JULY/AUGUST 2004 Volume 15, Number 4 © Copyright 2004 by the American Chemical Society
COMMUNICATIONS Poly(ethylene glycol)-Supported Enzyme Inactivators. Efficient Identification of the Site of Covalent Attachment to r-Chymotrypsin by PEG-TPCK Christine A. Schering, Boyu Zhong,‡ Jonathan C. G. Woo,† and Richard B. Silverman* Department of Chemistry, Department of Biochemistry, Molecular Biology, and Cell Biology, and the Drug Discovery Program, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208-3113. Received March 9, 2004; Revised Manuscript Received May 6, 2004
A new methodology utilizing an enzyme inactivator covalently attached to poly(ethylene glycol) (PEG) is described in which the PEG affords facile and mild quantification, isolation, and identification of the site of enzyme inactivation. As proof of concept, the known affinity labeling agent for R-chymotrypsin, N-tosyl-L-phenylalanine chloromethyl ketone (TPCK), was linked to PEG. The synthesis of the PEG-bound inactivator PEG-TPCK was carried out in good yields using standard solution-phase chemistry. Inactivation of R-chymotrypsin with PEG-TPCK was monitored via UV-vis spectroscopy in aqueous conditions, which resulted in less than 3% remaining activity, indicating that 97% of the R-chymotrypsin was covalently modified with PEG-TPCK. The MALDI-TOF mass spectrum showed only one new peak that was distinct in shape and corresponded to the mass of PEG-TPCK-Rchymotrypsin. Following proteolytic digestion, the PEG-TPCK-peptide was easily discernible from the rest of the digest in a HPLC trace because of its characteristic prolonged retention time and broad polymer shape. MALDI-TOF MS was used to determine the mass of the PEGylated peptide. Without prior removal of the PEG, the amino acid site to which PEG-TPCK covalently bound was determined via Edman sequencing. In comparison to other methods, the PEG-supported inactivator system is significantly cheaper and safer than the synthesis of radiolabeled compounds; furthermore, isolation of the PEGylated peptide is milder and more selective than standard affinity binding columns. Edman sequencing provides an exact determination of the site of inactivator covalent attachment without extensive, tedious LC-MS analysis of a complex peptide mixture. The method described here could be applied to a variety of enzymes as an alternative to current techniques.
The assessment of enzyme inactivators is an important tool to understand the catalytic mechanism and active * To whom correspondence should be addressed. E-mail:
[email protected]. † Current address: Lilly Research Laboratories, Indianapolis, IN 46285. ‡ Current address: Johnson & Johnson Pharmaceutical Research & Development, L.L.C., Titusville, NJ 08560.
site topology of enzymes. The traditional method to identify the site of covalent attachment involves synthesizing a radiolabeled inactivator (1). Because of the high cost of radiolabeled starting materials, the difficulty in synthesis, and many federal regulations, alternative methods are desirable. Site-directed mutagenesis has been used to determine the importance of an active-site residue, but this method lacks definitive support for
10.1021/bc049940p CCC: $27.50 © 2004 American Chemical Society Published on Web 06/09/2004
674 Bioconjugate Chem., Vol. 15, No. 4, 2004
Schering et al. Scheme 1. Synthesis of PEG-TPCKa
Figure 1. PEG-supported inactivator methodology.
