Activity Probes
DOI: 10.1002/ange.200501738
Design of a Mechanism-Based Probe for
Neuraminidase To Capture Influenza Viruses**
Chun-Ping Lu, Chien-Tai Ren, Yi-Ning Lai,
Shih-Hsiung Wu,* Wei-Man Wang, Jean-Yin Chen,
and Lee-Chiang Lo*
Influenza viruses, which cause upper respiratory tract prob-lems in humans, have long been a major threat to public health.[1]It is estimated that 10–20 % of the general
popula-[*] C.-P. Lu,[+] $Dr. C.-T. Ren,$Y.-N. Lai, Prof. S.-H. Wu[+] [++]
Institute of Biological Chemistry, Academia Sinica Taipei 115 (Taiwan)
Fax: (+ 886) 2-2653-9142 E-mail: [email protected] W.-M. Wang, J.-Y. Chen, Prof. L.-C. Lo
Department of Chemistry, National Taiwan University Taipei 106 (Taiwan)
Fax: (+ 886) 2-2362-1979 E-mail: [email protected] [+
] Institute of Biochemical Sciences, National Taiwan University Taipei 106, (Taiwan)
[++] Genomics Research Center, Academia Sinica
Taipei 115 (Taiwan)
[$] These authors contributed equally to this work.
[**] Influenza A virus (A/WSN/33) was a gift from Dr. Shin-Ru Shih, Chang Gung University, Taiwan) and polyclonal anti-FluA antibody was a gift from Dr. Hour-Young Chen (Center for Disease Control Taiwan). Japanese encephalitis virus (JEV, Taiwanese strain, RP-9) and the mouse monoclonal antibody specific for this virus were kindly provided by Dr. Yi-Ling Lin (IBMS Academia Sinica). Zanamivir was a gift from GlaxoWellcome Research and Develop-ment Ltd. (Stevenage, UK) and from Prof. Ching-Shih Chen (The Ohio State University, OH). We thank Prof. Yulin Lam for proof-reading the manuscript. This work was supported by the National Science Council (Grant nos. NSC 92-2113M-001-023 to S.-H.W. and NSC 93-3112-B-002-001 to L.-C.L.).
Supporting information for this article is available on the WWW under http://www.angewandte.org or from the author.
tion are affected in seasonal epidemics. Some devastating outbreaks recorded in history even claimed millions of lives worldwide.[2]Influenza viruses are typically spherical with a
diameter of about 100 nm.[3]The pathogenic properties of the
viruses have been extensively investigated.[4] Significant
advances in therapeutic treatments and in the detection of the viruses have been made in recent years as a result of this information. Among the limited number of proteins that are encoded by the viral RNA segments, two surface glycopro-teins, namely, hemagglutinin (HA) and neuraminidase (NA), have often been the focus of research. These two surface glycoproteins play important roles in the infection process; HA is responsible for the binding of viruses to the host cells, whereas NA is involved in the budding process.[5]Although
the surface antigens of the viruses mutate frequently to avert attacks from the immune system of the hosts, the critical catalytic activity of NA has to be maintained for successful infection and propagation. This special feature makes it an excellent candidate for research.[6]
NA (sialidase, N-acylneuraminosyl glycohydrolase, EC 3.2.1.18) is an exo-glycosidase that hydrolyzes the linkage of sialic acid residues, which are mostly found as terminal constituents of glycoconjugates. X-ray crystallographic infor-mation about the active site of NA has revealed important residues involved in the recognition and binding of sialic acid.[7] This has assisted in the development of several
reversible inhibitors of NA, such as zanamivir and oseltami-vir, which have been approved for the treatment of influ-enza.[8] Recently, McKimm-Breschkin et al. have
demon-strated that ligands derived from biotin-conjugated zanamivir were able to bind influenza virion on microtiter plates; this could in turn serve as a basis for a diagnostic method.[9]
Besides influenza viruses, many NA-containing microorgan-isms are pathogenic and it has been proposed that this enzyme plays an important role in the pathogenicity of these infections.[10]We therefore envision that the development of
activity probes that selectively form a covalent linkage with NA would be of great value, as demonstrated by the application for capturing the virus particles in this report.
