Enhanced inhibition of human anti-Gal antibody binding to mammalian cells by synthetic α-Gal epitope polymers

Article abstract of DOI:10.1021/ja990219h

Neoglycopolymers of polyacrylamide backbone conjugated with varying densities of Gala1-3Galβ1-4Glcβ trisaccharide epitopes (α-Gal epitopes) were designed and synthesized to study the inhibition of the binding of human natural anti-Gal antibodies to either

Full text of DOI:10.1021/ja990219h

8174
J. Am. Chem. Soc. 1999, 121, 8174-8181
Enhanced Inhibition of Human Anti-Gal Antibody Binding to
Mammalian Cells by Synthetic R-Gal Epitope Polymers
Jianq-Qiang Wang, Xi Chen, Wei Zhang, Sima Zacharek, Yongsheng Chen, and
Peng George Wang*
Contribution from the Department of Chemistry, Wayne State UniVersity, Detroit, Michigan 48202
ReceiVed January 22, 1999
Abstract: Neoglycopolymers of polyacrylamide backbone conjugated with varying densities of GalR1-3Galâ1-
4Glcâ trisaccharide epitopes (R-Gal epitopes) were designed and synthesized to study the inhibition of the
binding of human natural anti-Gal antibodies to either R-Gal-containing glycoproteins or R-Gal antigens on
the surface of mammalian cells. An inhibition ELISA using mouse laminin and a flow cytometry assay using
pig kidney cells (PK15) were established to determine the binding affinity of the synthesized polymers. In
comparison to the R-Gal monomer (GalR1-3Galâ1-4GlcNHAcâ), the R-Gal polymers dramatically enhanced
the inhibition of human anti-Gal antibodies (IgG, IgM, and IgA) binding to mouse laminin or mammalian
cells. Increases of 7.8 × 103- and 5.0 × 104 -fold in inhibitory potential of polymer 7C to IgA and IgM (with
IC50s of 7.0 and 5.6 nM respectively) were observed over the monomer in inhibition ELISA. The results also
indicated that binding enhancement of R-Gal polymers is greater for anti-Gal IgA and IgM than for IgG. Such
amplified binding differences among the three anti-Gal isotypes can be utilized to selectively inhibit or remove
a particular isotype of anti-Gal antibodies. Moreover, it was demonstrated through the flow cytometry assay
that certain R-Gal polymers are effective in inhibition of anti-Gal antibody (in human serum) binding to pig
kidney (PK15) cells. Thus, such synthetic carbohydrate polymers may find practical applications in cell
xenotransplantations.
Introduction
Xenotransplantation,1 which was considered as a means of
overcoming the shortage of human organs, tissues, and cells
for transplantation, has been hampered because of the severe
immunological rejection by the human body. Recent studies
indicated that the initial step of this rejection is the recognition
of human natural anti-Gal antibody to carbohydrate epitopes
bearing a GalR1-3Galâ terminus2 (termed as R-Gal epitope or
R-Gal) on the surface of animal cells. Trisaccharides GalR1-
3Galâ1-4Glcâ-R (1) and GalR1-3Galâ1-4GlcNAcâ-R′ (2)
and pentasaccharide GalR1-3Galâ1-4GlcNAcâ1-3Galâ1-
4Glcâ-R′′ (3) (Figure 1) have been identified as the major R-Gal
epitopes, which are abundantly expressed on the cells of most
mammals with the exception of humans, apes, and other Old
World primates.3 Conversely, anti-Gal antibodies (known as
anti-Gal) are the most abundant human natural antibodies in
blood serum, constituting 1-2% of total serum IgG and 3-8%
Figure 1. The structures of major R-galactosyl epitopes.
of total IgM.4 Interestingly, each individual subclass of anti-
Gal plays a quite distinctive role in the human immune rejection
of xenografts.2b,3b In the initial stage of the rejection (termed
hyperacute rejection or HAR), anti-Gal IgG binds to R-Gal
epitopes expressed on the surface of xenograft cells triggering
the antibody-dependent cell-mediated cytotoxicity by human
blood monocytes and macrophages. The IgM isotype of anti-
Gal is believed to be responsible for the complement activation
that leads to complement-mediated lysis of the xenograft cells.
Various approaches have been studied to prevent the hyperacute
rejection and to prolong xenograft cell survival in xenotrans-
planation based on elimination or reduction of the interaction
between R-Gal and anti-Gal. These methods include encapsula-
tion of cells for the cellular replacement therapy,5 depletion of
* To whom correspondence should be addressed. Tel: (313) 993-6759.
Fax: (313) 577-5831. E-mail: pwang@chem.wayne.edu.
(1) (a) Platt, J. L. Nature 1998, April, supplement to vol. 392, 11. (b)
Nature (Special Issue) 1998, January, 391, 309. (c) Rother, R. P.; Squinto,
S. P. Cell 1996, 36, 185.
(2) (a) Galili, U. Sci. Med. 1998, 5 (5), 28. (b) Cooper, D. K. C. Clin.
Trans. 1992, 6, 178. (c) Cooper, D. K. C.; Good, A. H.; Koren, E.; Oriol,
R.; Malcolm, A. J.; Ippolito, R. M.; Neethling, F. A.; Ye, Y.; Romano, E.;
Zuhdi, N. Transplant Immunol. 1993, 1 (3), 198.
(3) (a) Cooper, D. K. C.; Koren, E.; Oriol, R. Immunol. ReV. 1994, 141,
31. (b) Galili, U. Immunol. Today 1993, 14, 480. (c) Sandrin, M. S.;
Vaughan, H. A.; McKenzie, I. F. C. Transplant. ReV. 1994, 8, 134.. (d)
Samuelsson, B. E.; Rydberg, L.; Breimer, M. E.; BA¨ cker, A.; Gustavsson,
M.; Holgersson, J.; Karlsson, E.; Uyterwaal, A.-C.; Cairns, T.; Welsh, K.
Immunol. ReV. 1994, 141, 151. (e) Galili, U. Springer Semin. Immunopathol.
1993, 15 (2-3), 155. (f) Galili, U. Immunol. Ser. 1992, 55, 355. (c) Galili,
U. Blood Cells 1988, 14 (1), 205. (g) Hamadeh, R.; Galili, U.; Zhou, P.;
Griffiss, J. Clin. Diagn. Lab. Immunol. 1995, 2 (2), 125.
(4) (a) Galili, U.; Anaraki, F.; Thall, A.; Hill-Black, C.; Radic, M. Blood
1993, 82 (8), 2485. (b) McMorrow, I. M.; Comrack, C. A.; Sachs, D. H.;
DerSimonian, H. Transplantation 1997, 64 (3), 501.
(5) (a) Zhou, D.; Vacek, I.; Sun, A. M. Transplantation 1997, 64 (8),
1112. (b) Wang, T.; Lac´ık, I.; Brissova´, M.; Anilkumar, A. V.; Prokop, A.;
Hunkeler, D.; Green, R.; Shahrokhi, K.; Powers, A. C. Nat. Biotechnol.
1997, 15, 358. (c) Lim, F.; Sun, A. M. Science 1980, 210, 908. (d) Goosen,
M. F. A. in Immunoisolation of pancreatic islets; Lanza, R. P., Chick, W.
L., Eds.; R. G. Landes Company: Austin, TX., 1994, pp 21-44.
10.1021/ja990219h CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/31/1999

Human Anti-Gal Antibody Binding
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8175
anti-Gal antibodies from the human recipients using R-Gal
epitope affinity column (anti-Gal immunoadsorption approach),6
and inhibition of anti-Gal antibody binding to xenograft cells
by infusing soluble synthetic R-Gal oligosaccharides (anti-Gal
neutralization approach),7 etc. In attempts to prevent the rejection
using neutralization approach, a number of R-Gal oligosaccha-
rides have been tested for the inhibition of anti-Gal binding to
the porcine endothelial cells. Results indicated8 that it would
require a millimolar level of R-Gal trisaccharide GalR1-
3Galâ1-4GlcNAc to achieve 90% inhibition of anti-Gal binding
to the porcine endothelial cells, which is not sufficient for
practical application. Thus, how to conquer low binding affinity
of R-Gal to anti-Gal antibody becomes a focal point in this
approach.
In biological systems the interactions between cell surface
carbohydrates and sugar-binding proteins are of polyvalent
nature. Although low affinity and poor specificity are intrinsic
in protein binding, carbohydrates function as very important
signaling molecules in a wide variety of biological recognition
events. Their high avidity can be attributed, at least in part, to
their multivalent presentation pattern on the cell surface.
