Design and synthesis of activity probes for glycosidases

Article abstract of DOI:10.1021/ol0265315

(graph presented) A new synthetic route was developed for the preparation of activity probe 1 for β-glucosidase in this study. The key glycosidation step begins with benzyl p-hydroxyphenylacetate. Benzylic functionalization for the construction of the tra

Full text of DOI:10.1021/ol0265315

ORGANIC
LETTERS
2002
Vol. 4, No. 21
3607-3610
Design and Synthesis of Activity Probes
for Glycosidases
,†
Charng-Sheng Tsai,† Yaw-Kuen Li,‡ and Lee-Chiang Lo*
Department of Chemistry, National Taiwan UniVersity, Taipei 106, Taiwan,
and Department of Applied Chemistry, The National Chiao-Tung UniVersity,
Hsin-Chu 300, Taiwan
lclo@ccms.ntu.edu.tw
Received July 16, 2002
ABSTRACT
A new synthetic route was developed for the preparation of activity probe 1 for â-glucosidase in this study. The key glycosidation step begins
with benzyl p-hydroxyphenylacetate. Benzylic functionalization for the construction of the trapping device was achieved at later stages. Probe
1 was shown to be able to label the target enzyme. This cassette-like design offers great flexibility for future alterations. It would allow the
synthetic scheme to expand to other glycosidase probes with different linker/reporter combinations.
Activity probes have been developed and found to be useful
in the mechanism-based labeling and selection of biocatalysts
with designated hydrolytic activities, including mutant â-
lactamases1 and catalytic antibodies,2 from libraries. More
recently, these class-selective probes have been demonstrated
to be a powerful tool in proteomics studies for identifying
certain hydrolase families such as protein tyrosine phos-
phatases3,4 and serine hydrolases.5-8 They helped simplify
the otherwise complicated protein band/spot patterns after
gel electrophoresis. Protein phosphorylation and glycosyla-
tion are two mechanisms commonly seen in biological
systems for regulating protein activities. The major hydrolytic
enzymes involved in these processes are phosphatases and
various glycosidases. Therefore, these two enzyme families
are often important targets in biomedical studies. Since we
have previously reported the probes for protein tyrosine
phosphatases adopting a cassette-like design,3,4 we now
extend the target activities to glycosidase families. The
trapping device, which makes use of the quinone methide
chemistry9 and is the core of the probe, was derived from a
p-hydroxymandelic acid derivative.10 However, due to dif-
ficulties in the glycosidation step with the p-hydroxymandelic
acid derivative, in this study we explored a new synthetic
route for the preparation of activity probe 1 and studied it
on a model â-glucosidase.
Activity probe 1 thus designed carries a glucose residue
serving as the recognition head. This recognition head is
connected through a â-glycosidic linkage to a p-hydroxy-
benzylic fluoride moiety. When the designated glycosidic
bond is cleaved by a â-glucosidase, it undergoes 1,6-
elimination to remove a fluoride and generate a reactive
quinone methide. This reactive quinone methide intermediate
† National Taiwan University.
‡ The National Chiao-Tung University.
(1) Soumillion, P.; Jespers, L.; Bouchet, M.; Marchand-Brynaert, J.;
Winter, G.; Fastrez, J. J. Mol. Biol. 1994, 237, 415-422.
(2) Janda, K. D.; Lo, L.-C.; Lo, C.-H. L.; Sim, M.-M.; Wang, R.; Wong,
C.-H.; Lerner, R. A. Science 1997, 275, 945-948.
(3) Lo, L.-C.; Wang, H.-Y.; Wang, Z.-J. J. Chin. Chem. Soc. 1999, 46,
715-718.
(4) Lo, L.-C.; Pang, T.-L.; Kuo, C.-H.; Chiang, Y.-L.; Wang, H.-Y.; Lin,
J.-J. J. Proteome Res. 2002, 1, 35-40.
(5) Liu, Y.; Patricelli, M. P.; Cravatt, B. F. Proc. Natl. Acad. Sci. U.S.A.
1999, 96, 14694-14699.
(6) Cravatt, B. F.; Sorensen, E. J. Curr. Opin. Chem. Biol. 2000, 4, 663-
668.
(7) Kidd, D., Liu, Y., Cravatt, B. F. Biochemistry 2001, 40, 4005-4015.
(8) Patricelli, M. P.; Giang, D. K.; Stamp, L. M.; Burbaum, J. J.
Proteomics 2001, 1, 1067-1071.
(9) Wakselman, M. NouV. J. Chim. 1983, 7, 439-447.
(10) Lo, L.-C.; Lo, C.-H. L.; Janda, K. D.; Kassel, D. B.; Raushel, F.
M. Bioorg. Med. Chem. Lett. 1996, 6, 2117-2120.
10.1021/ol0265315 CCC: $22.00 © 2002 American Chemical Society
Published on Web 09/21/2002