Chemical probes that are able to form covalent linkages with hydrolase subfamilies have proven to be a powerful tool in modern proteomics studies.[11]We have previously
devel-oped activity probes for hydrolases such as phosphatases and b-glucosidase.[12]The concept for the design of these activity
probes originated from suicide substrates, or mechanism-based inhibitors, of enzymes.[13]This approach is unique as the
probes themselves are also the substrates of the correspond-ing hydrolases. The probes can be selectively activated once the recognition head is cleaved by the targeting hydrolase, thereby leading to covalent modifications of the hydrolase.[14]
More importantly and closely related to this report, the mechanism-based approach has been successfully applied in the screening and selection of biocatalysts from phage-displayed libraries, such as in the selection of mutant lactamases and in the search for catalytic antibodies with b-galactosidase activity.[15] Earlier studies have shown that
compound 1 was a mechanism-based inhibitor of Clostridium perfringens NA.[16]We thus developed probe 2 as a
mecha-nism-based probe for NA. Probe 2 carries four structural units
in its design; a sialic acid recognition head, which is connected to an ortho-difluoromethylphenyl latent trapping device, a linker, and a biotin reporter group. The biotin reporter group
serves two functions. On one hand, it is used to visualize the labeled NA after Western blotting with streptavidin-conju-gated peroxidase chemiluminescence. On the other hand, it could be used to attach the probe to the microtiter plates through avidin–biotin interactions during the virus-capturing study. When the designated glycosidic bond is cleaved by NA, the released intermediate 3 would undergo 1,4-elimination with removal of a fluoride ion to generate a reactive quinone methide, 4. The highly reactive quinone methide intermediate 4 would alkylate nearby nucleophiles of the enzyme, thereby resulting in biotinylated adduct 5 (Scheme 1). Since the viral surface is spiked with NA, covalent attachment through NA would then lead to the capture of virus particles.
The synthesis of probe 2 begins with a commercially available N-acetylneuraminic acid 6 (Scheme 2). All the synthetic procedures were combinations of efforts from previous publications.[17] In brief, fully protected glycosyl
chloride 8 was prepared and used directly for coupling with 2-hydroxy-5-nitrobenzaldahyde in a CHCl3/H2O biphasic
system by using tetrabutylammonium bromide as the phase-transfer catalyst to give compound 9. The formyl group of compound 9 could be converted into the difluoromethyl moiety of the trapping device by the DAST reagent. The structure of the difluoromethylphenyl group in compound 10
Scheme 1. Mechanism for selective activation and alkylation of neura-minidase with probe 2.
was supported by its 1H NMR spectroscopic data which
showed a triplet at d = 6.86 ppm with a coupling constant of
2J
H,F=54.9 Hz, a typical value for H–F coupling. A triplet at
d = 109.9 ppm (1J
C,F=237.2 Hz) in the13C NMR spectrum of
10 further confirmed the presence of the CHF2moiety. The
nitro group of compound 10 was then reduced by catalytic hydrogenation to give amine 11. Attachment of the linker and biotin reporter group yielded the fully protected probe 14. Final deprotection under alkaline conditions offered the desired probe 2 after LH-20 purification.
Probe 2 was first tested for its ability to biotinylate NA obtained from Arthrobacter ureafaciens. Arthrobacter ureafa-ciens neuraminidase (0.8 U, Sigma) was incubated in the presence or absence of probe 2 (200 mm) at 4 8C in 100 mm ammonium acetate buffer (10 mL). Bovine serum albumin (BSA) was used as a negative control. Labeled samples were applied to a 10 % polyacrylamide gel, which was then subjected to sodium dodecylsulfate (SDS) PAGE. After electrophoresis, the proteins were transferred from the gel onto a poly-vinylidene fluoride (PVDF) mem-brane. The PVDF membrane was blocked, washed, and developed by using the enhanced chemilumines-cence (ECL). Western blotting pro-tocols as recommended by the sup-plier (Amersham Biosciences). The result of the Western blot analysis only shows biotinylated proteins (Figure 1). In this experiment, three biotinylated bands corresponding to isoenzymes at 88, 66, and 52 KDa were observed for NA (Lane 1).[18]
Probe 2 had no effect on BSA, as
shown by the lack of labeled bands (Lane 2). In the absence of the probe, neither NA nor BSA was biotinylated (Lanes 3 and 4). This result firmly supports the alkylation process de-scribed in Scheme 1. The actual label-ing site was not further determined in this study, because a number of bio-logical applications utilizing the same latent trapping device have already established the covalent-bond-form-ing feature.[15, 19] Moreover, Lee and
co-workers recently used this activity-probe approach to characterize the catalytic domain of b-galactosidases from Xanthomonas manihotis, Escher-ichia coli, and Bacillus circulans, with the identification of mainly single modifications on Arg, Glu, and Glu residues, respectively.[20]
We also compared the inhibitory effect of probe 2 and zanamivir on the activity of NA from influenza A virus (A/WSN/33, H1N1), as well as from Arthrobacter ureafaciens (AU), Clos-tridium perfringens (CP), and Vibro cholerae (VC; Figure 2). Both compounds were tested at 3.3 mm in the NA inhibi-tion assays with
