Heparin versus DNA: Chiral preferences in polyanion binding to self-assembled multivalent (SAMul) nanostructures

Article abstract of DOI:10.1021/jacs.5b04344

This communication presents simple cationic self-assembling multivalent (SAMul) first generation dendrons based on l or d lysine, which form identical nanoscale assemblies in terms of dimensions and charge densities but toward which DNA and heparin exhibit different chiral binding preferences. However, higher generation dendrons with larger hydrophilic head groups are bound identically by these polyanions, irrespective of chirality. We propose that well-organized chiral ligands on the surface of self-assembled nanostructures can exhibit enantioselective polyanion binding. This demonstrates that small structural changes can be amplified by self-assembly and impact on nanoscale binding.

Full text of DOI:10.1021/jacs.5b04344

Communication
pubs.acs.org/JACS
Heparin versus DNA: Chiral Preferences in Polyanion Binding to Self-
Assembled Multivalent (SAMul) Nanostructures
Stephen M. Bromfield and David K. Smith*
Department of Chemistry, University of York, Heslington, York YO10 5DD, U.K.
S
* Supporting Information
previously examined,13 finding applications in areas such as
enantiomeric purification14 and asymmetric catalysis.15 Individ-
ABSTRACT: This communication presents simple cati-
onic self-assembling multivalent (SAMul) first generation
dendrons based on L or D lysine, which form identical
nanoscale assemblies in terms of dimensions and charge
densities but toward which DNA and heparin exhibit
different chiral binding preferences. However, higher
generation dendrons with larger hydrophilic head groups
are bound identically by these polyanions, irrespective of
chirality. We propose that well-organized chiral ligands on
the surface of self-assembled nanostructures can exhibit
enantioselective polyanion binding. This demonstrates that
small structural changes can be amplified by self-assembly
and impact on nanoscale binding.
ual DNA strands can also selectively bind helically comple-
mentary chiral peptide nucleic acids.16 Similarly, heparin can
discriminate chiral substrates, for example in capillary electro-
phoresis, where it has been used as a mobile phase additive.17
However, work from Wang and Rabenstein suggested
enantiomeric peptides interacted with heparin in an identical
manner.18 There have been studies in which heparin has been
used to compete with DNA binding,19 and very recently, CD was
used to probe interactions between a synthetic ligand and these
polyanions.20 However, chiral polyanion discrimination remains
rare, as does chiral recognition using micelles in cases where self-
assembly is a prerequisite for effective multivalent binding across
the surface.
We decided to probe self-assembling chiral cationic ligands for
binding similar polyanions (Figure 1) and chose lysine as the
olyanions are of vital importance in biological systems,
Pwhere they dominate many environments (indeed, much of
biology can be considered a “polyanion world”), and as such,
selective recognition is crucial.1 In electrostatic binding, it is
frequently argued that selectivity is difficult, with charge density
being the dominant factor which controls binding.2 However,
key recent studies have focused on the anisotropic and
stoichiometric organization of charge, and its consequent ability
to achieve more selective binding than might initially have been
expected.3
Multivalent platforms are an ideal way of achieving
interactions with nanoscale biological targets,4 in order to
manipulate biological processes or intervene as therapeutics in
nanomedicine.5 An approach to increase the synthetic
accessibility of multivalent systems is to use a self-assembling
multivalent (SAMul) approach, which only requires the synthesis
of smaller “drug-like” monomer units that self-assemble to
multiply up the ligand numbers presented to the target for
binding.6 The first example was reported by Whitesides and co-
workers who assembled glycoclusters able to intervene in viral
adhesion.7 A number of other researchers have since built on this
approach.8 We adopted the SAMul approach to develop families
of molecules targeted at DNA9 and heparin,10 both of which have
therapeutic relevance.11,12 As a result, we became interested in
the factors that govern selectivity, but given that our SAMul
systems simply display cationic ligands, we were unsure of the
extent to which they could employ “precisely defined”
interactions. As such, we wanted to determine whether the
SAMul approach could achieve any polyanion discrimination.