Carbohydrates are typically expressed on the cell surface in
clusters; thus their overall binding capacity with protein
receptors (commonly with multiple binding sites) is enhanced
over the affinity of individual monovalent ligands through
cooperative multiple interactions. Therefore, it is of logical
choice to use multivalent ligands as potent synthetic inhibitors
to effectively block the recognition process. The multivalent
inhibitors of carbohydrate-binding proteins, bacteria, and viruses
have been well studied primarily by Lee,9a,b,g Whitesides,9c,10a,b
Roy,9d,e and others9f,10d in recent years. They have demonstrated
that the multivalent forms of carbohydrate ligands, either
polymers or dendrimers, often have amplified inhibitory effects
over their monovalent counterparts, although the levels of
enhancement vary.
to anti-Gal antibodies. The inhibition ELISA and flow cytometry
assays were established accordingly to evaluate these polymers.
Our results clearly demonstrated that the overall avidity of
certain R-Gal-containing polymers toward anti-Gal antibodies
was significantly enhanced by their “multivalent” or “polyvalent
effects”.
Results and Discussion
Synthesis of r-Gal-Containing Polymers (7A-7F). The
syntheses of R-Gal-containing polymers were achieved by the
reaction of preactivated poly [N-(acryloyloxy) succinimide]
(pNAS) (5) with an R-Gal trisaccharide derivative (6), followed
by capping the active esters with aqueous ammonia (Scheme
1).10 All R-Gal polymers (7A-7F) used in this study were
prepared from one single batch of pNAS, which was obtained
by polymerization of N-(acryloyloxy) succinimide (4). The
molecular weight of this “parent” polymer was determined by
gel filtration chromatography after its complete hydrolysis to
poly (acrylic acid) sodium salt. The average molecular weight
of the hydrolyzed polymer was Mw ) 252 kD with a relatively
narrow molecular weight distribution (Mw/MN ) 1.5). The
degree of polymerization is ∼1.8 × 103. By varying the ratio
of R-Gal trisaccharide to active esters in pNAS, we obtained a
series of polymers with different densities of R-Gal. The ratios
of R-Gal unit to acrylamide unit calculated from the integration
1
of trisaccharide signals and acrylamide signals in H NMR
spectra were 1:1.8 (7A), 1:2.5 (7B), 1:3.3 (7C), 1:6.4 (7D), 1:15
(7E), and 1:29 (7F), respectively. The degrees of functional-
ization of polymers with R-Gal trisaccharide were 36% (7A),
28% (7B), 23% (7C), 13% (7D), 6% (7E), and 3% (7F),
respectively.
The synthesis of R-Gal trisaccharide derivative11 {N-[O-R-
D-galactopyranosyl)-(1f3)-O-(â-D-galactopyranosyl)-(1f4)-1-
â-D-glucopyranosyl]-5-aminopentamide} 6 started with a gly-
cosylation reaction between donor 8 and acceptor 9 promoted
by NIS/TfOH in dichloromethane at -30 °C with 4 Å MS as
water scavenger. Trisaccharide derivative 10 was afforded in
90% yield with 20:1 (R/â) selectivity. A mild hydrogenation
condition with PtO2 in methanol selectively reduced the azido
group of 10 to primary amine, which was immediately reacted
with 5-chlorovaleryl chloride to give compound 11 in 87% yield.
Sequential debenzylation and acetylation reactions of 11
produced peracetylated trisaccharide 13 in 89% yield. Nucleo-
philic substitution of 13 with sodium azide in DMF gave 14 in
90% yield. The fully deprotected R-Gal trisaccharide 6 was
therefore obtained from 14 in 93% yield after deacetylation and
hydrogenation.
A shortcut to compound 13 was the chemo-enzymatic
synthesis11 illustrated in Scheme 2. Glycosylation of UDP-
glucose (16) and acceptor 17 produced trisaccharide 18 in 65%
yield using a fusion enzyme (GalE-GalT)12 as a catalyst. This
enzyme contained both uridine-5′-diphospho-galactose 4-epi-
merase (GalE) and R(1f3) galactosyltransferase (GalT) and was
recently cloned in our laboratory by in-frame fusion of the
Escherichia coli gene galE to the 3′-terminus of a truncated
bovine galT gene within a high-expression plasmid. It has dual
In this report, we synthesized a series of polymers with varied
densities of R-Gal epitopes to investigate their binding affinity
(6) (a) Xu, Y.; Lorf, T.; Sablinski, T.; Gianello, P.; Bailin, M.; Monroy,
R.; Kozlowski, T.; Awwad, M.; Cooper, D. K. C.; Sachs, D. H.
Transplantation 1998, 65 (2), 172. (b) Taniguchi, S.; Neethling, F. A.;
Korchagina, E. Y.; Bovin, N.; Ye, Y.; Kobayashi, T.; Niekrasz, M.; Li, S.;
Koren, E.; Oriol, R.; Cooper, D. K. C. ibid 1996, 62, 1379. (b) Kroshus, T.
J.; Bolman, R. M.; Dalmasso, A. P. ibid 1996, 62, 5. (c) Leventhal, J. R.;
John, R.; Fryer, J. P.; Witson, J. C.; Derlich, J.-M.; Remiszewski, J.;
Dalmasso, A. P.; Matas, A. J.; Bolman, R. M., III ibid 1995, 59 (2), 294.
(7) (a) Simon, P. M.; Neethling, F. A.; Taniguchi, S.; Goode, P. L.; Zopf,
D.; Hancock, W. W.; Cooper, D. K. C. Transplantation 1998, 65 (3), 346.
(b) Rieben, R.; Von Allmen, E.; Korchagina, E. Y.; Nydegger, U.; Neethling,
F. A.; Kujundzic, M.; Koren, E.; Bovin, N.; Cooper, D. K. C. Xenotrans-
plantation 1995, 2, 98. (c) Li, S.; Neethling, F.; Taniguchi, S.; Yeh, J.-C.;
Kobayashi, T.; Ye, Y.; Koren, E.; Cummings, R. D.; Cooper, D. K. C.
Transplantation 1996, 62 (9), 1324.
(8) Galili, U.; Matta, K. L. Transplantation 1996, 62 (2), 256.
(9) (a) Lee, Y. C.; Townsend, R. R.; Hardy, M. R.; Lo¨nngren, J.; Arnarp,
J.; Haraldson, M.; Lo¨nn, H. J. Biol. Chem. 1983, 258, 199. (b) Lee, Y. C.;
Lee, R. T.; Rice, K.; Ichikawa, Y.; Wong, T.-C. Pure. Appl. Chem. 1991,
63, 499. For some reviews on multivalency, see: (c) Mammen, M.; Chai,
S.-K.; Whitesides, G. M. Angew. Chem., Int. Ed. Engl. 1998, 37, 7 (20),
2754. (d) Zanini, D.; Roy, R. Carbohydrate mimics: Concepts and Methods;
Chapleur, Y., Ed.; Verlag Chemie: Weinheim, Germany, 1998; Chapter
20, pp 385-415. (e) Roy, R. Top. Curr. Chem. 1997, 187, 241. (f) Kiessling,
L. L.; Pohl, N. L. Chem. Biol. 1996, 3, 71. (g) Lee, Y. C.; Lee, R. T. Acc.
Chem. Res. 1995, 28, 321.
(10) (a) Pollak, A.; Blumenfeld, H.; Wax, M.; baughn, R. L.; Whitesides,
G. M. J. Am. Chem. Soc. 1980, 102, 6324. (b) Mammen, M.; Dahmann,
G.; Whitesides, G. M. J. Med. Chem. 1995, 38, 4179. (c) Nishida, Y.; Dohi,
H.; Uzawa, H.; Kobayashi, K. Tetrahedron Lett. 1998, 39, 8681. (d) Bovin,
N. V. Glycoconjugate J. 1998, 15, 431. (e) Li, J.; Zacharek, S.; Chen, X.;
Wang, J.; Zhang, W.; Janczuk, A.; Wang, P. G. Bioorg. Med. Chem. 1999,
in press. (f) Bovin, N. V.; Korchagina, E. Y.; Zemlyanukhina, T. V.;
Byramova, N. E.; Galanina, O. E.; Zemlyakov, A. E.; Ivanov, A. E.; Zubov,
V. P.; Mochalova, L. V. Glycoconjugate J. 1993, 10, 142.
(11) (a) Fang, J.; Li, J.; Xi, C.; Zhang, Y.; Wang, J.; Guo, Z.; Zhang,
W.; Yu, L.; Brew, K.; Wang, P. G. J. Am. Chem. Soc. 1998, 120, 6635. (b)
Fang, J.; Chen, X.; Zhang, W.; Wang, J.; Andreana, P. R.; Wang, P. G. J.
Org. Chem. 1999, 64, in press. (c) Vic, G.; Scigelova, M.; Hastings, J. J.;
Howarth, O. W.; Crout, D. H. G. Chem. Commun. 1996, 1473. (d) Vic, G.;
Tran, C. H.; Scigelova, M.; Crout, H. G. ibid 1997, 169. (e) Thiem, J.;
Wiemann, T. Synthesis 1992, 141.
