A cassette ligation strategy with thioether replacement of three Gly-Gly peptide bonds: total chemical synthesis of the 101 residue protein early pregnancy factor [psi(CH(2)S)28-29,56-57,76-77].

Article abstract of DOI:10.1021/jo016121e

The 101 residue protein "early pregnancy factor" (EPF), also known as human chaperonin 10, was synthesized from four functionalized, but unprotected, peptide segments by a sequential thioether ligation strategy. The approach exploits the differential reactivity of a peptide-NHCH(2)CH(2)SH thiolate with XCH(2)CO-peptides, where X = Cl or I/Br. Initial model studies with short functionalized (but unprotected) peptides showed a significantly faster reaction of a peptide-NHCH(2)CH(2)SH thiolate with a BrCH(2)CO-peptide than with a ClCH(2)CO-peptide, where thiolate displacement of the halide leads to chemoselective formation of a thioether surrogate for the Gly-Gly peptide bond. This rate difference was used as the basis of a novel sequential ligation approach to the synthesis of large polypeptide chains. Thus, ligation of a model bifunctional N(alpha)-chloroacetyl, C-terminal thiolated peptide with a second N(alpha)-bromoacetyl peptide demonstrated chemoselective bromide displacement by the thiol group. Further investigations showed that the relatively unreactive N(alpha)-chloroacetyl peptides could be "activated" by halide exchange using saturated KI solutions to yield the highly reactive N(alpha)-iodoacetyl peptides. These findings were used to formulate a sequential thioether ligation strategy for the synthesis of EPF, a 101 amino acid protein containing three Gly-Gly sites approximately equidistantly spaced within the peptide chain. Four peptide segments or "cassettes" comprising the EPF protein sequence (BrAc-[EPF 78-101] 12, ClAc-[EPF 58-75]-[NHCH(2)CH(2)SH] 13, ClAc-[EPF 30-55]-[NHCH(2)CH(2)SH] 14, and Ac-[EPF 1-27]-[NHCH(2)CH(2)SH] 15) of EPF were synthesized in high yield and purity using Boc SPPS chemistry. In the stepwise sequential ligation strategy, reaction of peptides 12 and 13 was followed by conversion of the N-terminal chloroacetyl functional group to an iodoacetyl, thus activating the product peptide for further ligation with peptide 14. The process of ligation followed by iodoacetyl activation was repeated to yield an analogue of EPF (EPF psi(CH(2)S)(28)(-)(29,56)(-)(57,76)(-)(77)) 19 in 19% overall yield.

Full text of DOI:10.1021/jo016121e

A Ca ssette Liga tion Str a tegy w ith Th ioeth er Rep la cem en t of
Th r ee Gly-Gly P ep tid e Bon d s: Tota l Ch em ica l Syn th esis of th e 101
Resid u e P r otein Ea r ly P r egn a n cy F a ctor [ψ(CH2S)28-29,56-57,76-77
]
Darren R. Englebretsen,*,† Bronwyn Garnham, and Paul F. Alewood*
Centre for Drug Design and Development, Institute for Molecular Bioscience, The University of
Queensland, Brisbane, Queensland 4072, Australia
p.alewood@mail.imb.uq.edu.au
Received September 15, 2001
The 101 residue protein “early pregnancy factor” (EPF), also known as human chaperonin 10, was
synthesized from four functionalized, but unprotected, peptide segments by a sequential thioether
ligation strategy. The approach exploits the differential reactivity of a peptide-NHCH2CH2SH
thiolate with XCH2CO-peptides, where X ) Cl or I/Br. Initial model studies with short functionalized
(but unprotected) peptides showed a significantly faster reaction of a peptide-NHCH2CH2SH thiolate
with a BrCH2CO-peptide than with a ClCH2CO-peptide, where thiolate displacement of the halide
leads to chemoselective formation of a thioether surrogate for the Gly-Gly peptide bond. This rate
difference was used as the basis of a novel sequential ligation approach to the synthesis of large
polypeptide chains. Thus, ligation of a model bifunctional NR-chloroacetyl, C-terminal thiolated
peptide with a second NR-bromoacetyl peptide demonstrated chemoselective bromide displacement
by the thiol group. Further investigations showed that the relatively unreactive NR-chloroacetyl
peptides could be “activated” by halide exchange using saturated KI solutions to yield the highly
reactive NR-iodoacetyl peptides. These findings were used to formulate a sequential thioether ligation
strategy for the synthesis of EPF, a 101 amino acid protein containing three Gly-Gly sites
approximately equidistantly spaced within the peptide chain. Four peptide segments or “cassettes”
comprising the EPF protein sequence (BrAc-[EPF 78-101] 12, ClAc-[EPF 58-75]-[NHCH2CH2-
SH] 13, ClAc-[EPF 30-55]-[NHCH2CH2SH] 14, and Ac-[EPF 1-27]-[NHCH2CH2SH] 15) of EPF
were synthesized in high yield and purity using Boc SPPS chemistry. In the stepwise sequential
ligation strategy, reaction of peptides 12 and 13 was followed by conversion of the N-terminal
chloroacetyl functional group to an iodoacetyl, thus activating the product peptide for further ligation
with peptide 14. The process of ligation followed by iodoacetyl activation was repeated to yield an
analogue of EPF (EPF ψ(CH2S)28-29,56-57,76-77) 19 in 19% overall yield.