Given the inherent chirality of DNA and heparin, we thought
chiral SAMul systems might express some sort of binding
preference. The binding of chiral ligands to DNA has been
Figure 1. Monomer units 1L/D and 2L/D self-assemble to generate
multivalent ligand arrays for polyanion binding.
chiral, cationic surface ligand owing to its biological ubiquity in
polyanion-binding proteins.21 Lysine has been shown to be
among the most effective heparin binding groups.22 We then
modified our previously reported synthetic approach to SAMul
ligands10 to incorporate lysine, synthesizing compounds 1L and
1D (Figure 1), and control compound 1, unmodified with a
hydrophobic chain, which is incapable of self-assembly. Second
Received: April 27, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b04344
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
generation target molecules were synthesized in a related
convergent manner. Full experimental methods and character-
ization can be found in the Supporting Information (SI).
Chiralities of 1L/1D and 2L/2D were examined by circular
dichroism (CD) spectroscopy and shown to be equal and
opposite.
The ability of 1L and 1D to self-assemble in water was
examined using a Nile Red encapsulation assay,23 and critical
micelle concentrations (CMCs) of 29 9 and 27 13 μM,
respectively, were derived (see SI), essentially identical. For 2L
and 2D these values were 25 8 and 20 6 μM, respectively.
Control compounds 1 and 2 did not self-assemble at
concentrations up to 1 mM. The morphologies of 1L and 1D
self-assemblies were characterized by transmission electron
microscopy (TEM). Both 1L and 1D appeared to form spherical
micellar assemblies with approximate diameters of 8 nm (Figure
2). Compounds 2L and 2D also formed micelles ca. 9−11 nm in
Table 1. DLS Data for 1L and 1D under Different Electrolytic
Conditions
diameter (nm)
compd
Tris
Tris + NaCl
zeta potential
1L
1D
2L
2D
(7.6 0.3)
(7.8 0.2)
N/A
(9.0 0.3)
(8.2 0.3)
(6.0 0.2)
(6.4 0.3)
(+33.0 1.4)
(+26.3 8.4)
(+29.0 1.3)
(+26.4 6.9)
N/A
effect.10b DLS suggested that in solution, the self-assemblies
formed by 2L and 2D, have diameters of about 6 nm, suggesting
the larger hydrophilic head groups limit the growth of larger
nanostructures, owing to their volume requirement.
To quantify their relative heparin binding abilities, the
compounds were tested using our recently reported Mallard
Blue (MalB) competition assay in which displacement of MalB
from heparin by addition of the nanoscale binder, as monitored
by UV−vis spectroscopy, indicates the extent of binding.25 The
DNA binding abilities were also examined using a competition
assay involving ethidium bromide (EthBr) as the indicator dye,
the fluorescence of which is quenched on displacement.26
Titration curves are shown in Figure 3. All assays had to be
performed on freshly prepared samples, owing to the relatively
fast degradation of the compounds.9c,27 The analysis of the data
from this kind of assay is challenging for several reasons.
Polyanions such as heparin and DNA have multiple binding sites
along the polymer chain. The first assemblies that bind to the
biopolymers therefore do so with significantly different affinity to
the last assemblies; as more cationic assemblies bind to the
polyanion, charge−charge repulsions between them mean the
binding becomes less effective. Furthermore, heparin and DNA
are polydisperse in terms of their constitution. This problem is
particularly acute for heparin, which consists of a number of
different saccharides, all of which have different patterns of
anionic functionalization.28 Hence, there are significantly
variable binding sites along the polymer. As such, it is difficult,
if not meaningless, to extract a simple “binding constant” or
thermodynamic parameters from these assays. For simple
comparison of relative binding performances, we calculated
charge excess values (CE50) at 50% dye displacement (Table 2).
This reflects how much cationic charge (relative to polyanion
charge) is required to displace 50% of the indicator dye from its
complex with the polyanion (a lower number indicates a greater
efficiency and “better” binding). The CE50 value reflects the
binding when ca. half of the biopolymer has been bound by
SAMul nanostructures.
