Rapid ligations with equimolar reactants in water with the potassium acyltrifluoroborate (KAT) amide formation

Article abstract of DOI:10.1021/ja5018442

The identification of fast, chemoselective bond-forming reactions is one of the major contemporary challenges in chemistry. We show that chemoselective amide-forming ligations of potassium acyltrifluoroborates (KATs) and O-carbamoylhydroxylamines proceed in the presence of all unprotected functional groups with a second-order rate constant of 20 M^-1 s^-1. PEG chains, lipids, biotin, and dyes were introduced onto an unprotected 31-mer peptide (a GLP-1 analogue) with equimolar ratios of reactants within minutes at 1 mM and within 1 h at 100 μM, even with M_w 20 000 PEG. This conjugation reaction provides a new approach to the synthesis of molecules such as protein-protein and protein-polymer conjugates.

Full text of DOI:10.1021/ja5018442

Communication
pubs.acs.org/JACS
Rapid Ligations with Equimolar Reactants in Water with the
Potassium Acyltrifluoroborate (KAT) Amide Formation
Hidetoshi Noda, Gabor Eros, and Jeffrey W. Bode*
́
̋
Laboratorium fur Organische Chemie, Department of Chemistry and Applied Biosciences, ETH−Zurich, 8093 Zurich, Switzerland
̈
̈
̈
S
* Supporting Information
Scheme 1. Chemoselective Amide-Formation with an
Equimolar Ratio of Reactants by KAT Ligation
ABSTRACT: The identification of fast, chemoselective
bond-forming reactions is one of the major contemporary
challenges in chemistry. We show that chemoselective
amide-forming ligations of potassium acyltrifluoroborates
(KATs) and O-carbamoylhydroxylamines proceed in the
presence of all unprotected functional groups with a
second-order rate constant of 20 M−1 s−1. PEG chains,
lipids, biotin, and dyes were introduced onto an
unprotected 31-mer peptide (a GLP-1 analogue) with
equimolar ratios of reactants within minutes at 1 mM and
within 1 h at 100 μM, even with Mw 20 000 PEG. This
conjugation reaction provides a new approach to the
synthesis of molecules such as protein−protein and
protein−polymer conjugates.
We have previously documented chemical protein synthesis8
using the α-ketoacid-hydroxylamine (KAHA) amide-forming
ligation, which proceeds chemoselectively under aqueous
conditions.9 Although this reaction works well for the assembly
of peptide segments, the comparatively slow reaction rate10
requires higher temperatures (40−60 °C), relatively long
reaction times (8−24 h) and reactant concentrations of at
least 5 mM, which is not always feasible for higher molecular
weight substrates. As part of our search for alternative amide-
forming reactions,11 we recently reported coupling of O-Bz
hydroxylamines and KATs in tBuOH/H2O, but our study was
limited to small organic molecules lacking unprotected
functional groups.12 KATs showed rapid amide formations
along with excellent stability, prompting us to consider
developing this reaction as a solution to the long-standing
problem of rapid, chemoselective conjugation of stoichiometric
quantities of substrates under mild aqueous conditions.
The O-Bz hydroxylamines used in our first report were not
sufficiently stable toward standard synthetic protocols13 and
could not be easily incorporated into peptides or other
synthetic targets. We therefore evaluated a panel of hydroxyl-
amine derivatives for stability and reactivity in the amide-
forming reaction (Table 1). This led to our selection of N,N-
diethylcarbamate 1i, a derivative that was almost as reactive as
the O-Bz substrates. Importantly, it was stable toward the
ligation conditions and unprotected primary amines, tolerant of
acidic (TFA) deprotection of Boc and related protecting
groups, and compatible to Fmoc-cleavage with piperidine in its
Boc-protected form.