In tr od u ction
successes, chemical protein synthesis methods remain
limited in the length of protein that may be obtained in
high yield and purity. In stepwise solid-phase syntheses
of longer peptides, the yields have invariably been low
due to the difficulties inherent in separating the target
protein from a large number of resin-bound peptide
byproducts formed during chain assembly and from side
products generated in the final cleavage step(s). Conver-
gent synthesis strategies require coupling of fully pro-
tected peptide segments. While these strategies are often
successful, they require high levels of expertise, are slow,
and can be complicated by unforseen difficulties in
purification and coupling of the poorly soluble fully
protected peptide segments.10
In past years solid-phase,1-5 solution-phase,6-9 and
convergent solid-phase10,11 chemical peptide synthesis
techniques have been used to synthesize reliably highly
pure, fully biologically active proteins.12,13 Despite these
† Present address: Chemistry Department, Institute of Fundamen-
tal Sciences, Massey University, Palmerston North, New Zealand.
(1) Gutte, B.; Merrifield, R. B. J . Biol. Chem. 1971, 246, 1922-1941.
(2) Clark-Lewis, I.; Aebersold, R.; Ziltener, H.; Schrader, J . W.; Hood,
L. E.; Kent, S. B. H. Science 1986, 231, 134-139.
(3) Kent, S. B. H. Annu. Rev. Biochem. 1988, 57, 957-989.
(4) Schneider, J .; Kent, S. B. H. Cell 1988, 54, 363-368.
(5) Kent, S. B. H.; Alewood, D.; Alewood, P.; Baca, M.; J ones, A.;
Schnolzer, M. In Innovations and Perspectives in Solid-Phase Peptide
Synthesis; Epton, R., Ed.; Intercept Ltd.: Andover, 1992; pp 1-22.
(6) Yakima, H.; Fujii, N. J . Am. Chem. Soc. 1981, 103, 5867-5871.
(7) Sakakibara, S. Bioploymers (Pept. Sci.) 1995, 37, 17-28.
(8) Nishiuchi, Y.; Inui, T.; Nishio, H.; Bo´di, J .; Kimura, T.; Tsuji, F.
I.; Sakakibara, S. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 13549-13554.
(9) Quibell, M.; Packman, L. C.; J ohnson, T. J . Am. Chem. Soc. 1995,
117, 11656-11668.
In addition to established chemical protein synthesis
techniques, the recently described chemical ligation
technologies represent an important advance in chemical
protein synthesis.14-19 In its simplest form, a chemical
(10) Lloyd-Williams, P.; Albericio, F.; Giralt, E. Tetrahedron 1993,
49, 11065-11133.
(13) Borgia, J . A.; Fields, G. B. Tibtech 2000, 18, 243-251.
(14) Schnolzer, M.; Kent. S. B. H. Science 1992, 256, 221-225.
(15) Baca, M.; Kent, S. B. H. Proc. Natl. Acad. Sci. U.S.A. 1993, 90,
11638-11642.
(11) Benz, H. Synthesis 1994, 337-358.
(12) Wilken, J .; Kent, S. B. H. Curr. Opin. Biotechnol. 1998, 9, 412-
426.
10.1021/jo016121e CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/23/2002
J . Org. Chem. 2002, 67, 5883-5890
5883

Englebretsen et al.
ligation strategy requires the assembly of two peptide
segments which have chemoselective, mutually reactive
functional groups at the appropriate N- and/or C-termini.
The specific reaction of these N- and C-terminal func-
tionalities results in covalent coupling of the two peptide
segments, giving a product14 which has the “ligation site”
peptide bond replaced by an amide bond isostere or
surrogate. There are a number of benefits of methods for
protein synthesis by chemoselective ligation. The unpro-
tected peptide segments, prepared by efficient SPPS
techniques, may be purified quickly by standard RP-
HPLC techniques. The high specificity of the ligation
reaction between the N- and C-terminal functional groups
means that side chain protection is unnecessary, simpli-
fying synthesis and purification of the peptide segments.
An added advantage of using unprotected peptides is that
high concentrations, in denaturing aqueous solutions, of
both reactants can be employed, thus enhancing the rate
of the bimolecular ligation reaction.15 The highly specific
chemoselective ligation reaction usually goes to comple-
tion and results in few or no unwanted side products.15
Finally, chemical ligation is readily undertaken in aque-
ous or aqueous-organic solutions, which allows conve-
nient monitoring by RP-HPLC and LC-MS analysis.
