A pyrrolysine analogue for site-specific protein ubiquitination

Article abstract of DOI:10.1002/anie.200904472

The art of stitching proteins: D-Cys-ε-Lys and its diastereomer L-Cys-ε-Lys read through the UAC codon (see scheme). As the resulting proteins can participate in native chemical ligation (NCL), this process provides a means to prepare proteins chemoselect

Full text of DOI:10.1002/anie.200904472

Communications
DOI: 10.1002/anie.200904472
Protein Engineering
A Pyrrolysine Analogue for Site-Specific Protein Ubiquitination**
Xin Li, Tomasz Fekner, Jennifer J. Ottesen, and Michael K. Chan*
Native chemical ligation (NCL)[1] is the most widely used
strategy for the convergent synthesis of proteins and large
peptides.[2] It involves the chemoselective reaction between a
coupling partner 1 armed with a C-terminal thioester and a
peptide segment 2 bearing an N-terminal cysteine residue
Scheme 2. Pyrrolysine (5) and analogues 6–8.
acid, is sufficiently flexible to enable a number of other lysine
derivatives to read through the amber stop codon.[5–9] We
previously described 6, a stable THF-based analogue of 5.[10]
We also introduced 7, which, owing to the presence of a
terminal alkyne functionality, can be used as a chemical
handle to label proteins through click chemistry.[11] To further
expand the range of available pyrrolysine analogues with
unique and useful reactivity, we decided to test whether the
d-cysteine-based analogue (S,S)-8 (d-Cys-e-Lys) could read
through the UAG codon. We focused our attention on this
cysteine isomer because our related readthrough studies of
simple pyrrolysine analogues for protein click chemistry
indicated that the presence of a lysine acyl substituent with
the analogous sense of chirality to that found in 5 has a
profound and beneficial influence on incorporation effi-
ciency.[12] For comparison purposes, however, we also
included in our studies the diastereomeric analogue (R,S)-8
(l-Cys-e-Lys).
Scheme 1. Cysteine-based NCL.
(Scheme 1) to generate a product with a native backbone.
Specifically, a reversible transthioesterification in the pres-
ence of an exogenous thiol[2] R2SH gives an intermediate 3,
which in turn undergoes irreversible intramolecular S!N
acyl transfer to give the final peptide 4. Herein, we introduce
a genetically encoded pyrrolysine analogue that places a
ligation handle directly into a recombinant protein. We used
NCL at this internal ligation site to generate a semisynthetic
ubiquitinated protein.
The cellular machinery for the incorporation of pyrroly-
sine (5, Scheme 2),[3,4] the 22nd genetically encoded amino
The target pyrrolysine analogue (S,S)-8 was prepared by
coupling the N,S-protected cysteine derivative (S)-9 with Boc-
Lys-OtBu (10)[11] to provide amide (S,S)-11 in excellent yield
(96%, d.r. > 99.9:0.1; Scheme 3). Full deprotection with
trifluoroacetic acid (TFA)/Et3SiH furnished (S,S)-8 as its
TFA salt. The diastereomer (R,S)-8 was prepared in an
analogous manner (see the Supporting Information).
[*] X. Li,[+] Dr. T. Fekner,[+] Prof. M. K. Chan
The Ohio State Biophysics Program, Departments of Chemistry and
Biochemistry, The Ohio State University
484 W 12th Avenue, Columbus, OH 43210 (USA)
Fax: (+1)614-292-6773
E-mail: chan@chemistry.ohio-state.edu
Homepage: http://www.chemistry.ohio-state.edu/~chan/
Prof. J. J. Ottesen
Department of Biochemistry, The Ohio State University
484 W 12th Avenue, Columbus, OH 43210 (USA)
[+] These authors contributed equally.
[**] This research was supported by a grant from the US National
Institutes of Health (GM061796) and an American Heart Associ-
ation Great Rivers Affiliate Predoctoral Fellowship (0815449D) to
X.L. We also thank the staff of the CCIC Mass Spectrometry and
Proteomics Facility at OSU for protein analysis by mass spectrom-
etry, Professor Michael Zhu (OSU) for providing Rattus norvegicus
CaM cDNA, and Professor Bing Hao (UConn) for the human
ubiquitin cDNA.
Scheme 3. Synthesis of the pyrrolysine analogue (S,S)-8: a) Boc-Lys-
OtBu (10), BOP, N-methylmorpholine, CH2Cl2, room temperature,
24 h, 96%; b) TFA, Et3SiH, CH2Cl2, room temperature, 3 h, 90%
(approximate yield). Boc=tert-butoxycarbonyl, BOP=(benzotriazol-1-
yloxy)tris(dimethylamino)phosphonium hexafluorophosphate, Trt=tri-
tyl (triphenylmethyl).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200904472.
9184
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 9184 –9187