(12) Chen, X.; Fan, H.; Zhang, Y.; Brew, K.; Wang, P. G. manuscript
in preparation.

8176 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Wang et al.
Scheme 1. Chemical Synthesis of R-Gal Polymers (7A-F)a
a Reagents: (a) NIS, TfOH, CH2Cl2, 4A MS, -30 °C, 90%; (b) PtO2, H2, MeOH; (c) 5-chlorovaleryl chloride, Et3N, CH2Cl2, 87% (b and c); (d)
Pd/C, H2, MeOH; (e) Py/Ac2O, DMAP, 89%; (f) NaN3, DMF, 70 °C, 90%; (g) NaOMe, MeOH; (h) PtO2/H2, H2O-MeOH, 93% (g and h); (i) AlBN,
PhH, reflux; (j) DMF, rt, 24 h; 65 °C, 6 h; rt, 24 h then NH3‚H2O, rt, 24 h.
Scheme 2. Enzymatic Chemical Synthesis of 13
functions both as an epimerase and as a galactosyltransferase
that allows the use of relatively inexpensive UDP-glucose
instead of the high-cost UDP-galactose as the donor. The
compound 13 was then obtained from 18 after peracetylation,
reduction, and reaction with 5-chlorovaleryl chloride sequen-
tially.
Scheme 3. Chemoenzymatic Synthesis of 6
In addition, a more efficient synthesis of 6 was also achieved
using this bifunctional fusion enzyme. Reaction of disaccharide
derivative 24 with UDP-glucose gave compound 6 in 52% yield.
Polylactose13 25 with lactose/acrylamide ratio of 1:2.5 was
synthesized as a negative control for bioassays. The synthetic
route was shown in Scheme 4. Compound 21 was obtained in
79% yield using the same transformation from 10 to 11 as
illustrated in Scheme 1. Nucleophilic substitution of chloride
in 21 with sodium azide gave 22 in 84% yield. The fully
deprotected lactose derivative 24 was then obtained from 22 in
(13) (a) Park, W. K. C.; Aravind, S.; Romanowska, A.; Renaud, J.; Roy,
R. Methods Enzymol. 1994, 247, 325. (b) Aravind, S.; Park, W. K. C.;
Brochu, S.; Roy, R. Tetrahedron Lett. 1994, 35 (42), 7739. (c) Roy, R.;
Tropper, F. D.; Romanowska, A. Bioconjugate Chem. 1992, 3, 256, and
the references therein.

Human Anti-Gal Antibody Binding
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8177
Scheme 4. Synthesis of Polylactose 25a
Table 2. Inhibition of R-Gal Polymers and Monomer against the
Binding of Human Serum (male, Blood Type: AB) to Mouse
Laminin
IC50 [µM]b
compound
ratioa
IgG
74 ( 11 306 ( 22
1:1.8 12( 3 1.14( 0.26
IgA
IgM
monomer 26
polymer 7A
polymer 7B
polymer 7C
polymer 7D
polymer 7E
268 ( 31
0.070 ( 0.001
0.035( 0.002
1:2.5 9.4( 0.9 0.12 ( 0.04
1:3.3 18 ( 5
1:6.4 52( 8
0.091( 0.003 0.031( 0.009
3.3( 0.7
>1000
N/I
0.38( 0.02
89( 27
N/I
1:15
>500
polylactose 25 1:2.5 N/I
a Reagents: (a) PtO2, H2, MeOH, then 5-chlorovaleryl chloride, Et3N,
79%; (b) NaN3, DMF, 70 °C, 84%; (c) NaOMe, MeOH; (d) PtO2/H2,
H2O-MeOH, 91% (two steps); (e) pNAS, DMF, rt, 24 h; 65 °C, 6 h;
rt, 24 h, then NH3‚H2O, 24 h.
a Ratio of the unit of R-Gal epitope (or lactose) to that of acrylamide
in polymer. b The concentration of inhibitor at 50% inhibition of the
binding between human serum (10 µL) and mouse laminin in inhibition
ELISA. Errors are reported as 2σ. N/I: no inhibition.
Table 1. Inhibition of R-Gal Polymers and Monomer against the
Binding of Purified Human Anti-Gal Antibody (Male, Blood Type:
AB) to Mouse Laminin
Table 3. Inhibition of R-Gal Polymers and Monomer against the
Binding of Human Serum (Male, Blood Type: AB) to Pig Kidney
(PK15) Cells
IC50 [µM]b
IC50[µM]b
compound
ratioa
IgG
IgA
IgM
277 ( 24
0.0068 ( 0.0016
compound
ratioa
IgG
IgA
IgM
monomer 26
polymer 7A
polymer 7B
polymer 7C
polymer 7D
polymer 7E
polymer 7F
69 ( 5
55 ( 16
monomer 26
polymer 7A
polymer 7B
polymer 7C
polylactose 25
∼1000
>1000
>1000
1/1.8 0.37 ( 0.08 0.016 ( 0.006
1/2.5 0.28 ( 0.05 0.0053 ( 0.0023 0.0063 ( 0.0015
1:1.8
1:2.5
1:3.3
1:2.5
595
5.2
2.2
>1000
N/I
407
774
N/I
63
1/3.3 2.8 ( 0.9
1/6.4 7.2 ( 2.7
1/15 60 ( 5
0.0070 ( 0.0032 0.0056 ( 0.0014
0.043 ( 0.017
0.86 ( 0.28
0.027 ( 0.018
165 ( 13
N/I
1/29 93 ( 16
a Ratio of the unit of R-Gal epitope (or lactose) to that of acrylamide
in polymer. b The concentration of inhibitor at 50% inhibition of the
binding between human sera and pig kidney cells in flow cytometry
assay. N/I: no inhibition.
polylactose 25 1/2.5 N/I
N/I
N/I
a Ratio of the unit of R-Gal epitope (or lactose) to that of acrylamide
in polymer. b The concentration of inhibitor at 50% inhibition of the
binding between purified anti-Gal antibody (16 µg/mL) and mouse
laminin in inhibition ELISA. Errors are reported as 2σ. N/I: no
inhibition.
numbers of binding sites from IgG to IgA to IgM. In human
serum, IgM exists as a pentamer with 10 equivalent binding
sites in one molecule; 80% of IgA exists as a monomer with
two binding sites and 20% as higher oligomers with a higher
number of binding sites; IgG exists as a monomer with two
binding sites. Therefore, the “multivalent” effect is certainly
more pronounced for antibodies (IgM, IgA) with more protein-
carbohydrate interaction sites.
91% yield after deacetylation and hydrogenation. Polylactose
25 with lactose/acrylamide ratio of 1:2.5 was prepared by
reaction of pNAS (5) and intermediate 24 using the same
procedure as in the preparation of polymer 7B.
Evaluation of r-Gal Epitope Polymers. The evaluation of
the binding affinities of synthesized R-Gal-containing polymers
7 to anti-Gal antibodies was accomplished by inhibition ELISA14
(enzyme-linked immunosorbent assay) with purified human
(male, blood type AB) anti-Gal antibody as the primary anti-
body and mouse laminin as a natural source of R-Gal. The
concentrations of R-Gal polymers at 50% inhibition (IC50) of
anti-Gal antibody binding to R-Gal epitopes on mouse laminin
were measured and the IC50 data were summarized in Table 1.
All of the IC50 data presented in the text are the net R-Gal
trisaccharide concentrations (micromolar) calculated from the
degree of functionalization and polymerization of each polymer,
to compare with the corresponding R-Gal monomer explicitly.
Results indicated that polymers 7A-7D inhibit the antibody/
antigen binding better than the R-Gal monomer, GalR1-
3Galâ1-4Glcâ-NHAc (26), in all cases. The efficacy can be
illustrated by the IC50 result for polymer 7C of 5.6 nM, which
is 5.0 × 104-fold better than its monomeric analogue 26 in
inhibiting anti-Gal IgM. Polymer 7B has an IC50 of 5.3 nM
with 104-fold enhancement over the R-Gal monomer 26 in
inhibiting anti-Gal IgA. The enhancement proved to be greater
for anti-Gal IgM and IgA than for anti-Gal IgG. For example,
the activity enhancement of polymer 7B is 246-, 1.0 × 104-,
and 4.4 × 104-fold toward anti-Gal IgG, IgA, and IgM,
respectively. This observation is consistent with the increasing
The effectiveness of R-Gal polymers varied not only with
antibody isotype but also with the density of R-Gal epitopes
conjugated to the polymer. There is a steady increase in
inhibition efficacy for IgG and IgA from polymer E to B with
increased R-Gal densities in the polymers (Table 1). The same
trend can be observed for IgM from polymer E to C. However,
polymer 7A, with a higher density of R-Gal residues, exhibited
a lower inhibition than polymer 7B for IgA binding and the
same level as 7B for IgG and IgM. This activity drop can be
rationalized by the thermodynamics of polyvalent interactions
in a biological system recently reviewed by Whitesides, et
al.9c,10b Negative cooperative interactions and the enthalpically
diminished binding can be caused by the unfavorable spatial
conformation of polymers at high ligand concentrations. The
enhancements of inhibition by introducing more R-Gal epitopes
in 7A were obviously overcome by the simultaneous formation
of the noncooperative spatial conformation of the polymer as
well as steric hindrance. Similar to the positive cooperative
multivalent interactions, this negative effect is more evident for
antibodies with more binding sites, such as IgA and IgM. Work
is in progress to elucidate the detailed physical mechanism of
this enhancement. As a control in the assays, polylactose did
not show any inhibitory activities up to 1 mM.
