Glutamic Acid Selective Chemical Cleavage of Peptide Bonds

Article abstract of DOI:10.1021/acs.orglett.6b00317

Site-specific hydrolysis of peptide bonds at glutamic acid under neutral aqueous conditions is reported. The method relies on the activation of the backbone amide chain at glutamic acid by the formation of a pyroglutamyl (pGlu) imide moiety. This activation increases the susceptibility of a peptide bond toward hydrolysis. The method is highly specific and demonstrates broad substrate scope including cleavage of various bioactive peptides with unnatural amino acid residues, which are unsuitable substrates for enzymatic hydrolysis.

Full text of DOI:10.1021/acs.orglett.6b00317

Letter
pubs.acs.org/OrgLett
Glutamic Acid Selective Chemical Cleavage of Peptide Bonds
Joseph M. Nalbone,† Neelam Lahankar,† Lyssa Buissereth, and Monika Raj*
Department of Chemistry, Seton Hall University, 400 South Orange Avenue, South Orange, New Jersey 07079, United States
S
* Supporting Information
ABSTRACT: Site-specific hydrolysis of peptide bonds at glutamic acid under
neutral aqueous conditions is reported. The method relies on the activation of the
backbone amide chain at glutamic acid by the formation of a pyroglutamyl (pGlu)
imide moiety. This activation increases the susceptibility of a peptide bond toward
hydrolysis. The method is highly specific and demonstrates broad substrate scope
including cleavage of various bioactive peptides with unnatural amino acid residues,
which are unsuitable substrates for enzymatic hydrolysis.
Scheme 1. Rationale for the Site-Specific Hydrolysis at
Glutamic Acid through Backbone Amide Activation
ydrolysis of peptide bonds at a specific residue is an
Himportant biochemical tool for correlating protein
structure with activity,1 designing new therapeutic agents,2,3
converting proteins into fragments that are more amenable for
sequencing, and various new bioanalytical and bioengineering
applications.4 These new methods require residue-specific
cleavage of a peptide bond into large fragments. The peptide
bond (i.e., the amide group), however, is highly stable toward
hydrolysis, and the half-life for nonselective cleavage at room
temperature and pH 4−8 is 500−1000 years.5 Few enzymes and
synthetic reagents are commonly used for selective hydrolysis of
peptide bonds.6 Most of these peptidases are residue-selective,
such as trypsin, which is selective for cleavage at Arg and Lys;
chymotrypsin cleaves at Phe, Trp, and Tyr; pepsin at Phe and
Leu; endopeptidase Glu-C at Glu; and endopeptidase Lys-C at
Lys. The selectivity of peptidases can be adjusted by varying the
digestion time and the degree of prior unfolding. Nonetheless,
peptidases are limited because they tend to produce short
fragments ill-suited for bioanalytical applications, and they
require tedious procedures of peptidase removal from the
protein digest.6 Moreover, these enzymes require narrow ranges
of temperature and pH.
under milder reaction conditions. Previous attempts utilized
thionyl chloride for selective cleavage at Glu but were
unsuccessful in achieving the desired goals.19
The existing chemical reagents, such as cyanogen bromide,7 2-
nitro-5-thiocyanobenzoic acid,8 and 2-iodosylbenzoic acid,9 for
cleaving peptide bonds often require harsh conditions. As a
result, side reactions and the lack of specificity of amide-bond
hydrolysis limits their scope in chemical biology and synthetic
applications.10 Recently, asparagine-selective cleavage of peptide
bonds by using diacetoxyiodobenzene has been reported.11 The
use of metals for cleavage of peptide bonds has been intensively
studied, but their practical use for protein analysis is still in its
early stages.12−18 New chemical reagents with high selectivity
and improved efficiency are highly desirable for many emerging
applications. For controlled and specific cleavage, a daunting
task, a chemical reagent must selectively bind only to one
particular amino acid in the peptide sequence and specifically
cleave the peptide bond at the binding site. Based on this
principle, we report a chemical methodology for site-selective
cleavage of peptide bonds at glutamic acid with high efficiency
The selective cleavage of a peptide bond at glutamic acid
entails the strong activation of the side-chain carboxylate of Glu
to generate intermediate (A) (Scheme 1). The nucleophilic
attack of the amide nitrogen on the activated intermediate leads
to the formation of the backbone pyroglutamyl (pGlu) imide
moiety (B).20 The formation of pGlu imide moiety (B) makes
the C−N bond prone to hydrolysis and leads to the cleavage of
the peptide chain into the N-terminal fragment (C) and pGlu
imide-containing C-terminal fragment (D) (Scheme 1).
