Peptide ligations accelerated by N-terminal aspartate and glutamate residues

Article abstract of DOI:10.1021/ol2017356

A novel application of intramolecular base catalysis confers enhanced reaction rates for aminolysis ligations between peptide thioesters and peptides bearing N-terminal aspartate or glutamate residues. The broad scope of this process and its application i

Full text of DOI:10.1021/ol2017356

ORGANIC
LETTERS
2011
Vol. 13, No. 18
4770–4773
Peptide Ligations Accelerated by
N-Terminal Aspartate and
Glutamate Residues
Gemma L. Thomas, Yves S. Y. Hsieh, Candy K. Y. Chun, Zheng-Li Cai,
Jeffrey R. Reimers, and Richard J. Payne*
School of Chemistry, The University of Sydney, NSW 2006, Australia
richard.payne@sydney.edu.au
Received June 28, 2011
ABSTRACT
A novel application of intramolecular base catalysis confers enhanced reaction rates for aminolysis ligations between peptide thioesters and
peptides bearing N-terminal aspartate or glutamate residues. The broad scope of this process and its application in the total synthesis of the
diabetes drug exenatide is demonstrated.
Ligationchemistryrepresentsanextremelyvaluabletool
for the convergent assembly of peptide fragments to
provide polypeptides and proteins for biological study.1
The most efficient and widely used ligation method, in-
troduced by Kent and co-workers, is native chemical
ligation.1a,2 This method facilitates the chemoselective
condensation of a peptide bearing an N-terminal cysteine
residue and a peptide containing a C-terminal thioester
moiety to form a native amide bond. The reaction mechan-
ism first involves a transthioesterification step to generate
an intermediate thioester which subsequently rearranges
via an SfN acyl shift.1a,2,3 To date, the method has
been successfully implemented in the preparation of
hundreds of complex protein targets, including those with
post-translational modifications.1a,e The recent use of
thiol-derived proteinogenic amino acids for ligationÀ
desulfurization applications has further expanded the
repertoire of this powerful technology.1k,4
Thiol-free peptide ligation methods have also been
developed5 and exploited in the synthesis of peptides and
proteins.5c,e,6 These serve as complementary tools to native
chemical ligation and have found utility in the synthesis of
several large peptide targets. One such method, which utilizes
(4) (a) Crich, D.; Banerjee, A. J. Am. Chem. Soc. 2007, 129, 10064.
(b) Haase, C.; Rohde, H.; Seitz, O. Angew. Chem., Int. Ed. 2008, 47, 6807.
(c) Chen, J.; Wan, Q.; Yuan, Y.; Zhu, J.; Danishefsky, S. J. Angew. Chem.,
Int. Ed. 2008, 47, 8521. (d) Yang, R. L.; Pasunooti, K. K.; Li, F. P.; Liu,
X. W.; Liu, C. F. J. Am. Chem. Soc. 2009, 131, 13592. (e) Harpaz, Z.;
Siman, P.; Kumar, K. S. A.; Brik, A. ChemBioChem 2010, 11, 1232. (f)
Chen, J.; Wang, P.; Zhu, J. L.; Wan, Q.; Danishefsky, S. J. Tetrahedron
2010, 66, 2277. (g) Rohde, H.; Seitz, O. Biopolymers 2010, 94, 551.
(5) (a) Blake, J. Int. J. Pept. Prot. Res. 1981, 17, 273. (b) Aimoto, S.
Biopolymers 1999, 51, 247. (c) Hojo, H.; Matsumoto, Y.; Nakahara, Y.;
Ito, E.; Suzuki, Y.; Suzuki, M.; Suzuki, A. J. Am. Chem. Soc. 2005, 127,
13720. (d) Chen, G.; Wan, Q.; Tan, Z. P.; Kan, C.; Hua, Z. H.;
Ranganathan, K.; Danishefsky, S. J. Angew. Chem., Int. Ed. 2007, 46,
7383. (e) Payne, R. J.; Ficht, S.; Greenberg, W. A.; Wong, C.-H. Angew.
Chem., Int. Ed. 2008, 47, 4411.
(1) (a) Dawson, P. E.; Kent, S. B. H. Annu. Rev. Biochem. 2000, 69,
92. (b) Davis, B. G. Chem. Rev. 2002, 102, 579. (c) Macmillan, D. Angew
Chem., Int. Ed. 2006, 45, 7668. (d) Hackenberger, C. P. R.; Schwarzer, D.