To assess the practical application of R-Gal polymers, we
tested intact human serum (male, blood type: AB) instead of
purified human anti-Gal antibody using the inhibition ELISA
(14) Galili, U.; LaTemple, D. C.; Radic, M. Z. Transplantation 1998,
65 (8), 1129.

8178 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Wang et al.
Figure 2. Flow cytometry demonstration of enhanced inhibition of R-Gal polymer B versus monomer against the binding of anti-Gal IgM in
human serum to pig kidney (PK15) cells: horizontal axis, fluorescence intensity; vertical axis, cell counts. Peaks: no staining (PK15 cells alone),
left black; maximum staining (PK15 cells + human serum + secondary Ab), right black; staining in the presence of polymer B, red; staining in the
presence of monomer, green. Concentrations of net trisaccharide for both polymer and monomer: 0.1 µM (I), 1 µM (II), 10 µM (III), and 100 µM
(IV).
(Table 2). The results also indicated the activity enhancement
of the R-Gal polymers as compared to the monomer 26. The
same trends were also observed in the interaction of R-Gal
polymers with different isotypes of the antibody and with the
varied densities of the R-Gal epitope conjugated to the polymer.
Interestingly, the IC50s observed with the purified antibodies
were consistently lower than the IC50s with human sera. One
explanation is that the purified antibodies were obtained from
affinity column immobilized with R-Gal trisaccharide similar
in structure to the epitope on the polymer. Therefore, subsets
of antibodies selected during the purification would bind most
tightly to the polymer. Since the purified antibodies from our
affinity column contained 89% of IgM, 50% of IgA, and 42%
of IgG of corresponding anti-Gal in human sera,4b,6a the
increased proportion of IgM in the purified anti-Gal would
contribute to the greater inhibitory effect of the R-Gal polymers.
To more realistically mimic cell transplantation situations,
we carried out a flow cytometry assay15 in which the inhibition
of synthetic polymers against the binding of human natural anti-
Gal antibodies in intact serum to R-Gal epitopes on pig kidney
cells (PK15, ATCC) was measured (Table 3). As expected,
results from flow cytometry analysis demonstrated that R-Gal
polymers exhibited enhanced inhibition activities in comparison
to the monomer 6. With IC50s of approximately 5.2 and 2.2
µM, respectively, high levels of inhibition were achieved with
polymers 7A and 7B against anti-Gal IgM binding to PK15 cells.
In sharp contrast, the R-Gal monomer 26 has no significant
inhibitory effect up to 1 mM, whereas polylactose did not exhibit
any inhibition up to 1 mM. This inhibition enhancement in anti-
Gal IgM binding is clearly illustrated by the downshift of the
fluorescence intensity peak of polymer B (red curve) relative
to that of the R-Gal monomer (green curve) shown in cytometry
histograms (Figure 2, where low fluorescence represents higher
inhibition of the anti-Gal binding). In all concentrations tested,
ranging from 0.10 µM to 100 µM, no inhibition was observed
with monovalent R-Gal trisaccharide. One can see that the
enhancement of inhibitory activity of the polymer is concentra-
tion-dependent. At a net trisaccharide concentration of 0.1 µM,
the enhancement of the polymer over the monomer was minimal,
while at concentration of 100 µM, the inhibition of anti-Gal
binding to PK15 cells by the polymer jumped to 68%.
magnitude) than the values for binding to laminin. This may
be ascribed to the lower density of epitopes on the cell
surface.16,4,17 Moreover, the concentration of human sera used
in the cell assay was 2.5 times more than that in laminin assay,
which would bind more R-Gal epitopes on the polymer. Other
factors such as non-R-Gal antigen-antibody binding18 may
contribute to the difference at the same time.
Conclusion
In summary, the synthetic R-Gal-conjugated polymers sig-
nificantly enhanced activities in the inhibition of human anti-
Gal antibody binding to mouse laminin glycoproteins and
mammalian PK15 cells. Such enhancement is greater for anti-
Gal IgA and IgM than for IgG. The amplified binding
differences among the three anti-Gal isotypes can be utilized
to selectively inhibit and remove particular isotype antibodies.
Moreover, it was demonstrated through flow cytometry analysis
that certain R-Gal polymers are effective in inhibiting anti-Gal
antibodies in human serum binding to pig kidney cells. Thus,
such synthetic carbohydrate polymers not only can serve as tools
in studying R-Gal/anti-Gal interactions but also may find
practical applications in cell xenotransplantations.
Experimental Section
1
General. H and 13C spectra were recorded on 400 MHz Varian
VXR400 NMR and 500 MHz Varian Unity spectrometers. Mass spectra
were run at the mass spectrometry facility at the University of
California, Riverside. Baker silica gel (40 µm) was used for column
chromatography and E. Merck precoated TLC plates for thin-layer
chromatography. Size-exclusion chromatography was performed on
Biogel P2 resin using distilled water as the eluent. Dialysis was
performed using Spectra/Por Molecularporous membrane (16 mm
cylinder diameter, molecular weight cutoff 14 kD) against deionized
water.
Poly[N-(acryloyloxy)succinimide] (5). A mixture of N-(acryloyl-
oxy)succinimide (4) (6.86 g, 40.5 mmol) and AIBN (40 mg, 0.006
equiv) in benzene (300 mL) was heated at 60 °C for 36 h. After the
solution was cooled to room temperature, a white precipitate was
formed. This precipitate was filtered and washed with dry tetrahydro-
furan (3 × 100 mL). Drying in vacuo afforded poly[N-(acryloyloxy)-
succinimide] (5) (6.21 g, 91%; Lit.10a,b 98%) as a white fluffy solid.
The polymer was taken up in dry THF (500 mL), vigorously stirred
Our results also demonstrated that IC50s of the polymers
against antibody binding to cells were much higher (orders of
(16) Galili, U.; Shohet, S. B.; Kobrin, E.; Stults, C. L. M.; Macher, B.
A. J. Biol. Chem. 1988, 263, 17755.
(17) (a) Timpl, R. Eur. J. Biochem. 1989, 180, 487. (b) Beck, K.; Hunter,
I.; Engel, J. FASEB J. 1990, 4, 148.
(15) (a) Melamed, M. R., Lindmo, T., Mendelsohn, M. L., Eds. Flow
Cytometry and Sorting, 2nd ed.; Wiley-Liss, New York, 1990. (b)
Jacquemin-Sablon, A., Ed. Flow Cytometry: New DeVelopments; Springer-
Verlag: Berlin, New York, 1993.
(18) Cooper, D. K. C. Xenotransplantation 1998, 5, 6.

Human Anti-Gal Antibody Binding
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8179
1
for 3 days, filtered, and dried in vacuo. H NMR (400 MHz, DMSO)
23.5, 21.9, 21.8 (m), 21.5. HRMS calcd for C63H76NO22ClNa+
1256.4445, found 1256.4476.
δ 2.09 (br., CH2), 2.80 (br., CH2CH2), 3.13 (br., CH).
Determination of Molecular Weight of Poly[N-(acryloyloxy)-
succinimide] (pNAS). To a solution of 5 (50 mg) in DMF (2 mL) was
added 1 N NaOH (10 mL). The mixture was heated at 65 °C for 24 h.
After cooling to room temperature, the solution was exhaustively
dialyzed against distilled H2O. The solution of poly(acrylic acid) sodium
salt thus obtained was analyzed by HPLC (Waters Ultrahydrogel Linear)
using the following standards: poly(acrylic acid) sodium salt of MW
20, 60, 140, and 225 kD. Found: Mw ) 252 kD, MN ) 172 kD, Mw/
MN ) 1.5.
