Site-specific protein immobilization by Staudinger ligation

Article abstract of DOI:10.1021/ja036712h

The Staudinger ligation between an azido-protein and a phosphinothioester-derivatized surface is demonstrated to be an effective means for the site-specific, covalent immobilization of a protein. Immobilization yields of >50% are obtained in <1 min, and immobilized proteins have >80% of their expected activity. No other method enables more rapid immobilization or a higher yield of active protein. Because azido-peptides and azido-proteins are readily attainable by synthesis, biosynthesis, or semisynthesis, the Staudinger ligation could be of unsurpassed utility in creating microarrays of functional peptides and proteins. Copyright

Full text of DOI:10.1021/ja036712h

Published on Web 09/04/2003
Site-Specific Protein Immobilization by Staudinger Ligation
†
‡
†
,†,‡
Matthew B. Soellner, Kimberly A. Dickson, Bradley L. Nilsson, and Ronald T. Raines*
Departments of Chemistry and Biochemistry, UniVersity of Wisconsin, Madison, Wisconsin 53706
Received June 16, 2003; E-mail: raines@biochem.wisc.edu
Microarrays in which proteins are immobilized to a surface
enable high-throughput experiments that require only small amounts
of analyte. Such protein “chips” can be used to detect protein-
Scheme 1
1
ligand, protein-protein, and antibody-antigen interactions. At-
taching proteins covalently2 rather than noncovalently produces
,3
4
more robust surfaces. Attaching proteins in a uniform (that is,
specific)2 rather than random manner can provide a substantial
,4
3
5
advantage in assay sensitivity. Here, we report a general method
for the covalent, uniform immobilization of peptides and proteins
that is both rapid and high-yielding. The method relies on a traceless
version of the Staudinger ligation in which an azide and phosphi-
6
nothioester react to form an amide (Scheme 1). The reaction is
known to occur in high yield at room temperature in aqueous or
wet organic solvents and is compatible with the unprotected
functional groups of proteinogenic amino acids.
Scheme 2
As a model protein to immobilize, we chose ribonuclease S′
(RNase S′), which is a truncated form of RNase S. The protease
subtilisin prefers to cleave a single peptide bond in RNase A. The
product of this cleavage, RNase S, consists of two tightly associated
fragments: S-peptide, which derives from residues 1-20 of RNase
7
A, and S-protein, which derives from residues 21-124. Although
neither fragment alone has any enzymatic activity, RNase S is as
active as intact RNase A. The first 15 residues of S-peptide (S15)
8
bind to S-protein with an affinity similar to that of S-peptide. We
reasoned that the immobilization of S15 followed by incubation
with S-protein would generate an active RNase S′ tethered to a
surface.
The S15 peptide was synthesized with an azido group in either
a side chain or its main chain. To install an azido group into a side
chain, the ꢀ-amino group of Lys1 was replaced with an azido group
3
to generate (N )Lys1 S15. To install an azido group into the main
chain, a poly(ethylene glycol) (PEG) fragment with a terminal azido
group was attached to the R-amino group of Lys1 to generate N -
3
PEG-S15.
slide was treated with an excess of PEG having succinimidyl ester
termini (NHS-C(O)-PEG-C(O)-NHS) in DMF for 2 h, and diphe-
5
nylphosphinomethanethiol in DMF for 2 h, resulting in a surface-
9
10
bound phosphinothioester (Scheme 2). Serial dilutions of (N
Lys1 S15 or N -PEG-S15 were spotted on this phosphinothio-
ester-derivatized slide. S-Protein in 0.10 M sodium phosphate buffer
pH 7.2) was spotted on the slide, which was washed thoroughly
3
)-
3
(
with buffer.
Two types of assays were used to detect the presence of
immobilized RNase S′. First, the ribonucleolytic activity of each
spot was determined with a fluorogenic substrate.11 This assay
revealed the amount of active RNase S′ on the surface. After activity
assays were performed, the slide was washed thoroughly, and
1
2
immobilized RNase S′ was visualized by immunostaining. This
assay revealed the total amount of RNase S′ on the surface and,
hence, the overall yield of the immobilization reaction. Together,
As a surface, we chose glass slides because of their compatibility
with standard array and detection equipment. An amine-derivatized
3
the assays indicated that coupling via a side chain (as with (N )-
Lys1 S15) proceeded in 51% yield and that the enzyme coupled
†
Department of Chemistry.
Department of Biochemistry.
‡
13
11790
9
J. AM. CHEM. SOC. 2003, 125, 11790-11791
10.1021/ja036712h CCC: $25.00 © 2003 American Chemical Society