Protein synthesis by chemical ligation was first exem-
plified by the synthesis of a fully active HIV-1 protease
analogue15 by thioester-forming ligation. A feature of the
thioester amide bond surrogate is that it may be used as
a probe to study the role of backbone amide bonds in the
activity of the protein. For example, enzymatic charac-
terization of a “chemically mutated” HIV-1 protease
analogue demonstrated the crucial importance of the
hydrogen bonding contribution from the amide NH of
Ile50 in promoting the activity of HIV-1 protease, a result
which was only obtainable by studying a protein synthe-
sized using ligation chemistry.16 We have also described
the synthesis of HIV-1 protease analogues by both
thioether and thioester ligation strategies.17 In this work
significant differences in the activity of the various
analogues were found and were related to graduated
disruption of the substrate-binding ability of the ligated
analogues compared to the native protein.17 The choice
of the thioether surrogate has been pursued in our work17
because of the superior chemical stability of the thioether
bond compared to that of the comparatively base-labile
thioester.15
any of the 20 natural amino acids, although Val, Ile, and
Pro show slow ligation rates.20 A more general version
of native chemical ligation which extends to sites other
than X-Cys has been reported.21 Native chemical ligation
is also used in the solid-phase chemical ligation method.19
Native chemical ligation has been recently extended
beyond chemical synthesis by the development of ex-
pressed protein ligation, in which recombinant proteins
bearing C-terminal thioester functional groups are used
in subsequent native ligation syntheses.22,23 Other meth-
ods of chemical ligation have been used to synthesize
active HIV-1 protease analogues containing a pseudo-
proline linkage,24,25 and peptide dendrimers containing
thiazolidine, oxime, and hydrazone linkages.26
Chemical protein synthesis via ligation strategies is
limited by the current chemistry employed in stepwise
solid-phase peptide synthesis. Peptide segments up to
∼60 amino acids may be synthesized in high yield and
purity,3 thus imposing a limit of about 120 amino acids
on the length of proteins resulting from ligation of two
segments. We are interested in further exploiting and
developing solution-phase sequential ligation strategies
to enable the assembly of longer proteins (e.g., 100-200
amino acids). In this report we describe the development
of a sequential ligation method for protein synthesis
based on chemoselective thioether formation. The fun-
damental principles were developed through studies of
small peptides, and the application of this strategy is
demonstrated by the synthesis of an analogue of the
101 amino acid protein “early pregnancy factor” EPF
[ψ(CH2S)28-29,56-57,76-77], also known as human chapero-
nin 10.27,28
Resu lts a n d Discu ssion
Con cep t a n d Str a tegy. In the sequential peptide
ligation strategy a target protein sequence is divided into
a series of readily synthesized peptide segments or
cassettes where each suitably functionalized peptide has
the potential to be “inserted” regioselectively into the
synthetic scheme (Scheme 1). Each segment (except those
at the N- and C-termini of the target protein) has at its
C-terminus a defined chemoselective functional group,
while at its N-terminus there is a masked functional
group which can be readily activated.56
The first step of protein synthesis is chemoselective
reaction of the functional groups at the C- and N-termini
of segments B and C, respectively, to give product BC,
with the N-terminus of segment B incorporating a
masked functional group. The masked N-terminal func-
Other ligation chemistries have been described. A
particularly elegant example is native chemical ligation,
where two peptide fragments, one bearing a C-terminal
glycine thioester and the other an N-terminal cysteine
residue, react to form a protein containing a native amide
bond at the segment junction.17,18 Native ligation has
been extended recently to X-Cys sites, where X can be
(20) Hackeng, T. M.; Griffin, J . H.; Dawson, P. E. Proc. Natl. Acad.
Sci. U.S.A. 1999, 96, 10068-10073.
(21) Canne, L. E.; Bark, S. J .; Kent, S. B. H. J . Am. Chem. Soc.
1996, 118, 5891-5896.
(16) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H.
Science 1994, 266, 776-779.
(22) Severinov, K.; Muir, T. W. J . Biol. Chem. 1998, 273, 16205-
16209.
(17) (a) Englebretsen, D. R.; Bergman, D.A.; Garnham, B. G.;
Brinkworth, R. I.; Fairlie, D. P.; Alewood, P. F. In Proceedings of the
14th American Peptide Symposium; Kaumaya, P., Ed.; Mayflower:
Kingswinford, 1995; pp 589-590. (b) Wilce, J .A.; Love, S. G.; Rich-
ardson, J . S.; Alewood, P. F.; Craik, D. J . J . Biol. Chem. 2001, 276,
25997-26003.
(23) Muir, T. W.; Sondhi, D.; Cole, P. A. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 6705-6710.
(24) Liu, C.-F.; Rao, C.; Tam, J . P. J . Am. Chem. Soc. 1996, 118,
7-312.
(25) Tam, J . P.; Yu, Q.; Mioa, Z. Bioploymers (Pept. Sci.) 1999, 51,
311-332.
(18) Muir, T. W.; Dawson, P. E.; Kent, S. B. H. In Methods in
Enzymology; Fields, G. B., Eds.; Academic Press: New York, 1997; pp
266-298.
(19) Canne, L. E.; Botti, P.; Simon, R. J .; Chen, Y.; Dennis, E. A.;
Kent, S. B. H. J . Am. Chem. Soc. 1999, 121, 8720-8727.
(26) Shao, J .; Tam, J . P. J . Am. Chem. Soc. 1995, 117, 3893-3899.
(27) Cavanagh, A. C.; Morton, H. Eur. J . Biochem. 1994, 222, 551-
560.
(28) Monzini, N.; Legname, G.; Marcucci, F.; Gromo, G.; Modena,
D. Biochim. Biophys. Acta 1994, 1218, 478-480.
5884 J . Org. Chem., Vol. 67, No. 17, 2002

Synthesis of the Early Pregnancy Factor [ψ(CH2S)28-29,56-57,76-77
]
SCHEME 1. Gen er a l P r in cip le of P r otein
Syn th esis by Sequ en tia l Liga tion a
ated in these products were not strictly peptide backbone
amide bond surrogates.
Thioether amide bond surrogates were first introduced
into short peptide analogues using either preformed
pseudodipeptide subunits or by direct formation of a
thiomethylene bond using on-resin procedures.37,38 Previ-
ously, we have used solution ligation procedures to
synthesize HIV-1 protease and transthyretin analogues
containing Gly-ψ(CH2S)-Gly thioether amide bond sur-
rogates.17a,b Earlier studies of short protected peptide
analogues have demonstrated that the methodology can
in principle be extended to ligation sites other than Gly-
Gly, shown for example by the synthesis of the hindered
Leu-Nle thioether amide surrogate.38 In this study we
describe in detail the conditions for the sequential
ligation of C-terminal 2-mercaptoethylamido functional-
ized peptides and NR-haloacetyl peptides to generate
multiple thioether linkages which contain a thioether
surrogate of a Gly-Gly peptide bond.
a
(i) Chemoselective reaction of the functional groups at the C-
and N-termini of cassettes B and C, respectively, with cassette B
also incorporating a masked functional group at its N-terminus.