Angewandte
Chemie
To evaluate the UAG codon readthrough efficiency for
the two diastereomers (S,S)-8 and (R,S)-8, we employed the
brightly emitting red fluorescent protein mCherry[13] as a
reporter.[12] In brief, the Lys55 codon of this protein was
mutated site specifically to UAG and inserted into the
plasmid pPylST, which harbors the pyrrolysine tRNA (PylT)
and synthetase (PylS) genes. Escherichia coli strain BL21-
(DE3) transformed with this plasmid was grown in the Terrific
Broth medium supplied with either (S,S)-8 or (R,S)-8 at
varying concentrations. The results of the mCherry read-
through assays demonstrate that the presence of either isomer
enables readthrough of the UAG codon, although (S,S)-8
serves as a much better substrate in terms of incorporation
efficiency (Figure 1).
synthetic chemistry techniques required limit its wider use as
a general tool by researchers in biochemistry and structural
biology.
Herein, we demonstrate that it is possible to generate a
site specifically ubiquitinated protein in a single ligation step
from two genetically encoded segments by taking advantage
of our pyrrolysine analogue (S,S)-8. This analogue is an
excellent mimic of the three key structural elements of the
ubiquitination site. Namely, its lysine functionality is identical
to the target lysine residue in the substrate protein, the d-Cys
residue replaces the terminal Gly residue of ubiquitin, and an
isopeptide bond between ubiquitin and the target lysine side
chain is retained. Thus, the replacement of the Gly76 residue
from ubiquitin with d-Cys is the only difference between the
semisynthetic product and a natively ubiquitinated protein.
For our model studies, we chose calmodulin (CaM), a
small 17-kDa protein that plays a central role in calcium
signaling in eukaryotes. The reversible ubiquitination of CaM
is catalyzed by E3-CaM (ubiquitin–calmodulin ligase, EC
6.3.2.21)[21–23] at Lys21[24] and leads to the production of
ubiquitinated CaM (Ub-CaM).[24,25] Instead of targeting CaM
for proteosome degradation,[26] the ubiquitination of CaM
modulates its regulatory activities.
To generate (S,S)-8-containing CaM ((S,S)-8-CaM),
Rattus norvegicus CaM(Lys21Pyl) cDNA was subcloned into
pPylST (Figure 2a). The recombinant protein (S,S)-8-CaM
produced in this way was purified by hydrophobic-interaction
chromatography as described previously.[11] A quantity of
0.9 mg of (S,S)-8-CaM could be isolated from 50 mL of the
culture supplied with 2.5 mm (S,S)-8. Significantly, MALDI-
TOF MS analysis of the purified product (Figure 2c) demon-
strated that the reactive Cys mimic remains intact throughout
expression in a cellular system.
Figure 1. Dose-dependent readthrough of (S,S)-8 and (R,S)-8.
The incorporation of 8 into a target protein provides a
chemical handle for branching through NCL at a specific site.
A ubiquitinated protein is perhaps the most important
example of a branched protein structure. It is generated by
ubiquitination, a special posttranslational modification in
which the C-terminal glycine residue (G76) of the small
protein ubiquitin is attached to the e-amino group of a lysine
residue in a substrate through an isopeptide bond, the
formation of which is catalyzed by a series of enzymes.[14,15]
Protein ubiquitination plays an important role in many
cellular processes, including protein degradation, cellular
signaling, cell division and differentiation, and protein
trafficking.[14–18] Biochemical and structural studies on ubiq-
uitination require the isolation or generation of homogenous
ubiquitinated proteins. However, isolation from an in vivo
source is usually low yielding, and the obtained protein may
contain other posttranslational modifications. On the other
hand, the in vitro reconstitution of ubiquitination with
purified proteins and enzymes often suffers from low
productivity and limited availability of the specific ubiquitin
ligases. In a pioneering study that highlights the use of NCL to
generate ubiquitinated proteins, the Muir research group used
NCL assisted by a ligation auxiliary to prepare ubiquitinated
histone 2B (H2B) from three separate pieces.[19,20] Unfortu-
nately, this method is convenient only for proteins that
undergo ubiquitination near the termini, and the advanced
The truncated Homo sapiens ubiquitin Ub75 (containing
residues 1–75) was produced as an Ub75/intein/CBD (chitin-
binding domain) fusion protein and purified by chitin-affinity
chromatography. On-column thiolysis was initiated with
sodium 2-mercaptoethane sulfonate (MESNa) to generate
the Ub75 thioester (Ub75-SR, R = CH2CH2SO3Na), which
was mixed with (S,S)-8-CaM in a 5:1 molar ratio to promote
NCL (Figure 2a). The reaction mixture was incubated at
room temperature overnight, and the ubiquitinated calm-
odulin product (Ub*-CaM) was separated from unreacted
Ub75-SR and (S,S)-8-CaM by anion-exchange chromatogra-
phy. Approximately 30% of the recombinant protein (S,S)-8-
CaM was converted into ubiquitinated CaM (Figure 2b). This
moderate yield is probably due to the NCL step rather than
heterogeneity of (S,S)-8-CaM as a result of alternative
readthrough, as no full-length CaM was observed on SDS-
PAGE gels in the absence of the pyrrolysine analogue (S,S)-8.
The identity of the ligation product was confirmed by
MALDI-TOF mass spectrometry and tandem mass spec-
trometry (Figure 2c; see also Figure S2 and Table S1 in the
Supporting Information).
One fundamental question is whether Ub*-CaM gener-
ated by NCL has the same functional properties as enzymati-
cally prepared Ub-CaM. CaM is known to bind to phosphor-
ylase kinase and increase its activity. The ubiquitination of
CaM has been reported to lead to decreased affinity for
Angew. Chem. Int. Ed. 2009, 48, 9184 –9187
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9185

Communications
phosphorylase kinase and a decrease in the
degree of activation of the enzyme.[24] To
explore this issue, we evaluated the ability of
wild-type (WT) CaM and NCL-derived Ub*-
CaM to promote phosphorylase kinase activ-
ity in a coupled assay system. In our hands,
the Ub*-CaM prepared through NCL exhib-
ited a decreased ability to modulate phos-
phorylase kinase activity when compared to
WT CaM. In particular, the degree of activa-
tion (e) by Ub*-CaM was 85% lower than
that observed with WT CaM (Figure 3a).
This value compares well to the 75%
decrease reported for enzymatically gener-
ated Ub-CaM.[24]
e ¼ ðactivityCaMꢀactivityno CaMÞ=activityno CaM
ð1Þ
Having demonstrated that Ub*-CaM
behaves similarly to Ub-CaM in its effect on
phosphorylase kinase activity, we sought to
use Ub*-CaM to explore the effect of ubiq-
uitination on the CaM-mediated regulation
of other enzymes. One intriguing target was
protein phosphatase 2B (PP2B), the only
known protein phosphatase regulated by
CaM. Such a study had the added appeal
that no previous investigations had been
performed, and thus it served as an excellent
opportunity to explore a fundamental bio-
chemical question. We assayed PP2B activ-
ities in the presence and absence of WT CaM
or Ub*-CaM. PP2B dephosphorylation was
activated by WT CaM as expected; however,
surprisingly, ubiquitination had no effect on
the ability of CaM to modulate PP2B activity
(Figure 3b). The fact that ubiquitination of
CaM has a different effect on these kinase
and phosphatase systems opens the possibil-
ity that ubiquitination can modulate the
CaM-mediated linkage between calcium sig-
naling and phosphorylation.
In summary, we have shown that the
simple lysine derivatives (S,S)-8 and (R,S)-8
are incorporated into a protein in response to
the UAG codon when the pyrrolysine-incor-
poration machinery is present. The resulting
protein incorporating (S,S)-8 was successfully
used to introduce ubiquitin through NCL. As
this approach does not require the isolation
of the key ligases needed for enzymatic
ubiquitination, it should enable detailed bio-
chemical studies of many ubiquitinated pro-
teins that it would not otherwise be possible
to characterize.
Figure 2. a) Overall reaction scheme for the generation of ubiquitinated CaM through
NCL mediated by the pyrrolysine analogue (S,S)-8. b) The process was monitored by SDS-
PAGE; the proteins were visualized by staining with Coomassie Blue. c) The identity of the
coupling partners and their product was confirmed by MALDI-TOF mass spectrometric
analysis.
Received: August 10, 2009
Published online: October 30, 2009
9186
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 9184 –9187