C O MMU N I C A T I O N S
same amount of peptide became attached to the slide in 1 min as
in 1 h (Figure 1C) or 8 h (data not shown). Accordingly, the
immobilization reaction has t1/2 < 1 min.
The Staudinger ligation is the most efficacious method known
for the site-specific, covalent immobilization of a protein. No other
approach enables more rapid immobilization or a higher yield of
6
19
active protein (cf.: refs 2-5). Azido-peptides and azido-proteins
are readily attainable, and the reactivity of the azido group is
orthogonal to that of biomolecules. Accordingly, the Staudinger
ligation could be of unsurpassed utility in creating microarrays of
functional peptides and proteins.
Acknowledgment. The authors are grateful to Dr. K. J.
Woycechowsky for insightful discussions. This work was supported
by Grant GM44783 (NIH). B.L.N. was supported by the Abbott
Laboratories Fellowship in Synthetic Organic Chemistry (University
of Wisconsin-Madison).
Supporting Information Available: Procedures and additional data
for syntheses and analyses reported herein (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(
1) For reviews, see: (a) Kodadek, T. Chem. Biol. 2001, 8, 105-115. (b)
Zhu, H.; Snyder, M. Curr. Opin. Chem. Biol. 2001, 5, 40-45. (c) Lee,
Y.-S.; Mrksich, M. Trends Biotechnol. 2002, 20, S14-S18.
Figure 1. Site-specific protein immobilization by Staudinger ligation. (A)
Yield and activity of immobilization by Staudinger ligation. A phosphi-
nothioester-derived slide was spotted for 8 h at room temperature with
dilutions of an azido-S15 in DMF/H2O (50:1). The slide was washed
thoroughly with DMF, and then with 0.10 M sodium phosphate buffer (pH
(
2) (a) Houseman, B. T.; Huh, J. H.; Kron, S. J.; Mrksich, M. Nat. Biotechnol.
2002, 20, 270-274. (b) Lesaicherre, M.-L.; Uttamchandani, M.; Chen,
G. Y. J.; Yao, S. Q. Bioorg. Med. Chem. Lett. 2002, 12, 2079-2083. (c)
Toepert, F.; Knaute, T.; Guffler, S.; Pir e´ s, J. R.; Matzdorf, T.; Oschkinat,
H.; Schneider-Mergener, J. Angew. Chem., Int. Ed. 2003, 42, 1136-1140.
3) MacBeath, G.; Schreiber, S. L. Science 2000, 289, 1760-1763.
4) Zhu, H.; Bilgin, M.; Bangham, R.; Hall, D.; Casamayor, A.; Bertone, P.;
Lan, N.; Jansen, R.; Bidlingmaier, S.; Houfek, T.; Mitchell, T.; Miller,
P.; Dean, R. A.; Gerstein, M.; Snyder, M. Science 2001, 293, 2101-
2105.
1
0
7
.2). As controls, wild-type S15 (5 µM) was spotted for 8 h, no peptide
was spotted, or the slide was treated with NH2OH before spotting for 8 h
(
(
with N3-PEG-S15. After incubation with S-protein (10 µg/µL) for 8 h, each
11
spot was assayed for ribonucleolytic activity and fluorescence after staining
with a primary antibody (to RNase A) and a secondary antibody conjugated
12
to Alexa Fluor 488. (B) Uniform versus random immobilization of protein.
A slide displaying NHS-esters was spotted with ribonuclease A (5.0 nM)
or phosphinothioester, an azido-S15 (5.0 nM), and S-protein (10 µg/µL).
(
5) Peluso, P.; Wilson, D. S.; Do, D.; Tran, H.; Venkatasubbaiah, M.; Quincy,
D.; Heidecker, B.; Poindexter, K.; Tolani, H.; Phelan, M.; Witte, K.; Jung,
L. S.; Wanger, P.; Nock, S. Anal. Biochem. 2003, 312, 113-124.
6) For leading references, see: (a) Soellner, M. B.; Nilsson, B. L.; Raines,
R. T. J. Org. Chem. 2002, 67, 4993-4996. (b) Nilsson, B. L.; Hondal, R.
J.; Soellner, M. B.; Raines, R. T. J. Am. Chem. Soc. 2003, 125, 5268-
11,12
Each spot was assayed as in panel A.
(C) Timecourse of immobilization
(
by Staudinger ligation. A phosphinothioester-derived slide was spotted with
N3-PEG-S15 (5.6 nM) for 1-60 min, and then quenched with NH2OH.
After incubation with S-protein (10 µg/µL), each spot was subjected to
5269.
1
2
(7) (a) Richards, F. M. Compt. Rend. Lab. Carlsberg, Ser. Chim. 1955, 29,
immunoassay as in panel A.
3
29-346. (b) Richards, F. M.; Vithayathil, P. J. J. Biol. Chem. 1959,
