20
Bioconjugate Chem. 2009, 20, 20–23
Single-Step Azide Introduction in Proteins via an Aqueous Diazo Transfer Stijn F. M. van Dongen, Rosalie L. M. Teeuwen, Madhavan Nallani,† Sander S. van Berkel, Jeroen J. L. M. Cornelissen, Roeland J. M. Nolte, and Jan C. M. van Hest* Institute for Molecules and Materials, Radboud University Nijmegen, Heyendaalseweg 135, 6525 AJ, Nijmegen, The Netherlands. Received October 9, 2008; Revised Manuscript Received November 19, 2008
The controlled introduction of azides in proteins provides targetable handles for selective protein manipulation. We present here an efficient diazo transfer protocol that can be applied in an aqueous solution, leading to the facile introduction of azides in the side chains of lysine residues and at the N-terminus of enzymes, e.g. horseradish peroxidase (HRP) and the red fluorescent protein DsRed. The effective introduction of azides was verified by mass spectrometry, after which the azido-proteins were used in Cu(I)-catalyzed [3 + 2] cycloaddition reactions. Azido-HRP retained its catalytic activity after conjugation of a small molecule. This modified protein could also be successfully immobilized on the surface of an acetylene-covered polymersome. Azido-DsRed was coupled to an acetylene-bearing protein allowing it to act as a fluorescent label, demonstrating the wide applicability of the diazo transfer procedure.
In recent years, with the advent of chemical biology the need to manipulate and modify proteins has increased considerably. One of the important limitations in selective chemical modification of proteins is the overabundance of functional groups present in these biomacromolecules. In order to introduce, site specifically, bioorthogonal reactive groups, multiple strategies have been developed. In many cases, the free thiol groups of cysteine residues are specifically targeted for conjugation because these functionalities are reactive and relatively rarely present in proteins (1-4). Unfortunately, electrophiles that readily react with thiols are, to a lesser extent, also susceptible to attack by amines or even alcohols, leading to loss of specificity. Furthermore, cysteines are often essential for the structural integrity and function of proteins and therefore cannot always be modified. To overcome these shortcomings, some strategies make use of functional groups that do not naturally occur and are deliberately introduced, enabling individual targeting of these groups without inadvertent cross-linking or other reactions (5-8). An example of this approach is the [3 + 2] transition metalcatalyzed Huisgen cycloaddition between an azide and an acetylene function (1, 9-11). Since “click” chemistry has been shown to work well for protein modification, efficient methods for the controlled introduction of azides or acetylenes into biomolecules have become of great interest (12-14). Although azido-modified amino acids can be introduced by single-site or multisite protein engineering replacement strategies (7, 13, 15), the degree of site specificity provided by these laborious methods is not always needed. This led us to investigate the possibility of applying the relatively mild and facile diazo transfer reaction to amines, in order to directly introduce azides at the lysine positions of proteins. Traditionally, however, the diazo transfer reaction is carried out on small organic molecules in organic solvents such as CH2Cl2, with trifluoromethanesulfonyl azide (TfN3) as reagent and catalyzed by Cu(II) (16-18). These conditions make this type of reaction unsuitable for protein modification. Recently, this method was improved by Goddard-Borger and Stick (19), who introduced * E-mail:
[email protected]. † Current address: Institute of Materials Research & Engineering (IMRE), Research Link 3, Singapore 117602.
Figure 1. Structures of DsRed, horseradish peroxidase (HRP), and imidazole-1-sulfonyl azide hydrochloride (1). The molecular surfaces of all lysine residues in the proteins are shown. The structure of DsRed was derived from the PDB file 2VAD and contains 22 lysine residues. The structure of HRP was derived from PDB file 7ATJ and contains 6 lysine residues.
imidazole-1-sulfonyl azide hydrochloride (1, see Figure 1) as a nonexplosive and shelf-stable alternative to TfN3. Particularly interesting is the fact that 1 can be applied in protic solvents. We therefore envisaged the use of compound 1 to perform the diazo transfer reaction on proteins and enzymes in aqueous solutions, residue-specifically introducing azido-moieties at their lysine residues or N-terminus. To study the applicability of this relatively simple method, we initially used the enzyme horseradish peroxidase (HRP, see Figure 1) as a model protein (20, 21). To functionalize HRP, it was dissolved in milliQ (MQ) pure water (200 µL, 2.5 mg/ mL) and an aqueous solution of K2CO3 (100 µL, 2 mg/mL) was added, along with 25 µL of Cu(II)SO4 · 5H2O in MQ (1 mg/mL). After mixing, transfer agent 1 in MQ was added (1.75 equiv relative to the amines in HRP) and the reaction was left stirring overnight, after which the enzyme was purified using centrifugal filter devices. ESI-TOF analysis of the thus treated enzyme showed that, under the applied experimental conditions, an average of four amines were transformed into azides (see Figure 2). A small amount of HRP possessed three or five azides (out of a total of seven), which implies a statistical mixture of functionalized compounds. The influence of the diazo transfer treatment and subsequent conjugation of a small molecule (butynol; see Supporting
10.1021/bc8004304 CCC: $40.75 2009 American Chemical Society Published on Web 12/19/2008
Communications
Bioconjugate Chem., Vol. 20, No. 1, 2009 21
Figure 3. Catalytic activities of HRP and butynol-conjugated azidoHRP. Equal amounts of both enzymes were incubated with ABTS and H2O2, and the formation of ABTS · + was measured over time. Comparison of the initial slopes suggests that the “clicked” HRP has retained 75% of its catalytic activity (see Supporting Information).