Figure 2. TEM images of 1L (left) and 1D (right) in the absence (top)
and presence of heparin (middle) and DNA (bottom).
diameter; however, accurate sizing was complicated by the
tendency of the sample to deteriorate under the electron beam
(see SI). In the presence of heparin, in each case the micelles
became structured in a pseudocrystalline array and had apparent
diameters of ca. 7 nm. Conceptually related structured arrays
have previously been reported by ourselves,10 and also by
Kostiainen and co-workers;24 the spherical polycations can be
considered to be packing with the linear polyanions in a
nanoscale ionic-lattice-like arrangement.
Nonassembling molecules 1 and 2 were unable to displace the
indicator dyes from their complexes. Control compound 2 was
slightly better than 1, perhaps as a consequence of it having
double the number of ligands. As such, in the absence of self-
assembly the individual lysine-based ligands cannot bind these
polyanions to any significant degree. In contrast, both 1L and 1D
were able to displace the indicator dyes in both assays; it is
therefore evident that the SAMul effect is responsible for binding
in each case.
Dynamic light scattering (DLS) measurements were used to
estimate the solution-phase dimensions (Table 1). In 10 mM
Tris HCl buffer, 1L and 1D formed aggregates ca. 7.7 nm in
diameter, in good agreement with TEM. On the addition of 150
mM NaCl, the aggregates enlarged, with diameters of ca. 8−9
nm. This is due to a combination of salt-mediated charge
screening at the micellar surface and an enhanced hydrophobic
The data further suggest that 1D (1.13 0.19) uses its charges
more efficiently than 1L (1.94 0.38) when binding heparin,
while the opposite is true when binding DNA, i.e., 1L (1.99
0.54) is more effective than 1D (3.51 0.37). These differences
are well beyond the error range and were repeated across
multiple runs. As such, they can be considered significant. The
SAMul nanostructures formed by 1L and 1D have identical sizes
B
DOI: 10.1021/jacs.5b04344
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Figure 3. Characterization of the binders using the MalB displacement assay (heparin binding ability, top) and EthBr displacement assay (DNA binding
ability, bottom).
we suggest that there is less differentiation between “primary”
and “secondary” binding sites on DNA, and a consistent enantio-
preference is expressed throughout the titration as a result.
Compounds 2L and 2D displace 50% of MalB at charge
Table 2. CE50 Data for Binders Interacting with Heparin and
DNA Obtained by Displacement Assays with MalB and EthBr,
Respectively
ligand
heparin
DNA
efficiencies of 1.07
0.20 and 1.28
0.26; no significant
1
no binding
1.94 0.38
1.13 0.19
no binding
1.07 0.20
1.28 0.26
no binding
1.99 0.54
3.51 0.37
no binding
2.15 0.31
2.03 0.25
difference between enantiomers (Table 2). Indeed, it is clear that
this is the case across the whole titration (Figure 3). It therefore
appears that despite the same chiral ligands being present in the
second generation binders, this chiral information is less
apparent to heparin. Similar trends were observed for DNA
binding; compounds 2L and 2D bind DNA with almost identical
efficiencies and identical titration curves, indicating chirality has
no influence. We suggest that increased crowding at the surface
of the dendritic structure on the smaller self-assemblies formed
by 2L and 2D, as a consequence of the larger, more sterically
hindered hydrophilic ligands, masks the differences in chiral
expression.
It should be noted that all self-assembling ligands achieved
effective multivalent polyanion binding at concentrations below
their CMC values. The presence of polyanions such as DNA can
lower the effective CMC; indeed, such effects are well-known for
ionic surfactants.30 Such a phenomenon would suggest that
multivalency-enhanced self-assembly is operating here, in
addition to self-assembly enhanced multivalency; i.e., these two
mechanisms reinforce one another. Taken overall, these data
suggest that, in some cases, these two biological polyanions
exhibit chiral preferences when presented with enantiomeric
versions of the same SAMul binding unit; noteworthy as it
contradicts observations that suggested such interactions
primarily depend on charge density.2,31
1L
1D
2
2L
2D
and charge densities, and their only difference is chirality,
suggesting that the spatial arrangement of the cationic charges is
important when binding these polyanions and indicating some
degree of chiral recognition.