2
34, 1459-1465. For a review, see: (c) Raines, R. T. Chem. ReV. 1998,
in this manner had 85% activity.15 Coupling via the main chain (as
98, 1045-1065.
(
8) (a) Potts, J. T., Jr.; Young, D. M.; Anfinsen, C. B. J. Biol. Chem. 1963,
14
with N
3
-PEG-S15) proceeded in 67% yield, and the enzyme
2
38, 2593-2594. (b) Connelly, P. R.; Varadarajan, R.; Sturtevant, J.;
15
Richards, F. M. Biochemistry 1990, 29, 6108-6114.
9) The PEG layer (M
≈ 3400) prevented the nonspecific attachment of
protein to the glass surface.
(10) Peptide concentrations were 5.64 µM-56.4 pM for (N
-PEG-S15 (M 2284.4).
coupled in this manner had 92% activity.
(
r
Control experiments indicate that binding occurred only by
Staudinger ligation (Figure 1A, row 3). First, S15 without an azido
group was spotted on a phosphinothioester-derivatized slide. After
incubation of the slide with S-protein, no immobilized peptide was
detected with the activity assay or immunoassay. Second, im-
mobilized peptide was necessary for S-protein to bind to the surface,
indicating that the phosphinothioester does not react with the
functional groups of proteinogenic amino acids. Finally, azido-
peptides did not react with a phosphinothioester-derivatized slide
)Lys1 S15 (M
3 r
1773.8), and 4.38 µM-43.8 pM for N
3
r
(
11) Enzymatic activity assays were performed as described (Kelemen, B. R.;
Klink, T. A.; Behlke, M. A.; Eubanks, S. R.; Leland, P. A.; Raines, R. T.
Nucleic Acids Res. 1999, 27, 3696-3701). Briefly, 0.10 M MES-NaOH
buffer (pH 6.0) containing NaCl (0.10 M) and 6-carboxyfluorescein-(dA)-
rU(dA) -6-tetramethylrhodamine (6 µM) was added to each spot. At
2
various times, an aliquot was removed from the slide, and its fluorescence
was determined.
(
12) Immunoassays were performed on each spot by adding rabbit anti-RNase
A IgG (2 µg/mL) for 30 min. The slide was then washed with PBS. Each
spot was incubated with an Alexa Fluor 488-conjugated secondary
antibody (1 µg/mL) for 30 min. The slide was then washed with PBS,
and its fluorescence was determined.
that had been treated with NH
2
OH, which rapidly cleaves thioesters.16
Uniform immobilization of a protein analyte yields higher activity
than does random immobilization. Amine-derivatized slides were
treated with NHS-C(O)-PEG-C(O)-NHS in DMF for 2 h, and then
intact RNase A (5.0 nM) in buffer for 8 h. Such immobilization
will occur randomly via enzymic amino groups. Coupling proceeded
in 45% yield,17 but the coupled RNase A had only 6% activity
2
(
13) Bound peptide, 4.1 fmol/mm (Figure 1A, row 1, column 4); spotted
2
peptide, 8.0 fmol/mm .
2
(14) Bound peptide, 4.2 fmol/mm (Figure 1A, row 2, column 4); spotted
2
peptide, 6.2 fmol/mm .
-8
(
15) These values are lower limits, as RNase S′ (K
d
) 10 M; ref 7b) could
dissociate to some extent during the timecourse of assays.
(16) Gregory, M. J.; Bruice, T. C. J. Am. Chem. Soc. 1967, 89, 2121-2126.
2
2
(
(
17) Bound peptide, 3.2 fmol/mm ; spotted peptide, 7.0 fmol/mm .
18) The side-chain amino groups of Lys7, Lys41, and Lys66 are known to be
important for catalysis (ref 7c).
(
Figure 1B). This low activity is likely due to RNase A being
18
attached through amino groups that are important for function,
which is an intrinsic disadvantage of random immobilization.2-5
(19) For a biosynthetic route, see: (a) Kiick, K. L.; Saxon, E.; Tirrell, D. A.;
Bertozzi, C. R. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 19-24. For the
basis of a semisynthetic route, see: (b) Muir, T. W. Annu. ReV. Biochem.
Immobilization by Staudinger ligation is remarkably rapid. A
2003, 72, 249-289.
3
subsaturating concentration of N -PEG-S15 was spotted for 1-60
min. After addition of S-protein, an immunoassay showed that the
JA036712H
J. AM. CHEM. SOC.
9
VOL. 125, NO. 39, 2003 11791