inactivator binding (2). Affinity techniques, such as biotin/avidin, vary in selectivity and many times require nonaqueous incubation systems and harsh cleavage conditions (3). Recently, mass spectrometry has been used to analyze digested inactivated enzymes; however, the quality and complexity of the digest greatly influence the degree of difficulty and methodological approach of the subsequent MS analysis, and it does not always provide convenient identification of the site of inactivator binding (3, 4). An alternative method that would retain the positive features but overcome the failings of current approaches would facilitate the assessment of enzyme inactivators. Poly(ethylene glycol) (PEG) is a water- and organicsoluble polymer utilized in chemistry because its homogeneous reaction conditions contribute to fast kinetics and high loading levels, while PEG’s propensity to precipitate in ether aids in facile purification during PEGsupported synthesis of small molecules and biopolymers (5-10). PEGylation, the biological technique of covalently attaching PEG to nucleophilic amino acids of enzymes and peptides, produces new pharmaceutical agents and organic catalysts with enhanced stability and performance (11-14). PEG’s ability to impart its physiochemical properties on other molecules in both chemical and biological systems could allow it to overcome the above-mentioned failings of current approaches to identify the enzyme inactivator covalent binding site (Figure 1). Synthesis of a PEG-supported inactivator through standard solutionphase organic synthesis would be simpler and cheaper than synthesis of a radiolabeled molecule, with the added benefit that the PEG-supported inactivator could be subsequently characterized via NMR and MALDI-TOF MS (15). Because of the homogeneous aqueous reaction conditions and the flexible PEG framework, the PEGbound inactivator should be capable of inactivating the enzyme similarly to that of the unmodified inactivator (16). The inactivation of the enzyme and the concurrent formation of the PEGylated proteins can be quantified by the same assays used on the unmodified inactivator. The inactivated PEGylated enzyme can be quickly purified by size-exclusion chromatography and identified by MALDI-TOF MS because of the drastically altered retention time, molecular weight, and broad peak shape of PEGylated peptides and proteins (17, 18). Furthermore, a variety of proteolytic enzymes are known to cleave PEGylated proteins (12). Once digested, PEG alters the properties of the PEGylated peptide from those of the unmodified peptides, aiding in HPLC purification and identification of the PEGylated peptide by MALDI-TOF MS (19). Finally, MALDI-TOF MS and Edman sequencing are both amenable for use with PEGylated peptides, allowing determination of the covalent attachment site
a Reagents and conditions: (a) benzyl chloroformate, Na CO , 2 3 95%; (b) PCl5, 75%; (c) L-Phe, acetone, Na2CO3, 84%; (d) oxalyl chloride, DMF, CH2Cl2; (e) diazomethane, ether, 0 °C; (f) HCl(g), ether, 0 °C, three steps 30%; (g) 30% HBr in acetic acid, 65%; (h) succinic anhydride, DIPEA, CHCl3, 85%; (i) PEG-NH2 (MW 5000 g/mol), HATU, DIPEA, CH2Cl2, DMF, 80% yield, 90% loading.
without prior cleavage of the PEG handle (12-17). A PEG-supported inactivator would behave as both a tag to identify any polymer covalently modified species as well as a handle to isolate the PEGylated moieties. Exact determination of the site of inactivator covalent attachment would be possible at low cost, in mild and standard conditions, without harsh cleavage conditions to remove the polymer or analysis of a complicated digestion mixture. This new methodology utilizing a PEG-supported inactivator was developed using the known affinity labeling agent for R-chymotrypsin, N-tosyl-L-phenylalanine chloromethyl ketone (TPCK) (20). TPCK is known to bind covalently to His 57, part of the catalytic triad of R-chymotrypsin, with both time and concentration dependence (20-22). An analogue of TPCK, S-1-[1-N-(4amino benzenesulfonyl)aminophenethyl] chloromethyl ketone (1, Scheme 1), was synthesized (23-24). Replacement of the methyl group with the isosteric amino group provided a handle to link the PEG. Monomethoxyamino PEG (MW 5000) and the TPCK analogue were linked via succinic anhydride, resulting in a stable and UV-active diamide linker (2, Scheme 1). As with TPCK, the time and concentration dependent inactivation of R-chymotrypsin with PEG-TPCK (2) was monitored by the cleavage of L-benzyltyrosine methyl ester until less than 3% of the activity remained (24 h), implying 97% linkage of the PEG-TPCK to the R-chymotrypsin. Dialysis (MWCO 10 000 Da) was used to remove unreacted PEG-TPCK and buffer salts prior to MALDI-TOF MS analysis. Comparison of the molecular weight of the inactivated R-chymotrypsin (31 060 Da, broad) with those of the control R-chymotrypsin (25 230 Da, sharp) clearly indicated that the PEG-TPCK (approximately 5400 Da) was covalently bound to the R-chymotrypsin. The peak shapes also supported complete covalent binding of PEG-TPCK to R-chymotrypsin; the control is a sharp peak as would be expected for a single protein whereas the inactivated enzyme is a uniquely shaped broad peak caused by the polydisperse PEG covalently bound to the enzyme.(Figure 2).
Communications
Figure 2. MALDI-TOF MS of R-chymotrypsin and PEGTPCK-R-chymotrypsin (positive mode, CHCA matrix). The unique peak shape and large mass difference clearly indicate covalent attachment of the polydisperse PEG-TPCK (MW ∼ 5450, g/mol) to R-chymotrypsin.
Figure 3. HPLC analysis (Vydac C8, A:H2O, 0.1% TFA, B: ACN, 0.1% TFA, 20-80% in 40 min) of digested (a) R-chymotrypsin control, (b) PEG-TPCK-R-chymotrypsin, and (c) purified PEG-TPCK-peptide. The prolonged retention time and unique peak shape aid in immediate identification of the modified peptide and facile isolation of the peak.
