Synthesis and applications of end-labeled neoglycopolymers

Article abstract of DOI:10.1021/ol0259239

(Matrix Presented) Neoglycopolymers that vary in length and contain a single fluorescent reporter group were synthesized using ring-opening metathesis polymerization (ROMP). The utility of these materials is demonstrated by the development of a cellular b

Full text of DOI:10.1021/ol0259239

ORGANIC
LETTERS
2002
Vol. 4, No. 14
2293-2296
Synthesis and Applications of
End-Labeled Neoglycopolymers
,†,‡
Robert M. Owen,† Jason E. Gestwicki,‡ Travis Young,† and Laura L. Kiessling*
Departments of Chemistry and Biochemistry, UniVersity of WisconsinsMadison,
Madison, Wisconsin 53706
kiessling@chem.wisc.edu
Received March 24, 2002
ABSTRACT
Neoglycopolymers that vary in length and contain a single fluorescent reporter group were synthesized using ring-opening metathesis
polymerization (ROMP). The utility of these materials is demonstrated by the development of a cellular binding assay for L-selectin, a cell
surface protein that plays a role in inflammation. The data reveal that these multivalent ligands interact with multiple copies of L-selectin.
Multivalent binding events are ubiquitous in biological
systems. For example, multivalent protein-carbohydrate
complexation events are important in cell adhesion, host-
pathogen interactions, and the immune response.1 Despite
the importance of multivalent binding, mechanistic informa-
tion is often lacking. One critical issue that arises when a
multivalent ligand interacts with a cell-surface protein is
whether it binds multiple copies of the target receptor.
Though this binding mode is often invoked with multivalent
ligands, it is difficult to determine the stoichiometry of these
receptor-ligand complexes.2 We sought to address this issue
by examining multivalent ligand binding to cell-surface
L-selectin.
are multivalent; they display multiple copies of sulfated
carbohydrates.4 L-Selectin is localized to regions on the cell-
surface,5 and dimerization of L-selectin enhances its ability
to bind to endothelial cells.6 These data implicate multiva-
lency in the interaction of L-selectin and its ligands at the
cell surface, but direct evidence has been difficult to obtain.
Synthetic multivalent displays have emerged as powerful
tools for investigating multivalent binding.7 Polymeric
ligands that incorporate both binding elements and a reporter
group are especially useful for visualizing and quantitating
binding interactions.8 A reporter moiety, such as a fluorescent
or biotin group, is typically introduced through the polym-
erization of an appropriately substituted monomer or through
L-Selectin is displayed on the cell-surface of leukocytes
where it participates in the recruitment of these cells to the
inflamed endothelium.3,4 The natural ligands for L-selectin
(4) (a) Yeh, J. C.; Hiraoka, N.; Petryniak, B.; Nakayama, J.; Ellies, L.
G.; Rabuka, D.; Hindsgaul, O.; Marth, J. D.; Lowe, J. B.; Fukuda, M. Cell
2001, 105, 957-969. (b) Hemmerich, S.; Rosen, S. D. Glycobiology 2000,
10, 849-856. (c) Yarema, K. J.; Bertozzi, C. R. Curr. Opin. Chem. Biol.
1998, 2, 49-61.
(5) Bruehl, R. E.; Springer, T. A.; Bainton, D. F. J. Histochem. Cytochem.
1996, 44, 835-844.
† Department of Chemistry.
‡ Department of Biochemistry.
(1) (a) Lee, Y. C.; Lee, R. T. Acc. Chem. Res. 1995, 28, 321-327. (b)
Kiessling, L. L.; Pohl, N. L. Chem. Biol. 1996, 3, 71-77. (c) Roy, R. Curr.
Opin. Struct. Biol. 1996, 6, 692-702. (d) Mammen, M.; Choi, S.-K.;
Whitesides, G. M. Angew. Chem., Int. Ed. 1998, 37, 2754-2794. (e) Feizi,
T. Immunol. ReV. 2000, 173, 79-88. (f) Sacchettini, J. C.; Baum, L. G.;
Brewer, C. F. Biochemistry 2001, 40, 3009-3015. (g) Bertozzi, C. R.;
Kiessling, L. L. Science 2001, 291, 2357-2364.
(2) (a) Gestwicki, J. E.; Strong, L. E.; Kiessling, L. L. Angew. Chem.,
Int. Ed. 2000, 39, 4567-4570. (b) Cairo, C. W.; Gestwicki, J. E.; Kanai,
M.; Kiessling, L. L. J. Am. Chem. Soc. 2002, 124, 1615-1619.