he ideal chemoselective conjugation reaction would allow
Trapid, covalent-bond formation between two unique but
chemically stable moieties under aqueous conditions using
equimolar amounts of the ligation partners, regardless of the
size of the substrate or number and nature of the unprotected
functional groups. Chemical reactions that come close to this
ideal, such as the Cu-catalyzed Click reaction of azides and
alkynes1 or the addition of thiols to electrophiles, have
transformed the practice of organic and bioorganic chemistry.2
Limitations of these methods include the presence of toxic
reagents, the formation of unnatural bonds, and relatively slow
reaction rates necessitating the use of an excess of one of the
reactants. The remaining challenge in this area is the
identification of faster ligations (second-order rate constants
>10 M−1 s−1) that form natural bonds under aqueous
conditions without added reagents or catalysts.3 Such feats of
bond construction, such as DNA ligation,4 are routinely
accomplished by specific biochemical enzymes and can, in
certain circumstances, be co-opted for synthetic applications.
Fast, selective, natural-bond forming chemical ligations are
currently unknown.
Here we show that the chemoselective amide-bond forming
reaction of potassium acyltrifluoroborates (KATs)5 and O-
carbamoylhydroxylamines tolerates all common unprotected
functional groups and forms amide bonds under aqueous
conditions with a second-order rate constant of 20 M−1 s−1
(Scheme 1). This reaction enables selective conjugations of
large molecules using equimolar amounts of reactants, opening
an avenue to the synthesis of much larger, structurally defined
molecules including multidomain proteins, protein−polymer
conjugates,6 and oligomeric biomolecules.7
In order to properly assess the kinetics and chemoselectivity
of this ligation, we prepared a β-alanine derived monomer
containing the O-diethylcarbamoyl-hydroxylamine and attached
this to the N-terminus of a model peptide. The sequence was
chosen to include unprotected amines (Lys), thiols (Cys), and
carboxylic acids (Glu/Asp), which are the functional groups
Received: February 26, 2014
Published: March 31, 2014
© 2014 American Chemical Society
5611
dx.doi.org/10.1021/ja5018442 | J. Am. Chem. Soc. 2014, 136, 5611−5614

Journal of the American Chemical Society
Communication
Table 1. Determination of Suitable Ligation Partners
Scheme 2. KAT ligation of 2 with model peptide 4
hydroxyl-
amine
relative
reactivity
stability to 2°
stability to 1°
stability to
a
b
c
d
amine
amine
acid
1a
1b
1c
1d
1e
1f
NR
NR
NR
14
−
−
−
−
−
−
−
−
−
−
−
−
100
117
124
78
no
no
no
no
no
no
no
no
yes
yes
no
no
no
yes
1g
1h
1i
87
yes
a
b
1:1 tBuOH/H2O, 23 °C, 1h. N-Boc derivatives of 1 were used as
c
substrates. 20% piperidine in DMF, 23 °C, 24 h. 2-Phenylethylamine
(10 equiv), DMF, 23 °C, 1h. 50% TFA in CH2Cl2, 23 °C, 3 h. NR;
no reaction.
d
most likely to be encountered in applications of a chemo-
selective ligation (Scheme 2A). Due to the fast reaction rates,
we performed studies on the second-order kinetics using a 1:1
mixture of KAT 2 and peptide 4 at 30 μM at 23 °C in 1:1
tBuOH/H2O. Under these dilute conditions, we could
accurately monitor the conversion to product and the
disappearance of the starting materials by HPLC (Scheme
2B). At higher concentrations (100 μM) the ligations were
complete within 1 h, but the reaction was too fast for accurate
monitoring of the initial rate. We also found that the ligations
are faster at lower pH and added oxalic acid for these studies;
other acids such as phosphoric acid are also suitable. We
determined a second-order rate constant of 20 M−1 s−1,
considerably faster than the majority of ligation reactions in
common use.14 It does not approach the reported rates of the
tetrazine−trans-cyclooctene ligation15 but has the distinct
advantage of using chemically stable, easily handled functional
groups and forming an amide bond.16 We anticipate that
further development of the reaction partners will lead to faster
rates and ligations under neutral conditions, but even at the
current stage of development the reaction is fast enough for
stoichiometric ligations at <50 μM with reasonable times.
