Antibody-catalysed glycosyl transfer reactions from in vitro immunization

Article abstract of DOI:10.1039/a704885h

In vitro immunization using a 5-membered ring iminocyclitol afforded antibodies that catalyse the hydrolysis of p-nitrophenyl glucopyranoside and p-nitrophenyl galactopyranoside with rate enhancements (k_cat/k_uncat) of greater than 10⁴.

Full text of DOI:10.1039/a704885h

View Article Online / Journal Homepage / Table of Contents for this issue
Antibody-catalysed glycosyl transfer reactions from in vitro immunization
Jaehoon Yu,*a So Young Choi,a Sungsoo Lee,a Hyun Joo Yoon,b Sunjoo Jeong,c Hansuh Mun,a Hokoon Parka
and Peter G. Schultz*d
a Division of Applied Science, Korea Institute of Science and Technology, PO Box 131, Cheongryang, Seoul 131-179, Korea
b Department of Microbiology, College of Natural Sciences, Inje University, Kimhae 621-749, Korea
c Department of Molecular Biology, College of Natural Sciences, Dankook University, Seoul140-714, Korea
d Howard Hughes Medical Institute, Department of Chemistry, University of California, Berkeley, USA
In vitro immunization using a 5-membered ring iminocycli-
tol afforded antibodies that catalyse the hydrolysis of
p-nitrophenyl glucopyranoside and p-nitrophenyl galacto-
pyranoside with rate enhancements (kcat/kuncat) of greater
than 104.
p-nitrophenyl-b-d-glucopyranoside (Scheme 1). Hapten 1 was
synthesized from 5-keto-d-fructose and benzhydrylamine in
three steps.† The NaBH4 dependent reductive amination of
5-keto-d-fructose with benzhydrylamine8 afforded, after re-
crystallization, the indicated diastereoisomer (Scheme 1) of the
corresponding N-benzhydryl azasugar. Removal of the protect-
ing group by catalytic hydrogenation, and alkylation of the
resulting secondary amine with 3-nitrobenzyl bromide, yielded
hapten 1. The nitro group was then reduced with H2 using
Lindlar catalyst, converted to the isothiocyanate and coupled to
the carrier proteins keyhole limpet hemocyanin (KLH) and
bovine serum albumin (BSA).9 Sixteen hybridomas antibody
specific for hapten 1 were generated by standard methods10 and
ascites from each cell line were generated. Antibodies were
purified using Protein-A affinity chromatography11 followed by
ion-exchange chromatography using a mono S column,‡ and
were tested using the following substrates: p-nitrophenyl-a- and
-b-d-glucopyranosides, p-nitrophenyl-a- and -b-d- galactopyr-
anosides and p-nitrophenyl-a- and b-mannopyranosides. HPLC
analysis of the reaction products indicated that none of the
antibodies showed any hydrolytic activity under a range of
buffer conditions.‡
However, encouraged by recent progress in immunization
methods, we decided to generate antibodies by in vitro
immunization, wherein carrier free hapten is mixed directly
with spleen cells in culture.11 Using this method, fourteen
hybridoma cell lines secreting antibodies specific for hapten 1
were identified by enzyme-linked immunosorbent assay and
ascites were generated. The ascites were purified and assayed as
described above. Two antibodies were found to be catalytic:
Ab24 and AB21 catalysed the hydrolysis of p-nitrophenyl-b-d-
glucopyranoside 3 and p-nitrophenyl-b-d-galactopyranoside 4,
respectively. These antibodies displayed saturation kinetics and
their Michaelis constants were determined in 10 mm 2-(4-mor-
The generation and characterization of antibodies that catalyse
glycosyl transfer reactions should provide insights into the
mechanisms of naturally occuring glycosidases, and may lead to
the development of glycosidases with novel specificities.1–3
Previous efforts to generate antibodies with glycosidase activity
made use of cyclic acetals as model substrates and antibodies
generated against charged and conformationally restricted
analogues of the oxocarbonium ion intermediate.1–3 Rate
accelerations of the order of 102–103-fold were reported for
these catalytic antibodies. More recently, an antibody that
catalyses the hydrolysis of p-nitrophenyl galactopyranoside
with a rate enhancement (kcat/kuncat) of 7 3 104 was isolated by
screening a library of antibodies specific for a five-membered
ring iminocyclitol using a suicide substrate.4 We have inde-
pendently used a similar positively-charged twist boat hapten to
directly generate an antibody that catalyses the cleavage of
p-nitrophenyl-b-d-glucopyranoside 3 with a rate enhancement
(kcat/kuncat) of ca. 2 3 104. Antibodies were produced by in vitro
immunization with free hapten, suggesting this might be a
general method for generating catalytic antibodies, especially
for unstable or toxic immunogens.5
The pyranoside ring is known to have a half-chair or twist-
boat conformation in the transition state for glycosidic bond
cleavage.6 Another distinctive feature of the transition state is
the development of partial positive charge at the anomeric
position. Based on these structural and electronic character-
istics, hapten 1 was used as a transition state analogue7 to elicit
antibodies that catalyse the cleavage of the b-glucosidic bond of
HO
‡
O
HO
HO
HO
HO
HO
O
HO
HO
O
+
HO
H+
H2O
OH
HO
+
HO
O
NO2
O
NO2
HO
O2N
OH
H
X
HO
OH
N
HO
1 X = NO2
OH
2 X = NHC(S)–BSA or –KLH
Scheme 1
Chem. Commun., 1997
1957