(3) (a) Springer, T. A. Cell 1994, 76, 301-314. (b) Steeber, D. A.;
Tedder, T. F. Immunol. Res. 2000, 22, 299-317.
(6) (a) Galustian, C.; Childs, R. A.; Yuen, C. T.; Hasegawa, A.; Kiso,
M.; Lubineau, A.; Shaw, G.; Feizi, T. Biochemistry 1997, 36, 5260-5266.
(b) Li, X.; Steeber, D. A.; Tang, M. L. K.; Farrar, M. A.; Perlmutter, R.
M.; Tedder, T. F. J. Exp. Med. 1998, 188, 1385-1390.
(7) (a) Liang, R.; Yan, L.; Loebach, J.; Ge, M.; Uozumi, Y.; Sekanina,
K.; Horan, N.; Gildersleeve, J.; Thompson, C.; Smith, A.; Biswas, K.; Still,
W. C.; Kahne, D. Science 1996, 274, 1520-1522. (b) Kiessling, L. L.;
Strong, L. E. Top. Organomet. Chem 1998, 1, 199-231. (c) Lundquist, J.
J.; Toone, E. J. Chem. ReV. 2002, 102, 555-578. (d) Houseman, B. T.;
Mrksich, M. Top. Curr. Chem. 2002, 218, 1-44. (e) Lindhorst, T. K. Top.
Curr. Chem. 2002, 218, 201-235.
(8) Reviewed in: Bovin, N. V. Glycoconjugate J. 1998, 15, 431-446.
10.1021/ol0259239 CCC: $22.00 © 2002 American Chemical Society
Published on Web 06/11/2002

random conjugation to the polymer backbone.9 Although
useful for many purposes, these strategies provide limited
control over the number and location of the introduced
functionality. Polymers possessing a single end-label, how-
ever, can be selectively immobilized on surfaces to provide
novel materials for physical10 or biological studies.11 We
envisioned that multivalent ligands varying in valency but
containing a single fluorophore per ligand could provide
unique mechanistic information. If multivalent binding
occurs, saturating concentrations of ligands differing only
in their valency would give rise to different cell surface
fluorescence intensities (Figure 1). In contrast, monovalent
developed by the Grubbs group also has been used to
synthesize end-labeled polymers. Ruthenium alkylidene-
initiated reactions conducted in the presence of chain-transfer
agents provide symmetrically terminated materials.14 Alter-
natively, a terminal aldehyde can be installed by treatment
of 3 with molecular oxygen.15 We have shown that a
substituted enol ether can be used to produce specifically,
end-labeled polymers, such as 4.16
Here, we demonstrate the effectiveness of this strategy for
the synthesis of labeled polymers varying in length and
terminal functionality.
Termination Efficiency. Because our objective was to
generate multivalent ligands of different lengths, the capping
process must be efficient. To identify effective termination
agents, enol ether derivatives possessing unique functional
groups, including masked acid (6, 13), ketone (9), and
masked amine (10) groups, were synthesized (Scheme 1).
Scheme 1. Synthetic Routes to Acid-, Ketone-, and
Amine-Substituted Capping Agents
Figure 1. End-labeled materials can be used to reveal multivalent
interactions. The number of labels bound depends on the number
of binding sites occupied per multivalent ligand (parts a-c). The
binding of end-labeled multivalent ligands at saturation (Bmax) will
vary according to valency (equation shown in part 1d, B is moles
of bound ligand, F is the free multivalent ligand concentration).
binding interactions would yield identical fluorescence
intensities at saturation.
ROMP12 provides a means to generate end-labeled materi-
als varying in valency. When ROMP is living,13 an active
carbene (e.g., 3) remains at the polymer terminus after
consumption of the monomer (Figure 2). The metal alky-
Figure 2. Termination of ruthenium carbene-initiated ROMP
reactions with an enol-ether provides an end-labeled material (4).
The functional groups chosen provide the means to install
reporter groups under mild conditions using commercially
available reagents. The capping agents were assembled from
simple building blocks (Scheme 1). Enol ether 6 was obtained
in excellent yield from the known 2-trimethylsilyl ethyl
(TMSE) ester (5) of 4-pentenoic acid.17 Ozonolysis of 5
followed by Wittig reaction yielded the desired vinyl ether
6 as a 1:3 cis:trans mixture. Although cis-enol ethers have
been shown to react more quickly with ruthenium carbene
lidene 3 can undergo further transformations. For example,
Schrock and co-workers treated a polymer possessing a
terminal molybdenum carbene with aldehydes to provide
materials with unique redox and luminescence properties.10
The highly functional group tolerant ruthenium initiator 2
(9) (a) Ticha, M.; Kocourek, J. Carbohydr. Res. 1991, 213, 339-343.
