8 into the binding sites is less
tight than in the case of the monovalent ligand complex. It
may also be that the spacer needs to be so close to the protein
surface that both species are dehydrated strongly, leading to
an endothermic enthalpy contribution but a gain in entropy.
The change in entropy for the complexation of 8 to Av is
more than twice the value found for the complexation of 4
to Av, suggesting that the restrictions in orientation of the
spacer upon complexation are more than compensated by the
proposed dehydration of both spacer and protein surface. The
data in Table 1 suggests that, in order to induce intramolecular
complexation to Av, a divalent ligand with a longer spacer might
be more appropriate. This aspect is currently under investigation.
6 N. M. Green, Methods Enzymol., 1990, 184, 51.
7 (a) D. M. Taylor, H. Fukushima and H. Morgan, Supramol. Sci.,
1995, 2, 75; (b) D. S. Wilbur, P. M. Pathare, D. K. Hamlin and S. A.
Weerawarna, Bioconjugate Chem., 1997, 8, 819; (c) K. J. Hamblett,
B. B. Kegley, D. K. Hamlin, M.-K. Chyan, D. E. Hyre, O. W. Press,
D. S. Wilbur and P. S. Stayton, Bioconjugate Chem., 2002, 13, 588.
8 (a) R. H. Kramer and J. W. Karpen, Nature., 1998, 395, 710; (b) P. I.
Kitov, H. Shimizu, S. W. Homans and D. R. Bundle, J. Am. Chem.
Soc., 2003, 125, 3284.
9 (a) R. J. Jansen, A. E. Rowan, R. de Gelder, H. W. Scheeren and
R. J. M. Nolte, Chem. Commun., 1998, 121; (b) R. J. Jansen, R. de
Gelder, A. E. Rowan, H. W. Scheeren and R. J. M. Nolte, J. Org.
Chem., 2001, 66, 2643.
10 N. M. Green, Biochem. J., 1965, 94, 23c.
11 (a) W. Blokzijl and J. B. F. N. Engberts, Angew. Chem., Int. Ed. Engl.,
1993, 32, 1545; (b) S. Otto and J. B. F. N. Engberts, Org. Biomol.
Chem., 2003, 1, 2809.
12 K. A. Lyssenko, D. G. Gov, A. N. Kravchenko, I. E. Chikunov, O. V.
Lebedev and N. N. Makhova, Mendeleev Commun., 2004, 1.
13 W. A. Hendrickson, A. Pahler, J. L. Smith, Y. Satow, E. A. Merritt
and R. P. Phizackerley, Proc. Natl. Acad. Sci. USA, 1989, 86,
2190.
14 (a) A. Mulder, T. Auletta, A. Sartori, S. Del Ciotto, A. Casnati, R.
Ungaro, J. Huskens and D. N. Reinhoudt, J. Am. Chem. Soc., 2004,
126, 6627; (b) A. Mulder, J. Huskens and D. N. Reinhoudt, Org.
Biomol. Chem., 2004, 2, 3409.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 3 9 3 – 2 3 9 5
2 3 9 5