2’-(4-methylumbelliferyl)-a-d-N-acetyl-neuraminic acid (MUNANA, Sigma) as the substrate.[21]All
samples were measured in duplicate and fluorometric deter-minations were performed with a fluorometer (ThermoLab systems, Sweden). The excitation wavelength was 355 nm and the emission wavelength was 460 nm. Probe 2 inhibited all four NA activities (H1N1, AU, CP, VC) with IC50 values of
1.7, 0.68, 0.08, and 0.53 mm, respectively. These results indicate that probe 2 interferes with the activity of NA isolated from various sources at the active site, regardless of the variations the enzymes might have. On the other hand,
Scheme 2. Synthesis of probe 2 for neuraminidase. Conditions: a) MeOH, IR-120 (H+) resin, 92 %; b) AcCl,
AcOH; c) 2-hydroxy-5-nitrobenzaldehyde, Cs2CO3, Bu4NBr, H2O/CHCl3, 67 % for two steps; d) DAST,
CH2Cl2, 47 %; e) H2, 5 % Pd/C, MeOH, 95 %; f) succinic anhydride, TEA, CH2Cl2, 94 %; g) EDCI, HOBt, 13,
DIEA, DMF, 75 %; h) Na2CO3, MeOH; aqueous Na2CO3, 52 %. TFA = trifluoroacetic acid; TEA =
triethyl-amine; EDCI = 3-(3-dimethylaminopropyl)-1-ethylcarbodiimide; HOBt = 1-hydroxy-1H-benzotriazole; DIEA =N,N-diisopropylethylamine; DMF = N,N-dimethylformamide; DAST = (diethylamino)sulfur trifluor-ide.
Figure 1. Western blot analysis of probe 2 labeling the Arthro-bacter ureafaciens neu-raminidase.
Figure 2. Inhibition study of NA activities from A. ureafaciens (AU), C. perfringens (CP), V. cholerae (VC), and influenza A virus (Inf A) with probe 2 (gray), zanamivir (white), and the control (black). Virus, suspended in 32.5 mm b-morpholinoethanesulfonic acid buffer (pH 6.5) was incubated with either probe 2 or zanamivir (3.3 mm) at room temperature for 45 min and then incubated with MUNANA at 37 8C for 30 min. The residual NA activity is expressed as the percentage of activity relative to that obtained in the absence of reagent. Assays for AU, CP, and VC were similarly carried out, except in 80 mm acetate buffer (pH 5.0).
zanamivir effectively inhibited viral NA activity with an IC50
value of 2–3 nm, but it displayed weak inhibitory effects on the three bacterial NA activities, a result suggesting that the original strong binding interactions could be greatly compro-mised by possible variations in the active site. It must be emphasized that although one of the targets in this study, the influenza virus, is the same as one in the study of McKimm-Breschkin et al.,[9]the concept and approach of the current
strategy offers a versatile alternative for future applications. The labeling event in this study was a result of an activation step forming the reactive quinone methide and did not rely on the strong noncovalent binding which is a critical requirement in the previous approach, a fact which makes the current approach a more general one in targeting NA activities. It is especially worth noting that the advantage of the current approach becomes more prominent with the advent of zanamivir-resistant viruses.[22]
Having established the efficacy of probe 2 to biotinylate NA and thus influenza virus A virions, we next studied the capturing performance by utilizing the covalent-bond-form-ing feature. The tests were carried out by a modified ELISA method as described previously.[9] Briefly, a
streptavidin-coated 96-well ELISA plate (NUNC Immobilizer) was saturated with probe 2. BSA–biotin conjugate provided a negative control. After 1 hour of incubation, the plate was blocked with 0.1 % BSA/phosphate-buffered saline (PBS) for 1 hour and washed with PBS. Serial twofold dilutions of influenza A virus (A/WSN/33) were added and incubated for 1 hour at room temperature. After another wash, the captured viruses were detected by treatment with a polyclonal anti-FluA antibody, followed by a goat anti-rabbit IgG horseradish peroxidase (HRP) conjugate and a NeA-Blue tetramethylbenzidine substrate (TMB, Clinical Science Prod-ucts, Inc.). The results indicated that probe 2 bound to the microplate wells could successfully capture influenza virus A and the intensity of the responding signal was proportional to the number of virus particles present (Figure 3). The wells loaded with BSA–biotin conjugate gave a negative response. The possibility of any noncovalent bindings could be ruled out by the modest IC50 value (1.7 mm) on the viral NA. More
importantly, when the same procedure was applied to a mixture of influenza virus and a non-NA-containing Japanese encephalitis virus, only influenza virus was selectively
cap-tured and detected on the plate, a result strongly supporting the theory that the capture of virus particles was both probe and NA dependent. This conclusion was further supported by the experiment during which probe 2 failed to biotinylate influenza viruses that were preincubated with zanamivir, which effectively blocked the active site of NA on the viral surface. The results represent the first example of the use of a covalent-bond-forming mechanism for the capture of influ-enza virus particles. In addition, such covalent interactions between captured virus particles and the probe are known to be tolerant to harsh conditions,[15]thus making this
method-ology amenable for further manipulations.