(b) Kamitakahara, H.; Suzuki, T.; Nishigori, N.; Suzuki, Y.; Kanie, O.;
Wong, C. H. Angew. Chem., Int. Ed. 1998, 37, 1524-1528.
(10) (a) Albagli, D.; Bazan, G. C.; Schrock, R. R.; Wrighton, M. S. J.
Am. Chem. Soc. 1993, 115, 7328-7334. (b) Albagli, D.; Bazan, G. C.;
Schrock, R. R.; Wrighton, M. S. J. Phys. Chem. 1993, 97, 10211-10216.
(11) Gestwicki, J. E.; Cairo, C. W.; Mann, D. A.; Owen, R. M.; Kiessling,
L. L. Anal. Biochem., in press.
(14) (a) Gibson, V. C.; Okada, T. Macromolecules 2000, 33, 655-656.
(b) Morita, T.; Maughon, B. R.; Bielawski, C. W.; Grubbs, R. H.
Macromolecules 2000, 33, 6621-6623.
(15) (a) Gordon, E. J. Ph.D. Dissertation, University of Wisconsin,
Madison, WI, 1998. (b) Biagini, S. C. G.; Davies, R. G.; Gibson, V. C.;
Giles, M. R.; Marshall, E. L.; North, M. Polymer 2001, 42, 6669-6671.
(16) Gordon, E. J.; Gestwicki, J. E.; Strong, L. E.; Kiessling, L. L. Chem.
Biol. 2000, 7, 9-16.
(12) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18-29.
(13) Ivin, K. J.; Mol J. C. Olefin Metathesis and Metathesis Polymeri-
zation; Academic Press: London, 1997.
(17) Krafft, M. E.; Cheung, Y. Y.; Wright, C.; Cali, R. J. Org. Chem.
1996, 61, 3912-3915.
2294
Org. Lett., Vol. 4, No. 14, 2002

2,18 the isomers were not separated because an excess of the
mixture (15 equiv) was used in the termination step (vide
infra). Capping agents 9, 10, and 13, were all generated from
known amine 7.19 The ketone-containing capping agent 9
could be synthesized in a single step from 7 and the
N-hydroxysuccinimide (NHS) ester of levulinic acid (8).20
Similarly, compound 10 could be accessed from the reaction
of 7 with 2-trimethylsilylethyl N-succinimidyl carbonate
(Teoc-OSu).21 Preparation of the ethylene glycol-based
capping agent (13) required desymmetrization of the diacid
11 to give the diester 12. The benzyl ester was converted to
the NHS ester via the acid. The enol ether was installed under
standard amide bond forming conditions to afford capping
agent 13. The efficiencies of 6, 9, 10, and 13 in terminating
ROMP reactions were then assessed.
To test the effectiveness of our capping strategy, monomer
14 was polymerized using ruthenium initiator 2, and the
reaction was terminated by the addition of a capping agent.
Methyl ester 14 was used as a monomer because the resulting
polymer (15) could be readily separated from excess capping
agent by silica gel chromatography. Each of the capping
agents was used in a 15-fold excess as a mixture of cis:
trans isomers. The termination reaction was monitored by
1H NMR spectroscopy. Complete consumption of the
propagating carbene by each terminating agent occurred
within 3 h. After purification of the resulting polymers, the
capping efficiencies were determined by comparing the
integration of the signal due to the polymer phenyl protons
relative to a diagnostic signal from the capping agent (the
trimethylsilyl group for compounds 6, 10, and 13 or the
protons R to the ketone for 9). The termination efficiencies
were excellent for 6 and 10, good for 9, and moderate for
compound 13 (Table 1). The lower efficiencies for 9 and 13
of the carbamate-containing capping agent 10, which is less
Lewis basic than an amide,23 is more efficient.