To implement this methodology of site-selective cleavage of
peptides at glutamic acid, model hexapeptide Fmoc-Val-Ala-Glu-
Arg-Phe-Ala-NH2 (1a) was synthesized. For the formation of the
Received: January 30, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b00317
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Optimization of the Reaction Conditions on Peptide
a
1a for the Formation of pGlu Imide Moiety 2a
b
entry reagent (equiv) base (equiv)
additive time (h) conv (%)
1
2
3
4
5
DSC (40)
DIEA (40)
DIEA (20)
DIEA (20)
DIEA (40)
DIEA (20)
48
24
48
48
24
50
20
40
60
99
PyBrop (20)
PyBrop (20)
PyBrop (40)
PyBrop (20)
c
DMAP
a
Reaction conditions: peptide (1 equiv) was reacted with DSC/
PyBrop (20−40 equiv) and DIEA (20−40 equiv) in DMF at room
b
temperature. Conversion to 2a was calculated from the absorbance at
c
254 nm using HPLC. A small crystal of DMAP was added to the
reaction mixture. The entry in bold represents the optimized reaction
conditions. DIEA = N,N-diisopropylethylamine, DMAP = 4-(N,N-
dimethylamino)pyridine.
Figure 1. HPLC chromatogram of pGlu imide formation using
optimized reaction conditions in Table 1, entry 5, at t = 0 h (top), at t
= 17 h (middle), after hydrolysis under 0.1 M phosphate buffer (pH 7.5)
at 25 °C, and at t = 48 h (bottom). Insets show the MS corresponding to
retention times at 13.4 min (1a, top), 13.7 min (2a, middle), 22.7 min
(3a, bottom), and 5.6 min (4, bottom).
pGlu imide moiety, various acylation reagents were screened
(Table 1). In principle, such activation of the Glu side-chain
carboxylate could lead to not only the desired 5-membered ring
(B) through path a but also a six-membered ring (B′) (path b,
Scheme 1).20 However, our studies suggest high specificity for
the formation of the five-membered pGlu imide moiety (B),
which is supported by NMR analysis (Supporting Information).
Importantly, cleavage of peptides at glutamic acid (see below)
provides the chemical proof for the formation of the five-
membered ring (B, Scheme 1).
pGlu imide moiety containing C-terminal fragment 4 = 5.6 min
(Figure 1c) (Figure S2, Supporting Information).
To determine the effect of side-chain functionality of the
amino acid adjacent to Glu in the activation of the backbone
chain and hydrolysis, various peptides, Fmoc-Val-X-Glu-Arg-
Phe-Ala-NH2 (1a−j), with different amino acids at the X position
were explored (Table 2). As a result, fragment 4 was obtained in
To optimize the cyclization at glutamic acid, various reaction
conditions were explored on model hexapeptide 1a (Table 1). In
initial studies, N,N′-disuccinimidyl carbonate (DSC) was used
for the formation of the pGlu imide moiety, but a large excess of
reagents and longer reaction times were needed for the 50%
conversion to the desired cyclized peptide 2a (entry 1, Table 1).