In summary, we have designed and synthesized a mech-anism-based activity probe 2 for neuraminidase, which uses a latent quinone methide as the trapping device and forms a biotinylated adduct with Arthrobacter ureafaciens neuramin-idase in a model study. By taking advantage of the essential role played by the NA activity in the life cycle of the influenza virus, we evaluated the interaction between probe 2 and the virus. The covalent-labeling event led to diminished NA activities. Furthermore, it serves as the basis for capturing influenza virus particles on microplate wells. This novel approach of capturing the influenza virus, which provides a stronger interaction between virus and the stationary phase, will certainly offer opportunities for developing new applica-tions, such as rapid screening of antibodies against this group of viruses, and the development of sensitive and rapid diagnostic methods.
Received: December 15, 2004 Revised: May 20, 2005
Published online: October 7, 2005
.
Keywords: activity probes · hydrolases · influenza viruses · mechanism-based labeling · sialic acids[1] C. Liu, M. C. Eichelberger, R. W. Compans, G. M. Air, J. Virol. 1995, 69, 1099; K. Subbarao, A. Klimov, J. Katz, H. Regnery, W. Lim, H. Hall, M. Perdue, D. Swayne, C. Bender, J. Huang, M. Hemphill, T. Rowe, M. Shaw, X. Xu, K. Fukuda, N. Cox, Science 1998, 279, 393.
[2] J. S. Oxford, Rev. Med. Virol. 2000, 10, 119; N. J. Cox, K. Subbarao, Annu. Rev. Med. 2000, 51, 407.
[3] R. G. Webster, W. J. Beam, O. T. Gorman, T. M. Chambers, Y. Kawaoka, Microbiol. Rev. 1992, 56, 152.
[4] C. M. Carr, P. S. Kim, Science 1994, 266, 234; M. Tashiro, R. Rott, Semin. Virol. 1996, 7, 237; D. A. Steinhauer, Virology 1999, 258, 1; G. R. Whittaker, Exp. Rev. Mol. Med. 8 February 2001, can be found under http://www-ermm.cbcu.cam.ac.uk/01002447h.htm; T. M. Tumpey, A. Garcia-Sastre, J. K. Taubenberger, P. Palese, D. E. Swayne, C. F. Basler, Proc. Natl. Acad. Sci. USA 2004, 101, 3166.
[5] J. J. Skehel, D. C. Wiley, Annu. Rev. Biochem. 2000, 69, 531. [6] G. Taylor, Curr. Opin. Struct. Biol. 1996, 6, 830.
[7] J. N. Varghese, P. M. Colman, J. Mol. Biol. 1991, 221, 473; J. N. Varghese, J. L. McKimm-Breschkin, J. B. Caldwell, A. A. Kortt, P. M. Colman, Proteins Struct. Funct. Genet. 1992, 14, 327. [8] M. von Itzstein, W.-Y. Wu, G. B. Kok, M. S. Pegg, J. C. Dyason,
B. Jin, T. V. Phan, M. L. Smythe, H. F. White, S. W. Oliver, P. M. Colman, J. N. Varghese, D. M. Ryan, J. M. Woods, R. C. Bethell, Figure 3. ELISA-based detection of influenza A (A/WSN/33) captured
with probe 2 bound to streptavidin-coated microplate wells and detected with an anti-FluA antibody-HRP conjugate. PFU stands for plague-forming unit.