View Article Online
pholino)ethanesulfonic acid (MES), 100 mm NaCl, 0.02%
NaN3 buffer, pH 4.5, at 37 °C; rate constants for the uncatalysed
reaction were determined under the same conditions.‡ Antibody
Ab21 catalysed the hydrolysis of p-nitrophenyl-b-d-galactopyr-
anoside 4 (and not 3) with a kcat of 0.035 h21 and Km of 310 mm
Footnotes and References
†
5-Keto-d-fructose was synthesized by incubating d-fructose with
Gluconobactor cerinus (ATCC catalogue number IFO 3267). Hapten 1 was
purified on silica gel using CH2Cl2–MeOH (9:1 v/v) as eluent (Rf = 0.37);
d
H (300 MHz, CDCl3) 8.1 (s, 1 H), 8.0 (d, 1 H, J 3), 7.6 (d, 1 H, J 8), 7.45
(kcat/kuncat
=
2.5 3 104); antibody Ab24 catalysed the
(dd, 1 H, J 8, 3), 4.0–3.8 (m, 4 H), 3.6–3.3 (m, 8 H including 4H from 4
OHs), 3.1 (m, 1 H), 2.7 (m, 1 H); dC (75 MHz, [2H6]DMSO) 147.6 (3A
phenyl), 143.4 (1A phenyl), 135.3 (6A phenyl), 129.3 (4A phenyl), 122.9 (2A
phenyl), 121.6 (4A phenyl), 77.0, 76.1 (C3 and C4), 72.6 (C benzyl), 67.4,
62.0 (C2 and C5), 60.3, 57.4 (C1 and C6); m/z 299 (MH+).
hydrolysis of p-nitrophenyl-b-d-glucopyranoside 3 (and not 4)
with a kcat of 0.02 h21 and Km of 160 mm (kcat/kuncat = 2.2 3
104). Antibodies purified from different preparations of ascites
showed the same specific activities. Moreover, the catalytic
activity could be stoichiometrically titrated by absorbing
antibodies on increasing amounts of protein A agarose and
analysing the activity of the filtrate versus the amount of bound
antibody. The isolation of two antibodies, one specific for
substrate 3 and the other for substrate 4, also argues against
enzymic contamination.
‡ Antibodies were purified using a SP-Separose® (HiLoadTM 16/10)
column (Pharmacia). For the mobile phase, buffer A (50 mm MES, pH 5.5)
and buffer B (50 mm MES, 1.0 m NaCl, pH 5.5) were co-eluted with a linear
gradient of 0–50% B over 40 min and a flow rate of 2 ml min21. Assays
were carried out by mixing 90 ml of antibody (2–3 mg ml21) and 10 ml of
substrate (20 mm), each in reaction buffer, and then incubating at 37 °C for
3 days. The reaction was analysed by HPLC using a C18 microsorb column
(4.6 mm 3 15 cm, 5 mm), eluting with 50% aqueous MeOH at 0.8 ml min21
and monitoring a 315 nm. The rate was determined when less than 1% of the
substrate was converted to product. The data were fitted directly to the
Michaelis–Menten equation, v = (Vmax [S])/([S] + Km), using the nonlinear
curve fitting program from the Kaleidagraph software suite. The second
order rate constants for the H+, OH2 and AcOH catalysed reactions and the
The antibodies were inhibited by hapten 1 with a Ki value of
60 mm for the Ab21-catalysed hydrolysis of glucopyranoside 3,
and a Ki value of 380 mm for the Ab24-catalysed hydrolysis of
galactopyranoside 4.12 For both antibodies the rate acceleration
does not correlate with the relative affinities of the antibody for
substrate and hapten 1. This behaviour suggests that a structural
feature of the hapten (i.e. an amino or hydroxy group) results in
the presence of a catalytic group, such as aspartic or glutamic
acid, in the antibody combining site which can function as a
general acid in catalysis but does not contribute substantially to
the overall binding energy for the hapten. Indeed, diazoaceta-
mide treatment of antibody Ab21 led to a 97% reduction in
catalytic activity; 60% of the catalytic activity was retained
when modification was carried out in the presence of 2 mm
hapten 1.13 The relatively low affinities of the antibodies for
hapten 1 resulting from the in vitro immunization likely reflect
the absence of affinity maturation that normally occurs during