â-cyanoethyl group of compound 8 was removed by treat-
ment with DBU to form R-fluoroacid 9 in good yields (95%).
Since the free acid of compound 9 was not stable, it was
stored as its triethylammonium salt.
Scheme 1
The complete skeleton of the probe was constructed by
amide bond formation between R-fluorocarboxylate 9 and
the TFA salt of dioxaoctanediamine derivative of biotin
1012,13 in the presence of DCC/HOBt (Scheme 3). The fully
Scheme 3a
serves as the trapping device to alkylate nearby nucleophiles
on the biocatalysts, resulting in the biotinylated enzyme
(Scheme 1). The biotin reporter group in probe 1 not only is
useful for highly sensitive secondary detection but also offers
the advantage of enriching the labeled proteins in proteomic
studies.
The synthesis of probe 1 begins with benzyl p-hydroxy-
phenylacetate 3 (Scheme 2). It was reacted with glucosyl-
a (a) DCC, HOBt, TEA, DMF (83%); (b) Na2CO3, MeOH (92%).
Scheme 2a
protected probe 11 was then subjected to final deacetylation
by treatment with Na2CO3/MeOH to give probe 1 in high
yields.
In the preliminary study, â-glucosidase cloned from
FlaVobacterium meningosepticum was employed.14,15 This
enzyme belongs to the family 3 hydrolases. The catalytic
mechanisms of this â-glucosidase and other family 1
hydrolases have been extensively studied.16,17 To demonstrate
that probe 1 was a substrate of â-glucosidase and the
concomitant elimination from the hydrolyzed intermediate
really occurred, we performed time course experiments by
incubating probe 1 (6.1 mM) with enzyme (0.66 µM) in
phosphate buffer (50 mM, pH 7.0). The fluoride signals were
monitored with 19F NMR. As shown in Figure 1, at time
zero, the two sets of doublets around -168 ppm represent
the signals of probe 1, which exists as a pair of diastereomers
(shown in inset). Both diastereomers of probe 1 underwent
hydrolysis by the enzyme in a comparable rate to release
inorganic fluoride, supporting the 1,6-elimination depicted
in Scheme 1. The new singlet at -122 ppm results from
inorganic fluoride. Probe 1 did not produce any activation/
elimination in the absence of â-glucosidase.
a (a) AgOTf, CH2Cl2 (67%); (b) H2, Pd/C (91%); (c) DCC,
DMAP, HOCH2CH2CN, CH2Cl2 (95%); (d) NBS, CCl4, hν (92%);
(e) AgNO3, acetone/H2O (55%); (f) DAST, CH2Cl2 (85%); (g)
DBU, CH2Cl2 (95%).
bromide 2 in the presence of AgOTf to give â-glucoside 4
in 67% yield. The â-configuration of the glycosidic linkage
was indicated by the coupling constant between H-1 and H-2
(J ) 7.6 Hz). The conversion of the benzyl protecting group
in compound 4 to a â-cyanoethyl group was achieved by
hydrogenation followed by coupling with DCC/DMAP/
HOCH2CH2CN to give â-cyanoethyl ester 5 in 86% yield
for two steps. The benzylic functionalization was introduced
first by bromination with NBS under irradiation,11 followed
by hydrolysis with AgNO3 in acetone/H2O to give benzylic
alcohol intermediate 7 in 51% yield for two steps. The newly
formed hydroxyl group was converted to fluoride by reaction
with DAST to offer the protected half unit 8, covering the
recognition head and the trapping device, in 85% yield. The
(12) Sigal, G. B.; Mammen, M.; Dahmann, G.; Whitesides, G. M. J.
Am. Chem. Soc. 1996, 118, 3789-3800.
(13) Braun, M.; Camps, X.; Vostrowsky, O.; Hirsch, A.; Endress, E.;
Bayerl, T. M.; Birkert, O.; Gauglitz, G. Eur. J. Org. Chem. 2000, 7, 1173-
1182.
(14) Li, Y.-K.; Chu, S.-H.; Sung, Y.-H. J. Chin. Chem. Soc. 1998, 45,
603-610.
(15) Li, Y.-K.; Lee, J.-A. Enzyme Microb. Technol. 1999, 24, 144-
150.
(16) Kempton, J. B.; Withers, S. G. Biochemistry 1992, 31, 9961-9969.
(17) Li, Y.-K.; Chir, J.; Chen, F.-Y. Biochem. J. 2001, 355, 835-
840.
(11) Briggs, J. C.; Haines, A. H.; Taylor, R. J. K. J. Chem. Soc., Perkin
Trans. I 1995, 27-32.
3608
Org. Lett., Vol. 4, No. 21, 2002