Figure 2. Mass spectrometry results (ESI-TOF): (a) unreacted HRP; (b) azido-HRP; (c) enlargement of (b). The mass difference between (a) and (b) is four times 26, corresponding to four diazo transfers. A closer inspection of (b), as shown in (c), reveals that tri- and pentaazido-substituted HRP molecules are also present, indicating that azidoHRP is a statistical mixture of transfer products.
Information) on the catalytic activity of HRP was assayed using 2,2′-azinobis(3-ethyl-benzothiazoline-6-sulfonic acid) (ABTS) as substrate, which is converted by HRP into the strongly absorbing product ABTS · + using H2O2. A comparison was made with the catalytic activity of the untreated enzyme. The results are shown in Figure 3. It can be concluded that the activity of the enzyme is only slightly reduced by the introduction of the azido functions and the subsequent conjugation. To test whether azido-HRP can be used for bioconjugation studies, experiments were performed to link this modified enzyme to the surface of polymersomes (22-24). Polymersomes are robust spherical vesicles with an average diameter of ca. 200 nm, created via the self-assembly of amphiphilic block copolymers (25, 26). For our experiments, polymersomes derived from polystyrene40-b-poly(L-isocyanoalanine(2-thiophen3-yl-ethyl)amide)50 (PS-PIAT) admixed with 10 wt % PS40poly(ethylene glycol)67-acetylene were used (24), designed for exterior modification via the acetylene moieties presented on their surfaces. The compounds were incubated in a phosphate buffer (20 mM, pH 7.4) with azido-HRP for 60 h in the presence of a Cu(I) catalyst. Hereafter, they were washed using centrifugal filter devices until no residual HRP activity could be detected in the flowthrough. To test the resulting aggregates for HRP activity, they were incubated with ABTS in the presence of H2O2. The progress curves shown in Figure 4 indicate that azidoHRP can be successfully immobilized on the polymeric surfaces
Figure 4. Progress curve for polymersomes with and without surfaceconjugated HRP. The polymersomes were incubated with ABTS in the presence of H2O2, and the formation of ABTS · + was measured as a function of time. After a short initial incubation period (i.e., 8 min), azido-HRP-conjugated polymersomes produced increasing amounts of ABTS · + (black dots). Polymersomes that were incubated with untreated HRP did not show any activity (gray dots).
using “click” chemistry. Electron micrographs of the (clicked) polymersomes are shown in Figure 5. To further evaluate the scope of our method, we also investigated the diazo transfer reaction on DsRed (see Figure 1), an engineered red fluorescent protein from Discosoma striata (27). To this end, DsRed was treated with transfer agent 1 in the presence of K2CO3 and Cu(II)SO4 · 5H2O in MQ (0.5 equiv of 1 was used relative to the amines in DsRed). Analysis by ESI-TOF revealed that under these experimental conditions an average of 4 amines were converted to the corresponding azides, as shown in Figure 6. The fluorescent azido-DsRed was subsequently reacted with a biochemically engineered elastinlike protein (ELP, MW 39.8 kDa) fitted with an acetylene moiety at its N-terminus. The Cu(I)-catalyzed conjugation successfully labeled the ELP with the fluorescent protein marker, as was demonstrated by SDS-PAGE analysis (Figure 7). As an additional demonstration, azido-DsRed was “clicked” to propargylic poly(ethylene glycol)45 and analyzed in the same fashion (Figure 7) (28). In summary, we may conclude that the diazo-compound 1 is a robust and facile transfer reagent for the introduction of azides in proteins. It was demonstrated that the catalytic activity of
22 Bioconjugate Chem., Vol. 20, No. 1, 2009
Communications
Figure 5. TEM and SEM images of polymersomes containing GOx (top row) and of vesicles containing GOx that have azido-HRP immobilized on their surfaces (bottom row). The black scale bars represent 1 µm. Figure 7. SDS-PAGE analysis of DsRed conjugation reactions. Proteins were visualized with Coomassie staining (top) or Via fluorescence of DsRed (bottom): (1) acetylene-ELP; (2) azido-DsRed; (3) PEGylation of azido-DsRed. A signal with an increase in mass of ∼2 kDa can be seen, corresponding to one PEG addition; (4) reaction of azido-DsRed with acetylene-ELP. A signal with a mass difference of ∼40 kDa can be seen, corresponding to ELP-conjugated DsRed.