N-[O-(r-D-galactopyranosyl)-(1f3)-O-(2,4,6-tri-O-acetyl-â-D-ga-
lactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-1-â-D-glucopyranosyl]-5-
chloropentamide (12). A solution of N-[O -(2,3,4,6-Tetra-O-benzyl-
R-D-galactopyranosyl)-(1f3)-O-(2,4,6-tri-O-acetyl-â-D-galactopyrano-
syl)-(1f4)-2,3,6-tri-O-acetyl-1-â-D-glucopyranosyl]-5-chloropentamide
(11) (8.39 g, 6.80 mmol) in methanol (200 mL) was hydrogenated at
room temperature and 4 atm pressure in the presence of palladium on
activated carbon (10%) (2 g) overnight. The reaction mixture was
diluted with methanol (100 mL), filtered through a pad of Celite
(caution: Pd/C can easily cause fire), washed with methanol (200 mL),
and concentrated to give debenzylated product 12 (5.96 g, 100%, Rf )
O-(2,3,4,6-Tetra-O-benzyl-r-D-galactopyranosyl)-(1f3)-O-(2,4,6-
tri-O-acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-1-â-D-
glucopyranosyl azide (10). To a solution of O-(2,4,6-tri-O-acetyl-â-
D-galactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-1-â-D-glucopyranosyl azide
(9) (0.27 g, 0.43 mmol) and phenyl 2,3,4,6-tetra-O-benzyl-1-thio-â-D-
galactopyranoside (8) (0.41 g, 0.65 mmol) in dichloromethane (15 mL)
was added 4 Å molecular sieves (0.8 g). The mixture was stirred for 1
h at room temperature and then cooled to -30 °C. N-Iodosuccinimide
(NIS, 95%; 0.25 g, 0.65 mmol) and trifluoromethanesulfonic acid
(TfOH; 0.02 mL) were added. The reaction mixture was stirred for 1
h at -30 °C. Triethylamine was added to adjust the pH to 8.0. The
mixture was filtered through a pad of Celite and washed with
chloroform. The combined solution was successively washed with
aqueous sodium thiosulfate and water. After concentration, the residue
was chromatographed on a silica gel column (hexane/ethyl acetate: 2:1,
3:2, sequentially) to give the product 10 (0.42 g, 86%, Rf ) 0.50, 1:1
hexane/ethyl acetate) as a white foamy solid. 1H NMR (500 MHz,
CDCl3) δ 1.82 (s, 3H), 1.93 (s, 3H), 2.01 (s, 3H), 2.07 (s, 3H), 2.08 (s,
3H), 2.09 (s, 3H), 3.49 (d, J ) 6.4 Hz, 2H), 3.64-3.82 (m, 6H), 3.97-
4.12 (m, 3H), 4.31 (d, J ) 7.8 Hz, 1H), 4.37-4.51 (m, 4H), 4.60-
4.73 (m, 4H), 4.80-4.93 (m, 3H), 5.04-5.12 (m, 2H), 5.19 (m, 2H),
5.43 (d, J ) 3.0 Hz, 1H), 7.22-7.34 (m, 20H). 13C NMR (125.7 MHz,
CDCl3) δ 170.4, 170.2, 170.1, 169.5, 169.4, 168.8, 138.6, 138.5 (m),
138.0, 128.3 (m), 128.2 (m), 128.1 (m), 127.9 (m), 127.7 (m), 127.6
(m), 127.5, 127.4, 101.1, 95.0, 87.7, 78.4, 75.7, 75.3, 75.2, 74.9, 74.8,
73.6, 73.3, 73.2, 72.9, 72.6, 71.1, 71.0, 70.5, 69.8, 68.5, 64.8, 61.9,
61.3, 20.7 (m), 20.6 (m), 20.5, 20.4. MS (m/e) 1164 (M + Na+). HRMS
calcd for C58H67N3O21Na+ 1164.4165, found 1164.4117.
1
0.25, 8:1 chloroform/methanol). H NMR (500 MHz, CDCl3) δ 1.75
(m, 4H), 2.04(1) (s, 3H), 2.04(4) (s, 3H), 2.08 (s, 3H), 2.09 (s, 3H),
2.11 (s, 3H), 2.16 (s, 3H), 2.20 (m, 2H), 3.52 (t, J ) 6.5 Hz, 2H), 3.55
(m, 1H), 3.64-3.92 (m, 9H), 4.09 (m, 2H), 4.21 (m, 1H), 4.41 (m,
2H), 4.82 (dd, J1 ) J2 ) 9.0 Hz, 1H), 4.98 (m, 2H), 5.20 (dd, J1 ) J2
) 9.0 Hz, 1H), 5.27 (m, 1H), 5.45 (m, 1H), 6.39 (d, J ) 9.0 Hz, 1H).
13C NMR (125.7 MHz, CDCl3) δ 173.4, 172.1, 171.9, 171.4, 171.3,
170.7, 170.4, 101.4, 97.9, 78.6, 76.5, 75.2 (m), 73.1, 71.6 (m), 71.3,
70.7, 70.5, 69.3, 66.5, 62.8, 62.6, 62.1, 45.1, 36.2, 32.4, 23.1. HRMS
calcd for C35H52NO22ClNa+ 896.2567, found 896.2530.
N-[O-(2,3,4,6-Tetra-O-acetyl-r-D-galactopyranosyl)-(1f3)-O-
(2,4,6-tri-O-acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-
1-â-D-glucopyranosyl]-5-chloropentamide (13) (from 12). N-[O -(R-
D-galactopyranosyl)-(1f3)-O-(2,4,6-tri-O-acetyl-â-D-galactopyranosyl)-
(1f4)-2,3,6-tri-O-acetyl-1-â-D-glucopyranosyl]-5-chloropentamide (12)
(6.04 g, 6.91 mmol) was dissolved in a mixture of pyridine (100 mL)
and acetic anhydride (70 mL). 4-(Dimethylamino)pyridine (100 mg)
was added, and the solution was stirred at room temperature for 3 h.
After removal of pyridine and acetic anhydride, the residue was
dissolved in ethyl acetate (300 mL) and the solution was successively
washed with HCl (0.5 N aqueous, 50 mL), water (100 mL), and
saturated aqueous sodium bicarbonate solution. After concentration,
the residue was chromatographed on a silica gel column (hexane/ethyl
acetate: 1:1, 1:2, sequentially) to give the product 13 (6.39 g, 89%, Rf
) 0.5, 1:3 hexane/ethyl acetate) as a white solid. 1H NMR (300 MHz,
CDCl3) δ 1.73 (m, 4H), 1.91 (s, 3H), 2.01 (m, 6H), 2.02 (s, 3H), 2.03
(s, 3H), 2.04 (s, 3H), 2.06 (s, 3H), 2.10 (s, 6H), 2.12 (s, 3H), 2.18 (m,
2H), 3.49 (t, J ) 6.0 Hz, 2H), 3.77 (m, 4H), 4.14 (m, 6H), 4.35 (m,
2H), 4.78 (dd, J1 ) J2 ) 9.9 Hz, 1H), 5.03-5.29 (m, 7H), 5.41 (d, J
) 3.0 Hz, 1H), 6.22 (d, J ) 9.0 Hz, 1H). 13C NMR (75.5 MHz, CDCl3)
δ 172.5, 171.2, 170.3, 170.2 (m), 170.1 (m), 169.8, 169.7, 169.2, 168.6,
100.7, 93.5, 77.9, 75.6, 74.5, 72.9, 72.2, 70.9, 70.8, 69.6, 67.6, 67.1,
66.8, 66.5, 64.6, 61.9, 61.1, 61.0, 44.3, 35.5, 31.7, 22.4, 20.6-20.4
(m). HRMS calcd for C43H60NO26ClNa+ 1064.2990, found 1064.3017.
N-[O-(2,3,4,6-Tetra-O-benzyl-r-D-galactopyranosyl)-(1f3)-O-
(2,4,6-tri-O-acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-
1-â-D-glucopyranosyl]-5-chloropentamide (11). A solution of O
-(2,3,4,6-tetra-O-benzyl-R-D-galactopyranosyl)-(1f3)-O-(2,4,6-tri-O-
acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-1-â-D-glucopy-
ranosyl azide (10) (9.08 g, 7.96 mmol) in methanol (200 mL) was
hydrogenated for 2 h at room temperature and 4 atm pressure in the
presence of platinum(IV) oxide hydrate (380 mg). The solid was filtered
off, and the filtrate was concentrated under reduced pressure to give a
white solid. Then anhydrous dichloromethane (300 mL) was added to
dissolve the solid and the mixture was cooled to -78 °C. Triethylamine
(3.32 mL, 23.88 mmol) and 5-chlorovaleryl chloride (1.27 mL, 11.94
mmol) were added sequentially to the reaction mixture. The mixture
was then warmed to room temperature and stirred for 3 h. After
concentration, the residue was dissolved in ethyl acetate (300 mL) and
then washed successively with HCl (0.5 N aqueous, 50 mL), water
(100 mL), and saturated aqueous sodium bicarbonate solution. After
solvent removal, the residue was chromatographed on a silica gel
column (hexane/ethyl acetate: 1:1, 2:3, sequentially) to afford product
11 (8.54 g, 87%, Rf ) 0.45, 1:1 hexane/ethyl acetate) as a white foamy
solid. 1H NMR (400 MHz, CDCl3) δ 1.78 (m, 4H), 1.85 (s, 3H), 1.96
(s, 3H), 2.05 (s, 3H), 2.08 (s, 3H), 2.10 (s, 3H), 2.11 (s, 3H), 2.22 (m,
2H), 3.53 (m, 5H), 3.70 (m, 1H), 3.76 (m, 1H), 3.81 (m, 4H), 4.00-
4.17 (m, 4 H), 4.33 (m, 1H), 4.42 (m, 2H), 4.53 (m, 2H), 4.66 (m,
1H), 4.73 (m, 2H), 4.84 (m, 2H), 4.94 (m, 1H), 5.10 (m, 2H), 5.29 (m,
2H), 5.46 (d, J ) 3.2 Hz, 1H), 6.29 (d, J ) 9.2 Hz, 1H), 7.26-7.40
(m, 20H). 13C NMR (100.6 MHz, CDCl3) δ 173.7, 172.5, 171.6, 171.5,
171.4, 170.5, 169.9, 139.7 (m), 139.1, 129.5 (m), 129.4 (m), 129.3
(m), 129.2 (m), 128.8 (m), 128.7 (m), 128.6 (m), 128.5 (m), 101.9,
96.1, 79.5, 79.1, 76.7, 76.5, 76.3, 75.9, 75.7, 74.8, 74.4, 74.3, 74.0,
73.4, 72.2, 72.1, 71.5, 70.9, 69.5, 65.9, 63.2, 62.5, 45.5, 36.7, 32.9,
N-[O-(2,3,4,6-Tetra-O-acetyl-R-D-galactopyranosyl)-(1f3)-O-
(2,4,6-tri-O-acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-
1-â-D-glucopyranosyl]-5-azidopentamide (14). A mixture of N-[O
-(2,3,4,6-Tetra-O-acetyl-R-D-galactopyranosyl)-(1f3)-O-(2,4,6-tri-O-
acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-1-â-D-glucopy-
ranosyl]-5-chloropentamide (13) (6.31 g, 6.06 mmol) and sodium azide
(0.79 g, 12.12 mmol) in DMF (20 mL) was heated at 70 °C for 12 h.
After removal of DMF, the residue was dissolved in ethyl acetate (300
mL) and the solution was filtered through a pad of Celite and washed
with ethyl acetate (100 mL). The combined filtrate was concentrated,
and the residue was chromatographed on a silica gel column (hexane/
ethyl acetate: 1:1, 1:2, sequentially) to give the product 14 (5.71 g,
1
90%, Rf ) 0.45, 1:3 hexane/ethyl acetate) as a white solid. H NMR
(400 MHz, CDCl3) δ 1.58 (m, 2H), 1.65 (m, 2H), 1.94 (s, 3H), 2.04(2)
(s, 3H), 2.04(5) (s, 3H), 2.05(3) (s, 3H), 2.05(8) (s, 3H), 2.07 (s, 3H),
2.09 (s, 3H), 2.13 (s, br., 6H), 2.16 (s, 3H), 3.28 (t, J ) 6.4 Hz, 2H),
3.80 (m, 4H), 4.16 (m, 6H), 4.38 (m, 2H), 4.81 (dd, J1 ) J2 ) 9.6 Hz,
1H), 5.06-5.44 (m, 6H), 5.44 (d, J ) 3.0 Hz, 1H), 6.24 (d, J ) 9.6
Hz, 1H). 13C NMR (100.6 MHz, CDCl3) δ 173.6, 172.4, 171.5, 171.4
(m), 171.3 (m), 170.9, 170.8, 170.3, 169.7, 101.9, 94.5, 79.0, 76.8,
75.6, 74.0, 73.3, 72.0, 71.9, 70.7, 68.7, 68.2, 67.9, 67.6, 65.7, 63.0,
62.3, 62.1, 52.1, 36.8, 29.3, 23.3, 21.9, 21.8 (m), 21.7 (m), 21.6 (m),
21.5. HRMS calcd for C43H60N4O26ClNa+ 1071.3393, found 1071.3362.

8180 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Wang et al.
N-[O-r-D-Galactopyranosyl)-(1f3)-O-(â-D-galactopyranosyl)-
(1f4)-1-â-D-glucopyranosyl]-5-azidopentamide (15). To a solution
of N-[O -(2,3,4,6-Tetra-O-acetyl-R-D-galactopyranosyl)-(1f3)-O-(2,4,6-
tri-O-acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-1-â-D-glu-
copyranosyl]-5-azidopentamide (14) (1.01 g, 0.96 mmol) in absolute
methanol (500 mL) was added sodium methoxide to adjust the pH value
of the solution to 9. The mixture was stirred for 5 h and then neutralized
with Dowex 50WX2-100 (H+) resin. The resin was filtered and washed
with methanol. The combined methanol solution was concentrated to
give product 15 (0.57 g, 93%, Rf ) 0.75, 8:3:3 2-propanol/water/ethyl
1-â-D-glucopyranosyl]-5-chloropentamide (13) (from 19). Compound
13 (0.19 g, 81%) was prepared from azide (19) (0.22 g, 0.23 mmol)
using the same procedure described in the preparation of 11. Spectra
data was shown in the preparation of 13 from 12.
N-[O-r-D-Galactopyranosyl)-(1f3)-O-(â-D-galactopyranosyl)-
(1f4)-1-â-D-glucopyranosyl]-5-aminopentamide (6) (from 24). Com-
pound 6 (73 mg, 52%) was prepared from 24 (103 mg, 0.23 mmol)
using the same procedure described in the preparation of 18. Spectra
data was shown in the preparation of 6 from 15.
N-[O-(2,3,4,6-tera-O-acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-
O-acetyl-1-â-D-gluco pyranosyl]-5-chloropentamide (21). Compound
21 (1.56 g, 79%, Rf ) 0.45, 1:2 hexane/ethyl acetate, white foamy
solid) was prepared from 20 (1.74 g, 2.73 mmol) using the same
procedure described in the preparation of 11. 1H NMR (500 MHz,
CDCl3) δ 1.72 (m, 4H), 1.93 (s, 3H), 2.01 (s, 3H), 2.02 (s, br., 6H),
2.04 (s, 3H), 2.08 (s, 3H), 2.13 (s, 3H), 2.18 (m, 2H), 3.49 (t, J ) 6.4
Hz, 2H), 3.72 (m, 2H), 3.86 (t, J ) 6.8 Hz, 1H), 4.05 (m, 3H), 4.41 (d,
J ) 11.6 Hz, 1H), 4.45 (d, J ) 8.0 Hz 1H), 5.07 (dd, J1 ) 8.0 Hz, J2
) 10.0 Hz, 1H), 5.18 (t, J ) 9.5 Hz, 1H), 5.26 (t, J ) 9.0 Hz, 1H),
5.32 (d, J ) 3.5 Hz, 1H), 6.31 (d, J ) 8.8 Hz, 1H). 13C NMR (125.7
MHz, CDCl3) δ 173.2, 171.9, 171.0 (2C), 170.8, 170.7, 169.9, 169.6,
101.5, 78.7, 76.6, 75.2, 73.1, 71.7 (2C), 71.4, 69.7, 67.3, 62.6, 61.6,
45.0, 36.2, 32.4, 23.1, 21.5, 21.4, 21.3, 21.2 (m), 21.1. HRMS calcd
for C31H44NO18ClNa+ 776.2145, found 776.2135.
N-[O-(2,3,4,6-tera-O-acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-
O-acetyl-1-â-D-glucopyranosyl]-5-azidopentamide (22). Compound
22 (1.04 g, 84%, Rf ) 0.40, 1:2 hexane/ethyl acetate, white solid) was
prepared from 21 (1.23 g, 1.63 mmol) using the same procedure
described in the preparation of 14. 1H NMR (400 MHz, CDCl3) δ 1.58
(m, 2H), 1.66 (m, 2H), 1.96 (s, 3H), 2.04 (s, br., 6H), 2.05 (s, 3H),
2.06 (s, 3H), 2.11 (s, 3H), 2.15 (s, 3H), 2.21 (m, 2H), 3.28 (t, J ) 6.4
Hz, 2H), 3.75 (m, 2H), 3.87 (t, J ) 6.8 Hz, 1H), 4.12 (m, 3H), 4.42 (d,
J ) 11.6 Hz, 1H), 4.46 (d, J ) 8.0 Hz 1H), 4.82 (t, J ) 9.2 Hz, 1H),
4.94 (dd, J1 ) 3.2 Hz, J2) 10.4 Hz, 1H), 5.10 (dd, J1 ) 7.6 Hz, J2)
10.4 Hz, 1H), 5.21 (t, J ) 9.2 Hz, 1H), 5.29 (dd, J1 ) 8.8 Hz, J2)
10.0 Hz, 1H), 5.34 (dd, J1 ) 3.6 Hz, J2) 1.0 Hz, 1H), 6.31 (d, J ) 8.8
Hz, 1H). 13C NMR (100.6 MHz, CDCl3) δ 173.7, 172.4, 171.5, 171.4,
171.3, 171.2, 170.5, 170.1, 102.0, 79.0, 77.1, 75.5, 73.4, 72.1, 72.0,
71.7, 70.0, 67.7, 63.0, 61.9, 52.1, 36.8, 29.3, 23.3, 22.0, 21.9, 21.8
(2C), 21.7, 21.6. HRMS calcd for C31H44N4O18Na+ 783.2548, found
783.2547.