PEG-TPCK-R-chymotrypsin (900 µM) was denatured, reduced, alkylated, and then subsequently digested with R-chymotrypsin. The digestion was stopped by allowing the mixture to run through a PD-10 desalting column. The PEG-TPCK-peptide peak in the HPLC (Figure 3b) was easily identified because it appears as a distinct broad peak as the result of the polydispersity of the PEG not present in the digested control (Figure 3a). It is known that PEGylated peptides and proteins elute as polydisperse peaks following the nonPEGylated moieties because of an increase in size and decrease in polarity (18, 25). This new peak was isolated by HPLC (Figure 3c) and characterized by MALDI-TOF MS to be a PEGylated peptide because of the distinct polydispersed peak centered around 7400 g/mol (Figure 4). Gas-phase Edman sequencing identified only two peptides, offset by three amino acids because of partial cleavage after Thr 54; the sequence of one peptide was VVTAAXC′GVTTSD- - -A and the sequence of the other was AAXC′GVTTSDVVVAGEF. Both sequences correlate to the known primary sequence of R-chymotrypsin with the indeterminable amino acid (X) corresponding to His 57, the known site of TPCK covalent binding to R-chymotrypsin. The expected molecular weight for each sequence through Phe 71 is 2017.98 g/mol and 1718.79 g/mol, respectively. If the molecular weights of the inactivator (MW 450.92 g/mol) and PEG (MW 5000 g/mol) are added to these, then the expected molecular weights of the PEG-TPCK-peptides would be 7468.89 and 7169.71 g/mol. These masses agree with the broad peak seen by MALDI-TOF MS (Figure 4). No other free or PEG-TPCK-
Bioconjugate Chem., Vol. 15, No. 4, 2004 675
Figure 4. MALDI-TOF MS on the purified PEG-TPCKpeptide (negative mode, DHB matrix). The distinct shape is indicative of a PEGylated peptide.
bound peptide sequences were detected by Edman sequencing or mass spectrometry. This new methodology utilizing an enzyme inactivator covalently bound to PEG enabled facile, mild, and selective isolation and identification of the site of PEG-TPCK covalent attachment to R-chymotrypsin, the same residue to which unmodified TPCK binds. The polydisperse nature of the PEG aided in visualization and subsequent isolation of the PEGylated enzyme and peptide. Inexpensive starting materials were used to synthesize PEGTPCK. The inactivation occurred in aqueous conditions and thereby could be easily monitored and purified via standard techniques. The PEGylated enzyme was fragmented into small peptides via proteolytic digestion. HPLC was used to easily identify and then purify the PEG-TPCK-peptide from the digested mixture. The peptide sequence was determined by Edman sequencing without prior cleavage of the PEG, indicating without ambiguity the site of PEG-TPCK attachment. Because of its facile and mild nature, this method could be applied to a variety of enzymes instead of currently used techniques. ACKNOWLEDGMENT
This research was supported by NIH (GM32634). Edman sequencing was performed by the Protein Separations and Analysis Laboratory at Purdue University. Supporting Information Available: Experimental procedures and spectrometric data. This material is available free of charge via the Internet at http://pubs.acs.org. LITERATURE CITED (1) Suketa, Y., and Inagai, T. (1975) Active site directed inactivators of mouse submaxillary renin. Biochemistry 14 (14), 3188-3194. (2) Delaire, M., Labia, R., Samama, J. P., and Masson, J. M. (1992) Site-directed mutagenesis at the active site of Escherichia coli TEM-1 beta-lactamase. Suicide inhibitor-resistant mutants reveal the role of arginine 244 and methionine 69 in catalysis. J. Biol. Chem. 267 (29), 20600-20606. (3) Hubalek, F., Pohl., and Edmondson, D. E. (2003) Structural Comparison of Human Monoamine Oxidases A and B: Mass Spectrometry Monitoring of Cysteine Reactivities. J. Biol. Chem. 278 (31), 28612-28618. (4) McCarter J. D., Burgoyne D. L., Miao S., Zhang S., Callahan J. W., and Withers S. G. (1997) Identification of Glu-268 as the catalytic nucleophile of human lysosomal beta-galactosidase precursor by mass spectrometry. J. Biol. Chem. 272 (1), 396-400.