The yields for the conversion from 1a to 2a were calculated from
HPLC data (Table 1). Next, bromotris(pyrrolidino)phosphon-
ium hexafluorophosphate (PyBrOP) was used for the activation
since it is one of the strongest coupling reagents and provides the
corresponding acyl bromide. Initial reaction with uncrystallized
PyBrop gave pyrrolide 2a′ as a side product with a mass 53 Da
higher than the starting peptide 1a (for details, see Figure S1,
Supporting Information).21−23 With recrystallized PyBrOP in
hand, various other reaction conditions, such as base, additive,
and time, were screened on a model hexapeptide, Fmoc-Val-Ala-
Glu-Arg-Phe-Ala-NH2 (1a) (entries 2−5, Table 1). The progress
of the reaction was monitored by injecting the sample into an
analytical HPLC after regular intervals of time (Figure 1). At time
= 0 h, a sharp peak with retention time tR = 13.4 min
corresponding to the model hexapeptide 1a was observed, as
analyzed by mass spectrometer (MS) (Figure 1a). After 17 h, one
sharp peak with a retention time tR = 13.7 min was observed,
indicating the formation of the pGlu imide moiety 2a, as analyzed
by MS (Figure 1b). After the modified peptide was incubated in
phosphate buffer (pH 7.5) for 48 h, the sharp peak 2a at 13.7 min
disappeared and two new peaks appeared with retention times of
5.6 and 22.7 min (Figure 1c). MS analysis of these peaks
corresponded to the cleavage products at the N-terminal side of
glutamic acid, with N-terminal fragment 3a = 22.7 min, and the
Table 2. Glu-Selective Amide Bond Cleavage of Fmoc-Val-X-
a
Glu-Arg-Phe-Ala-NH2 (1a−j)
b
entry
substrate
X
conv (%)
1
1a
1b
1c
1d
1e
1f
Ala
Gly
Arg
Met
Asn
His
Phe
Tyr
Ile
99
99
99
80
90
90
90
90
65
99
2
3
4
5
6
7
1g
c
8
1h
c
9
1i
10
1j
Asp
a
Reaction conditions: peptide (1a−j, 1 equiv) was reacted with
PyBrop, DIEA (20 equiv), and a crystal of DMAP in DMF at room
temperature followed by hydrolysis with 0.1 M phosphate buffer (pH
b
7.5) at 25 °C for 48 h. Conversion to N-terminal fragment, Fmoc-V-
X-OH (3a−j), was calculated from the absorbance at 254 nm using
c
HPLC. Hydrolysis for 5 days.
B
DOI: 10.1021/acs.orglett.6b00317
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
all cases after hydrolysis, with the HPLC conversion ranging
from 65 to 99%. The results indicated that unprotected peptides
with X = Gly (1b), Arg (1c), Met (1d), Asn (1e), His (1f), and
Phe (1g) proceeded cleanly in a manner similar to that of 1a
(entries 2−7, Table 2). In contrast, substrates that contain X =
Tyr (1h) and Ile (1i) with bulky side groups gave the cleaved
products in good yields but required longer time (5 days) for
cleavage (entries 8 and 9, Table 2).
In the case of peptide 1j, containing Asp along with Glu,
cleavage of the backbone peptide bond was observed only at the
Glu residue (entry 10, Table 2). The contrasting reaction
between Asp and Glu is due to the unlikely formation of the
constrained 4-membered ring 5 at Asp, compared to the
kinetically favorable 5-membered ring 2j at Glu (Scheme 2 and
Figure S3, Supporting Information). Therefore, this peptide
bond cleavage method is highly Glu selective.
amino acid residues. Peptides 9 and 10, which are made up of D-
amino acids and a mixture of L- and D-amino acids, respectively,
were cleaved successfully under the reaction conditions at Glu
with ease and high yields (entries 3 and 4, Table 3). This
chemical cleavage of unnatural D-amino acid residues containing
peptides is a huge advantage over the conventional enzymatic
method, where enzymes do not recognize and cleave these
modified peptides. The conversion of natural L-amino acids to
unnatural D-amino acids is a well-known mutation responsible
for various age related disorders such as cataracts and Alzheimer’s
disease.24 Thus, this method can be used as a diagnostic tool to
determine different types of mutations in proteins and their role
in the progression of diseases.
Peptide 11, comprising an intramolecular disulfide bridge,
afforded the cleavage product at glutamic acid with an intact
disulfide bond (entry 5, Table 3). Thus, this methodology can be
used to determine the position of disulfide pairing in a peptide
chain, which is in contrast to other chemical reagents.25,26
Peptide 12, containing a serine residue with a reactive
hydroxymethyl group at the side chain, remained unreacted
under the reaction conditions, and cleavage was observed only at
Glu (entry 6, Table 3). This and Table 2 demonstrated the high
specificity of this methodology toward glutamic acid. Next, this
methodology was successfully applied for the scission of three
bioactive peptides: 13, a putative coproporphyrinogen III
oxidase fragment, 16, and Aβ (10−19)11 19, a fragment of
Alzheimer’s disease associated amyloid-β peptide (entries 1−3,
Table 4, and Figure S4).