1
acetate) as a white solid. H NMR (300 MHz, D2O) δ 1.48 (m, 2H),
1.56 (m, 2H), 2.21 (t, J ) 6.5 Hz, 2H), 3.20 (t, J ) 6.5 Hz, 2H), 3.30
(m, 1H), 3.51-3.73 (m, 12H), 3.80 (m, 2H), 3.87 (m, 2H), 4.37 (d, J
) 8.0 Hz, 1H), 4.83 (d, J ) 9.5 Hz, 1H), 4.99 (d, J ) 4.0 Hz, 1H). 13
C
NMR (75.5 MHz, CDCl3) δ 178.4, 103.5, 96.1, 79.8, 78.8, 77.8, 77.0,
75.8, 75.7, 72.1, 71.5, 70.2, 69.9, 69.8, 68.8, 65.5, 61.7, 61.6, 60.7,
51.4, 35.8, 28.1, 22.9. HRMS calcd for C23H40N4O16H+ 629.2517, found
629.2504.
N-[O-r-D-Galactopyranosyl)-(1f3)-O-(â-D-galactopyranosyl)-
(1f4)-1-â-D-glucopyranosyl]-5-aminopentamide (6) (from 15). A
solution of N-[O-R-D-Galactopyranosyl)-(1f3)-O-(â-D-galactopyrano-
syl)-(1f4)-1-â-D-glucopyranosyl]-5-azidopentamide (15) (0.54 g, 0.86
mmol) in methanol (35 mL) was hydrogenated at room temperature
and 4 atm pressure in the presence of platinum dioxide (60 mg). The
reaction mixture was diluted with methanol (50 mL) and filtered through
a pad of Celite. The Celite pad was washed with methanol (50 mL).
The combined methanol solution was concentrated to gave 6 [0.52 g,
100%, Rf ) 0.15, 7:3:2 2-propanol/NH4OH (27% aqueous solution)/
1
water] as a white solid. H NMR (500 MHz, D2O) δ 1.43 (m, 4H),
2.17 (t, J ) 7.5 Hz, 2H), 2.62 (m, 2H), 3.27 (m, 2H), 3.46-3.68 (m,
12H), 3.75 (m, 1H), 4.00 (m, 2H), 4.33 (d, J ) 8.0 Hz, 1H), 4.79 (d,
J ) 8.5 Hz, 1H), 4.95 (d, J ) 3.5 Hz, 1H). 13C NMR (125.7 MHz,
CDCl3) δ 179.0, 103.4, 96.0, 79.7, 78.6, 77.7, 76.9, 75.8, 75.6, 72.0,
71.4, 70.1, 69.9, 69.7, 68.8, 65.4, 61.6, 61.5, 60.5, 40.7, 36.3, 29.6,
23.2. HRMS calcd for C23H42N2O16H+ 603.2613, found 603.2629.
Chemoenzymatic Synthesis of Trisaccharide Derivative 13. (A)
O-r-D-Galactopyranosyl)-(1f3)-O-(â-D-galactopyranosyl)-(1f4)-1-
â-D-Glucopyranosyl azide (18). To a mixture of 1-azido-â-D-lactoside
(17) (0.20 g, 0.54 mmol), UDP-glucose (0.33 g, 0.54 mmol), bovine
serum albumin (BSA) (0.1%), MnCl2 (10 mM) in Tris-HCl (100 mM,
pH ) 7.0, 27 mL) was added fusion enzyme GalE-GalT (∼5 units).
The reaction mixture was shaken gently for 2 days at room temperature
and then passed through chloride anion exchange column (Dowex-Cl
1 × 8-200). The elute was concentrated and then purified with gel
permeation chromatography (Bio-Gel P2) to give trisaccharide 18 (0.19
g, 65%, Rf ) 0.25, 7:1:2 2-propanol/NH4OH (27% aqueous solution)/
N-[O-(â-D-galactopyranosyl)-(1f4)-1-â-D-glucopyranosyl]-5-azi-
dopentamide (23). Compound 23 (0.56 g, 91%, Rf ) 0.75, 5:1:1
2-propanol/H2O/EtOAc, white solid) was prepared from 22 (1.01 g,
1.33 mmol) using the same procedure described in the preparation of
1
15. H NMR (500 MHz, D2O) δ 1.49 (m, 2H), 1.54 (m, 2H), 2.22 (t,
1
water) as a white solid. H NMR (500 MHz, D2O) δ 3.13 (t, J ) 9.0
J ) 7.0 Hz, 2H), 3.20 (t, J ) 6.0 Hz, 2H), 3.30 (m, 1H), 3.50-3.67
(m, 8H), 3.77 (m, 2H), 4.31 (d, J ) 7.5 Hz, 1H), 4.84 (d, J ) 9.0 Hz
1H). 13C NMR (100.6 MHz, CDCl3) δ 178.5, 103.5, 79.7, 78.4, 77.0,
76.0, 75.8, 73.1, 72.1, 71.6, 69.2, 61.7, 60.5, 51.4, 35.8, 28.1, 22.9.
N-[O-(â-D-galactopyranosyl)-(1f4)-1-â-D-glucopyranosyl]-5-ami-
nopentamide (24). Compound 24 (0.51 g, 100%, Rf ) 0.20, 7:3:2
2-propanol/NH3H2O/H2O, white solid) was prepared from 23 (0.54 g,
1.16 mmol) using the same procedure described in the preparation of
Hz, 1H), 3.46-3.84 (m, 15H), 4.00 (d, J ) 2.5 Hz, 1H), 4.01 (t, J )
6.5 Hz, 1H), 4.34 (d, J ) 8.0 Hz, 1H), 4.59 (d, J ) 8.5 Hz, 1H), 4.96
(d, J ) 3.5 Hz, 1H). 13C NMR (125.7 MHz, CDCl3) δ 102.7, 95.3,
89.8, 77.8, 77.0, 76.6, 74.9, 74.3, 72.4, 70.7, 69.5, 69.2, 69.0, 68.1,
64.7, 60.9, 60.8, 59.8. HRMS calcd for C18H31N3O15Na+ 552.1652,
found 552.1639.
(B) O-(2,3,4,6-Tetra-O-acetyl-r-D-galactopyranosyl)-(1f3)-O-
(2,4,6-tri-O-acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-
1-â-D-glucopyranosyl azide (19). Azid 19 (0.26 g, 92%, Rf ) 0.70,
1:3 hexane/ethyl acetate, white solid) was prepared from O-R-D-
Galactopyranosyl)-(1f3)-O-(â-D-galactopyranosyl)-(1f4)-1-â-D-Glu-
copyranosyl azide (18) (0.16 g, 0.36 mmol) using the same procedure
1
6 from 15. H NMR (500 MHz, D2O) δ 1.35 (m, 2H), 1.48 (m, 2H),
2.18 (t, J ) 6.5 Hz, 2H), 2.56 (t, J ) 7.0 Hz, 2H), 3.26 (m, 1H), 3.37
(dd, J1) 8.0 Hz, J2 ) 10.0 Hz), 3.47-3.65 (m, 8H), 3.75 (m, 2H),
4.28 (d, J ) 7.5 Hz, 1H), 4.80 (d, J ) 9.0 Hz 1H). 13C NMR (100.6
MHz, CDCl3) δ 179.2, 104.0, 80.3, 78.9, 77.6, 76.6, 76.3, 73.7, 72.6,
72.1, 69.7, 62.3, 61.0, 40.9, 36.5, 30.6, 23.4. HRMS calcd for
C17H32N2O11Na+ 463.1904, found 463.1913.
A Typical Procedure for Preparation of r-Gal Polymers (Poly-
mers 7A-7F) and Polylactose (25). A solution of N-[O -R-D-
Galactopyranosyl)-(1f3)-O-(â-D-galactopyranosyl)-(1f4)-1-â-D-glu-
co- pyranosyl]-5-aminopentamide (6) (72 mg, 119.6 µmol) in DMSO/
DMF (1:1, 2 mL) containing diiospropyl ethylamine (30 µL) was added
to a stirred solution of pNAS 5 (100 mg, 590 µmol) in DMF (12 mL).
The mixture was stirred at room temperature for 24 h, heated at 65 °C
for 6 h, and then stirred continuously at room temperature for 24 h.