(ii) The masked N-terminal functional group of the resulting
ligated product, BC, is transformed to the activated intermediate
*BC.( iii) Ligation of the N, C terminal bifunctional cassette
peptide A with activated peptide *BC yielding the peptide ABC
which contains two ligation sites and a masked functional group
at its N-terminus. Further repetition of the process results in
elongation of the protein chain.
Syn th esis of [N-(2-Mer ca p toeth yla m id o)]-F u n c-
tion a lized P ep tid es. Peptide segments were assembled
on a linker derived from 2-aminoethanethiol in order to
introduce a C-terminal N-(2-mercaptoethyl)amide moi-
ety.39 C-Terminal N-(2-mercaptoethylamide)-functional-
ized peptides have been synthesized using linkers gen-
erated by on-resin procedures.40,41 However, in our
approach we preferred to introduce a fully characterized
linker to a well-behaved aminomethyl-polystyrene resin.5
A clear parallel exists with Boc aminoacyl-OCH2Pam-
polystyrene resins, which incorporate a purified, char-
acterized chemical “handle” or linker onto a well-
characterized resin5,42 and are the resins of choice over
in-situ derivatized “Merrifield resins” which retain un-
desirable resin-bound functional groups. Accordingly, we
synthesized the linker 4-[NR-Boc-2-aminoethylmercapto]-
methylphenoxyacetic acid (Boc-AMPA 1) in two steps.39
The Boc-AMPA linker was subsequently attached to an
aminomethyl-polystyrene resin and used to assemble
C-terminally 2-mercaptoethylamido functionalized pep-
tides using Boc chemistry with in situ neutralization.43
Selected peptides were functionalized at their N-terminus
using the anhydrides of either chloroacetic and bromo-
acetic acid. The peptides were liberated from the resin
and side chain protecting groups simultaneously removed
using standard treatment with HF in the presence of
p-cresol scavenger. After purification by preparative
RP-HPLC, all purified peptides58 showed single peaks by
RP-HPLC analysis, while ES-MS analysis showed the
expected molecular ion masses.
tional group of BC is then transformed to the activated
intermediate *BC. Subsequent ligation of the bifunc-
tional cassette A with activated peptide *BC yields the
peptide ABC which contains two ligation sites and a
masked functional group at its N-terminus. Further
repetition of the process results in elongation of the
protein chain. In this strategy the major requirement for
successful sequential ligation is the incorporation of an
unreactive but readily transformed masked N-terminal
functional group. In this report we describe how this may
be readily achieved via thioether ligation chemistry.
Th ioeth er Liga tion . Thioether linkages are gener-
ated by simple nucleophilic displacement, by a thiol-
bearing peptide, of the halogen from the NR-haloacetyl
group of a second peptide. Peptide and protein syntheses
which incorporate thioether ligation strategies were
rendered practical by the discovery that NR-chloro- and
bromo-acetylated peptides survive treatment by liquid
HF29-31 used during the cleavage of peptides synthesized
via Boc SPPS.57 NR-Haloacetylated peptides may also be
synthesized via Fmoc SPPS.32 NR-Haloacetylated peptides
have been coupled to cysteine-bearing peptides in the
synthesis of cyclic peptides and peptide oligomers,29,30
peptide-protein immunogens,33 a protein construct,34
a
(36) Nefzi, A.; Sun, X.; Mutter, M. Tetrahedron Lett. 1995, 36, 229-
230.
peptide-porphyrin template complex,35 a template-as-
sembled synthetic protein,36 and a designed transmem-
brane protein.32 However, the thioether linkages gener-
(37) Spatola, A. F.; Darlak, K. Tetrahedron 1988, 44, 821-833.
(38) Paladino, J .; Thurieau, C.; Morris, A. D.; Kucharczyk, N.;
Rouissi, N.; Regoli, D.; Fauchere, J .-L. Int. J . Pept. Protein Res. 1993,
42, 284-293.
(29) Lindner, L.; Robey, F. A. Int. J . Pept. Protein Res.1987, 30, 794-
800.
(30) Robey, F. A.; Fields, R. L. Anal. Biochem. 1989, 177, 373-377.
(31) Ivanov, B. B.; Robey, F. A. Pept. Res. 1996, 9, 305-307.
(32) Englebretsen, D. R.; Choma, C. T.; Robillard, G. T. Tetrahedron
Lett. 1998, 39, 4929-4932.
(33) Kolodny, N.; Robey, F. Anal. Biochem. 1990, 187, 136-140.
(34) Muir, T. W.; Williams, M. J .; Ginsberg, M. H.; Kent, S. B. H.
Biochemistry 1994, 33, 7701-7708.
(35) Choma, C. T.; Kaestle, K.; Akerfeldt, K. S.; Kim, R. K.; Groves,
J . T.; DeGrado, W. F. Tetrahedron Lett. 1994, 35, 6191-6194.
(39) Englebretsen, D. R.; Garnham, B. G.; Bergman, D. A.; Alewood,
P. F. Tetrahedron Lett. 1995, 36, 8871-8874.
(40) Rietman, B. H.; Smulders, R. H. P. H.; Eggen, I. F.; Van Vliet,
A.; Van De Werken, G.; Tesser, G. I. Int. J . Pept. Protein Res. 1994,
44, 199-206.
(41) Winger, T. M.; Ludovice, P. J .; Chaikof, E. L. Bioconjugate
Chem. 1995, 6, 323-326.
(42) Mitchell, A. R.; Kent, S. B. H.; Engelhard, M.; Merrifield, R. B.
J . Org. Chem. 1978, 43, 2845-2852.
(43) Schno¨lzer, M.; Alewood, P. F.; J ones, A.; Alewood, D.; Kent, S.