Considering the titration curves for 1L and 1D more fully, it is
worth noting that for heparin binding, when only small amounts
of SAMul nanostructures are present, there is little difference in
affinity between 1L and 1D, as reflected by the initially
overlapping binding profiles in the titration curve in Figure 3.
This would suggest that the “primary” binding sites on heparin
from which MalB can be displaced show no preferential binding
between enantiomers and bind based on charge density.
However, as more of the SAMul system is added, and about
one-third of the MalB dye has been displaced, the differences
between 1L and 1D become significant and the titration curves
diverge (as also reflected in the CE50 value at 50% displacement
of dye). We therefore suggest that for effective binding to these
“secondary” binding sites on heparin, more than charge density
matters, and chiral organization is significant. We have reported
previously that the stoichiometry of binder−heparin complexes
is important in determining the degree of binding;29 however,
this is the first time enantiodiscrimination has been observed. For
DNA binding, however, significant differences between 1L and
1D exist throughout the titration. The distribution of phosphate
anions down the DNA backbone is regular in this polymer
(unlike the high variability of anionic sites in heparin). As such,
In summary, SAMul is a useful approach for achieving some
selectivity in polyanion binding. Although the effects are quite
small, this is a very challenging problem, and this is a rare example
of enantioselective binding in molecular recognition between
nanoscale surfaces with different biomolecules exhibiting
different preferences. The appropriate organization of the
multivalent ligand array through self-assembly is important in
C
DOI: 10.1021/jacs.5b04344
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
20300. (d) Barnard, A.; Posocco, P.; Fermeglia, M.; Tschiche, A.;
transcribing the binding selectivity onto the nanoscale surface.
Innate differences between the molecular recognition potential
of heparin and DNA may have biological significance in the
polyanion rich biological environment, and we suggest these
results are important in understanding binding in biorelevant
conditions.
Calderon, M.; Pricl, S.; Smith, D. K. Org. Biomol. Chem. 2014, 12, 446−
455. (e) Tschiche, A.; Staedtler, A. M.; Malhotra, S.; Bauer, H.; Bottcher,
̈
C.; Sharbati, S.; Calderon, M.; Koch, M.; Zollner, T. M.; Barnard, A.;
Smith, D. K.; Einspanier, R.; Schmidt, N.; Haag, R. J. Mater. Chem. B
2014, 2, 2153−2167.
(10) (a) Rodrigo, A. C.; Barnard, A.; Cooper, J.; Smith, D. K. Angew.
Chem., Int. Ed. 2011, 50, 4675−4679. (b) Bromfield, S. M.; Posocco, P.;
Chan, C. W.; Calderon, M.; Guimond, S. E.; Turnbull, J. E.; Pricl, S.;
Smith, D. K. Chem. Sci. 2014, 5, 1484−1492.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.5b04344.
■
S
(11) Srinivas, R.; Samanta, S.; Chaudhuri, A. Chem. Soc. Rev. 2009, 38,
3326−3338.
(12) Bromfield, S. M.; Wilde, E.; Smith, D. K. Chem. Soc. Rev. 2013, 42,
9184−9185.
Synthetic methods, characterization data, assay methods,
CMC and anion binding data, compound degradability,
and TEM images (PDF)
(13) Qu, X. G.; Trent, J. O.; Fokt, I.; Priebe, W.; Chaires, J. B. Proc.
Natl. Acad. Sci. U. S. A. 2000, 97, 12032−12037.
(14) Michaud, M.; Jourdan, E.; Villet, A.; Ravel, A.; Grosset, C.; Peyrin,
E. J. Am. Chem. Soc. 2003, 125, 8672−8679.
AUTHOR INFORMATION
Corresponding Author
*david.smith@york.ac.uk
■
(15) Roelfes, G. Mol. BioSyst. 2007, 3, 126−135.