Angew. Chem., Int. Ed. 2008, 47, 10030. (e) Dirksen, A.; Dawson, P. E.
Curr. Opin. Chem. Biol. 2008, 12, 760. (f) Kent, S. B. H. Chem. Soc. Rev.
2009, 38, 338. (g) Kan, C.; Danishefsky, S. J. Tetrahedron 2009, 65, 9047.
(h) Payne, R. J.; Wong, C.-H. Chem. Commun. 2010, 46, 21. (i)
Tiefenbrunn, T. K.; Dawson, P. E. Biopolymers 2010, 94, 95. (j) Yuan,
Y.; Chen, J.; Wan, Q. A.; Wilson, R. M.; Danishefsky, S. J. Biopolymers
2010, 94, 373. (k) Kumar, K. S. A.; Brik, A. J. Pept. Sci. 2010, 16, 524.
(2) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H.
Science 1994, 266, 776.
(6) (a) Yuan, Y.; Chen, J.; Wan, Q.; Tan, Z. P.; Chen, G.; Kan, C.;
Danishefsky, S. J. J. Am. Chem. Soc. 2009, 131, 5432. (b) Tan, Z. P.;
Shang, S. Y.; Halkina, T.; Yuan, Y.; Danishefsky, S. J. J. Am. Chem.
Soc. 2009, 131, 5424.
(3) Hackeng, T. M.; Griffin, J. H.; Dawson, P. E. Proc. Natl. Acad.
Sci. U.S.A. 1999, 96, 10068.
r
10.1021/ol2017356
Published on Web 08/10/2011
2011 American Chemical Society

the inherent nucleophilicity of N-terminal amines for the
generation of peptide bonds via reaction with C-terminal
peptide thioesters as N-acyl donors, is direct aminolysis
ligation.5e A significant benefit of this approach, compared
with other thiol-free methods, is that reactions proceed
without epimerization of the C-terminal amino acid of the
peptide thioester.5aÀd However, as these reactions rely on
intermolecular formation of an amide bond, they tend to be
significantly slower than native chemical ligation and are not
chemoselective in the presence of the ε-amino side chain of
lysine.5e Nonetheless, such ligation reactions are high yielding
with a number of amino acids at the ligation junction, and
side chain protection of lysine ensures regioselectivity.5d,e,6
Herein, we report that the rates of aminolysis ligations
between peptide thioesters and peptides bearing N-terminal
aspartate (Asp) or glutamate (Glu) residues are enhanced as
a result of intramolecular base catalysis facilitated by the
side chain carboxylate moiety. The synthetic utility of this
reaction is explored, together with its application in the total
synthesis of the type II diabetes drug exenatide (Byetta).
Initial studies focused on the reaction of model peptides
1À4 with peptide thioester 5 (Scheme 1) under mixed
solvent/buffer conditions [4:1 v/v N-methylpyrrolidinone
Scheme 1. (A) Aminolysis Ligation Reactions between Peptides
1À4 and Peptide Thioester 5. (B) Kinetics of the Direct Aminolysis
Peptide Ligation between Peptide Thioester 5 and Peptides 1À4
(R2 = (CH2)2CO2Et)a
a Key: blue diamond = peptide 1, maroon circle = peptide 2, red
triangle = peptide 3, green square = peptide 4.
(NMP):1.25 M Gn HCl, 200 mM HEPES, 2 vol % PhSH,
3
pH 7.5].5e After 72 h, the reaction mixtures were purified by
HPLC to afford ligation products 6À9. Excellent yields of 6
(quant) and 7 (92%) were obtained when peptides 1 and 2,
bearing N-terminal Asp and Glu residues, respectively, were
reacted with 5. In contrast, reaction of 5 with 3 and 4
(containing sterically similar N-terminal Asp- and Glu-
methyl esters, respectively) provided the corresponding liga-
tion products 8and 9in poor yields (17À37% isolated yield).
