Palladium-mediated site-selective Suzuki-Miyaura protein modification at genetically encoded aryl halides

Article abstract of DOI:10.1039/c0cc04970k

Site-specific genetic incorporation of unnatural p-halophenylalanine amino acid residues as 'tags' coupled with Pd(0)-mediated Suzuki-Miyuara 'modification' has been enabled by discovery of an effective small molecule palladium scavenger.

Full text of DOI:10.1039/c0cc04970k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 1698–1700
www.rsc.org/chemcomm
COMMUNICATION
Palladium-mediated site-selective Suzuki–Miyaura protein modification
at genetically encoded aryl halidesw
Christopher D. Spicer and Benjamin G. Davis*
Received 15th November 2010, Accepted 8th December 2010
DOI: 10.1039/c0cc04970k
Site-specific genetic incorporation of unnatural p-halophenyl-
alanine amino acid residues as ‘tags’ coupled with Pd(0)-
mediated Suzuki–Miyuara ‘modification’ has been enabled by
discovery of an effective small molecule palladium scavenger.
however, to employ genetically encoded coupling partners. In
doing so, this highly selective carbon–carbon bond formation
could have potential in vivo applications. However, efficient
cross-couplings at genetically encoded aryl halides remain an
unsolved problem despite being first envisaged over a decade
1
3
The modification of proteins is a key process in living cells,
greatly increasing diversity in protein functionality and structure.
Yet our ability to model and mimic such modifications is
limited by the relatively small number of bio-orthogonal and
ago.
A number of developments in the field of synthetic biology
have allowed the genetic incorporation of such haloaryl amino
acids through use of auxotrophic strains, and the expansion of
the genetic code through amber stop codon suppression, and
1
,2
site-selective synthetic methodologies that are available.
Traditionally, protein modification has been carried out at
4
,14–17
codon expansion.
In particular, the incorporation of
3
4
nucleophilic cysteine or lysine residues. However, in recent
years a number of reactions have been developed, that allow
site-selective modification at unconventional residues.
Amongst the most attractive of these modifications are transition
p-iodophenylalanine (pIPhe) has been well documented in a
range of proteins, with high fidelity and protein yields
1
approaching those of the wild-type.
8–20
We report here the incorporation of pIPhe into the model
maltose binding protein (MBP) using UAG stop-codon
suppression, and the use of this protein to explore Suzuki–
Miyaura couplings on a protein surface as part of a general
5
metal-catalysed reactions. The reactive groups associated
with such reactions are not commonly found in proteins
and, through manipulation of metal and ligands, high
conversions under mild conditions and with high functional
group tolerance can be achieved.
2
1
tag-and-modify strategy (Scheme 1). We also demonstrate
the use of a scavenger, 3-mercaptopropionic acid (3-MPrAc),
to remove palladium from samples allowing ready and accurate
monitoring and optimization of these protein reactions. We
finally demonstrate the first Suzuki–Miyaura coupling at a
genetically incorporated aryl halide, resulting in high conversion
under ambient conditions.
We have previously demonstrated the utility of transition
metals in protein modification. For instance, Ru-catalysis in
protein olefin metathesis was achieved at privileged allyl
6
–8
chalcogenide containing amino acid residues.
In addition,
a convenient, water-soluble palladium catalyst was recently
developed for Suzuki–Miyaura cross-couplings at a cysteine-
linked aryl halide. This catalyst’s use in small molecule synthesis
In this study, His-tagged MBP was used as a model protein,
2
2,23
due to its high stability and solubility.
used protein, especially as a fusion construct. Importantly,
both MBP and His are sequences that are widely used in
MBP is a widely
9
was also demonstrated. Although related cross-couplings
have previously been reported on proteins containing genetically
incorporated coupling partners, they have suffered from low
conversions (2% for Heck and 25% for Sonogashira
6
protein constructs. Using the pDBÁHisÁMBP vector as a starting
point, an amber stop codon was installed at sites corresponding
to residue E13 by site-directed mutagenesis to generate the
1
0,11
12
couplings),
or the requirement of high temperatures.
ꢀ
Our initial report on Suzuki–Miyaura coupling was the first
instance where cross-coupling had proceeded to full conversion
on a protein substrate. To demonstrate the suitability of cross-
coupling on proteins, the aryl halide was installed chemically.
For more general application of such couplings it is desirable,
mutant plasmid pMBPÁE13 . This site was chosen as the
crystal structure of MBP indicated that this residue was
Chemistry Research Laboratory, University of Oxford, 12 Mansfield
Road, Oxford, OX1 3TA, UK. E-mail: Ben.Davis@chem.ox.ac.uk;
Fax: +44 (0)1865 275674; Tel: +44 (0)1865 275652
w Electronic supplementary information (ESI) available: Full experi-
mental details including raw LC-MS data and protein expression and
purification. See DOI: 10.1039/c0cc04970k
Scheme 1 Tag-and-modify strategy utilizing genetically incorporated
p-halo-Phe as the ‘tag’ and Suzuki–Miyaura cross coupling as the
‘modify’ steps, respectively.
1
698 Chem. Commun., 2011, 47, 1698–1700
This journal is c The Royal Society of Chemistry 2011

View Article Online
2
solvent accessible. This plasmid was then co-transformed
3
to palladium catalysis, but is potentially relevant to all work
utilising transition metals, or other Lewis acidic species, and
may hinder future work in metal-catalyzed protein modification.
We therefore looked to identify scavengers that might
co-ordinate the palladium more strongly than the protein,
and therefore remove it from the protein surface. In addition
to allowing reaction analysis, such scavengers could in future
be used for reaction purification via dialysis or size-exclusion
chromatography. This scavenging therefore makes such
reactions applicable to therapeutic proteins. Scavenging
experiments were initially undertaken on the readily available
protein equine myoglobin which is commonly used as a
reference in LC-MS analysis.
2
into E. coli with a pEVOL plasmid. This second plasmid
0
tyr
codes for a mutant tRNA/tRNA synthetase pair that when
expressed allows for suppression of the UAG amber codon at
ꢀ
site E13 with a tRNA charged with pIPhe such that this ‘stop’
2
codon is instead translated as pIPhe. The resulting protein
0
ꢀ
product MBP–E13 (1) was expressed in good yields
À1
(14 mg L ) and high fidelity (Fig. S1, ESIw; in the absence
of pIPhe no protein expression was observed).
With this site-selectively tagged protein in hand we examined
its reactivity with Pd under various conditions. However, for
certain reactions dramatic losses of signal intensity were
observed, such that precise reaction monitoring by mass
spectrometry was rendered difficult. To determine the potential
sources of this signal loss, either loss of protein during reaction
EDTA, a common hexadentate chelating ligand, was the
2
4
first potential scavenger to be identified. DTT was also
chosen, as it contains two nucleophilic thiol residues and is
also compatible with proteins as shown by its use as a reducing
agent. Additionally, cysteine was seen as a potential scavenger
and has been used previously to greatly reduce palladium
(
through degradation or precipitation) or simply signal
suppression (through reduced ionization or the nonspecific
co-ordination of multiple metal cations to the protein surface
generating many adduct isoforms, each at small quantities that
are undetectable) we first investigated the product of reactions
with Pd. SDS-PAGE showed no significant loss of protein at
2
5
contaminants in small molecule synthesis. However, none
of these ligands showed any positive effects on the protein MS
trace over a range of concentrations. Amongst the other
ligands considered, 3-mercaptopropionic acid (3-MPrAc) has
previously been used conjugated to magnetic nanoparticles to
5
00 equiv. of palladium (see ESIw). Similar observations were
made with other proteins equine myoglobin (Fig. 1a–d) and
the model protein Np276 (Fig. 1e), from Nostoc punctiforme.
In the latter, the direct observation of small number adduct
isoforms (corresponding to mono- and di-co-ordination) at
low Pd concentrations strongly supported the second (metal
suppression) hypothesis.
2
6
co-ordinate palladium. We speculated that this material
could form a 6-membered chelate to a metal centre. Gratifyingly,
it was found that when this ligand was used as a scavenger, a
strong re-emergence of the protein ion series was observed
À1
This problem of non-specific metal binding to proteins has
5
been previously discussed by Francis and Antos. It is
(Fig. 1a–d). A scavenger concentration of 4.4 mmol mL
À1
(3 equiv. w.r.t. Pd), added as a dilute solution (5 mL mL ) in
water to ensure accurate titration, was found to be optimal for
important to note that this is a problem that is not specific
Fig. 1 (a) Addition of 500 equiv. palladium catalyst 3 to equine myoglobin, followed by addition of the scavenger 3-MPrAc, (b) total ion
chromatogram of pure protein (left), protein in the presence of 500 equiv. Pd (middle), and after titration with 3-MPrAc (right), (c) corresponding
ion series at time of protein elution, (d) corresponding deconvoluted mass chromatogram, (e) deconvoluted mass spectrum of Np276 in the
presence of Pd showing discrete mono- and di-coordination of Pd.
This journal is c The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 1698–1700 1699

View Article Online
acids as coupling partners opens up the possibility of installing
hydrophobic groups onto a protein surface, without the need
5
,9
for organic solvents or denaturing conditions.
We are now working towards unlocking the potential of
Suzuki–Miyaura couplings, and investigating their scope for
modification of proteins. We are particularly interested in the
installation of hydrophobic groups, and the formation of
bi-aryls as a method for fluorogenic protein labelling. We
thank Prof. P.G. Schultz for donation of pEVOL (pIPhe)
plasmid, Justin M. Chalker, Drs Huiwang Ai, Mark
Batchelor, John Porter and Rikki Alexander for helpful
discussions, and UCB and BBSRC for funding. BGD is a
Royal Society Wolfson Research Merit Award recipient and is
supported by an EPSRC LSI Platform grant.
Notes and references
1
2
3
4
5
6
E. M. Sletten and C. R. Bertozzi, Angew. Chem., Int. Ed., 2009, 48,
974.
C. P. R. Hackenberger and D. Schwarzer, Angew. Chem., Int. Ed.,
008, 47, 10030–10074.
J. M. Chalker, G. J. L. Bernardes, Y. A. Lin and B. G. Davis,
Chem.–Asian J., 2009, 4, 630–640.
A. J. de Graaf, M. Kooijman, W. E. Hennink and
E. Mastrobattista, Bioconjugate Chem., 2009, 20, 1281–1295.
J. M. Antos and M. B. Francis, Curr. Opin. Chem. Biol., 2006, 10,
6
2
ꢀ
Scheme 2 Coupling of furan-3-boronic acid to MBP–E13 . (a) t = 0,
b) t = 1 h, (c) t = 2 h.
(
À1
myoglobin (1 mg mL ) (see ESIw). A diad of 3-MPrAc with a
propane linker, 2, was also synthesised, and showed similar
scavenging ability at double the number of equivalents.
Now in a position to more readily investigate Suzuki–
Miyaura couplings directly on proteins, this system was
applied to MBP. Initial experiments showed optimal conditions
to be highly dependent on both palladium and boronic acid
concentrations. We chose to use furan-3-boronic acid as our
coupling partner, as we had previously observed a high
2
53–262.
J. M. Chalker, Y. A. Lin, O. Boutureira and B. G. Davis, Chem.
Commun., 2009, 3714–3716.
7 Y. A. Lin, J. M. Chalker, N. Floyd, G. J. L. Bernardes and
B. G. Davis, J. Am. Chem. Soc., 2008, 130, 9642–9643.
8
Y. A. Lin, J. M. Chalker and B. G. Davis, J. Am. Chem. Soc., 2010,
32, 16805–16811.
9 J. M. Chalker, C. S. C. Wood and B. G. Davis, J. Am. Chem. Soc.,
009, 131, 16346–16347.
1
2
9
10 K. Kodama, S. Fukuzawa, H. Nakayama, T. Kigawa, K. Sakamoto,
T. Yabuki, N. Matsuda, M. Shirouzu, K. Takio, K. Tachibana and
S. Yokoyama, ChemBioChem, 2006, 7, 1577–1581.
coupling rate and efficiency in model reactions (Scheme 2).
Gratifyingly, at 50 equiv. of Pd, and 680 equiv. of boronic
acid, 50% conversion was observed at 37 1C after 1 hour
11 K. Kodama, S. Fukuzawa, H. Nakayama, K. Sakamoto,
T. Kigawa, T. Yabuki, N. Matsuda, M. Shirouzu, K. Takio,
S. Yokoyama and K. Tachibana, ChemBioChem, 2007, 8, 232–238.
(
Scheme 2b). After 2 hours, complete conversion to the cross-
coupled product was observed (Scheme 2c). In the absence of
either palladium or boronic acid no reaction was observed,
consistent with the proposed coupling mechanism. In addition, in
the absence of 3-MPrAc no discernable protein peak could
be observed by LC-MS, indicating the critical use of the
scavenger during reaction analysis (see ESIw).
1
2 E. Brustad, M. L. Bushey, J. W. Lee, D. Groff, W. Liu and
P. G. Schultz, Angew. Chem., Int. Ed., 2008, 47, 8220–8223.
13 N. Sharma, R. Furter, P. Kast and D. A. Tirrell, FEBS Lett., 2000,
67, 37–40.
4
4 A. J. Baldwin, J. A. J. Arpino, W. R. Edwards, E. M. Tippmann
1
and D. D. Jones, Mol. BioSyst., 2009, 5, 764–766.
15 T. L. Hendrickson, V. d. Crecy-Lagard and P. Schimmel, Annu.
´
Rev. Biochem., 2004, 73, 147–176.
6 K. Kirshenbaum, I. S. Carrico and D. A. Tirrell, ChemBioChem,
In conclusion, we have developed a method for scavenging
palladium from protein surfaces using the ligand 3-MPrAc.
This scavenging was essential for LC-MS analysis of the
Suzuki–Miyaura cross-coupling. We then used this technique
to undertake and optimize the first reported full-conversion
Suzuki–Miyaura coupling on a protein at a genetically
incorporated amino acid. This reaction did not require elevated
temperatures or mixed solvents, and proceeded to completion
in 2 hours. This site-specific reaction is an addition to the ‘tool-kit’
for protein modification, and also allows the formation of
structures previously difficult to produce by conventional
methods. For example, the furano-phenylalanine created here
on MBP has not been previously incorporated by any other
means, and exhibits a unique spectral signature which may
prove useful for protein tracking. The work presented here
generalizes Suzuki–Miyaura couplings on proteins and makes
it more applicable as a methodology for chemical post-
translational modification. In particular, the use of boronic
1
2
002, 3, 235–237.
17 D. R. Liu, T. J. Magliery, M. Pastrnak and P. G. Schultz, Proc.
Natl. Acad. Sci. U. S. A., 1997, 94, 10092–10097.
1
1
2
8 J. Guo, E. M. Charles III, H. S. Lee, D. Groff and P. G. Schultz,
Angew. Chem., Int. Ed., 2009, 48, 9148–9151.
9 J. Xie, L. Wang, N. Wu, A. Brock, G. Spraggon and P. G. Schultz,
Nat. Biotechnol., 2004, 22, 1297–1301.
0 T. S. Young, I. Ahmad, J. A. Yin and P. G. Schultz, J. Mol. Biol.,
2
010, 395, 361–374.
2
2
1 B. G. Davis, Pure Appl. Chem., 2009, 81, 285–298.
2 J. D. Fox, R. B. Kapust and D. S. Waugh, Protein Sci., 2001, 10,
622–630.
2
3 M. B. Zwick, L. L. C. Bonnycastle, K. A. Noren, S. Venturini,
E. Leong, C. F. Barbas, C. J. Noren and J. K. Scott, Anal.
Biochem., 1998, 264, 87–97.
24 Y. Urawa, M. Miyazawa, N. Ozeki and K. Ogura, Org. Process
Res. Dev., 2003, 7, 191–195.
2
ˇ
5 M. Prashad, Y. Liu and O. Repic, Adv. Synth. Catal., 2003, 345,
533–536.
6 L. M. Rossi, L. L. R. Vono, F. P. Silva, P. K. Kiyohara,
E. L. Duarte and J. R. Matos, Appl. Catal., A, 2007, 330, 139–144.
2
1
700 Chem. Commun., 2011, 47, 1698–1700
This journal is c The Royal Society of Chemistry 2011