(16) (a) Corradini, R.; Sforza, S.; Tedeschi, T.; Marchelli, R. Chirality
2007, 19, 269−294. (b) Menchise, V.; De Simone, G.; Tedeschi, T.;
Corradini, R.; Sforza, S.; Marchelli, R.; Capasso, D.; Saviano, M.;
Pedone, C. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 12021−12026.
(17) (a) Stalcup, A. M.; Agyei, N. M. Anal. Chem. 1994, 66, 3054−
3059. (b) Nishi, H.; Kuwahara, Y. J. Biochem. Biophys. Methods 2001, 48,
89−102.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Meg Stark (Department of Biology, York) for TEM
and Marcelo Calderon (FU Berlin) for some DLS. This research
was funded by University of York, BBSRC (DTA scheme), and a
UoY Postdoctoral Fellowship for S.M.B.
(18) Wang, J.; Rabenstein, D. L. Biochemistry 2006, 45, 15740−15747.
(19) (a) Xu, Y.; Szoka, F. C. Biochemistry 1996, 35, 5616−5623.
(b) Ruponen, M.; Yla-Herttuala, S.; Urtti, A. Biochim. Biophys. Acta,
Biomembr. 1999, 1415, 331−341. (c) Ainalem, M. L.; Bartles, A.; Muck,
J.; Diaz, R. S.; Carnerup, A. M.; Zink, D.; Nylander, T. PLoS One 2014, 9,
e92692.
(20) Zsila, F. Chirality 2015, DOI: 10.1002/chir.22471.
(21) Capila, I.; Linhardt, R. J. Angew. Chem., Int. Ed. 2002, 41, 390−
412.
(22) (a) Mecca, T.; Consoli, G. M. L.; Geraci, C.; La Spina, R.;
Cunsolo, F. Org. Biomol. Chem. 2006, 4, 3763−3768. (b) Choi, S.;
Clements, D. J.; Pophristic, V.; Ivanov, I.; Vemparala, S.; Bennett, J. S.;
Klein, M. L.; Winkler, J. D.; DeGrado, W. E. Angew. Chem., Int. Ed. 2005,
44, 6685−6689.
(23) Stuart, M. C. A.; van de Pas, J. C.; Engberts, J. B. F. N. J. Phys. Org.
Chem. 2005, 18, 929−934.
REFERENCES
■
(1) (a) Jones, L. S.; Yazzie, B.; Middaugh, C. R. Mol. Cell. Proteomics
2004, 3, 746−769. (b) Kayitmazer, A. B.; Seeman, D.; Minsky, B. B.;
Dubin, P. L.; Xu, Y. Soft Matter 2013, 9, 2553−2583.
(2) (a) Szelke, H.; Schubel, S.; Harenberg, J.; Kramer, R. Bioorg. Med.
Chem. Lett. 2010, 20, 1445−1447. (b) Li, D.; Wagner, N. J. J. Am. Chem.
Soc. 2013, 135, 17547−17555.
̈
̈
(3) (a) Chen, K.; Xu, Y.; Rana, S.; Dubin, P. L.; Rotello, V. M.; Sun, L.;
Guo, X. Biomacromolecules 2011, 12, 2552−2561. (b) Wang, S.; Chen,
K.; Li, L.; Guo, X. Biomacromolecules 2013, 14, 818−827.
(4) Fasting, C.; Schalley, C. A.; Weber, M.; Seitz, O.; Hecht, S.; Koksch,
B.; Dernedde, J.; Graf, C.; Knapp, E. W.; Haag, R. Angew. Chem., Int. Ed.
2012, 51, 10472−10498.
(5) Uhlenheuer, D. A.; Petkau, K.; Brunsveld, L. Chem. Soc. Rev. 2010,
39, 2817−2826.
(6) Barnard, A.; Smith, D. K. Angew. Chem., Int. Ed. 2012, 51, 6572−
6581.
(7) Kingery-Wood, J. E.; Williams, K. W.; Sigal, G. B.; Whitesides, G.