B. H. Int. J . Pept. Protein Res. 1992, 40, 180-193.
J . Org. Chem, Vol. 67, No. 17, 2002 5885

Englebretsen et al.
F IGURE 1. Graph of the pH dependence of the reaction of ClAc-IGGF and BrAc-IGGF with LPGKWKPKMI-[NH(CH2)2-SH]. In
both cases the dependence of the rate on pH corresponds to pH dependent increase of ionization of the peptide thiol (pKa 8.2).
Va r ia bles Affectin g th e Ra te of th e Th ioeth er
Liga tion Rea ction . Surprisingly, there are few reports
of detailed physicochemical studies on the reaction of
thiols with haloacetyl compounds in aqueous environ-
ments.44 We thus carried out a short study to determine
the effect of pH and the halide leaving group on the rate
of the thioether formation by ligating the model peptides
LPGKWKPKMI-[NH(CH2)2-SH] 2 and X-Ac-IGGF (X )
Cl, 3; X ) Br, 4) to give product LPGKWKPKMIG-ψ-
(CH2S)-GIGGF 5. The peptide sequence chosen for this
study, LPGKWKPKMIGGIGGF, HIV-1 PR [38-53], with
the GG ligation site underlined, was selected as a model
as both the starting peptide segments (2, 3, and 4) and
ligated product 5 were all well resolved and thus conve-
niently monitored by RP-HPLC analysis. In these experi-
ments a peptide concentration of 2 mM was selected as
this represented a typical concentration used for ligating
long peptide segments.14,15,17
Ligation of LPGKWKPKMI-[NH(CH2)2-SH] 2 with
ClAc-IGGF 3 or BrAc-IGGF 4 at several pH values
(Figure 1) shows that, for a given halogen leaving group,
the rate of formation of ligation product 5 increased with
increasing pH. This result was expected on the basis that
the concentration of thiolate anion, the active nucleophilic
species, would increase with pH. Figure 1 also shows that
at identical pH values (7.9 or 8.5) the initial reaction rate
of BrAc-IGGF 4 was approximately an order of magni-
tude greater than that of ClAc-IGGF 3.
The differences in reaction rates of R-chloro- and
R-bromo-acetylated peptides were further explored in a
competition experiment at pH 8.5. To distinguish be-
tween the relative rate of displacement of chloride versus
bromide, ClAc-IGGS 6 (used instead of ClAc-IGGF 3
to improve starting material and product resolution by
RP-HPLC analysis) and BrAc-IGGF 4 were reacted
with thiolated peptide 2. The product ratio of 97:3
(by RP-HPLC integration) for LPGKWKPKMIG-ψ(CH2S)-
G-IGGF 5 versus LPGKWKPKMIG-ψ(CH2S)-GIGGS 7
(derived from bromide and chloride displacement, re-
spectively) showed the high chemoselectivity of thiolated
peptide 2 for the NR-bromoacetyl over the NR-chloroacetyl
peptide. Unexpectedly, a similar competition study be-
tween BrAc-IGGF 4 and IAc-IGGS 8 for 2 at pH values
of 8.5, 7.0, and 6.0 showed no significant difference in
reactivity toward the thiolated peptide, with approxi-
mately equal ratios of 5 and 7 (from bromide and iodide
displacement, respectively) being observed (data not
shown). Importantly, at pH 7.0 the ligation reaction for
both haloacetyl peptides 4 and 8 was complete after 2 h.
Ch em oselective Con tr ol of Th ioeth er Liga tion . To
investigate whether chemoselective control was achiev-
able during the ligation of an NR-bromoacetyl peptide
with another peptide bearing both an NR-chloroacetyl
group and a C-terminal 2-mercaptoethylamido group,
ClAc-LPGKWKPKMI-[NH(CH2)2-SH] 9 was reacted with
BrAc-IGGF 4 at pH 8.5. Selective displacement of bro-
mine by thiolate was observed to give ClAc-LPGKWK-
PKMI-G-ψ(CH2S)-G-IGGF 10 as the sole product after 1
h. Other possible cyclic or oligomeric products resulting
from thiol displacement of chloride were not observed by
RP-HPLC and ES-MS analyses. We envisaged that the
NR-chloroacetyl group of the ligated product 10 may act
as a “masked” functional group under the right condi-
tions. While in principle NR-chloroacetylated peptides
could react further with another C-terminal thiol-peptide
under more forcing conditions (e.g., higher pH and/or
longer reaction time) to give a peptide with two thioether
ligation sites, in practice the reaction of chloroacetylated
peptides with thiol functions was quite slow, even at pH
9.5 (Figure 1). Accordingly, we investigated altering the
reactivity of NR-chloroacetyl peptides by substituting the
chlorine atom with either a bromine or iodine atom, and
thus increasing the reactivity of the peptide for subse-
quent ligation with another thiolated peptide.
Activa tion of ClAc-IGGF 7 by Ha lid e Exch a n ge.
The conversion of ClAc-IGGF 3 to IAc-IGGF, 11, was
readily achieved within 90 min using saturated KI
solution (∼7.5 M) over the pH range 2 to 7. Conversion
could also be achieved using saturated KI in 6 M urea.
Denaturants, acting as chaotropes, such as urea are often
essential if high mM concentrations of peptide segments
are to be achieved for chemoselective ligations. ES-MS
analysis of RP-HPLC fractions from these experiments
also showed that no hydrolysis of the iodoacetyl to a
hydroxyacetyl group had occurred. The rate of conversion
of ClAc-IGGF 3 to IAc-IGGF 11 was much greater than
conversion of 3 to BrAc-IGGF 4 (<20% after 5 h) using
either 6 M Gu‚HBr or saturated KBr solutions, and
(44) Lindley, H. Biochem. J . 1962, 82, 418-425.
5886 J . Org. Chem., Vol. 67, No. 17, 2002

Synthesis of the Early Pregnancy Factor [ψ(CH2S)28-29,56-57,76-77
]
SCHEME 2. Syn th esis of EP F by Sequ en tia l Th ioeth er Liga tion a
a
(i) Ligation of ClAc-EPF(58-75)-[NH(CH2)2SH] 13 and BrAc-EPF(78-101) 12 to give ClAc-EPF(58-75)-G-ψ(CH2S)-G-(78-101) 16.
(ii) Conversion of 16 to IAc-EPF(58-75)-G-ψ(CH2S)-G-(78-101) 17. (iii) Ligation of ClAc-EPF(30-55)-[NH(CH2)2SH] 14 and IAc-EPF(58-
75)-G-ψ(CH2S)-G (78-101) 17 to give ClAc-EPF(30-55)-G-ψ(CH2S)-G-(58-75)-G-ψ(CH2S)-G-(78-101) 18. (iv) Simultaneous exchange
and ligation of Ac-EPF(1-27)-[NH(CH2)2SH] 15 with ClAc-EPF(30-55)-G-ψ(CH2S)-G-(58-75)-G-ψ(CH2S)-G-(78-101) 18 yielding ligated
EPF: Ac-EPF(1-27)-G-ψ(CH2S)-G-(30-55)-G-ψ(CH2S)-G-(58-75)-G-ψ(CH2S)-G-(78-101) 19.
therefore chloride to bromide exchange was not investi-
gated further. It should be noted, however, that in 6 M
Gu‚HCl the rate of the reverse reaction (conversion of a
bromoacetyl peptide to chloroacetyl) was significant (data
not shown). Thus in further experiments only urea was
used as denaturant.
F IGURE 2. Amino acid sequence of EPF. In the sequential
ligation synthesis of this protein from four peptide cassettes,
amide bonds between the underlined Gly-Gly dipeptides were
replaced by a thioether surrogate.
Syn t h esis of E a r ly P r egn a n cy F a ct or (E P F )
[ψ(CH2S)28-29,56-57,76-77]. The preliminary reactivity stud-
ies underpinned the principles of successful sequential
thioether ligation. Thus a bromoacetyl group was highly
reactive toward thiolate at pH 8.5, while an accompany-
ing chloroacetyl functionality was relatively unreactive.
However, the chloroacetyl group could be converted
quantitatively within 90 min to the reactive iodoacetyl
group, which could then react selectively with a thiol in
the presence of another chloroacetyl group. On the basis
of these simple principles, we undertook to establish
sequential peptide ligation through the synthesis of an
analogue of the 101 residue “early pregnancy factor”
(EPF)27,28 that contained three thioether ligation sites.
EPF, an extracellular cytokine,27 is identical in se-
quence to the intracellular “folding” protein human
chaperonin 10 and provided a convenient synthetic target
as it possesses three approximately equidistantly dis-
tributed Gly-Gly sites (Figure 2), suitable for replacement
by the thioether amide bond surrogate. Inspection of the
crystal structures of two chaperonin 10 homologues45,46
indicated that the selected Gly-Gly residues lay in
exposed loops and that replacement of the peptide bonds
by thioether surrogates would not interfere with critical
intra- or intermolecular hydrogen bonds required to
maintain structure. In addition, activity assays using the
rosette inhibition test were available for EPF,47 so it
would be possible to assess the impact of the thioether
amide bond surrogates on activity of the synthetic EPF
in direct comparison to the native protein.
The route chosen to synthesize EPF [ψ-
(CH2S)28-29,56-57,76-77] is shown in Scheme 2. Four specif-
ically functionalized peptide segments [BrAc-EPF-
(78-101) 12; ClAc-EPF(58-75)-[NHCH2CH2SH] 13; ClAc-
EPF(30-55)-[NHCH2CH2SH] 14; and Ac-EPF(1-27)-
[NHCH2CH2SH] 15 were prepared and purified in good
yield. The assembly of EPF [ψ(CH2-S)28-29,56-57,76-77] 19
was then performed by sequential segment ligation
(45) Hunt, J . F.; Weaver, A. J .; Landry, S. J .; Gierasch, L.; Deisen-
hofer, J . Nature 1996, 379, 37-45.
(46) Xu, Z.; Horwich, A. L.; Sigler, P. B. Nature 1997, 388, 741-
750.
(47) Morton, H.; Hegh, V.; Clunie, G. J . A. Nature 1974, 249, 459-
460.
J . Org. Chem, Vol. 67, No. 17, 2002 5887

Englebretsen et al.
F IGURE 3. ES-MS mass spectrum of EPF [ψ(CH2S)28-29,56-57,76-77]. Inset: mathematical reconstruction of the mass spectrum
showing appearance of ligated EPF as a singly charged species.
reactions. For each step we employed a small excess of
the thiol-containing peptide segment and followed the
ligation by LC-MS analysis. All reactions shown in
Scheme 2 were complete within 4 h. To avoid minor losses
of material we combined the final halo-exchange and
ligation steps when reacting 15 and ClAc-EPF(30-55)-
G-ψ(CH2S)-G-(58-75)-G-ψ(CH2S)-G-(78-101) 18. The li-
gated EPF 19 was purified to yield 11.7 mg of protein
with a cumulative yield of 19% based on BrAc-EPF -(78-
101). RP-HPLC analysis showed a single peak48 and ES-
MS analysis (Figure 3) confirmed the high purity of the
protein with good agreement between the observed
molecular weight of the ligated protein (10853 ( 2 Da)
and the calculated value (10852.8 Da, average isotope
mass). Preliminary assays for EPF activity using a
rosette inhibition test indicated that the ligated protein
possessed full biological activity.47,48 In addition, ES-MS
analysis under native ionization conditions showed that
the ligated protein appeared to associate into heptamers
in a fashion similar to native chaperonin 10.48
Early sequential-type ligation methods focused on
laborious semisynthetic strategies using specially tailored
recombinant proteins.49,50 Later, so-called “convergent
chemical ligation” strategies use a combination of ligation
chemistries for protein synthesis; examples include a
“tethered dimer” HIV-1 protease51 and a heterodimeric
cMyc-Max protein construct containing two N-termini.52
Most recently sequential native ligation methods have
been used in a model synthesis of a 95 amino acid protein
module,19 to insert a short fluorophore-bearing synthetic
peptide between two recombinant protein domains53 and
to synthesize a three zinc finger protein.54
We have shown by the synthesis of fully active48 EPF
[ψ(CH2S)28-29,56-57,76-77] the general utility of a novel
sequential thioether ligation strategy for chemical protein
synthesis. An interesting feature of this sequential liga-
tion strategy is that each component peptide segment
could, in principle, act as an interchangeable cassette.
This approach would be particularly useful for chemical
syntheses of variant proteins where each of the analogue-
containing cassette peptides would be inserted during the
synthesis of the full length protein, thereby avoiding the
need for multiple full length syntheses.
Con clu sion
We have described a novel thioether cassette strategy
for the chemical ligation of unprotected peptide segments.
The demonstration of “tunable” N-haloacetyl-peptide
chemistry opens up many possibilities for future chemose-
lective ligation approaches. The successful synthesis of
the fully active “early pregnancy factor” analogue, EPF
[ψ(CH2S)28-29,56-57,76-77] with three thioether peptide bond
replacements should make this approach a useful addi-
tion to the armamentarium of chemical protein synthesis
methods. Further, the method should be compatible with
other ligation chemistries (e.g., native chemical ligation)
and other strategies (e.g., solid-phase chemical ligation).
Exp er im en ta l Section
Mater ials an d Meth ods. Diisopropylcarbodiimide, 2-amino-
ethanethiol, and R-bromoacetic acid were supplied by Aldrich
(Milwaukee). 4-Hydroxymethylphenoxyacetic acid was sup-
plied by Sigma (St. Louis, MO). R-Iodoacetic acid was pur-
chased from BDH Chemicals Ltd. (Poole, UK). R-Chloroacetic
acid was purchased from Ajax Chemicals (Sydney, Australia).
Boc-L-amino acid-O-CH2-Pam resins (Pam: phenylacetamido-
methyl) were purchased from ABI Inc. (Foster City, CA).
Aminomethyl polystyrene resin [100-200 mesh, 1% DVB
cross-linked, amine substitution 0.83 mmol/g] and Boc-L-
(48) Love, S.; Englebretsen, D. R.; Garnham, B. G.; Cavanagh, A.
C.; H Morton, H.; Alewood, P. F. In. The NATO Series on Molecular
Aggregates; Kluwer Academic Publishers: Dordrecht, The Netherlands,
1998; pp 171-181
(49) Gaertner, H. F.; Offord, R. E.; Cotton, R.; Timms, D.; Camble,
R.; Rose, K. J . Biol. Chem. 1994, 269, 7224-7230.
(50) Gaertner, H. F.; Offord, R. E.; Cotton, R.; Timms, D.; Camble,
R.; Rose, K. Bioconjugate Chem. 1994, 5, 333-338.
(51) Baca, M.; Muir, T. W.; Schnolzer, M.; Kent, S. B. H. J . Am.
Chem. Soc. 1995, 117, 1881-1887.
(52) Canne, L. E.; Ferre´-D’Amare´, Y.; Burley, S. K.; Kent, S. B. H.
J . Am. Chem. Soc. 1995, 117, 2998-3007.
(53) Cotton, G. J .; Ayers, B.; Xu, R.; Muir, T. W. J . Am. Chem. Soc.
1999, 121, 1100-1101.
(54) Beligere, G. S.; Dawson, P. E. Biopolymers (Pept. Sci.) 1999,
51, 363-369.
5888 J . Org. Chem., Vol. 67, No. 17, 2002

Synthesis of the Early Pregnancy Factor [ψ(CH2S)28-29,56-57,76-77
]
amino acids were from the Peptide Institute (Osaka, J apan).
Dichloromethane, diisopropylethylamine, N,N-dimethylforma-
mide, di-tert-butyl pyrocarbonate, and trifluoroacetic acid were
obtained from Auspep (Melbourne, Australia). RP-HPLC grade
acetonitrile was purchased from Millipore-Waters (Sydney,
Australia). HBTU (2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyl-
uronium hexafluorophosphate) was purchased from Richelieu
Biotechnologies (Quebec, Canada). Deionized water was used
throughout and was prepared by a Milli-Q water purification
system. Screw-cap glass peptide synthesis reaction vessels (20
mL) with sintered glass filter frit were obtained from Embell
Scientific Glassware (Queensland, Australia). An all-Kel-F
apparatus was used for HF cleavage. Argon, helium, and
nitrogen (all ultrapure grade) were from BOC gases (Queens-
land, Australia). 1H NMR spectra were recorded on a 300 MHz
instrument in CDCl3 and chemical shifts are reported in parts
per million (ppm) downfield from (CH3)4Si. Thin-layer chro-
matography was performed using silica gel plates. Unless
otherwise noted in the text, analytical RP-HPLC was per-
formed using Vydac C4 or C18 columns (5 µm, 0.46 cm × 15
cm) with linear gradients of 5-85% B over 40 min at a flow
rate of 1 mL/min (A ) 0.1% aqueous TFA; B ) 90% CH3CN,
10% H2O, 0.09% TFA) and monitoring at 214 nm.