In order to easily prepare KATs for this ligation, we took
advantage of the strong electron-withdrawing nature of the
potassium acyltrifluoroborate moiety and devised a convenient
approach for attaching PEG chains of various molecular
weights,17 biotin,18 and dyes19 by nucleophilic aromatic
substitution (Scheme 3). This proved particularly convenient
for the preparation of PEG KATs 6a−6c; the desired product
could be separated from excess 2 by dialysis to give pure
PEGylated reagents without any additional purification. Lipid-
derived KAT 6d was prepared by a modification of our previous
method.20 Ongoing work in our laboratories and others will
Scheme 3. Syntheses of KAT 6 via SNAr of 2
a
6d was prepared by a modification of the reported procedure. See
Supporting Information for details.
further simplify the preparation of other KATs in the near
future.
The use of this ligation for synthetic conjugations also
requires a means to prepare complex molecules containing the
requisite hydroxylamine functionality. Given our interest in
fully synthetic constructs such as PEGylated peptides21 and
synthetic protein−protein conjugates, we first targeted the
preparation of peptides containing the hydroxylamine as a side-
chain functional group. We selected a 31-amino acid residue
analogue of the antidiabetic peptide GLP-1 as a first target, as
conjugations to these peptides are a critical and troublesome
step in the manufacture and development of new treatments.22
Several PEGylated variants are in development and a lipidated
form, liraglutide, is an approved and marketed drug.23
Improved methods for conjugating molecules to fully
unprotected peptides or proteins are in great demand for
5612
dx.doi.org/10.1021/ja5018442 | J. Am. Chem. Soc. 2014, 136, 5611−5614

Journal of the American Chemical Society
Communication
Scheme 4. Chemoselective Ligations of Unprotected GLP-1 Analogue with Equimolar Reactants
both synthetic and semisynthetic therapeutics, particularly as
some of the PEGs and other conjugates can be even more
expensive than the peptide or protein they are meant to modify.
The ideal conjugation reaction would allow the use of 1:1
stoichiometry of the two reagents, proceed rapidly under
aqueous conditions, tolerate all unprotected functional groups,
and operate without toxic reagents or byproducts. We were
pleased to find that the KAT ligation fulfills all of these criteria
and offers the further advantage of forming a stable,
biocompatible amide bond.
work on the preparation of proteins bearing the hydroxyl-
amines and KAT moieties, by both synthetic and semisynthetic
approaches, is in progress and is aided by the high stability of
both of these functional groups to aqueous conditions, HPLC
purification, and standard reagents.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures and spectroscopic data for all new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
Synthetic GLP-1 analogue 724 (1.0 equiv) was allowed to
react with KATs 6a-6f (1.0 equiv) at either 1 mM or 100 μM in
H2O (Scheme 4). In all cases, clean conversion to the desired
product was observed in <1 h, even when Mw 20 000 PEG
reagent 6c was used. KAT ligations are accelerated in aqueous
solvents and are not affected by the nature or presence of
organic cosolvents. We added DMSO as a cosolvent in the
reactions with 6d and 6e due to the low solubilities of these
KATs. Both ligations proceeded smoothly with essentially the
same rate as other cases. The presence of water, however, is
essential; under strictly anhydrous conditions the rate of the
ligation slows dramatically.
The synthesis of large molecules (Mw >5000) bearing many
types of diverse functional group is a frontier of synthetic
organic chemistry.25 The limits on size, complexity, and
stoichiometry arise from the inherently slow reaction rates or
poor starting material stability of most organic reactions
suitable for selective bond formation in the presence of
unprotected functional groups.
AUTHOR INFORMATION
Corresponding Author
bode@org.chem.ethz.ch
■
Notes
The authors declare the following competing financial
interest(s): ETH Zurich has filed a patent application for the
̈
use of this technology for PEGylation.
ACKNOWLEDGMENTS
■
G.E. was supported by a Novartis Postdoctoral Fellowship. We
thank the LOC MS Service for analyses and Safwan Aroua,
Jessada Mahatthananchai, and Aaron Dumas for helpful
discussions. This work was supported by ETH Zurich and an
̈
ETH Research Grant (ETH-43 13−2).
REFERENCES
■
The KAT ligation offers a natural-bond forming ligation fast
enough (20 M−1 s−1) for synthetic approaches to protein
engineering, biomolecule conjugates,26 functional polymers,
and the formation of oligomers of complex molecules. Further
(1) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int. Ed. 2002, 41, 2596.
(2) Finn, M. G.; Fokin, V. V. Chem. Soc. Rev. 2010, 39, 1231.
(3) Kalia, J.; Raines, R. T. Curr. Org. Chem. 2010, 14, 138.
5613
dx.doi.org/10.1021/ja5018442 | J. Am. Chem. Soc. 2014, 136, 5611−5614

Journal of the American Chemical Society
Communication
(4) For a recent review on this topic: Gorska, K.; Winssinger, N.
Angew. Chem., Int. Ed. 2013, 52, 6820.
(5) Molander, G. A.; Raushel, J.; Ellis, N. M. J. Org. Chem. 2010, 75,
(26) (a) Hemantha, H. P.; Brik, A. Bioorg. Med. Chem. 2013, 21,
3411. (b) Haj-Yahya, M.; Fauvet, B.; Herman-Bachinsky, Y.; Hejjaoui,
M.; Bavikar, S. N.; Karthikeyan, S. V.; Ciechanover, A.; Lashuel, H. A.;
Brik, A. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 17726.
4304.
(6) (a) Broyer, R. M.; Grover, G. N.; Maynard, H. D. Chem.
Commun. 2011, 47, 2212. (b) Jung, B.; Theato, P. Adv. Polym. Sci.
2013, 253, 37. (c) Kochendoerfer, G. G.; Chen, S.-Y.; Mao, F.;
Cressman, S.; Traviglia, S.; Shao, H.; Hunter, C. L.; Low, D. W.; Cagle,
E. N.; Carnevali, M.; Gueriguian, V.; Keogh, P. J.; Porter, H.; Stratton,
S. M.; Wiedeke, M. C.; Wilken, J.; Tang, J.; Levy, J. J.; Miranda, L. P.;
Crnogorac, M. M.; Kalbag, S.; Botti, P.; Schindler-Horvat, J.; Savatski,
L.; Adamson, J. W.; Kung, A.; Kent, S. B. H.; Bradburne, J. A. Science
2003, 299, 884.
(7) (a) Hutchins, B. M.; Kazane, S. A.; Staflin, K.; Forsyth, J. S.;
Felding-Habermann, B.; Smider, V. V.; Schultz, P. G. Chem. Biol. 2011,
18, 299. (b) Hudak, J. E.; Barfield, R. M.; de Hart, G. W.; Grob, P.;
Nogales, E.; Bertozzi, C. R.; Rabuka, D. Angew. Chem., Int. Ed. 2012,
51, 4161.
(8) (a) Dawson, P. E.; Kent, S. Annu. Rev. Biochem. 2000, 69, 923.
(b) Raibaut, L.; Ollivier, N.; Melnyk, O. Chem. Soc. Rev. 2012, 41,
7001. (c) Hemantha, H. P.; Narendra, N.; Sureshbabu, V. V.
Tetrahedron 2012, 68, 9491.
(9) (a) Pattabiraman, V. R.; Ogunkoya, A. O.; Bode, J. W. Angew.
Chem., Int. Ed. 2012, 51, 5114. (b) Ogunkoya, A. O.; Pattabiraman, V.
R.; Bode, J. W. Angew. Chem., Int. Ed. 2012, 51, 9693.
(10) Pusterla, I.; Bode, J. W. Angew. Chem., Int. Ed. 2012, 51, 513.
(11) (a) Bode, J. W.; Sohn, S. S. J. Am. Chem. Soc. 2007, 129, 13798.
(b) Schafer, G.; Matthey, C.; Bode, J. W. Angew. Chem., Int. Ed. 2012,
̈
51, 9173.
(12) Dumas, A. M.; Molander, G. A.; Bode, J. W. Angew. Chem., Int.
Ed. 2012, 51, 5683.
(13) Knowles, D. A.; Mathews, C. J.; Tomkinson, N. C. O. Synlett
2008, 2769.
(14) Dirksen, A.; Dawson, P. E. Bioconjugate Chem. 2008, 19, 2543.
(15) (a) Blackman, M. L.; Royzen, M.; Fox, J. M. J. Am. Chem. Soc.
2008, 130, 13518. (b) Taylor, M. T.; Blackman, M. L.; Dmitrenko, O.;
Fox, J. M. J. Am. Chem. Soc. 2011, 133, 9646.
(16) (a) Radzicka, A.; Wolfenden, R. J. Am. Chem. Soc. 1996, 118,
6105. (b) Kalia, J.; Raines, R. T. Angew. Chem., Int. Ed. 2008, 47, 7523.
Forming an amide bond makes Staudinger ligation a popular choice
for in vivo study in spite of its slow kinetics: Patterson, D. M.;
Nazarova, L. A.; Prescher, J. A. ACS Chem. Biol. 2014, 9, 592.
(17) For reviews on PEGylation on pharmaceuticals: (a) Harris, J.
M.; Chess, R. B. Nat. Rev. Drug. Discovery 2003, 2, 214. (b) Kang, J. S.;
DeLuca, P. P.; Lee, K. C. Expert Opin. Emerg. Drugs 2009, 14, 363.
(18) For a review on biotinylation: Trippier, P. C. Chem. Med. Chem.
2013, 8, 190.
(19) For a review on use of dyes in chemical biology: Gonca̧ lves, M.
S. T. Chem. Rev. 2009, 109, 190.
(20) Dumas, A. M.; Bode, J. W. Org. Lett. 2012, 14, 2138.
(21) For selected examples of site-specific PEGylation: (a) Deiters,
A.; Cropp, T. A.; Summerer, D.; Mukherji, M.; Schultz, P. G. Bioorg.
Med. Chem. Lett. 2004, 14, 5743. (b) Scheck, R. A.; Francis, M. B. ACS
Chem. Biol. 2007, 2, 247. (c) Debets, M. F.; van Berkel, S. S.;
Schoffelen, S.; Rutjes, F. P. J. T.; van Hest, J. C. M.; van Delft, F. L.
Chem. Commun. 2010, 46, 97. (d) Serwa, R.; Majkut, P.; Horstmann,
B.; Swiecicki, J.-M.; Gerrits, M.; Krause, E.; Hackenberger, C. P. R.
Chem. Sci. 2010, 1, 596. (e) Dumas, A.; Spicer, C. D.; Gao, Z.;
Takehana, T.; Lin, Y. A.; Yasukohchi, T.; Davis, B. G. Angew. Chem.,
Int. Ed. 2013, 52, 3916.
(22) Payne, R. W.; Murphy, B. M.; Manning, M. C. Pharm. Dev.
Technol. 2011, 16, 423.
(23) Drucker, D. J.; Dritselis, A.; Kirkpatrick, P. Nat. Rev. Drug
Discovery 2010, 9, 267.
(24) For the structure comparison with related peptides, see
Supporting Information.
(25) (a) Hawker, C. J. Science 2005, 309, 1200. (b) Pattabiraman, V.
R.; Bode, J. W. Nature 2011, 480, 471.
5614
dx.doi.org/10.1021/ja5018442 | J. Am. Chem. Soc. 2014, 136, 5611−5614