Comparison of facially amphiphilic biaryl dendrimers with classical amphiphilic ones using protein surface recognition as the tool

Article abstract of DOI:10.1021/ja0622406

Facially amphiphilic biaryl dendrimers are compared with the more classical benzyl ether amphiphilic dendrimers for molecular recognition, using protein binding as the probe. The protein used for the proposed study is chymotrypsin (ChT). A generation-depe

Full text of DOI:10.1021/ja0622406

Published on Web 06/27/2006
Comparison of Facially Amphiphilic Biaryl Dendrimers with
Classical Amphiphilic Ones Using Protein Surface
Recognition as the Tool
Akamol Klaikherd, Britto S. Sandanaraj, Dharma Rao Vutukuri, and
S. Thayumanavan*
Contribution from the Department of Chemistry, UniVersity of Massachusetts,
Amherst, Massachusetts 01003
Received April 2, 2006; E-mail: thai@chem.umass.edu
Abstract: Facially amphiphilic biaryl dendrimers are compared with the more classical benzyl ether
amphiphilic dendrimers for molecular recognition, using protein binding as the probe. The protein used for
the proposed study is chymotrypsin (ChT). A generation-dependent binding affinity was observed with the
benzyl ether dendrimers, while the affinities were independent of generation in the case of the biaryl
dendrimers. Similarly, although the ligands incorporated in both dendrons are the same, the biaryl dendrimers
are able to bind more proteins compared to the benzyl ether dendrimers. For example, G3-dendron of
biaryl dendrimer can bind six molecules of chymotrypsin, whereas G3-analogue of benzyl ether dendrimers
can bind only three molecules of chymotrypsin. This result is consistent with our hypothesis that the internal
layers of the facially amphiphilic biaryl dendrons are solvent-exposed and accessible for recognition. In
addition, the systematic size differences in dendrons were also used to gain insights into the substrate
selectivity that the enzyme gains upon binding to a ligand scaffold.
Introduction
only the peripheries of the dendrimers that are decorated with
the ligand functionalities.4 This is mainly because it is only the
The globular shape of dendrimers combined with the fact
that these molecules can be obtained with a high degree of
control in molecular weight have made them attractive candi-
dates for supramolecular chemistry.1,2 One of the major
advantages that dendrimers provide is the ability to display
multiple copies of ligand functionalities to bind to a receptor,
which takes advantage of features such as polyvalent interac-
tions.3 In cases where such a polyvalency is investigated, it is
peripheral functionalities that are thought to be fully solvent-
exposed and therefore can make available multiple copies of a
ligand for recognition. We have recently reported an amphiphilic
dendrimer in which every repeat unit within the dendrimer
backbone contains a hydrophilic and a hydrophobic functional-
ity.5 We had suggested that, in contrast to the classical
amphiphilic dendrimers,6-8 the biaryl ones adopt a conformation
(4) (a) Kensinger, R. D.; Yowler, B. C.; Benesi, A. J.; Schengrund, C.-L.
Bioconjugate Chem. 2004, 15, 349-358. (b) Wolfenden, M. L.; Cloninger,
M. J. J. Am. Chem. Soc. 2005, 127, 12168-12169. (c) Sashiwa, H.;
Shigemasa, Y.; Roy, R. Macromolecules 2001, 34, 3905-3909. (d)
Majoros, I. J.; Thomas, T. P.; Mehta, C. B.; Baker, J. R., Jr. J. Med. Chem.
2005, 48, 5892-5899. (e) Thoma, G.; Katopodis, A. G.; Voelcker, N.;
Duthaler, R. O.; Streiff, M. B. Angew. Chem., Int. Ed. 2002, 41, 3195-
3198. (f) Zanini, D.; Roy, R. Bioconjugate Chem. 1997, 8, 187-192. (g)
Gitsov, I.; Lin, C. Curr. Org. Chem. 2005, 9, 1025-1051 (h) Woller, E.
K.; Cloninger, M. J. Biomacromolecules 2001, 2, 1052-1054.
(1) For some general references to dendrimers, see: (a) Newkome, G. R.;
Moorefield, C. N.; Vo¨gtle, F. Dendrimers and Dendrons: Concepts,
Syntheses, Applications, 2nd ed.; Wiley-VCH: Weinheim, 2001. (b) Fre´chet,
J. M. J.; Tomalia, D. A. Dendrimers and Other Dendritic Polymers; John
Wiley & Sons: New York, 2001. (c) Grayson, S. M.; Fre´chet, J. M. J.
Chem. ReV. 2001, 101, 3819-3868. (d) Tomalia, D. A.; Fre´chet, J. M. J.
J. Polym. Sci., Part A: Polym. Chem. 2002, 40, 2719-2728. (e) Fre´chet,
J. M. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 3713-3725. (f)
Bosman, A. W.; Janssen, H. M.; Meijer, E. W. Chem. ReV. 1999, 99, 1665-
1688. (g) Svenson, S.; Tomalia, D. A. AdV. Drug DeliVery ReV. 2005, 57,
2106-2129. (h) Cloninger, M. J. Curr. Opin. Chem. Biol. 2002, 6, 742-
748. (i) Majoral, J.-P.; Caminade, A.-N. Acc. Chem. Res. 2004, 37, 341-
348. (j) Fischer, M.; Vo¨gtle, F. Angew. Chem., Int. Ed. 1999, 38, 884-
905. (k) Gestermann, S.; Hesse, R.; Windisch, B.; Vo¨gtle, F. Stimulating
Concepts in Chemistry; Wiley-VCH: Weinheim, 2000.
(2) For reviews on supramolecular aspects of dendrimers, see; (a) Fre´chet, J.
M. J. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 4783-4787. (b) Newkome,
G. R.; He, E.; Moorefield, C. N. Chem. ReV. 1999, 99, 1689-1746. (c)
Zeng, F.; Zimmerman, S. C. Chem. ReV. 1997, 97, 1681-1712. (d) Smith,
D. K.; Diederich, F. Top. Curr. Chem. 2000, 210, 183-227. (e) Ong, W.;
Go´mez-Kaifer, M.; Kaifer, A. E. Chem. Commun. 2004, 1677-1683. (f)
Smith, D. K. Chem. Commun. 2006, 34-44.
(5) (a) Vutukuri, D. R.; Basu, S.; Thayumanavan, S. J. Am. Chem. Soc. 2004,
126, 15636-15637. (b) Bharathi, P.; Zhao, H.; Thayumanavan, S. Org.
Lett. 2001, 3, 1961-1964.
(6) Amphiphilic dendrimers based on convergent growth: (a) Hawker, C. J.;
Wooley, K. L.; Fre´chet, J. M. J. J. Chem. Soc., Perkin Trans. 1 1993, 1287-
1297. (b) Jayaraman, M.; Fre´chet, J. M. J. J. Am. Chem. Soc. 1998, 120,
12996-12997. (c) Liu, M.; Kono, K.; Fre´chet, J. M. J. J. Controlled Release
2000, 65, 121-131. (d) Pesak, D. J.; Moore, J. S. Macromolecules 1997,
30, 6467-6482. (e) Luman, N. R.; Smeds, K. A.; Grinstaff, M. W. Chem.s
Eur. J. 2003, 9, 5618-5626.
(7) Amphiphilic dendrimers based on divergent growth: (a) Newkome, G. R.;
Moorefield, C. N.; Baker, G. R.; Saunders: M. J.; Grossman, S. H. Angew.
Chem., Int. Ed. Engl. 1991, 30, 1178-1180. (b) Newkome, G. R.;
Moorefield, C. N.; Baker, G. R.; Johnson, A. L.; Behera, R. K. Angew.
Chem., Int. Ed. Engl. 1991, 30, 1176-1178. (c) Kuzdzal, S. A.; Monnig,
C. A.; Newkome, G. R.; Moorefield, C. N. J. Chem. Soc., Chem. Commun.
1994, 2139-2140. (d) Pan, Y.; Ford, W. T. Macromolecules 1999, 32,
5468-5470. (e) Newkome, G. R.; Young, J. K.; Baker, G. R.; Potter, R.
L.; Audoly, L.; Cooper D.; Weis, C. D.; Morris K.; Johnson, C. S., Jr.
Macromolecules 1993, 26, 2394-2396.
(3) (a) Mammen, M.; Choi, S. K.; Whitesides, G. M. Angew. Chem., Int. Ed.
1998, 37, 2755-2794. (b) Gestwicki, J. E.; Cairo, C. W.; Strong, L. E.;
Oetjen, K. A.; Kiessling, L. L. J. Am. Chem. Soc. 2002, 124, 14922-
14933. (c) Badjic, J. D.; Nelson, A.; Cantrill, S. J.; Turnbull, W. B.;
Stoddart, J. F. Acc. Chem. Res. 2005, 38, 723-732. (d) Kitov, P. I.; Bundle,
D. R. J. Am. Chem. Soc. 2003, 125, 16271-16284. (e) Ro¨ckendorf, N.;
Lindhorst, T. K. Top. Curr. Chem. 2001, 217, 98-135.
9
10.1021/ja0622406 CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 9231-9237
9231

A R T I C L E S
Klaikherd et al.
in such a way that all the hydrophilic functionalities throughout
the molecule’s backbone are presented on the solvent-exposed
exterior in water. In such a conformation, it is thought that all
the hydrophobic functionalities are presented in the interior of
the assembly. If this were the case, the number of possible ligand
functionalities that can be presented within a dendrimer
backbone is nearly twice as much as that of the classical
dendrimers, within a generation. For example in a G-3 dendron,
while one could have 8 copies of a ligand in the classical
dendrimers, it is possible to have 15 copies of the ligand in our
dendrimers, if our structural hypothesis is correct. Could we
then use a ligand-protein interaction as a probe to test this
structural hypothesis? We address this issue by estimating the
binding ratio of these two classes of dendrimers with the protein,
R-chymotrypsin (ChT). Similarly, using dynamic light scattering
studies, we have previously shown that our biaryl dendrimers
aggregate to form particles ranging from 10 to 40 nm. Thus, a
large number of ligand functionalities are preorganized on the
surface of the particle in the biaryl dendrimers in solution. On
the other hand, the Fre´chet-type classical amphiphilic dendrimers
do not form such aggregates (vide infra). Therefore, in the latter
case, there is a systematic, incremental change in the number
of ligand functionalities presented on a unimolecular dendrimer
surface in the latter case. By comparing the two scaffolds, it is
interesting to ask what is the critical size needed for an efficient
binding that compares to a scaffold that contains numerous
ligands on a 10-40 nm particle? This paper concerns the
comparison of our biaryl-based amphiphilic dendrimers with
the classical Fre´chet-type benzyl ether amphiphilic dendrimers8a,9
to gain insights into such structural requirements for recognition.
Simple anionic functionalities such as carboxylic acids have
been previously used to recognize the cationic patch of ChT
that surrounds the active site of the protein.10 In our own
previous work, we had used amphiphilic homopolymers con-
taining carboxylate units to bind to ChT. In that work, our
objective was to identify the effect of the polymer-protein
interactions upon the structure and function of the enzyme. This
work by us and related work by others have established that
poly-carboxylate scaffolds are effective in binding and modify-
ing the enzymatic activity of ChT. In the present work with
dendrimers: (i) We utilize the binding ability to protein surfaces
as a probe to understand and compare the conformational
features of two classes of dendrimers. That is, others and we
have previously used artificial molecular scaffolds to influence
the properties of a biological macromolecule. In the present case,
we use the influence of binding on the behavior of the biological
macromolecule to understand the properties of the artificial
scaffold. (ii) Since we can increase the number of ligands highly
systematically in the fully monodisperse dendrimer scaffold, the
molecular level details of understanding the poly(ligand)-
protein interactions become better. For this purpose, we compare
the results from the two dendritic scaffolds with linear polymers,
whenever appropriate.
Structures of both our facially amphiphilic dendrimers 1-4,
and the Fre´chet-type benzyl ether dendrimers 5-7 are shown
in Chart 1. Syntheses and characterization of all new molecules
are available in the Supporting Information. Note that the biaryl
dendrimers are based on arylalkyl ether connectivity, and
therefore the benzyl ether dendrimers are appropriate scaffolds
for comparison.
Results and Discussion
We first estimated the binding affinity and the binding ratio
of dendrons vs chymotrypsin using the enzymatic activity of
the protein against chromogenic substrates such as S1. The
enzyme cleaves the amide bond at the C-terminal of phenyl-
alanine in S1 to afford p-nitroaniline, which can be monitored
using its distinct absorption spectrum at 405 nm. Note that the
determination of binding affinities and binding ratios were
achieved using well-established protocols for chymotrypsin with
such substrates.10i,11 Change in enzymatic activity upon binding
to the dendrimer was determined by adding an S1 stock solution
to a preincubated ChT-dendrimer solution. The experiments
were carried out with different concentrations of dendrimer,
while the ChT concentration was kept constant at 3.2 µM. A
control experiment was performed under identical conditions
without addition of dendrimer. The binding constant and ratio
of the ChT/dendron were obtained by plotting the activity of
ChT against dendron concentrations. No inhibition effect was
observed with the small molecule G0 dendron (1) (Figure 1).
The activity of ChT decreased significantly with 2, with 94%
inhibition observed at 48 µM and then saturated at higher
concentrations. Dendrimers 3 and 4 exhibited 96% inhibition
at 9.6 µM and 1.9 µM concentrations, respectively (Figure 1).
The concentration of dendrons used in this study is well above
the critical micellar concentration (cmc). The binding ratios
obtained for each of these dendrons are also shown in Table 1.
Since the binding obtained here is based on electrostatic
interactions (vide infra), we have also qualitatively estimated
binding ratios using gel electrophoresis, the results of which
were consistent with the enzymatic assay.
(8) Linear hybrid amphiphilic dendrimers: (a) Gitsov, I.; Fre´chet, J. M. J. J.
Am. Chem. Soc. 1996, 118, 3785-3786. (b) Lambrych, K. R.; Gitsov, I.
Macromolecules 2003, 36, 1068-1074. (c) Ihre, H.; Padilla De Jesus, O.
L.; Fre´chet, J. M. J. J. Am. Chem. Soc. 2001, 123, 5908-5917. (d) Gitsov,
I.; Lambrych, K. R.; Remnant, V. A.; Pracitto, R. J. Polym. Sci. Part A:
Polym. Chem. 2000, 38, 2711-2727. (e) Fre´chet, J. M. J.; Gitsov, I.;
Monteil, T.; Rochat, S.; Sassi, J. F.; Vergelati, C.; Yu, D. Chem. Mater.
1999, 11, 1267-1274.
The binding ratio of the dendrimer to protein might seem
high at first sight. For example, the G-3 dendron 7 binds to the
protein in a 1:3.3 ratio. The size of a G-3 dendron is about
2-3 nm, while the size of ChT itself is about 4 nm. Arranging
three 4 nm particles three-dimensionally around another 2-3
nm particle with only a few contact points is plausible. First
we will compare the binding ratios between the two classes of
dendrons to investigate whether the internal carboxylates in the
biaryl dendrons are available for recognition with a macromo-
lecular receptor such as ChT. The binding ratios of dendrons 2
and 5 with ChT are 0.7 and 0.4, respectively, as shown in Table
(9) (a) Hawker, C. J.; Fre´chet, J. M. J. J. Am. Chem. Soc. 1990, 112, 7638-
7647. (b) Hawker, C. J.; Fre´chet, J. M. J. J. Am. Chem. Soc. 1992, 114,
8405-8413.
(10) (a) Verma, A.; Rotello, V. M. Chem. Commun. 2005, 301-312. (b) Park,
H. S.; Lin, Q.; Hamilton, A. D. J. Am. Chem. Soc. 1999, 121, 8-13. (c)
Jain, R. K.; Hamilton, A. D. Org. Lett. 2000, 2, 1721-1723. (d) Arvizo,
R. R.; Verma, A.; Rotello, V. M. Supramol. Chem. 2005, 17, 155-161.
(e) Fischer, N. O.; McIntosh, C. M.; Simard, J. M.; Rotello, V. M. Proc.
Natl. Acad. Sci. U.S.A. 2002, 99, 5018-5023. (f) Hong, R.; Emrick, T.;
Rotello, V. M. J. Am. Chem. Soc. 2004, 126, 13572-13573. (g) Fischer,
N. O.; Verma, A.; Goodman, C. M.; Simard, J. M.; Rotello, V. M. J. Am.
Chem. Soc. 2003, 125, 13387-13391. (h) Verma, A.; Simard, J. M.; Rotello,
V. M. Langmuir 2004, 20, 4178-4181. (i) Hong, R.; Fischer, N. O.; Verma,
A.; Goodman, C. M.; Emrick, T.; Rotello, V. M. J. Am. Chem. Soc. 2004,
126, 739-743.
(11) Sandanaraj, B. S.; Vutukuri, D. R.; Simard, J. M.; Klaikherd, A.; Hong,
R.; Rotello, V. M.; Thayumanavan, S. J. Am. Chem. Soc. 2005, 127,
10693-10698.
9
9232 J. AM. CHEM. SOC. VOL. 128, NO. 28, 2006

Facially and Classical Amphiphilic Biaryl Dendrimers
A R T I C L E S
Chart 1 a
a (a) Schematic representation of facially amphiphilic biaryl dendrimers and classical amphiphilic dendrimers of G3-generation; (b) chemical structures
of amphiphilic biaryl dendrimers 1-4; (c) chemical structures of classical amphiphilic dendrimers 5-7
1. Similarly, at higher generations, the biaryl dendrimers 3 and
4 bind more ChT molecules relative to the dendrimer 6 and 7.
This could be simply explained by the fact that the number of
carboxylic units presented in the biaryl dendrons 3 and 4 is
higher than that in the dendrons 6 and 7. To further illustrate
this feature, we have tabulated the number of carboxylic acid
groups involved per ChT molecule with each of the dendrons.
The trend among the biaryl dendrons 2-4 is very similar to
that observed with the benzyl ether dendrons 5-7. Note that
we have taken in to account the internal carboxylic acid moieties
in addition to the peripheral functionalities in this estimate.
Therefore, this result is taken to suggest that the carboxylates
in internal layers of dendrimer are available for recognition.
We also find that the number of carboxylic acid groups that
are needed for binding ChT decreases with generation in both
dendrons. The reason for this so-called dendritic effect is not
clear to us at this time.
The binding affinities of 2-4 toward ChT are also shown in
Table 1. First of all, it is clear that a critical number of covalently
tethered carboxylic acid functionalities is needed for any
9
J. AM. CHEM. SOC. VOL. 128, NO. 28, 2006 9233

A R T I C L E S
Klaikherd et al.
Figure 1. Inhibition study: Concentration dependent assay studies of ChT (3.2 µM); (a) amphiphilic biaryl dendrimers and (b) classical dendrimers. Assay
studies were carried out with various concentrations of dendron in the presence of 5 mM sodium phophate buffer pH 7.4. The solutions were incubated for
1 h. The percentage activity of ChT was plotted against the concentration of dendron (µM).
Table 1. Comparison of Number of Carboxylic Group, Biding
Ratio, and Binding Constant of Biaryl Dendrimers and Classical
ligands in the nonaggregated Fre´chet-type dendrons, we have
Dendrimers
With the slow increments in the preorganized carboxylate
the opportunity to ask the following: what is the critical number
no. of carboxylic
acid groups
binding ratio
no. of carboxylic
groups per ChT
dissociation
of carboxylate moieties that need to be preorganized in a ligand
scaffold to recognize the surface of ChT with a micromolar
binding affinity? That is, which of the Fre´chet-type dendrimers
matches the Kd observed with the particles formed from
dendrimers 2-4? It is important to point out that the micromolar
binding affinity was also observed with the recently reported
polymer nanoparticles.11 Therefore, it is reasonable to assume
this binding affinity as the ceiling for these types of molecules.
The G-1 dendron, 5, compared to 2 has a lower binding affinity
toward ChT. One could attribute this to the need for a critical
number of carboxylate groups that are needed for the binding.
However, note that, although the G-2 dendron 6 (containing
four carboxylate units) exhibits much higher affinity than
dendron 5, it is still much lower than our that of G-1 dendron
2 (containing three carboxylate units). Dendron 7, with eight
carboxylate units, seems to approach the micromolar binding
affinities of 2-4. This result suggests that this dendron
represents the size and number of ligands at which the binding
affinity reaches a maximum and then plateaus.
compd
(no. of ChT per dendron)
constant Kd (M)
1
2
3
4
5
6
7
1
3
7
15
2
4
n/a
0.7
2.0
5.7
0.4
1.2
3.3
n/a
4.4
3.5
2.6
5.0
3.3
2.4
n/a
5.78 × 10-6
1.47 × 10-6
6.02 × 10-7
2.03 × 10-2
1.81 × 10-5
3.61 × 10-6
8
significant binding of the protein. The biaryl 1 building block
unit which contains only one carboxylic acid functionality did
not exhibit any binding with the protein. Dendron 5, which
contains two carboxylic acid units, seems to be capable of
binding ChT, but with a very low binding affinity. On the other
hand, dendron 2 that contains three carboxylic acid groups shows
a micromolar binding affinity.
Although the difference in binding affinities is quite large
among 1, 2, and 5, the differences in binding affinities among
dendrons 2-4 are relatively small. This could be due to the
following feature of the biaryl dendrimers. We have shown that
amphiphilic biaryl dendrimers show aggregation in water to form
a micelle-type assembly approximately 10-40 nm in size.
Therefore, a large number of carboxylic units are preorganized
and presented to the protein. Once it is assembled into such a
nanoparticle, it is likely that the protein does not show any
difference in binding affinity with generation. For example,
when comparing 2 and 5, both dendrons bind ChT with a
binding ratio that requires more than one dendron per protein.
But, the binding affinity of 2 is much greater, since the
carboxylates are preorganized due to aggregation. Also, if our
hypothesis was true, generation dependence would be observed
with the Fre´chet-type dendrons 5-7, since these molecules do
not aggregate and therefore do not preorganize any more ligands
through aggregation. Indeed, generation dependence on binding
affinity was observed with 5-7, as shown in Table 1. We have
performed DLS studies on an aqueous solution of dendrons 5-7
to confirm that these molecules do not aggregate. These
dendrons do not exhibit any evidence of aggregation.
It is also interesting to compare the present results with our
previous results obtained with an amphiphilic homopolymer.
The binding affinity of dendron 4 is the same as that of the
amphiphilic homopolymer (7 × 10-7 M). However, the binding
ratio is higher with the polymer (1:10, polymer/ChT) than that
of dendron 4. It should also be taken into account that each
polymer chain contains approximately 74 carboxylic acid units
(degree of polymerization (DP) is 74). Therefore, if we compare
the binding ratio with respect to the number of carboxylic acid
units, the number of carboxylates required to bind ChT in
dendrimers is lesser than that in the linear polymer. These results
could be due to the inherent difference in shape and confor-
mational nature of the two macromolecules. This comparison
further illustrates the availability of the internal carboxylate
moieties for binding in the biaryl dendrimers.
Nature of Dendrimer-Protein Interaction. When studying
the interaction of an artificial molecular scaffold with protein,
it is important to understand the consequence of the binding
9
9234 J. AM. CHEM. SOC. VOL. 128, NO. 28, 2006

Facially and Classical Amphiphilic Biaryl Dendrimers
A R T I C L E S
Figure 2. CD studies: (a) amphiphilic biaryl dendrimers, (b) classical amphiphilic dendrimers; 3.2 µM ChT incubated with 5.0 µM dendrons. The solutions
were incubated in 5 mM sodium phosphate buffer pH 7.4 for 1 h. The spectra were compared with the ChT native and ChT thermo denature spectra.
Figure 3. Effect of ionic strength: (a) amphiphilic biaryl dendrimers, (b) classical amphiphilic dendrimers; the ChT was incubated with dendrons in the
presence of varied concentrations of NaCl. The data were normalized to account for the enhanced activity of ChT due to the salt effect.
events upon the structure of the protein. For example, we should
investigate whether the observed inhibition is due to partial or
complete denaturation of the proteins. To probe these possibili-
ties, circular dichroism (CD) experiments were carried out. The
CD spectrum of native ChT shows two characteristic peaks at
232 and 204 nm.12 Thermal denaturation of ChT results in the
loss of the peak at 232 nm, and a blue shift is observed for the
peak at 204 nm, as could be seen in Figure 2. The far-UV CD
spectra of all dendrimer/ChT complexes illustrated no significant
change in the CD spectrum. These results suggest that the
binding event does not result in denaturation of the protein.
Since the interaction is noncovalent, based on electrostatics,
and since the binding event does not denature the protein, it
should be possible to recover the enzymatic activity of the
protein by screening the electrostatic interaction in solution. A
common way of doing this is to increase the ionic strength of
the solution.10h,11 To investigate the possibility of the release
of ChT from the dendrimer surface, enzymatic activities of
dendrimer/ChT complexes were studied at different salt con-
centrations ranging from 0.01 to 0.5 M NaCl. Control experi-
ments were conducted using the same concentration of ChT
and NaCl without the dendrons for each data point. Each %
activity of ChT in Figure 3a and 3b was normalized to this
control experiment. The binding and inhibition effect of the
ChT-dendron complex were strongly dependent upon ionic
strength, as shown in Figure 3. The activity of ChT was
dramatically recovered by increasing the salt concentration from
0.01 to 0.2 M. Thus, it is clear that the preliminary binding of
the dendrimer to protein is based on electrostatic interaction
and is reversible. The efficiency of binding lessened due to
disruption of the electrostatic interaction by increasing the ionic
strength in solution.
From the studies so far, it is clear that both classes of
dendrimers interact with the protein through electrostatics and
that this interaction is reversible. However, the binding affinity
and binding ratio of this interaction are different for these two
classes of dendrimers. These have been attributed to the fact
that the biaryl dendrimers contain a carboxylate moiety in each
repeating unit and that the biaryl dendrimers are preorganized
into a micelle-type assembly. We were interested in finding out
whether this difference affects the function of the dendrimer-
protein complex. Also, the systematic differences in each
generation of the dendron provide us an opportunity to gain
certain insights into the nature of the interaction between ChT
and multivalent ligand scaffolds in general. For example,
(12) (a) Celej, M. S.; D’Andrea, M. G.; Campana, P. T.; Fidelio, G. D.; Bianconi,
M. L. Biochem. J. 2004, 378, 1059-1066. (b) Sreerama, N. R.; Woody,
W. Anal. Biochem. 2000, 287, 252-260.
9
J. AM. CHEM. SOC. VOL. 128, NO. 28, 2006 9235

A R T I C L E S
Klaikherd et al.
Figure 4. Substrate selectivity: (a) Activity of ChT (3.2 µM) with different substrates; (b and c) normalized activity of ChT (3.2 µM) with different
substrates. Condition: 3.2 µM ChT was incubated in 75 µM 1; other samples were incubated in 25 µM 2, 10.7 µM 3, 5 µM 4, 37.5 µM 5, 18.8 µM 6, and
9.4 µM 7 (all dendrons have the identical number of COOH groups ) 75 µM). (d) Normalized activity of ChT (3.2 µM) with 200 µM 5. Other samples,
ChT was incubated in 98 µM 6 and 47 µM 7, respectively. All solutions were incubated in the presence of 5 mM sodium phosphate buffer pH 7.4 for 1 h.
recently, it has been shown that an electrostatically based
complex of artificial molecular scaffolds with ChT results in
an enzymatic activity that is dependent on the charge of the
chromogenic substrate.10f,11 In those studies, it has been shown
that the negatively and positively charged substrates (S1 and
S3) exhibit significant inhibition and hyperactivity, respectively,
in the presence of the negatively charged ligand scaffolds. While
these results are attributed to the repulsive or attractive interac-
tion between the substrate and the ligand, the inhibition observed
with neutral substrates has been attributed to steric hindrance
of the active site. Dendrons, under study here, provide a unique
opportunity to test the latter part of the hypotheses. This is
because the Fre´chet-type dendrimers 5-7 represent systematic,
small increments in size of the ligand scaffold, while our biaryl
dendrimers 2-4 represent a rather large increase from 7 due to
aggregation. To investigate this structure-property relationship,
we carried out enzymatic assays with three different substrates
S1-S3. Results of these studies are outlined in Figure 4.
First of all, it should be noted that, in order to understand
the influence of the dendritic molecule upon the enzymatic
activity, we carried out the necessary control experiments with
each of the substrates under identical conditions, in the absence
of the dendrons. The relative activities shown in Figure 4b-d
are normalized to these controls. Nonetheless, the inherent
relative activities of the native ChT toward each of the substrates
S1-S3 are shown in Figure 4a.
Figure 4b shows the activity of ChT against three substrates
in the presence of compounds 1-4. In this case, a 3.2 µM
solution of ChT was incubated with 75 µM, 25 µM, 10.7 µM,
and 5 µM solutions of 1-4, respectively, Different concentra-
tions of dendrons are taken here to ensure that the relative
amounts of carboxylate units vs the number of protein molecules
are similar. Activities of each of these dendron-ChT combina-
tions were examined against each of the three substrates S1-
S3. While compound 1 does not affect any change in the
enzymatic activity, compounds 2-4 are able to exhibit a
significant difference in reactivity toward the three substrates.
This is understandable, because compound 1 exhibits no
discernible binding toward the enzyme and therefore does not
exert any difference in activity. On the other hand, compounds
2-4 exhibit large inhibition, minor inhibition, and hyperactivity
against substrates S1-S3, respectively. However, it is interesting
to find that there is no difference in magnitude of selectivity
among the dendrons 2-4. This is perhaps understandable, since
the amount of aggregation in all these three dendrons is
sufficient to exert the maximum effect on substrate selectivity.
On the other hand, since the size of the dendrons change more
systematically with the Fre´chet-type dendrons, we investigated
the effect of the dendron generation on the substrate selectivity
of the dendron/ChT complex. When maintaining the relative
ratio of the carboxylates and the protein molecule the same as
above, the results obtained are shown in Figure 4c. It is
9
9236 J. AM. CHEM. SOC. VOL. 128, NO. 28, 2006

Facially and Classical Amphiphilic Biaryl Dendrimers
A R T I C L E S
Summary
intriguing to note that there seems to be a dendritic effect upon
substrate selectivity. However, it is important that we rule out
the alternate possibility, which involves the fact that there is a
larger variation in the binding affinity of dendrons 5-7.
Therefore, the results in Figure 4c could be a simple manifesta-
tion of the bound vs unbound enzymatic activity. Indeed, when
we increased the concentration of the dendrons to ensure that
most of the proteins are in the bound form (using the binding
affinities shown in Table 1), the substrate selectivities obtained
for dendrons 6 and 7 were very similar to those for compounds
2-4 (Figure 4d). On the other hand, the substrate selectivity
for the dendron 5 increases but does not reach the maximum
due to the very low binding affinity of the dendron. Further
increase of the concentration of 5, to ensure that most of the
proteins are bound, results in a precipitation from the buffer
solution.
The above results show that the size of the ligand scaffold
does not have a significant effect on the substrate selectivity of
the enzyme. One could hypothesize that the selectivity is then
driven mainly by electrostatics. It is also important to point out
another intriguing result in the case of the enzymatic activity
in the presence of dendrons 2-7. As mentioned above, the
inhibition of the enzymatic activity against the neutral substrate
S2 has been attributed mainly to the steric effect that the ligand
scaffolds could provide for the substrate to access the active
site. If this were the case, one would expect the relative activity
of the enzyme to decrease with respect to S2, when moving
from dendron 6 to 7, since the latter dendron is bigger. However,
this was not the case. If anything, there was a slight increase in
the relative activity in the presence of 7. Similarly, one would
expect dendrons 2-4 compared to 6 and 7 to have much lower
activity against neutral substrates due to the larger aggregate
size. Once again, we found this not to be the case. Therefore,
the argument that the activity against the neutral substrate is
indicative of the steric crowding by the ligand scaffold does
not seem to be correct. The binding likely causes an inherent
change in the enzymatic activity, in which steric crowding could
be a minor contributor. But, there is also some other factor
involved; an allosteric change in the enzyme that is not
discernible by absorption or CD spectroscopy is an example of
a possibility. However, since we do not have any spectroscopic
evidence for such an effect, this proposed possibility remains
speculative at this time.
We have used biomolecular recognition as the probe to
investigate the properties of the recently reported facially
amphiphilic biaryl dendrimers by comparing these molecules
with the corresponding amphiphilic benzyl ether dendrons. From
these systematic studies, we show that: (i) the carboxylic acid
functionalities present in the internal layers as well as the
periphery are available for molecular recognition in the biaryl
dendrons. (ii) Preorganization of multiple copies of ligands
through aggregation in the biaryl dendrimers results in high
binding affinities, even for small dendrons. (iii) A G-3 dendron
containing eight copies of the ligands in a classical, benzyl ether
dendron is the critical size to approach the binding affinities
exhibited by the large preorganized particles for ChT. (iV) The
systematic variations in the dendron sizes also allowed us to
show that attributing the observed inhibition with neutral
substrate S2 to steric effect is not correct. Gaining fundamental
insights regarding the relative recognition capabilities of the
facially amphiphilic biaryl dendrons will open up new pos-
sibilities for these amphiphilic assemblies in applications such
as targeted drug delivery.
Experimental Section
R-Chymotrypsin (ChT) from bovine pancreas (E.C. 3.4.21.1) and
all other chemicals were purchased from Sigma Aldrich chemical
company. Dendrons 1-4 were synthesized using our previously
published procedure.5a Dendrons 5-7 were synthesized by following
literature protocols.8a,9
Enzymatic Activity Assay: Enzymatic hydrolysis was monitored
using a microplate reader (EL808IU, Bio-Tek Instruments, Winoosk,
VT). The reactions were carried out in 5 mM sodium phosphate buffer
pH 7.4 with 3.2 µM ChT in either a varied concentration of dendrons
or specified. The enzymatic hydrolysis was taken by adding 16 µL of
26 mM substrate to 184 µL of preincubated ChT-dendron supernatant.
Hydrolysis of substrate was monitored for 10-30 min at 405 nm. The
assays were recorded in triplicate, and the average values were reported
with less than 10% standard deviation.
Circular Dichroism (CD): Far-UV CD spectra were obtained from
190 to 240 nm in a Jasco J-720 (Jasco Instruments, Tokyo, Japan)
spectrophotometer with a 1 mm path length quartz cuvette. ChT (3.2
µM) was incubated with dendrons (5 µM) in 5 mM sodium phosphate
pH 7.4 for 1 h. Three scans were averaged at a rate of 20 nm min-1
with a sample interval of 0.2 nm and an 8 s response. The temperature
was fixed at 25 °C.
Next, to identify whether unbound carboxylates present in
the dendrimers could be the reason for recruiting the substrates
closer to the active site and therefore an increase in activity for
substrate S3, we varied the relative concentration of the
dendrimer vs ChT (all concentrations above the binding affinity
and CMC). The idea is that this increase will afford an increased
number of free (unbound) carboxylates in dendrimers. We found
that the selectivity is independent of the concentration of the
dendrimer. This is similar to what we observed with our
amphiphilic polymers as well. These results suggest that the
unbound carboxylate is not a primary source of the observed
selectivity.
Acknowledgment. Support from NIGMS of the National
Institute of Health is gratefully acknowledged (GM-65255).
Supporting Information Available: Syntheses and procedure
for the estimation of binding constants. This material is available
free of charge via the Internet at http://pubs.acs.org.
JA0622406
9
J. AM. CHEM. SOC. VOL. 128, NO. 28, 2006 9237