an immune response.14 In addition, antibodies generated against
carbohydrates typically show relatively low affinities. It is of
interest that both b-glucosidase and b-galacosidase activities
were generated using hapten 1. The five-membered ring in
hapten 1 is flattened, mimicking a boat or twist-boat conforma-
tion of the pyranoside ring in the transition state. Either hapten
1 has some conformational flexibility or its geometry must map
onto the critical recognition elements for both substrates and/or
transition states.
+
psuedo-first order rate constant for the reaction in water are: kH = 1.1 3
1023 m21 h21, kOH = 0.29 m21 h21, kAcOH = 1.2 3 1026 m21 h21, and
2
kAH2O = 8.0 3 1027 h21 for substrate 3; kH = 3.8 3 1023 m21 h21
,
+
kOH = 0.58 m21 h21, kAcOH = 2.0 3 1026 m21 h21 and kAH2O = 1.6 3
1026 h21 for substrate 4.
2
1 J.-L. Reymond, K. D. Janda and R. A. Lerner, Angew. Chem., Int. Ed.
Engl., 1991, 30, 1711.
2 H. Suga, N. Tanimoto, A. J. Sinskey and S. Masamune, J. Am. Chem.
Soc., 1994, 116, 11197.
3 J. Yu, L. C. Hsieh, L. Kochersperger, S. Yonkovich, J. C. Stephans,
M. A. Gallop and P. G. Schultz, Angew. Chem., Int. Ed. Engl., 1994, 33,
339.
4 K. Janda, L.-C. Lo, M.-M. Shin, R. Wang, C.-H. Wong and
R. A. Lerner, Science, 1997, 275, 945.
5 M. Stahl, B. Goldie, S. P. Bourne and N. R. Thomas, J. Am. Chem. Soc.,
1995, 117, 5164.
6 M. L. Sinnott, Chem. Rev., 1990, 90, 1171; S. Rosenberg and
J. F. Kirsch, Biochemistry, 1981, 20, 3196; L. E. H. Smith, L. H. Mohr
and M. A. Raftery, J. Am. Chem. Soc., 1973, 95, 7497; B. A. Malcolm,
S. Rosenberg, M. J. Corey, J. S. Allen, A. de Baetselier and J. F. Kirsch,
Proc. Natl. Acad. Sci. USA, 1989, 86, 133; R. Kuroki, L. H. Weaver and
B. W. Matthews, Science, 1993, 262, 2030; T. L. Amyes and
W. P. Jencks, J. Am. Chem. Soc., 1989, 111, 7888; S. G. Withers and
I. P. Street, J. Am. Chem. Soc., 1988, 110, 8551; V. L. Schramm,
B. A. Horenstein and P. C. Kline, J. Biol. Chem., 1994, 269, 18259.
7 K.-C. Liu, J. Org. Chem., 1991, 56, 6280.
8 A. B. Reitz and E. W. Baxter, Tetrahedron Lett., 1990, 31, 6777.
9 A. C. Braisted and P. G. Schultz, J. Am. Chem. Soc. 1994, 116, 2211.
10 E. Harlow and D. Lane, Antibodies: A Laboratory Manual, Cold Spring
Harbor Press, New York, 1988.
11 U. B. Fredriksson, L. G. Fagerstam, A. W. G. Cole and T. Lundgreb,
Protein A-Sepharose C1-4B Affinity Purification of Ig G Monoclonal
Antibodies, Pharmacia AB, Uppsala, Sweden, 1986.
12 I. H. Segel, Enzyme Kinetics, Wiley, New York, 1975.
13 A. L. Grossberg and D. Pressman, J. Am. Chem. Soc., 1960, 82,
5478.
Finally, Lerner4 reported that direct immunization of a
related hapten afforded catalytic antibodies with only 100-fold
rate enhancements; we isolated no catalytic antibodies from
direct immunization. This suggests that in vitro immunization,
like antibody phage display methods, may complement conven-
tional approaches for producing catalytic antibodies.
This work was partially supported by the Korea Institute of
Science (Grant No E14100-1996) and the Ministry of Health
and Welfare of Korea (Grant No HMP-96-D-1-1024). This
work was also supported by the Director, Office of Energy
Research, Office of Basic Energy Sciences, Division of
Materials Sciences, and also by the Division of Energy
Biosciences of the U.S. Department of Energy under Contract
No. DE-AC03-76SF00098. P. G. S. is a Howard Hughes
Medical Institute Investigator and a W. M. Keck Foundation
Investigator.
14 C. A. Borrebaeck, Trends Biotechnol., 1986, 4, 147 and references cited
therein.
Received in Corvallis, OR, USA, 7th July 1997; 7/04885H
1958
Chem. Commun., 1997