ACKNOWLEDGMENT This work was funded in part by the Dutch National Research School Combination - Catalysis (NRSC-C) and NWO-ACTS. We thank RJRW Peters for the synthesis of R-Me, ω-dipropargylic poly(ethylene glycol)45. Supporting Information Available: Materials, equipment and protocols, synthesis of imidazole-sulfonylazide, diazo transfer conditions, preparation of polymersomes, 1,3-dipolar cycloaddition reaction conditions, enzyme activity assays, and ESI-TOF analysis. This material is available free of charge via the Internet at http://pubs.acs.org.
LITERATURE CITED Figure 6. Mass spectrometry results (ESI-TOF) for (un)functionalized DsRed: (a) unreacted DsRed; (b) azido-DsRed. The broad signal shown in b contains approximate masses for mono-, di-, tri-, and tetra-azidoDsRed, along with a small signal for the unfunctionalised protein. This statistical mixture of transfer products, caused by the 22 lysine residues present in DsRed, slightly compromises the accuracy of the deconvolution results. The symmetry of the signal suggests that penta and higher degrees of azido-DsRed are also present.
enzymes, such as HRP, after the diazo transfer reaction and subsequent coupling to a small molecule was not substantially reduced. This was further verified by the immobilization of the azido-HRP on the surface of a polymersome, after which the catalytic activity of the enzyme was also retained. To further demonstrate the versatility of the diazo transfer reaction, the fluorescent protein DsRed was equipped with azides and subsequently conjugated to an acetylene-bearing elastin-like protein via Cu(I)-catalyzed triazole formation. These examples show that the diazo-transfer reaction is a suitable technique for the selective modification of proteins under ambient conditions. It should be noted, however, that the present method may not be suitable for proteins containing lysines that are essential to their biological activity.
(1) Dirks, A. J. T., van Berkel, S. S., Hatzakis, N. S., Opsteen, J. A., van Delft, F. L., Cornelissen, J. J. L. M., Rowan, A. E., van Hest, J. C. M., Rutjes, F. P. J. T., and Nolte, R. J. M. (2005) Preparation of biohybrid amphiphiles via the copper catalysed Huisgen [3 + 2] dipolar cycloaddition reaction. Chem. Commun. 33, 4172-4174. (2) Reulen, S. W. A., Brusselaars, W. W. T., Langereis, S., Mulder, W. J. M., Breurken, M., and Merkx, M. (2007) Protein-liposome conjugates using cysteine-lipids and native chemical ligation. Bioconjugate Chem. 18, 590–596. (3) Heredia, K. L., Bontempo, D., Ly, T., Byers, J. T., Halstenberg, S., and Maynard, H. D. (2005) In situ preparation of protein ”Smart” polymer conjugates with retention of bioactivity. J. Am. Chem. Soc. 127, 16955–16960. (4) Thordarson, P., Le Droumaguet, B., and Velonia, K. (2006) Well-defined protein-polymer conjugates-synthesis and potential applications. Appl. Microbiol. Biotechnol. 73, 243–254. (5) Tilley, S. D., and Francis, M. B. (2006) Tyrosine-selective protein alkylation using π-allylpalladium complexes. J. Am. Chem. Soc. 128, 1080–1081. (6) Van Hest, J. C. M., Kiick, K. L., and Tirrell, D. A. (2000) Efficient incorporation of unsaturated methionine analogues into proteins in vivo. J. Am. Chem. Soc. 122, 1282–1288. (7) Wang, L., and Schultz, P. G. (2005) Expanding the genetic code. Angew. Chem., Int. Ed. 44, 34–66.