V. J. Hotham, J. M. Cameron, C. R. Penn, Nature 1993, 363, 418; S. M. Cheer, A. J. Wagstaff, Drugs 2002, 62, 71.
[9] J. L. McKimm-Breschkin, P. M. Colman, B. Jin, G. Y. Krippner, M. McDonald, P. A. Reece, S. P. Tucker, L. Waddington, K. G. Watson, W.-Y. Wu, Angew. Chem. 2003, 115, 3226; Angew. Chem. Int. Ed. 2003, 42, 3118.
[10] R. Schauer, Trends Biochem. Sci. 1985, 10, 357; A. P. Corfield, Glycobiology 1992, 2, 509; H. B. Tang, E. Dimango, R. Bryan, M. Gambello, B. H. Iglewski, J. B. Glodberg, A. Prince, Infect. Immun. 1996, 64, 37.
[11] B. F. Cravatt, E. J. Sorensen, Curr. Opin. Chem. Biol. 2000, 4, 663; D. A. Campbell, A. K. Szardenings, Curr. Opin. Chem. Biol. 2003, 7, 296; D. A. Jeffery, M. Bogyo, Curr. Opin. Biotechnol. 2003, 14, 87; A. E. Speers, B. F. Cravatt, ChemBioChem 2004, 5, 41.
[12] L.-C. Lo, H.-Y. Wang, Z.-J. Wang, J. Chin. Chem. Soc. 1999, 46, 715; L.-C. Lo, T.-L. Pang, C.-H. Kuo, Y.-L. Chiang, H.-Y. Wang, J.-J. Lin, J. Proteome Res. 2002, 1, 35; C.-S. Tsai, Y.-K. Li, L.-C. Lo, Org. Lett. 2002, 4, 3607.
[13] M. Wakselman, Nouv. J. Chim. 1983, 7, 439; R. B. Silverman, Mechanism-Based Enzyme Inactivation: Chemistry and Enzy-mology, Vols. 1 and 2, CRC, Boca Raton, FL, 1988.
[14] L.-C. Lo, C.-H. L. Lo, K. D. Janda, D. B. Kassel, F. M. Raushel, Bioorg. Med. Chem. Lett. 1996, 6, 2117.
[15] P. Soumillion, L. Jespers, M. Bouchet, J. Marchand-Brynaert, G. Winter, J. Fastrez, J. Mol. Biol. 1994, 237, 415; K. D. Janda, L.-C. Lo, C.-H. L. Lo, M.-M. Sim, R. Wang, C.-H. Wong, R. A. Lerner, Science 1997, 275, 945.
[16] P.-A. Driguez, B. Barrere, B. Chantegrel, C. Deshayes, A. Doutheau, G. Quash, Bioorg. Med. Chem. Lett. 1992, 2, 1361. [17] R. Kuhn, P. Lutz, P. L. MacDonald, Chem. Ber. 1966, 99, 611; P. Meindl, H. Tuppy, Monatsh. Chem. 1965, 96, 802; F. Baum-berger, A. Vasella, Helv. Chim. Acta 1986, 69, 1927; S. Halazy, V. Berges, A. Ehrhard, C. Danzin, Bioorg. Chem. 1990, 19, 330; G. Sabatino, M. Chinol, G. Paganelli, S. Papi, M. Chelli, G. Leone, A. M. Papini, A. De Luca, M. Ginanneschi, J. Med. Chem. 2003, 46, 3170.
[18] M. Iwamori, Y. Ohta, Y. Uchida, Y. Tsukada, Glycoconjugate J. 1997, 14, 67.
[19] M. Ichikawa, Y. Ichikawa, Bioorg. Med. Chem. Lett. 2001, 11, 1769; J. R. Betley, S. Cesaro-Tadic, A. Mekhalfia, J. H. Rickard, H. Denham, L. J. Partridge, A. Plueckthun, G. M. Blackburn, Angew. Chem. 2002, 114, 801; Angew. Chem. Int. Ed. 2002, 41, 775; Q. Zhu, X. Huang, G. Y. J. Chen, S. Q. Yao, Tetrahedron Lett. 2003, 44, 2669.
[20] M. Kurogochi, S. Nishimura, Y. C. Lee, J. Biol. Chem. 2004, 279, 44 704.
[21] M. Potier, L. Mameli, M. Belishle, L. Dallaire, S. B. Melancom, Anal. Biochem. 1979, 94, 287.