In addition, reaction of 5 with peptides containing sterically
equivalent N-terminal Asn and Gln residues provided poor
yields of the desired product, together with a cyclic imide
byproduct of the peptide fragment.
moieties)3,7 and (2) generation of an unsymmetrical
anhydride followed by an intramolecular OfN acyl
shift.8
Scheme 2A shows activation free energies ΔGq and
reaction free energies ΔG calculated for putative path-
ways for the ligation reaction between model peptide 10
and peptide thioester 11. Calculations were performed
Scheme 2. (A) Calculated Reaction and Activation Energies
(kcal molÀ1).a (B) Transition-State Structures TSIIa (Bond
˚
Lengths in A)
Kinetic studies of ligation reactions between 5 and pep-
tides 1À4 were performed to rationalize the observed
difference in reaction efficiencies (Scheme 1). The data
was fitted to a unimolecular rate equation (see the Support-
ing Information). A greater than 9-fold increase in the
reaction rate was observed when 5 was reacted with Asp-
or Glu-containing peptides 1or 2(t1/2:1=13.1h, 2=9.5h)
compared to the analogous N-terminal Asp- and Glu-
methyl ester-containing peptides 3 and 4 (t1/2: 3 = 190 h,
4 = 85 h). The rate enhancement observed for N-terminal
Asp- and Glu-containing peptides therefore appears to be a
direct result of the carboxylate side chain.
Two likely mechanistic pathways are postulated for
this rate enhancement, both of which invoke the side
chain carboxylate: (1) intramolecular base catalysis
(previously proposed for the acceleration of acyl
shift-based transformations with peptides bearing N-
terminal or neighboring Asp, Glu, His, or carboxylate
(7) (a) Kemp, D. S.; Carey, R. I. J. Org. Chem. 1993, 58, 2216.
(b) Brik, A.; Ficht, S.; Yang, Y.-Y.; Bennett, C. S.; Wong, C.-H. J. Am.
Chem. Soc. 2006, 128, 15026. (c) Lutsky, M.-Y.; Nepomniaschiy, N.;
Brik, A. Chem. Commun. 2008, 1229. (d) Dheur, J.; Ollivier, N.; Melnyk,
O. Org. Lett. 2011, 13, 1560.
a CH3CH2SH, pKa = 10.61; PhSH, pKa = 6.5.
Org. Lett., Vol. 13, No. 18, 2011
4771

using density functional theory (DFT) by GAUSSIAN-
099 using the B3LYP10 density functional, the auc-cc-
pVDZ basis set,11 explicit solvation, and the PCM
implicit aqueous solvation model12 with solution-phase
thermodynamic corrections (see the Supporting Infor-
mation for additional reactions and all optimized
structures).13
For R = COOCH3 the transition state is early (CN amide
bond unformed) and not ionic (TSIIa, see Scheme 2B and
the Supporting Information).
Experimentally, the observed rate difference im-
plies a less pronounced difference in transition state
energies (ca.1.3 kcal molÀ1) than that calculated
(12 kcal molÀ1) for pathway IIa. Alternative calcula-
tions, involving simultaneous attack of OHÀ from
the solvent on the NH2 hydrogens at the transition
state, parametrized for pH 7.5 (IIb, Scheme 2A, see
the Supporting Information for TS structures) reveal
this mechanism to be significantly more favorable
(ΔGqIIb = 25 kcal molÀ1) for R = COOCH3, consis-
tent with the observed rate differences. In the case of
R = COOÀ, additional OHÀ catalysis provides a less
favorable pathway than IIa (IIa: ΔGqIIa = 24 kcal molÀ1
The first reaction stage for aminolysis ligations
(reaction I, Scheme 2A) is thiolate exchange from alkyl
thioester 11 to phenylthio ester 12.14 The thiolate ex-
change is calculated to proceed at pH 7.5 via nucleo-
philic attack of thiophenolate on the carbonyl over a
transition state at ΔGq = 26 kcal molÀ1, followed by
I
protonation of the displaced alkyl thiolate, in agree-
ment with recent calculations for native chemical
ligation.15 The importance of this step was verified
experimentally by conducting ligation reactions of 1
and 2 with 5 in the absence of thiophenol which showed
dramatically slower reaction rates (see the Supporting
Information).
vs IIb: ΔGq = 27 kcal molÀ1).
IIb
Taken together, these calculations suggest that peptides
bearing R = COOÀ are more likely to react via pathway
IIa, with the observed rate acceleration facilitated via
intramolecular base catalysis by the free side chain carbox-
ylate. In the absence of a carboxylate-containing side
chain, reactions are subject to a less effective intermolecu-
lar base catalysis (OHÀ) process IIb.