Scheme 2. Reactivity of Asp vs Glu toward the Backbone
Activation for Peptide 1j
Table 4. Glu-Selective Cleavage of Bioactive Peptides
a
yield
entry
substrate
(%)
1
2
3
Fmoc-Met-Gly-His-Gln-Glu-His-Leu-Pro-Tyr- NH2 (13)
Fmoc-Leu-Pro-Arg-Leu-Gln-Glu-Ala-Trp-Gln- NH2 (16)
79
75
80
To determine the substrate scope of this method, it was further
evaluated (Table 3). The reaction was applied for the hydrolysis
Fmoc-Tyr-Glu-Val-His-His-Gln-Lys-Leu-Val-Phe-NH2
(19)
4
Fmoc-Ala-Gly-Leu-Pro-Glu-Lys-Tyr-NH2 (22)
82
Table 3. Substrate Scope of Glu-Selective Amide Bond
Hydrolysis
a
Yield of N-terminal fragment with Fmoc group.
a
b
yield
Finally, this methodology was evaluated on a bioactive peptide,
amyloid A protein fragment (Homo sapiens) 22, with a proline
residue next to Glu (entry 4, Table 4). Treatment of the peptide
22 with Pybrop for 17 h followed by hydrolysis under neutral
aqueous buffer conditions cleaved the peptide at the Pro-Glu site
(eq 4, Figure S4). This is in contrast to hydrolysis by proteases
since the location of proline at a neighboring position nearly
blocks the cleavage completely independent of the amino acid
residue.20,18 Thus, this method can potentially be used to
selectively cleave a broad range of peptides/proteins at glutamic
acid independent of the surrounding amino acid residues.
Interestingly, peptides 19 and 22 with a free side-chain lysine
only generated the kinetically favorable five-membered pGlu
moiety (for details, see Figure S5, Supporting Information). High
specificity, broad substrate scope, and easy purification
demonstrate the widespread use of this methodology and its
ability to determine the structure of unknown peptides/proteins.
Site-selective hydrolysis of unreactive peptide bonds under
mild and metal-free reaction conditions has been developed. The
methodology utilizes the activation of a backbone amide chain to
cleave the peptide bond specifically at glutamic acid. The
chemical reagents can be easily removed after the cleavage, unlike
proteases. Disulfide bonds are stable toward the reaction
entry
substrate
(%)
c
1
Fmoc-Ala-Val-Arg-Glu-Val-Ala-Phe-Glu-Arg-Phe-Gly-Phe-
NH2 (7)
80
c
2
Fmoc-Arg-Ala-Gly-Ala-Glu-Val-Arg-Phe-Ala-Glu-Ala-Phe-
Gly-NH2 (8)
85
3
4
5
Fmoc-D-Val-D-Ala-D-Glu-D-Arg-D-Phe-D-Ala- NH2 (9)
Fmoc-D-Val-Ala-Glu-D-Arg-D-Phe-Ala-NH2 (10)
80
80
75
Fmoc-Cys-Gly-Arg-Arg-Ala-Cys-Gly-Glu-Phe-Ala-Gly-NH2,
disulfide bond (11)
6
Fmoc-Arg-Ala-Glu-Ala-Gly-Ser-Gly-Phe-NH2 (12)
90
a
Reaction conditions: peptide (1 equiv) was reacted with PyBrop (20
equiv), DIEA (20 equiv), and a crystal of DMAP in DMF followed by
cleavage with 0.1 M phosphate buffer (pH 7.5) at 25 °C for 48 h
b
unless otherwise noted. Yield of N-terminal fragment with Fmoc
group. Three fragments were detected in the HPLC trace for cleavage
at both Glu residues.
c
of longer 12-mer and 13-mer peptides (7 and 8) with two
glutamic acid residues at internal positions. Peptides 7 and 8
underwent effortless cleavage at both glutamic acids and
delivered three fragments in high yield (80−85%) (entries 1
and 2, Table 3, and Supporting Information). Next, we extended
the current conditions to the scission of peptides with unnatural
C
DOI: 10.1021/acs.orglett.6b00317
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(13) Parac, T. N.; Ullmann, G. M.; Kostic,
1999, 121, 3127.
(14) Zhu, L.; Qin, L.; Parac, T. N.; Kostic, N. M. J. Am. Chem. Soc.
́
N. M. J. Am. Chem. Soc.
conditions; thus, this methodology can be used to determine the
position of disulfide pairing in peptides. This method exhibits
broad substrate scope, including the cleavage of peptides at the
Pro-Glu site, as such kind of site is resistant to enzymatic
degradation. Since the hydrolysis of mutated peptides with
unnatural amino acid residues such as D-amino acid, which are
unsuitable substrates for enzymes, proceeded with ease under the
reaction conditions, this methodology can be used to determine
the mutations responsible for various diseases. This technology is
highly specific for hydrolysis of peptides at one particular residue,
which is one of the key requirements for semisynthesis and
bioengineering of fusion proteins. These studies lay the
groundwork for further studies aimed at developing artificial
chemical proteases for the cleavage of target proteins responsible
for various diseases and exploring this reaction for biotechno-
logical applications. Work in this direction is currently underway
in our laboratory.