M. J. Am. Chem. Soc. 1992, 114, 7303−7305.
(8) (a) Kim, B. S.; Hong, D. J.; Bae, J.; Lee, M. J. Am. Chem. Soc. 2005,
127, 16333−16337. (b) Dreher, M. R.; Simnick, A. J.; Fischer, K.; Smith,
R. J.; Patel, A.; Schmidt, M.; Chilkoti, A. J. Am. Chem. Soc. 2008, 130,
(24) Mikkila, J.; Rosilo, H.; Nummelin, S.; Seitsonen, J.; Ruokolainen,
J.; Kostiainen, M. A. ACS Macro Lett. 2013, 2, 720−724.
(25) (a) Bromfield, S. M.; Barnard, A.; Posocco, P.; Fermeglia, M.;
Pricl, S.; Smith, D. K. J. Am. Chem. Soc. 2013, 135, 2911−2914.
(b) Bromfield, S. M.; Posocco, P.; Fermeglia, M.; Pricl, S.; Rodríguez-
́
Lopez, J.; Smith, D. K. Chem. Commun. 2013, 49, 4830−4832.
(26) (a) Cain, B. F.; Baguley, B. C.; Denny, W. A. J. Med. Chem. 1978,
21, 658−668. (b) Boger, D. L.; Fink, B. E.; Brunette, S. R.; Tse, W. C.;
Hedrick, M. P. J. Am. Chem. Soc. 2001, 123, 5878−5891.
́
(27) Gillies, E. R.; Dy, E.; Frechet, J. M. J.; Szoka, F. C. Mol.
Pharmaceutics 2005, 2, 129−138.
(28) Rabenstein, D. L. Nat. Prod. Rep. 2002, 19, 312−331.
687−694. (c) Muller, M. K.; Brunsveld, L. Angew. Chem., Int. Ed. 2009,
̈
(29) Bromfield, S. M.; Posocco, P.; Fermeglia, M.; Tolosa, J.; Herreros-
48, 2921−2924. (d) Rosenzweig, B. A.; Ross, N. T.; Tagore, D. M.;
Jayawickramarajah, J.; Saraogi, I.; Hamilton, A. F. J. Am. Chem. Soc. 2009,
131, 5020−5021. (e) Nalluri, S. K. M.; Voskuhl, J.; Bultema, J. B.;
Boekema, E. J.; Ravoo, B. J. Angew. Chem., Int. Ed. 2011, 50, 9747−9751.
(f) Dane, E. L.; Ballok, A. E.; O’Toole, G. A.; Grinstaff, M. W. Chem. Sci.
2014, 5, 551−557. (g) Montalvo, G. L.; Zhang, Y.; Young, T. M.;
Costanzo, M. J.; Freeman, K. B.; Wang, J.; Clements, D. J.; Magavern, E.;
Kavash, R. W.; Scott, R. W.; Liu, D. H.; DeGrado, W. F. ACS Chem. Biol.
2014, 9, 967−975.
Lop
́ ́
ez, A.; Pricl, S.; Rodriguez-Lopez, J.; Smith, D. K. Chem. - Eur. J.
2014, 20, 9666−9674.
(30) Li, D.; Wagner, N. J. J. Am. Chem. Soc. 2013, 135, 17547−17555.
(31) (a) Konop, A. J.; Colby, R. H. Langmuir 1999, 15, 58−65.
(b) Chiappisi, L.; Hoffmann, I.; Gradzielski, M. Soft Matter 2013, 9,
3896−3909.
(9) (a) Jones, S. P.; Gabrielson, N. P.; Pack, D. W.; Smith, D. K. Chem.
Commun. 2008, 4700−4702. (b) Jones, S. P.; Gabrielson, N. P.; Wong,
C. H.; Chow, H. F.; Pack, D. W.; Posocco, P.; Fermeglia, M.; Pricl, S.;
Smith, D. K. Mol. Pharmaceutics 2011, 8, 416−429. (c) Barnard, A.;
Posocco, P.; Pricl, S.; Calderon, M.; Haag, R.; Hwang, M. E.; Shum, V.
W. T.; Pack, D. W.; Smith, D. K. J. Am. Chem. Soc. 2011, 133, 20288−
D
DOI: 10.1021/jacs.5b04344
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX