A Polymer Scaffold for Protein Oligomerization

Article abstract of DOI:10.1021/ja037646m

We report the design and synthesis of well-defined polymers for the noncovalent oligomerization of proteins. The reported scaffolds, which were generated by atom-transfer radical polymerization (ATRP), take advantage of the well-characterized interaction

Full text of DOI:10.1021/ja037646m

Published on Web 01/23/2004
A Polymer Scaffold for Protein Oligomerization
Byron R. Griffith,† Benjamin L. Allen,§ Alan C. Rapraeger,§ and Laura L. Kiessling*,†,‡
Departments of Chemistry, Pathology and Laboratory Medicine, and Biochemistry, UniVersity of
Wisconsin-Madison, Madison, Wisconsin 53706
Received July 30, 2003; E-mail: kiessling@chem.wisc.edu
Protein clustering is critical for information transfer in biology.
For example, it is implicated in growth factor-mediated cell
signaling1,2 and integrin-facilitated cell adhesion.3 Few approaches
exist, however, for testing the consequences of protein clustering.
Here we report the design, synthesis, and biological activity of a
well-defined polymer for the oligomerization of proteins. The
reported scaffold uses the interaction between a Ni2+ complex and
a sequence of histidine residues (i.e., a “His tag”).4 We present a
general method for the oligomerization of a protein bearing a His
tag.
Figure 1. (Left) Succinimide ester-substituted polymer generated by ATRP
has a degree of polymerization (DP) of 114 and a polydispersity index (PDI)
of 1.3. Su is defined as succinimide. (Right) NTA derivative possessing an
amino group and diglycine linker that was employed in the conjugation
reaction.
Our studies build on observations that saccharide-functionalized
polymers can oligomerize concanavalin A.5,6 Given the importance
of protein oligomerization in biology, we sought a general, modular
method to assemble multiple copies of a protein. Proteins that
possess a His tag can readily be produced, and the Ni2+ complex-
His tag interaction has been widely applied.7-11 We envisioned that
a polymer possessing multiple copies of a Ni2+ complex could be
used to oligomerize proteins.
Our polymer targets were designed to possess several key
characteristics. Specifically, we sought to generate polymers of
various lengths and narrow polydispersity indices (PDIs). Such
compounds would allow examination of how the extent of protein
oligomerization influences a response. Second, we wanted a means
to determine the relative loading of metal chelator on the polymer
backbone. Atom-transfer radical polymerization (ATRP) can be
used to generate modifiable scaffolds (e.g., 1) of various lengths
(Figure 1).12 Moreover, ATRP can afford materials with narrow
PDIs.13 Finally, polymers bearing a unique end group can be
generated. Thus, we investigated the suitability of ATRP for
generating polymers capable of assembling proteins.
To append a moiety that would bind to proteins bearing an
oligohistidine tag, we synthesized nitriloacetic acid (NTA) deriva-
tive 2 as the Ni2+-chelating group (Figure 1). Benzyl protecting
groups on the NTA derivative provide a resolved NMR signal for
the determination of relative coupling efficiencies. A diglycine
linker was added to the NTA-containing nucleophile to facilitate
efficient coupling of the metal chelating unit to the polymer
backbone.
To generate the target polymers, compound 1 was end-function-
alized using TBSCl.14 The protected NTA derivative was conjugated
to the backbone under standard conditions. To vary the “density”
of chelator groups, either 0.75 or 2.0 equiv of 2 were added to the
coupling reaction. Excess ethanolamine was added to quench any
remaining reactive sites (Figure 2). The ester protecting groups were
removed via hydrolysis. Ni2+ was added under basic conditions,
and the polymers were subjected to extensive dialysis to afford the
final products, 4a and 4b.
Figure 2. Scheme for synthesis of polymers presenting Ni2+-chelating
groups. L ) water or His. DIC ) diisopropylcarbodiimide. R1/R2 represents
a random distribution of R1 and R2.
critical for many signaling processes including morphogenesis and
tumor angiogenesis.1,2,15 The oligosaccharide heparan sulfate chains
of proteoglycans (or the polysaccharide heparin) are critical,16 as
they can oligomerize FGFs and present multiple copies of them to
FGF receptors (FGFRs).17,18 We sought to determine whether our
synthetic polymer would activate FGF-mediated signaling.