End-Labeled Polymers as Cell-Surface Probes. Given
that end-labeled polymers could be generated efficiently,
these materials were used to investigate ligand binding to
L-selectin-displaying cells. We found previously that mul-
tivalent ligands presenting 3,6-disulfogalactose residues bind
to L-selectin.16,24 We envisioned that fluorescent derivatives
could be used to assay binding16,25 and to assess the
importance of multivalent interactions. A series of polymers
in which the degree of polymerization was varied was
generated by controlling the ratio of monomer 1626 to
ruthenium alkylidene initiator 2. The polymerization reactions
were terminated by the addition of 13 to afford electrophilic
polymers that could be sequentially functionalized. Treatment
with the amine-bearing 3,6-disulfogalactose derivative (17)27
yielded the desired polymers (18a-c) (Scheme 2). Hydroly-
Scheme 2. Synthesis of Labeled
3,6-Disulfogalactose-substituted Polymers 20a-c for L-Selectin
Binding Assay. The Degree of Polymerization (DP30) Was
1
Determined from the H NMR Spectra
Table 1. Capping Efficiencies for the Polymerization of
Alkene 14
capping efficiencya
sis of the terminal ester and conjugation of fluorescein
cadaverine to the resulting amine group provided the target
labeled ligands (19a-c).16 The interaction of these materials
with Jurkat cells displaying L-selectin was then assessed.
Jurkat cells were exposed to polymers 19a-c or a
fluorescently labeled anti-L-selectin antibody (Figure 3A)
entry
capping agent
1
2
3
4
6
9
10
13
>95%/93%
80%/86%
>95%/92%
64%/68%
a Efficiencies are an average of a minimum of two separate experiments
and were determined for monomer-to-catalyst ratios of 15:1/50:1.
(21) Schute, R. E.; Rich, D. H. Synthesis 1987, 4, 346-349.
(22) Choi, T. L.; Chatterjee, A. K.; Grubbs, R. H. Angew. Chem., Int.
Ed. 2001, 40, 1277-1279.
(23) Grana, A. M.; Mosquera, R. A. J. Comput. Chem. 1999, 20, 1444-
may arise from their amide groups, which may chelate to
the ruthenium alkylidene species.22 The termination reaction
1454.
(24) Gordon, E. J.; Strong, L. E.; Kiessling, L. L. Bioorg. Med. Chem.
1998, 6, 1-7.
(18) Hillmyer, M. A.; Grubbs, R. H. Macromolecules 1995, 28, 8662-
8667.
(19) Wasserman, H. H.; Cook, J. D.; Vu, C. B. J. Org. Chem. 1990, 55,
1701-1702. Typical cis:trans ratios obtained ranged from 1:2 to 1:4.
(20) Lee, J. H.; Baker, T. J.; Mahal, L. K.; Zabner, J.; Bertozzi, C. R.;
Wiemer, D. F.; Welsh, M. J. J. Biol. Chem. 1999, 274, 21878-21884.
(25) Wittmann, V.; Datta, A. K.; Koeller, K. M.; Wong, C. H. Chem.
Eur. J. 2000, 6, 162-171.
(26) Strong, L. E.; Kiessling, L. L. J. Am. Chem. Soc. 1999, 121, 6193-
6196.
(27) Manning, D. D.; Hu, X.; Beck, P.; Kiessling, L. L. J. Am. Chem.
Soc. 1997, 119, 3161-3162.
Org. Lett., Vol. 4, No. 14, 2002
2295

demonstrating that neoglycopolymers can cluster proteins in
solution and in the cell.2,31 Compounds with more complex
binding epitopes32 may more effectively interact with
multiple copies of cell-surface L-selectin.
Our previous studies indicate that multivalent but not
monovalent ligands induce L-selectin shedding from white
blood cells.24 One explanation for this difference is that
polymeric ligands cluster multiple copies of L-selectin but
monovalent ligands do not. Because the ligands used in these
studies induce L-selectin downregulation,24 our results sug-
gest that the ability of these ligands to modulate cell-surface
levels of L-selectin depends on clustering.33 This finding
offers new opportunities for the design of L-selectin antago-
nists.34 Our results also suggest mechanisms by which natural
multivalent ligands may bind L-selectin.
In conclusion, end-labeled polymers are valuable probes
of multivalent ligand-receptor interactions. Our data reveal
that multivalent ligands varying in length with single end-
labels can be generated using ROMP. The investigations
described here highlight the utility of such materials for
probing the mechanisms underlying multivalent binding at
the cell surface. The ligands we have generated provide the
means to investigate mechanistic aspects of cell-surface
receptor-ligand interactions that have been inaccessible. We
anticipate that the strategies outlined can be used to
synthesize polymers that will be useful in a wide range of
materials science and in vitro and in vivo biological
applications.