676 Bioconjugate Chem., Vol. 15, No. 4, 2004 (5) Toy, P. H., and Janda, K. D. (2000) Soluble polymer bound cleavage reagents: a multipolymer strategy for the cleavage of tertiary amines from REM resin. Org. Lett. 2 (15), 22052207. (6) Mutter, M., Hagenmaier, H., and Bayer, E. (1971) New method of polypeptide synthesis. Angew. Chem. Int. 10 (11), 811-812. (7) Pillai, V. N. R., Mutter, M. Bayer, E., and Gatfield, I.,(1980) New, easily removable poly(ethylene glycol) supports for the liquid-phase method of peptide synthesis. J. Org. Chem. 45 (26), 5364-5370. (8) Tamada, K., Fujita, E., and Nishimura, S. I. (1998) High performance polymer supports for enzyme-assisted synthesis of glycoconjugates. Carbohydr. Res. 305 (3-4), 443-461. (9) Bonora, G. M. (1987) Use of poly(ethylene glycol) (PEG) as a soluble polymeric support in the oligonucleotide synthesis. A proposal for a new liquid-phase method. Gazz. Chim. Ital. 117 (6), 379-80. (10) Gravert, D. J., and Janda, K. D. (1997) Organic Synthesis on Soluble Polymer Supports: Liquid-Phase Methodologies. Chem. Rev. 97 (2), 489-509. (11) Roberts, M. J., Bentley, M. D., and Harris, J. M. (2002) Chemistry for peptide and protein PEGylation. Adv. Drug Delivery Rev. 54 (4), 459-476. (12) Veronese, F. M. (2001) Peptide and protein PEGylation. A review of problems and solutions. Biomaterials 22 (5), 405417. (13) Delgado, C., Francis, G. E., and Fisher, D. (1992) The uses and properties of PEG-linked proteins. Crit. Rev. Ther. Drug Carrier Syst. 9 (3-4), 249-304. (14) Matsushima, A., Okada, M., and Inada, Y. (1984) Chymotrypsin modified with poly(ethylene glycol) catalyzes peptide synthesis reaction in benzene. FEBS Lett. 178 (2), 275-277. (15) Pillai, V. N. R., and Mutter, M. (1981) Conformational studies of poly(oxyethylene)-bound peptides and protein sequences. Acc. Chem. Res. 14 (4), 122-130. (16) Hooftman, G., Herman, S., and Schacht, E. J. (1996) Poly(ethylene glycol)s with reactive end groups. II. Practical
Schering et al. consideration for the preparation of protein-PEG conjugates. Bioact. Compatible Polym. 11 (2), 135-159. (17) Drioli, S., Adamo, I., Ballico, M., Morvan, F., and Bonora, G. M. (2002) Liquid-Phase synthesis and characterization of a conjugated chimeric oligonucleotide-PEG-peptide. Eur. J. Org. Chem. (20), 3473-3480. (18) Suzawa, T., Nagamura, S., Saito, H., Ohta, S., Hanai, N., and Yamasaki, M. (2000) Synthesis and HPLC analysis of enzymatically cleavable linker consisting of poly(ethylene glycol) and dipeptide for the development of immunoconjugate. J. Control. Release 69 (1), 27-41. (19) Clark, R., Olson, K., Fuh, G., Marian, M., Mortensen, D., Teshima, G., Chang, S., Chu, H., Mukku, V., Canova-Davis, E., Somers, T., Cronin, M., Winkler, M., and Wells, J. A. (1996) Long-acting growth hormones produced by conjugation with poly(ethylene glycol). J. Biol. Chem. 271 (36), 2196921977. (20) Schoellmann, G., and Shaw, E. (1963) Direct evidence for the presence of histidine in the active center of chymotrypsin. Biochemistry 2, 252-255. (21) Ong, E. B., Shaw, E., and Schoellmann, G. (1965) The identification of the histidine residue at the active center of chymotrypsin. J. Biol. Chem. 240 (2), 694-698. (22) Ke´dzy, F. J., Thomson, A., and Bender, M. L. (1967) Reaction of chymotrypsin and L-1-chloro-3-tosylamido-4phenyl-2-butanone. J. Am. Chem. Soc. 89 (4), 1004-1009. (23) Jameson, G. W., and Elmore, D. T. (1974) Affinity chromatography of bovine trypsin. A rapid separation of bovine alpha- and beta-trypsin. Biochem. J. 141 (2), 555-565. (24) Popenoe, E. A., and du Vigneaud, V. (1954) Synthesis of L-phenylalanyl-L-glutaminyl-L-asparagine. J. Am. Chem. Soc. 76, 6202-6203. (25) Snider, J., Neville, C., Yuan, L. C., and Bullock, J. (1992) Characterization of the heterogeneity of poly(ethylene glycol)modified superoxide dismutase by chromatographic and electrophoretic techniques. J. Chromatogr. 599 (1-2), 141-55.
BC049940P