We employed an FGF-8b fusion protein that possesses a His
tag. To determine whether multivalent 4a can assemble multiple
copies of FGF-8b, we carried out chemical cross-linking studies
on a mixture of polymer 4a and His-tagged FGF-8b.14 The amount
of FGF-8b dimer increases with increasing concentrations of
polymer 4a until a maximum value is reached. At higher concentra-
tions of polymer 4a, the amount of dimer decreases because each
polymer binds to fewer copies of FGF-8b. These data indicate that
functionalized polymers generated by ATRP can be used to cluster
a target protein.
We next tested whether such polymers could promote FGF-8b
signaling using a cell proliferation assay. A heparan sulfate-deficient
cell line (BaF3 cells) transfected with a single FGF receptor
(FGFR4) was employed. In the absence of FGF-8b and a sulfated
polysaccharide such as heparin, these cells undergo apoptosis. They
can be rescued by treatment with soluble heparin and FGF-8b, but
both components are required. We hypothesized that if our polymer
could oligomerize FGF-8b, it might promote cell proliferation in
the absence of heparin. Cells treated with the polymer alone did
not proliferate. When BaF3 cells were treated with polymer 4a and
the FGF-8b fusion protein, however, they proliferated (Figure 3).
Similar results were obtained with polymer 4b. This proliferation
The ability of multivalent displays to cluster a protein of interest,
fibroblast growth factor 8b (FGF-8b), was assessed. FGFs are
† Department of Chemistry.
§ Department of Pathology and Laboratory Medicine.
‡ Department of Biochemistry.
9
1608
J. AM. CHEM. SOC. 2004, 126, 1608-1609
10.1021/ja037646m CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
the data suggest the polymers cluster FGF-8b and present it to the
FGFR, thereby activating signaling.
Our results indicate that ATRP can be used to synthesize
biologically active polymers capable of clustering target proteins.
We have demonstrated that polymers presenting Ni2+ complexes
can cluster His-tagged FGF-8b via noncovalent interactions. This
clustering results in FGF-mediated cell proliferation in the absence
of heparin or heparan sulfate. Thus, these polymers can be used to
dissect the role of FGF clustering in FGF signaling. Because our
strategy is modular, polymers of this type can be used to examine
the consequences of clustering any soluble protein bearing a His
tag. We anticipate that polymers of different lengths can be
employed to dissect how the extent of protein clustering influences
a range of cellular responses.
Acknowledgment. This research was supported by the NIH
(GM55984 to L.L.K., GM48850 to A.C.R.). B.R.G. was supported
by the NIH Biotechnology Training Program (GM08349) and
B.L.A. by a fellowship from the American Heart Association
(0215155Z).
Figure 3. Proliferation of BaF3 cells. Proliferation was measured using a
modified MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-
mide) assay. Cells were incubated at 37 °C for 48-72 h after treatment as
follows: 30 nM polymer 4a and FGF-8b ([); 3.0 µM FGF-8b alone (2).
The value obtained for FGF-8b (30 nM) and heparin (100 nM) is also
indicated (9). The errors represent the standard deviation of samples
performed in quadruplicate. Experiments performed on 3 separate days gave
similar results; the data shown are from experiments carried out on a single
day.
Supporting Information Available: Gel permeation chromatog-
raphy data, synthetic procedures and spectroscopic data, results from
FGF-8b cross-linking experiments, procedures used in the determination
of the Tm of FGF-8b, and procedures for the proliferation assay (PDF).
This material is available free of charge via the Internet at http://
pubs.acs.org.
was dependent on the presence of FGF-8b and the polymer (Figure
3), as cells treated with FGF-8b alone or polymer alone failed to
proliferate to any significant extent.
We hypothesized that multiple copies of FGF-8b could assemble
on the polymer and that it is the oligomerized FGF-8b that promotes
proliferation. If such a model is correct, the extent of proliferation
should depend on the relative concentrations of FGF-8b and the
polymer. If the polymer concentration is held constant, proliferation
activity should increase as the concentration of FGF-8b increases;
more copies of FGF-8b should bind to each polymer. The extent
of proliferation should reach a maximum when the polymer cannot
accommodate additional copies of FGF-8b. At very high concentra-
tions of FGF-8b, however, excess monomeric FGF-8b can begin
to compete with the polymer-bound FGF-8b for the FGFR. If
FGF-8b binds directly to the FGFR, the extent of proliferation will
decrease because monomeric FGF-8b does not activate proliferation
(Figure 3).
Using a fixed concentration (30 nM) of polymer 4a, we tested
the effect of FGF-8b concentration on proliferation. Activity begins
at an FGF-8b-to-polymer ratio of ca. 1:1 and reaches a maximum
when the ratio is 10:1. As is predicted for multivalent binding, once
this maximum is reached, higher concentrations of FGF-8b do not
afford increases in proliferation. These data indicate that polymer
4 may be used to explore the consequences of protein clustering.
The decrease in proliferation at ratios of 30:1 or greater likely
represents competition between the polymer-oligomerized FGF-
8b and monomeric FGF-8b for the FGFR binding sites.
References
(1) Ornitz, D. M., Bioessays 2000, 22, 108-112.
(2) Taipale, J.; Keski-Oja, J. FASEB J. 1997, 11, 51-59.
(3) van der Flier, A.; Sonnenberg, A. Cell Tissue Res. 2001, 305, 285-298.
(4) Hochuli, E.; Bannwarth, W.; Dobeli, H.; Gentz, R.; Stuber, D. Biotech-
nology 1988, 6, 1321-1325.
(5) Cairo, C. W.; Gestwicki, J. E.; Kanai, M.; Kiessling, L. L. J. Am. Chem.
Soc. 2002, 124, 1615-1619.
(6) Dimick, S. M.; Powell, S. C.; McMahon, S. A.; Moothoo, D. N.; Naismith,
J. H.; Toone, E. J. J. Am. Chem. Soc. 1999, 121, 10286-10296.
(7) Hochuli, E.; Dobeli, H.; Schacher, A. J. Chromatogr. 1987, 411, 177-
184.
(8) O’Shannessy, D. J.; O’Donnell, K. C.; Martin, J.; Brighamburke, M. Anal.
Biochem. 1995, 229, 119-124.
(9) Sigal, G. B.; Bamdad, C.; Barberis, A.; Strominger, J.; Whitesides, G.
M. Anal. Chem. 1996, 68, 490-497.
(10) Paborsky, L. R.; Dunn, K. E.; Gibbs, C. S.; Dougherty, J. P. Anal. Biochem.
1996, 234, 60-65.
(11) Dorn, I. T.; Neumaier, K. R.; Tampe, R. J. Am. Chem. Soc. 1998, 120,
2753-2763.
(12) Godwin, A.; Hartenstein, M.; Muller, A. H. E.; Brocchini, S. Angew.
Chem., Int. Ed. 2001, 40, 594-597.
(13) Matyjaszewski, K.; Xia, J. Chem. ReV. 2001, 101, 2921-2990.
(14) See the Supporting Information.
(15) Loo, B. M.; Salmivirta, M. J. Biol. Chem. 2002, 277, 32616-32623.
(16) Rapraeger, A. C.; Krufka, A.; Olwin, B. B. Science 1991, 252, 1705-
1708.
(17) Spivakkroizman, T.; Lemmon, M. A.; Dikic, I.; Ladbury, J. E.; Pinchasi,
D.; Huang, J.; Jaye, M.; Crumley, G.; Schlessinger, J.; Lax, I. Cell 1994,
79, 1015-1024.
(18) Kwan, C. P.; Venkataraman, G.; Shriver, Z.; Raman, R.; Liu, D. F.; Qi,
Y. W.; Varticovski, L.; Sasisekharan, R. J. Biol. Chem. 2001, 276, 23421-
23429.
(19) Chi, Y. H.; Kumar, T. K. S.; Wang, H. M.; Ho, M. C.; Chiu, I. M.; Yu,
C. Biochemistry 2001, 40, 7746-7753.
An alternative mechanism of action for polymers 4a and 4b is
that they stabilize FGF-8b under cellular conditions, as has been
reported for FGF-1.19,20 We therefore measured the melting
temperature (Tm) of FGF-8b.14 The value determined, 60 °C,
indicates that FGF-8b is stable under the assay conditions. Together
(20) Arunkumar, A. I.; Kumar, T. K. S.; Kathir, K. M.; Srisailam, S.; Wang,
H. M.; Leena, P. S. T.; Chi, Y. H.; Chen, H. C.; Wu, C. H.; Wu, R. T.;
Chang, G. G.; Chiu, I. M.; Yu, C. Protein Sci. 2002, 11, 1050-1061.
JA037646M
9
J. AM. CHEM. SOC. VOL. 126, NO. 6, 2004 1609