Three possible mechanisms were considered for
the second-stage amide bond formation (reaction II,
Scheme 2A) between peptide 10 and peptide thioester 12
to afford 13; namely IIa-direct aminolysis, IIb-direct ami-
nolysis facilitated by OHÀ interacting with the transition
state,16 and IIc-reaction via an anhydride intermediate.
These were considered (where appropriate) for reaction
with both an N-terminal aspartate residue (R = COOÀ)
and the corresponding methyl ester (R = COOCH3). Given
Given the enhanced reaction rates of aminolysis
ligations at N-terminal Asp and Glu residues observed
above, we next moved to explore the synthetic utility of
the transformation for reactions at thioesters bearing a
range of C-terminal residues. To this end, peptides 1
and 2 were reacted with peptide thioesters (5 and
15À21) bearing a representative range of C-terminal
amino acids (Scheme 3). Ligation of 1 and 2 to thioe-
sters with C-terminal glycine and alanine residues
provided the desired ligation products in excellent
yields (86% to quant, entries 1À4, Scheme 3). Ligation
reactions to a peptide thioester bearing a C-terminal
methionine residue also proceeded smoothly to afford
the desired products in 70À74% isolated yields (entries
5 and 6). Reaction of these peptides with thioesters
possessing C-terminal aromatic residues (Tyr and Phe)
provided the desired ligation products in good yields
(65À84%, entries 7À10). Unfortunately, reaction of 1
and 2 with a peptide thioester containing a C-terminal
Asn residue led to a diminished yield of the desired
product (entries 11 and 12). This was due to the
formation of a five-membered cyclic imide byproduct,
resulting from facile intramolecular attack of the re-
active C-terminal thioester by the primary amide side
chain of Asn under the ligation conditions. This was
not observed when the same peptides were reacted with
a peptide thioester bearing a C-terminal glutamine
residue which provided excellent yields of the desired
ligation product (70% to quant, entries 13 and 14).
Finally, ligation of 1 and 2 to 21 bearing a sterically
encumbered valine residue also provided the desired
ligation products in synthetically useful yields (55À58%,
entries 15 and 16). It should be noted that no epimerization
was detected in the products of any of the reactions
studied here.
the high activation barrier (ΔGq = 30 kcal molÀ1) and
IIc
endothermic nature (ΔGIIc = 11 kcal molÀ1) of anhydride
14 formation in IIc, this is not considered to be a likely
pathway. Furthermore, the operation of such a mechanism
would generate a cyclic succinimide (and adducts thereof) in
reactions of peptides bearing N-terminal Glu residues;
however, these byproducts were not observed.
In the case of the direct aminolysis process IIa, a
significantly faster reactionrateiscalculatedforR=COOÀ
compared to R = COOCH3 (ΔGqIIa = 24 kcal molÀ1 vs
36 kcal molÀ1).17 A typical transition state (TSIIa for
R = COOÀ) for reaction pathway IIa is shown in
Scheme 2B and is late: it is subsequent to the formation
of the new CN bond and ionic scission of the CS bond
and involves attack by the resulting thiolate on a proton of
the produced secondary amide cation. This process is
assisted by intramolecular base catalysis by interaction of
the side chain carboxylate with the other amide proton.
(8) Thomas, G. L.; Payne, R. J. Chem. Commun. 2009, 4260.
(9) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B. et al. Gaussian 09,
Revision A.02; Gaussian, Inc., Pittsburgh, PA, 2009.
(10) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(11) Kendall, R. A.; Dunning, T. H., Jr.; Harrison, R. J. J. Chem.
Phys. 1992, 96, 6796.
(12) Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027.
(13) Ho, J.; Klamt, A.; Coote, M. L. J. Phys. Chem.A 2010, 114,
13442.
(14) Hupe, D. J.; Jencks, W. P. J. Am. Chem. Soc. 1977, 99, 451.
(15) Wang, C.; Guo, Q.-X.; Fu, Y. Chem. Asian J. 2011, 6, 1241.
(16) Garel, J.; Tawfik, D. S. Chem.;Eur. J. 2006, 12, 4144.
(17) Buildup of the phenylthio ester is observed experimentally,
suggesting that ΔGqIIa is actually g Gq .