1994, 116, 5218.
(15) Hoyer, D.; Cho, H.; Schultz, P. G. J. Am. Chem. Soc. 1990, 112,
3249.
(16) Schepartz, A.; Cuenoud, B. J. Am. Chem. Soc. 1990, 112, 3247.
(17) Kita, Y.; Nishii, Y.; Higuchi, T.; Mashima, K. Angew. Chem., Int. Ed.
2012, 51, 5723.
́ ́
(18) Milovic, N. M.; Kostic, N. M. J. Am. Chem. Soc. 2003, 125, 781.
(19) Battersby, A. R.; Reynolds, J. J. J. Chem. Soc. 1961, 524.
(20) Tofteng, A. P.; Sørensen, K. K.; Conde-Frieboes, K. W.; Hoeg-
Jensen, T.; Jensen, K. J. Angew. Chem., Int. Ed. 2009, 48, 7411.
(21) Montalbetti, C. A. G. N.; Falque, V. Tetrahedron 2005, 61, 10827.
(22) Alsina, J.; Barany, G.; Albericio, F.; Kates, S. Lett. Pept. Sci. 1999, 6,
243.
(23) Coste, J.; Frerot, E.; Jouin, P. J. Org. Chem. 1994, 59, 2437.
(24) Maeda, H.; Takata, T.; Fujii, N.; Sakaue, H.; Nirasawa, S.;
Takahashi, S.; Sasaki, H.; Fujii, N. Anal. Chem. 2015, 87, 561.
(25) Edman, P.; Begg, G. Eur. J. Biochem. 1967, 1, 80.
(26) Gorman, J. J.; Wallis, T. P.; Pitt, J. J. Mass Spectrom. Rev. 2002, 21,
183.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.6b00317.
Experimental procedures and full spectroscopic data for all
new compounds (PDF)
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: monika.raj@shu.edu.
Author Contributions
†J.M.N. and N.L. contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
M.R. acknowledges the University Research Council (URC) for
financial support. J.M.N. thanks the NASA NJ Space Grant
Consortium for fellowship. We thank Ryan Cohen at Merck for
structural analysis by NMR. M.R. thanks Prof. Paramjit S. Arora
at NYU for useful discussions.
REFERENCES
■
(1) Bakhtiar, R.; Thomas, J. J.; Siuzdak, G. Acc. Chem. Res. 2000, 33,
179.
(2) Chae, P. S.; Kim, M. S.; Jeung, C. S.; Lee, S. D.; Park, H.; Lee, S.;
Suh, J. J. Am. Chem. Soc. 2005, 127, 2396.
(3) Lee, T. Y.; Suh, J. Chem. Soc. Rev. 2009, 38, 1949.
(4) Wallace, C. J. A. Protein Engineering by Semisynthesis; CRC Press:
Boca Raton, FL, 2000.
(5) Thorner, J.; Emr, S. D.; Abelson, J. N. Methods Enzymol. 2000, 326,
200.
(6) Arnau, J.; Lauritzen, C.; Petersen, G. E.; Pedersen, J. Protein
Expression Purif. 2006, 48, 1.
(7) Gross, E. Methods Enzymol. 1967, 11, 238.
(8) Degani, Y.; Patchornik, A. Biochemistry 1974, 13, 1.
(9) Mahoney, W. C.; Smith, P. K.; Hermodson, M. A. Biochemistry
1981, 20, 443.
(10) Walker, J. M. The Protein Protocols Handbook; Humana Press:
Totowa, NJ, 2002.
(11) Tanabe, K.; Taniguchi, A.; Matsumoto, T.; Oisaki, K.; Sohma, Y.;
Kanai, M. Chem. Sci. 2014, 5, 2747.
(12) Kaminskaia, N. V.; Johnson, T. W.; Kostic,
1999, 121, 8663.
́
N. M. J. Am. Chem. Soc.
D
DOI: 10.1021/acs.orglett.6b00317
Org. Lett. XXXX, XXX, XXX−XXX