NH3‚H2O (concentrated aqueous, 1.5 mL) was added dropwise to the
reaction mixture followed by stirring at room temperature for 24 h.
1
described in the transformation from 12 to 13. H NMR (500 MHz,
CDCl3) δ 1.92 (s, 3H), 2.01 (s, 3H), 2.03 (s, 6H), 2.04 (s, 6H), 2.09 (s,
3H), 2.10 (s, 6H), 2.12 (s, 3H), 3.70 (m, 1H), 3.79 (m, 3H), 3.99-4.17
(m, 6H), 4.43 (m, 2H), 4.62 (d, J ) 8.5 Hz, 1H), 4.82 (t, J ) 8.5 Hz,
1H), 5.06 (m, 1H), 5.14 (t, J ) 11.5 Hz, 1H), 5.20 (m, 3H), 5.30 (m,
1H), 5.41 (m, 1H). 13C NMR (125.7 MHz, CDCl3) δ 171.1, 170.9 (2C),
170.8 (2C), 170.6, 170.4, 170.2 (2C), 169.4, 101.8, 94.1, 88.3, 76.2,
75.5, 73.5, 73.2, 71.6, 71.5, 70.3, 68.3, 67.8, 67.5, 67.1, 65.3, 62.4,
61.9, 61.7, 21.4 (m), 21.3 (m), 21.2 (m), 21.1. MS (FAB) C38H51N3O25
988 (M + K+).
(C) N-[O-(2,3,4,6-Tetra-O-acetyl-r-D-galactopyranosyl)-(1f3)-
O-(2,4,6-tri-O-acetyl-â-D-galactopyranosyl)-(1f4)-2,3,6-tri-O-acetyl-

Human Anti-Gal Antibody Binding
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8181
The resulting mixture was then dialyzed exhaustively against distilled
H2O to yield a polymer containing R-Gal epitopes as a white powder.
The density of R-Gal in this polymer is 1:6.4 (polymer 7D) calculated
from the integration of trisaccharide signal and acrylamide signal in
1H NMR spectra. Using different ratios of 6 to 5, polymers 7A-7F
were obtained in the ratio of 1:1.8, 1:2.5, 1:3.3, 1:6.4, 1:15, and 1:29,
respectively. Polylactose 25 (1:2.5) was prepared by reaction of 24
with pNAS 5 using the same procedure as above.
sera with secondary antibody as maximum staining (0% inhibition) was
used. The percent inhibition was calculated with eq 1:
(M - S)/(M - B) ) % inhibition
(1)
S was the OD450 reading of the sample with different concentrations of
R-Gal inhibitors (mouse laminin + R-Gal polymer + purified anti-Gal
Ab or human sera + 2nd Ab). B was the OD450 value of the background
staining (mouse laminin + 2nd Ab), while M was the OD450 value of
the maximum staining (mouse laminin + purified anti-Gal Ab or human
sera + 2nd Ab). The percent inhibition versus the concentration of
inhibitors was plotted for each R-Gal compound. IC50 was calculated
from the curve obtained.
Bioassays for the Inhibition of Anti-Gal Antibody Binding to
Mouse Laminin and Pig Kidney (PK15) Cells by Synthetic r-Gal
Polymers. (A) Materials. Pig kidney (PK15) cell line was purchased
from the American Type Culture Collection (ATCC). Bovine serum
albumin (BSA), human serum (male, type AB), mouse laminin,
peroxidase-conjugated goat anti-human IgG, IgM, or IgA and FITC-
conjugated goat anti-human IgG, IgM, or IgA antibodies were purchased
from Sigma. NHS-activated sepharose was purchased from Pharmacia
Biotech. Other reagents were all from commercially available sources.
(D) Flow Cytometry Assay with Pig Kidney (PK15) Cells.
Polymers 7A, 7B, and 7C were analyzed with a flow cytometry assay
to determine their inhibitory potential against the binding of human
anti-Gal antibodies to PK15 cells. PK15 cells were detached by versene
solution (PBS containing 0.68 mM of EDTA and 1.1 mM of D-glucose)
and then resuspended and blocked with RPMI 1640 serum-free medium
(SFM) containing 1% BSA for 30 min at 0 °C. Cells (6 × 105) were
added into polystyrene tubes (Falcon, 6 mL) and incubated with a series
of cocktails (100 µL) for 60 min on ice. The cocktails were made by
incubating 50 µL of heat-inactivated human sera with 50 µL of varying
concentrations of R-Gal monomer or polymers for 60 min at 4-7 °C.
The cells were separated and washed with RPMI extensively (3 × 1
mL). FITC-conjugated goat anti-human IgG, IgM, or IgA antibody
(secondary antibody, 80 µg/mL, 200 µL) was then added, and the
resulting mixture was incubated for 30 min on ice. After washing with
RPMI (3 × 1 mL), the cells were suspended in PBS (400 µL, pH 7.4)
and subjected to fluorescence analysis performed on a FACScan
apparatus (Becton-Dickinson) with 448 nm argon laser equipment. Data
were processed using PC-LYSYS Software, version 1.1 (1994), and
the means of fluorescence intensity were calculated. PK15 cells with
the secondary antibody were used as a background control. PK15 cells
with human sera and secondary antibody were used as maximum
staining (0% inhibition). The percent inhibition was calculated with
eq 2:
(B) Isolation of Polyclonal anti-Gal Antibody from Human
Serum. Polyclonal anti-Gal antibody was isolated from human sera
using an R-Gal [immobilized trisaccharide (R-Gal1-3Galâ1-4Glc-
NHCO(CH2)4NH2 6) on sepharose beads] affinity chromatography
column.6a,6b,19 Human serum was heated in a 56 °C water bath for 30
min to inactivate the complement and then passed through the column.
After an extensive washing with phosphate buffered saline (PBS buffer,
pH7.4), the bound anti-Gal antibody was eluted with a glycine-HCl
buffer (pH 2.8). Monitored with UV detector (280 nm), the elute
containing antibodies was immediately adjusted to pH 7.2 using 0.1
M NaOH. The resulting antibody solution was stored as frozen aliquots
in PBS buffer.
(C) ELISA Inhibition Assay with Mouse Laminin. An ELISA
was conducted using mouse laminin, a basement membrane glycopro-
tein containing 50-70 R-Gal epitopes per molecule, as the solid-phase
antigen. The purified human polyclonal anti-Gal antibodies (32 ng/
mL) or human sera (2.5-fold dilute) was first incubated with varying
concentrations of R-Gal compounds for 1.5 h at room temperature with
gentle shaking. An aliquot (50 µL) of the mixture was then added to
each microtiter plate well precoated with mouse laminin (50 µL/well
of 10 µg/mL in 0.1 N Na2CO3-NaHCO3 buffer, pH ) 9.5). After
incubation for 1.5 h at room temperature, unbound antibodies were
washed out with PBS-tween (pH ) 7.4, 0.5% tween, 5 × 200 µL). A
secondary antibody (1:1000 peroxidase-conjugated goat anti-human IgG
or IgM or IgA; 50 µL/well) was introduced, and the incubation was
allowed to proceed for 1 h at room temperature. After being washed
with PBS-tween buffer (5 × 200 µL), standard substrate (3,3′,5,5′-
tetramethylbenzidine: H2O2, 9:1; 100 µL/well) was added. The
enzymatic oxidation reaction produced a blue stain in each well. The
staining reaction was stopped by adding 1 N H2SO4 (100 µL/well).
Readings of optical absorption were taken at 450 nm (BioRad
Microplate Reader, model 3550-UV). PBS with the secondary antibody
was used as a background control. Purified anti-Gal antibody or human
(M - S)/(M - B) ) % inhibition
(2)
S was the mean of fluorescence intensity with different concentrations
of R-Gal compounds (PK15 cells + R-Gal polymer + human sera +
2nd Ab). B was the mean of fluorescence intensity for background
staining (PK15 cells + 2nd Ab), while M was the mean of fluorescence
intensity for maximum staining (PK15 cells + human sera + 2nd Ab).
The percent inhibition versus the concentration of inhibitor was plotted
for each R-Gal compound. IC50 was calculated from the curve obtained.
Acknowledgment. We thank NIH (GM 54074 & GM
54073) for the financial support of this work.
Supporting Information Available: Spectral data for
compounds 5-7, 10-15, 18-19, and 21-26. This material is
available free of charge via the Internet at http://pubs.acs.org.
(19) (a) Cuatrecasas, P. Nature 1970, 228, 1327. (b) Kohn, J.; Wilchek,
M. Enzyme Microb. Technol. 1982, 4, 161. (c) Jost, J. P., Saluz, H. P.,
Eds. A Laboratory Guide to in Vitro Studies of Protein-DNA Interactions;
Birkha¨user Verlag: Basel, Switzerland, 1992.
JA990219H