Figure 2. Gels showing the â-glucosidase-mediated activation of
the probe 1, resulting in biontinylated modification of the enzyme.
Reaction mixtures were separated by 8% SDS-PAGE. Coomassie
blue-stained (left) and ECL-developed (right) gels are presented.
Figure 1. Time course study for the hydrolysis of probe 1 by
â-glucosidase as monitored with 19F NMR at 0, 0.5, 3.5, and 17.5
h: the two sets of doublets (enlarged in inset) at -168 ppm are
the signal for the starting material; the singlet at -122 ppm
represents the inorganic fluoride.
they have rarely been utilized as labeling probes. Activity
probes can selectively and covalently label a target protein
without reducing activity when they modify nonessential
residues. Besides, an activity probe could serve in a wide
range of applications depending on the property of the
reporter groups. To achieve these goals, it would need
additional tests and possible alterations to function well in
various systems. For example, the prototype probe developed
in this study is ready to use for rapid screening of glycosi-
dases from numerous microbial sources. However, it pro-
duced cross-labeling when activated in the presence of a
mixture of proteins in a preliminary study (data not shown)
and might not be suitable for proteomics applications in its
current format. This result is different from that of tyrosine
phosphatase probe,4 in which case no cross-labeling was
observed, indicating that the structure of the biocatalysts also
plays an important role in determining the performance of
this type of probe.
In summary, we have accomplished a new synthetic route
for the activity probe of â-glucosidase. Benzyl p-hydroxy-
phenylacetate was used as the starting point to overcome
the difficulties in the glycosidation step. Benzylic function-
alization for the trapping device was successfully introduced
at later stages. Probe 1 developed in this study was able to
label a model â-glucosidase. Although only a single probe
was prepared and examined in this study, its cassette-like
design provides great flexibility for future alterations. It not
only allows various sugar units and different anomeric
configurations to be constructed in the future to offer
important intermediate 9 but it could also incorporate a large
number of linker/reporter combinations to meet demands in
various applications. We are currently studying the effect
of a trapping device2,22 together with the linker/reporter
combination, including the length and hydrophilic/hydro-
phobic properties, to improve their performance in proteo-
mics applications. These results will be reported in due
course.
Labeling of the enzyme by activity probe 1 was next
examined. Two parallel labeling experiments were carried
out and analyzed (Figure 2). On the left-hand side, lanes 1
and 2 show the Coomassie blue stained gel, indicating the
relative amounts of the protein loaded. The gel on the right
(lanes 3 and 4) was visualized with streptavidin-conjugated
peroxidase chemiluminescence after blotting onto a nitrocel-
lulose membrane. It clearly shows the labeling of â-glucosi-
dase by the probe 1 (lane 4). The slightly smeared band with
higher molecular weights (lanes 2 and 4) suggests possible
multiple modifications. On the contrary, there was no labeling
in the absence of probe 1 (lane 3). In separate labeling
experiments with a higher enzyme concentration (6.6 µM),
we observed the formation of a protein precipitation within
5 min of mixing. The residual activity was found to be less
than 10% after 30 min of incubation using p-nitrophenyl
glucoside as the assay substrate. This dramatic activity loss
may be attributed to the active site labeling or, more likely,
the severe aggregation of the multiply labeled proteins.
Although the degree of modification was not determined in
this study, the observation of multiple labeling accompanied
by the loss of activity is consistent with the previous study
on the phosphotriesterase probe bearing an identical trapping
device, in which case the multiply labeled proteins were
characterized by LC/MS.10 A similar phenomenon of multiple
labeling by the quinone methide intermediate was also
extensively studied on a new mechanism-based inhibitor of
â-glucosidase purified from Agrobacterium faecalis.18
It has to be stressed that while suicide substrates based
on the generation of quinone methide intermediates have
been reported for â-glucosidases and â-glucuronidase,18-21
(18) Zhu, J.; Withers, S. G.; Reichardt, P. B.; Treadwell, E.; Clausen,
T. P. Biochem. J. 1998, 332, 367-371.
(19) Halazy, S.; Berges, V.; Ehrhard, A.; Danzin, C. Bioorg. Chem. 1990,
18, 330-344.
(20) McCarter, J. D.; Withers, S. G. J. Am. Chem. Soc. 1996, 118, 241-
242.
(21) Azoulay, M.; Chalard, F.; Cesson, J.-P.; Florent, J.-C.; Monneret,
C. Carbohydr. Res. 2001, 332, 151-156.
(22) Ichikawa, M.; Ichikawa, Y. Bioorg. Med. Chem. Lett. 2001, 11,
1769-1773.
Org. Lett., Vol. 4, No. 21, 2002
3609

Acknowledgment. This work was supported by the
National Science Council (NSC 91-3112-B-002-013) (to
L.-C.L.). We also thank Professor Jing-Jer Lin for helpful
discussions.
13C NMR, for compounds 1, 4-9, and 11 as well as
conditions for the labeling of â-glucosidase with probe 1.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Supporting Information Available: Experimental
procedures and characterization, including copies of 1H and
OL0265315
3610
Org. Lett., Vol. 4, No. 21, 2002