Figure 3. (A) Treatment of Jurkat cells with (i) fluorophore-labeled
anti-L-selectin antibody, (ii) polymer 19a, (iii) 19b, or (iv) 19c.
(B) Results from the L-selectin binding assay. The binding of 19a-c
is reversed upon addition of unlabeled 3,6-disulfogalactose or sialyl
Lewis x-bearing polymers (data not shown). Concentrations refer
to 3,6-disulfogalactose residue concentration. The approximate
stoichiometry is defined as the number of copies of L-selectin that
interact simultaneously with a polymer of a given DP.28
toscreen for ligand binding. As anticipated,16 fluorescence
microscopy experiments indicate that 19a-c interact with
Jurkat cells displaying L-selectin. To quantify the amount
of bound ligand, Jurkat cells were treated with 19a-c and
washed, and the intensity of the fluorescence emission was
assessed (Figure 3B). The resulting data were fit to the
equation in Figure 1d to determine the dissociation constant
(Kd) for each compound.28 The Kd values for the polymers29
were dependent on the polymer length. Polymer 19c (degree
of polymerization,30 DP of 150) had a potency that was ca.
10-fold greater than the shortest polymer (19a, DP of 35)
on a saccharide residue basis. Although useful for compari-
son, this calculation underestimates the relative increase in
potency. The individual saccharide (3,6-disulfogalactose)
does not bind to L-selectinsa multivalent display is required.
Because the polymers possess a single fluorophore label,
their ability to bind multiple copies of L-selectin at the cell
surface could be determined. The concentration of polymer
required to saturate the fluorescence intensity (Bmax) was
measured. Dividing the average amount of L-selectin on a
Jurkat cell-surface by the calculated concentrations of bound
ligand provides an estimate of the relative stoichiometry of
the L-selectin-ligand complexes. By this analysis, polymer
19c (DP of 150) bound approximately five copies of
L-selectin while polymer 19a (DP of 35) bound only one or
two copies of L-selectin. These data indicate that compounds
19a-c engage in multivalent interactions with L-selectin and
that the stoichiometry of the resulting complex depends on
the valency of the ligand. This result is consistent with studies
Acknowledgment. This work was supported in part by
the NIH (GM 55984). The UW-Madison Chemistry NMR
facility is supported in part by NSF (CHE9629688 and
CHE9208463) and the NIH (RR08389). R.M.O thanks
Pharmacia Corporation and Eastman Chemical Company for
fellowships. J.E.G. was supported by the NIH Biotechnology
Training Program (GM08349). The authors thank R. J.
Hinklin, C. W. Cairo, and J. K. Pontrello for helpful
discussions and the laboratories of Profs S. Bednarek and
P. Friesen (UW-Madison) for generous use of equipment.
Supporting Information Available: Experimental pro-
cedures and characterization data for compounds 6, 9, 10,
12, 13, 15, 18, and 19. This material is available free of
charge via the Internet at http://pubs.acs.org.
OL0259239
(31) (a) Burke, S. D.; Zhao, Q.; Schuster, M. C.; Kiessling, L. L. J. Am.
Chem. Soc. 2000, 122, 4518-4519. (b) Gestwicki, J. E.; Strong, L. E.;
Kiessling, L. L. Chem. Biol. 2000, 7, 583-591. (c) Gestwicki, J. E.; Strong,
L. E.; Cairo, C. W.; Boehm, F. J.; Kiessling, L. L. Chem. Biol. 2002, 9,
163-169. (d) Gestwicki, J. E.; Kiessling, L. L. Nature 2002, 415, 81-84.
(32) (a) Sanders, W. J.; Gordon, E. J.; Dwir, O.; Beck, P. J.; Alon, R.;
Kiessling, L. L. J. Biol. Chem. 1999, 274, 5271-5278. (b) Manning, D.
D.; Bertozzi, C. R.; Pohl, N. L.; Rosen, S. D.; Kiessling, L. L. J. Org.
Chem. 1995, 60, 6254-6255.
(33) Gordon, E. J.; Sanders, W. J.; Kiessling, L. L. Nature 1998, 392,
30-31.
(34) Simanek, E. E.; McGarvey, G. J.; Jablonowski, J. A.; Wong, C. H.
Chem. ReV. 1998, 98, 833-862.
(28) See the Supporting Information.
(29) Kd values were calculated on a saccharide residue basis.
(30) Degree of polymerization (DP) was determined from the integration
of 1H NMR signals in the phenyl region against those in the alkene region.
2296
Org. Lett., Vol. 4, No. 14, 2002