Communications (8) Hahn, M. E., and Muir, T. W. (2005) Manipulating proteins with chemistry: a cross-section of chemical biology. Trends Biochem. Sci. 30, 26–34. (9) Huisgen, R. (1963) 1,3-Dipolare Cycloadditionen - Ruckschau und Ausblick. Angew. Chem., Int. Ed. 75, 60460+. (10) Lutz, J. F. (2007) 1,3-Dipolar cycloadditions of azides and alkynes: A universal ligation tool in polymer and materials science. Angew. Chem., Int. Ed. 46, 1018–1025. (11) Wang, Q., Chan, T. R., Hilgraf, R., Fokin, V. V., Sharpless, K. B., and Finn, M. G. (2003) Bioconjugation by copper(I)catalyzed azide-alkyne [3 + 2] cycloaddition. J. Am. Chem. Soc. 125, 3192–3193. (12) Strable, E., Prasuhn, D. E., Udit, A. K., Brown, S., Link, A. J., Ngo, J. T., Lander, G., Quispe, J., Potter, C. S., Carragher, B., Tirrell, D. A., and Finn, M. G. (2008) Unnatural amino acid incorporation into virus-like particles. Bioconjugate Chem. 19, 866–875. (13) Schoffelen, S., Lambermon, M. H. L., van Eldijk, M. B., and van Hest, J. C. M. (2008) Site-specific modification of Candida antarctica lipase B via residue-specific incorporation of a noncanonical amino acid. Bioconjugate Chem. 19, 1127–1131. (14) Van Kasteren, S. I., Kramer, H. B., Gamblin, D. P., and Davis, B. G. (2007) Site-selective glycosylation of proteins: creating synthetic glycoproteins. Nat. Protocols 2, 3185–3194. (15) Kodama, K., Fukuzawa, S., Nakayama, H., Kigawa, T., Sakamoto, K., Yabuki, T., Matsuda, N., Shirouzu, M., Takio, K., Tachibana, K., and Yokoyama, S. (2006) Regioselective carbon-carbon bond formation in proteins with palladium catalysis; New protein chemistry by organometallic chemistry. ChemBioChem 7, 134–139. (16) Vasella, A., Witzig, C., Chiara, J. L., and Martinlomas, M. (1991) Convenient synthesis of 2-azido-2-deoxy-aldoses by diazo transfer. HelV. Chim. Acta 74, 2073–2077. (17) Beckmann, H. S. G., and Wittmann, V. (2007) One-pot procedure for diazo transfer and azide-alkyne cycloaddition: Triazole linkages from amines. Org. Lett. 9, 1–4. (18) Pothukanuri, S., and Winssinger, N. (2007) A highly efficient azide-based protecting group for amines and alcohols. Org. Lett. 9, 2223–2225.
Bioconjugate Chem., Vol. 20, No. 1, 2009 23 (19) Goddard-Borger, E. D., and Stick, R. V. (2007) An efficient, inexpensive, and shelf-stable diazotransfer reagent: Imidazole1-sulfonyl azide hydrochloride. Org. Lett. 9, 3797–3800. (20) Adams, J. C. (1977) Technical considerations on use of horseradish-peroxidase as a neuronal marker. Neuroscience 2, 141. (21) Hanker, J. S., Yates, P. E., Metz, C. B., and Rustioni, A. (1977) New specific, sensitive and non-carcinogenic reagent for demonstration of horseradish-peroxidase. Histochem. 9, 789–792. (22) Felici, M., Marza´-Pe´rez, M., Hatzakis, N. S., Nolte, R. J. M., and Feiters, M. C. (2008) β-Cyclodextrin-appended giant amphiphile: aggregation to vesicle polymersomes and immobilisation of enzymes. Chem.-Eur. J. 14, 9914–9920. (23) Opsteen, J. A., Brinkhuis, R. P., Teeuwen, R. L. M., Lo¨wik, D. W. P. M., and van Hest, J. C. M. (2007) “Clickable” polymersomes. Chem. Commun. 30, 3136–3138. (24) Van Dongen, S. F. M., Nallani, M., Schoffelen, S., Cornelissen, J. J. L. M., Nolte, R. J. M., and van Hest, J. C. M. (2008) A block copolymer for functionalisation of polymersome surfaces. Macromol. Rapid Commun. 29, 321–325. (25) Discher, D. E., and Eisenberg, A. (2002) Polymer vesicles. Science 297, 967–973. (26) Vriezema, D. M., Hoogboom, J., Velonia, K., Takazawa, K., Christianen, P. C. M., Maan, J. C., Rowan, A. E., and Nolte, R. J. M. (2003) Vesicles and polymerized vesicles from thiophene-containing rod-coil block copolymers. Angew. Chem., Int. Ed. 42, 772–776. (27) Campbell, R. E., Tour, O., Palmer, A. E., Steinbach, P. A., Baird, G. S., Zacharias, D. A., and Tsien, R. Y. (2002) A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. U.S.A. 99, 7877–7882. (28) The small mass difference between lanes for DsRed is presumably caused by the influence of either Cu(II) or the introduced diazo-moieties; further investigations did not satisfactorily elucidate this. The fluorescent image still indicates that both bands are indeed DsRed signals. BC8004304