I
4772
Org. Lett., Vol. 13, No. 18, 2011

Scheme 3. Scope of Direct Aminolysis Ligation Reactions
between Peptides Containing N-Terminal Asp (1) and Glu (2)
Residues and Peptide Thioesters (5 and 15À21)a
Scheme 4. Total Synthesis of Exenatide (Byetta) via Glu-
Assisted Aminolysis Ligation Reaction
with methionine thioester 26,19b acidolytic cleavage from
the resin, and purification by HPLC afforded 24 in 20%
overall yield over 23 linear steps. With the requisite frag-
ments 23 and 24 in hand, ligation was performed under
our intramolecular base-catalyzed conditions. The reaction
progress was monitored by HPLCÀMS and, after 96 h, was
treated with piperidine followed by methanolic hydrazine to
remove the Fmoc carbamate and two side-chain trifluor-
oacetamide moieties. Gratifyingly, purification by HPLC
afforded exenatide 22 in a succinct three steps with an
impressive 57% overall yield.
In summary, we have demonstrated a significant rate
enhancement when peptide thioesters and peptides bearing
N-terminal Asp or Glu residues are reacted under amino-
lysis ligation conditions. This rate enhancement is likely
facilitated by intramolecular base catalysis by the side
chain carboxylate of Asp and Glu stabilizing the transition
state. The transformation was shown to be synthetically
useful with a range of peptide thioester coupling partners
and the utility of the methodology was further exemplified
in the efficient total synthesis of the 39 amino acid type II
diabetes drug exenatide. It is anticipated that these ami-
nolysis ligation reactions will find future application in the
assembly of a range of peptide and proteins, including
other peptide drugs.
a R = À(CH2)2CO2Et; (a) = 72 h, (b) = 96 h, (c) = 120 h reaction
time.
Having demonstrated the scope of the aminolysis ligation
reactions, we next sought to exploit the technology in a
ligation-based synthesis of the peptide therapeutic exenatide
22 (marketed as Byetta), a type II diabetes drug (Scheme 4).
Exenatide is a synthetic version of the 39 amino acid peptide
exendin-4, a structural analogue of the incretin hormone
glucagon-like peptide-1, found in the venom of the Gila
Monster desert lizard (Heloderma suspectum).18 We envi-
saged the synthesis of exenatide via disconnection between
Met14 and Glu15, thus enabling aminolysis between pep-
tide fragment 23 and peptide thioester 24. Fragment 23 was
synthesized by Fmoc-strategy SPPS from Rink amide resin
(Scheme 4, see the Supporting Information). Incorporation
of a trifluoroacetamide-protected Lys27 residue was neces-
sary to prevent unwanted aminolysis on the ε-amino group
during the ligation reaction. Peptide thioester 24, bearing
N-terminal Fmoc-protection and trifluoroacetamide side-
chain protection of Lys12, was synthesized via the use of a
side chain anchoring strategy (Scheme 4).19 Briefly, Fmoc-
Glu-OAll was immobilized onto Rink amide resin to pro-
vide 25 and the peptide elongated via standard Fmoc-
strategy SPPS (see the Supporting Information). At this
point, removal of the C-terminal allyl ester, thioesterification
Acknowledgment. We acknowledge the Australian Re-
search Council (DP0986632 and DP110102932) and an
Endeavour Research Fellowship (G.L.T.) for funding.
Note Added after ASAP Publication. The corrected
version of this Letter was reposted August 12, 2011.
(18) (a) Eng, J.; Kleinman, W. A.; Singh, L.; Singh, G.; Raufman,
J. P. J. Biol. Chem. 1992, 267, 7402. (b) Furman, B. L. Toxicon 2011,
DOI: 10.1016/j.toxicon.2010.12.016.
(19) (a) Wang, P.; Miranda, L. P. Int. J. Pept. Res. Ther. 2005, 11,
117. (b) Ficht, S.; Payne, R. J.; Guy, R. T.; Wong, C.-H. Chem.;Eur. J.
2008, 14, 3620. (c) Yang, Y. Y.; Ficht, S.; Brik, A.; Wong, C.-H. J. Am.
Chem. Soc. 2007, 129, 7690. (d) Ajish Kumar, K. S.; Harpaz, Z.;
Haj-Yahya, M.; Brik, A. Bioorg. Med. Chem. Lett. 2009, 19, 3870.
(e) Thompson, R. E.; Jolliffe, K. A.; Payne, R. J. Org. Lett. 2011, 13, 680.
Supporting Information Available. Experimental pro-
cedures and characterization of peptide products, com-
putational procedures, and optimized structures. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Org. Lett., Vol. 13, No. 18, 2011
4773