Angewandte
Chemie
[3] B. Hao, W. Gong, T. K. Ferguson, C. M. James, J. A. Krzycki,
M. K. Chan, Science 2002, 296, 1462 – 1466.
[4] G. Srinivasan, C. M. James, J. A. Krzycki, Science 2002, 296,
1459 – 1462.
[5] T. Mukai, T. Kobayashi, N. Hino, T. Yanagisawa, K. Sakamoto, S.
Yokoyama, Biochem. Biophys. Res. Commun. 2008, 371, 818 –
822.
[6] H. Neumann, S. Y. Peak-Chew, J. W. Chin, Nat. Chem. Biol.
2008, 4, 232 – 234.
[7] T. Yanagisawa, R. Ishii, R. Fukunaga, T. Kobayashi, K.
Sakamoto, S. Yokoyama, Chem. Biol. 2008, 15, 1187 – 1197.
[8] T. Kobayashi, T. Yanagisawa, K. Sakamoto, S. Yokoyama, J. Mol.
Biol. 2009, 385, 1352 – 1360.
[9] D. P. Nguyen, H. Lusic, H. Neumann, P. B. Kapadnis, A. Deiters,
J. W. Chin, J. Am. Chem. Soc. 2009, 131, 8720 – 8721.
[10] W. T. Li, A. Mahapatra, D. G. Longstaff, J. Bechtel, G. Zhao,
P. T. Kang, M. K. Chan, J. A. Krzycki, J. Mol. Biol. 2009, 385,
1156 – 1164.
[11] T. Fekner, X. Li, M. M. Lee, M. K. Chan, Angew. Chem. 2009,
121, 1661 – 1663; Angew. Chem. Int. Ed. 2009, 48, 1633 – 1635.
[12] X. Li, T. Fekner, M. K. Chan, unpublished results.
[13] N. C. Shaner, R. E. Campbell, P. A. Steinbach, B. N. Giepmans,
A. E. Palmer, R. Y. Tsien, Nat. Biotechnol. 2004, 22, 1567 – 1572.
[14] C. M. Pickart, M. J. Eddins, Biochim. Biophys. Acta Mol. Cell
Res. 2004, 1695, 55 – 72.
[15] C. M. Pickart, D. Fushman, Curr. Opin. Chem. Biol. 2004, 8,
610 – 616.
[16] R. L. Welchman, C. Gordon, R. J. Mayer, Nat. Rev. Mol. Cell
Biol. 2005, 6, 599 – 609.
[17] D. Mukhopadhyay, H. Riezman, Science 2007, 315, 201 – 205.
[18] W. Li, Y. Ye, Cell. Mol. Life Sci. 2008, 65, 2397 – 2406.
[19] C. Chatterjee, R. K. McGinty, J. P. Pellois, T. W. Muir, Angew.
Chem. 2007, 119, 2872 – 2876; Angew. Chem. Int. Ed. 2007, 46,
2814 – 2818.
[20] R. K. McGinty, J. Kim, C. Chatterjee, R. G. Roeder, T. W. Muir,
Nature 2008, 453, 812 – 816.
[21] H. P. Jennissen, M. Laub, Biol. Chem. Hoppe-Seyler 1988, 369,
1325 – 1330.
[22] M. Majetschak, M. Laub, C. Klocke, J. A. Steppuhn, H. P.
Jennissen, Eur. J. Biochem. 1998, 255, 492 – 500.
[23] M. Majetschak, M. Laub, H. E. Meyer, H. P. Jennissen, Eur. J.
Biochem. 1998, 255, 482 – 491.
[24] M. Laub, J. A. Steppuhn, M. Bluggel, D. Immler, H. E. Meyer,
H. P. Jennissen, Eur. J. Biochem. 1998, 255, 422 – 431.
[25] H. P. Jennissen, G. Botzet, M. Majetschak, M. Laub, R.
Ziegenhagen, A. Demiroglou, FEBS Lett. 1992, 296, 51 – 56.
[26] M. Laub, H. P. Jennissen, Biochim. Biophys. Acta Mol. Cell Res.
1997, 1357, 173 – 191.
Figure 3. a) Phosphorylase kinase activity with phosphorylase b as the
substrate and b) PP2B activity with p-nitrophenyl phosphate as the
substrate in the presence of WT CaM, (S,S)-8-CaM, Ub*-CaM, or
Ub75-SR. The activities are reported as normalized values against the
wild-type enzyme in the absence of other activators. The error bars
denote the standard deviations, each calculated from three independ-
ent measurements.
Keywords: native chemical ligation · protein engineering ·
.
protein modifications · pyrrolysine · ubiquitination
[1] P. E. Dawson, T. W. Muir, I. Clark-Lewis, S. B. Kent, Science
1994, 266, 776 – 779.
[2] S. B. Kent, Chem. Soc. Rev. 2009, 38, 338 – 351.
Angew. Chem. Int. Ed. 2009, 48, 9184 –9187
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9187