overall yield), mp 83-85 °C (uncorr.) Anal. Calcd for C16H23-
NO5S: C, 56.30; H, 6.74; N, 4.11. Found: C, 56.37; H, 6.81;
1
N, 4.03. H NMR (CDCl3): δ 7.25 (d, 2H, J ) 8.7 Hz), 6.87 (d,
2H, J ) 8.6 Hz), 4.65 (s, 2H), 3.67 (s, 2H), 3.26 (m, br, 2H),
2.52 (t, 2H), 1.45 (s, 9H).
F u n ction a liza tion of Am in om eth yl P olystyr en e w ith
4-Met h yl-[N-Boc-2-a m in oet h ylm er ca p t o]m et h ylp h en -
oxya cetic Acid . Aminomethyl polystyrene‚HCl resin (1.1
mmol amine) was swollen in DMF:DIEA 3:1 for 10 min. The
resin was collected by filtration and washed with DMF. The
AMPA linker 1 (1.5 mmol, 1.4 equiv) and HBTU (1.5 mmol)
were dissolved in 4 mL of DMF and DIEA (460 µL, 2.6 mmol)
was added. The activated linker solution was then added to
the resin. After mixing for 25 min, the ninhydrin assay
indicated complete substitution of resin-bound amine by the
linker.
P ep tid e Syn th esis. Peptides were synthesized manually
on a 0.50 mmol scale using HBTU activation of Boc-amino
acids with in situ neutralization chemistry as previously
described.41 The syntheses were performed on Boc-Phe-OCH2-
Pam, Boc-Ser(OBzl)-OCH2-Pam, and appropriately function-
alized aminomethyl polystyrene resins. The following amino
acid side chain protection was used: Boc-Lys(ClZ)-OH, Boc-
Ser(Bzl)-OH, Boc-Trp(CHO)-OH, Boc-Thr(Bzl)-OH, Boc-Arg-
(Tos)-OH, Boc-Asp(OChx)-OH and Boc-Glu(OChx)-OH, Boc-
Asn(Xan)-OH and Boc-Gln(Xan)-OH, Boc-Tyr(2BrZ)-OH. Each
residue was coupled for 10 min and coupling efficiencies were
determined by the quantitative ninhydrin reaction.55 Prior to
HF cleavage, the N-terminal Boc protecting group was re-
moved (100% TFA) followed by formyl group removal where
required (1.5 mL of ethanolamine in 25 mL of DMF/5% water
(2 × 30 min). The peptide-resins were washed with DMF and
then DCM and dried on a sinter with vacuum under a gentle
stream of nitrogen. Selected peptides were haloacetylated at
their N-terminus, as follows. The amino-peptide-resin was
washed with DMF, neutralized for 2 min with 20% DIEA/
DMF, washed with DMF, and haloacetylated for 10 min using
the desired haloacetic acid anhydride (5-10 equiv) dissolved
in DMF. After a final DMF wash the NR-haloacetyl peptide
resin was washed with DCM and dried as above. All dried
peptide-resins were cleaved using 10 mL of p-cresol:HF 1:9
Mass spectra were acquired on a PE Sciex API III triple
quadrupole mass spectrometer equipped with an ion spray
atmospheric pressure ionization source. Samples (10 µL) were
injected into a moving solvent (30 µL/min; 50/50 CH3CN/0.05%
TFA) coupled directly to the ionization source via a fused silica
capillary interface (50 µm i.d. × 50 cm length). Sample droplets
were ionized at a positive potential of 5 kV and entered the
analyzer through an interface plate and subsequently through
an orifice (100-120 µm diameter) at a potential of 80 V. Full
scan mass spectra were acquired over the mass range of 400
to 2000 daltons with a scan step size of 0.1 Da. Molecular
masses were derived from the observed m/z values using the
program MacSpec 3.3 (PE-Sciex Toronto, Canada). Calculated
average masses were determined using the MacBiospec pro-
gram (PE-Sciex Toronto, Canada).
Syn th esis of 4-[N-Boc-2-a m in oeth ylm er ca p to]m eth -
ylp h en oxya cetic Acid (AMP A lin k er ) 1. 4-Hydroxymeth-
ylphenoxyacetic acid (1.82 g, 10 mmol) and 2-aminoethanethiol
hydrochloride (1.14 g, 10 mmol) were dissolved in 15 mL of
TFA, and stirred at room temperature. Thin-layer chroma-
tography (n-butanol:acetic acid:water, 4:1:1) showed complete
reaction after 70 min. The mixture was diluted with 150 mL
of water and lyophilized. The residue was dissolved in 100 mL
of 0.1% aqueous TFA, filtered to remove insoluble material,
and lyophilized. The white fluffy powder (4-[2-aminoethyl-
mercapto]methylphenoxyacetic acid‚TFA salt, 3.07 g, 86%) was
dissolved in 22 mL of 0.1 M NaOH. Dioxane (25 mL) was added
and the solution was cooled to 0 °C. Di-tert-butyl pyrocarbonate
(2.25 g, 10.3 mmol, 1.2 equiv) was dissolved in 15 mL of
dioxane and added to the stirred solution. After 35 min the
solution was warmed to room temperature. After 55 min thin-
layer chromatography of the cloudy solution indicated incom-
plete reaction. Ten milliliters of a 0.1 M NaOH solution was
then added, whereupon the solution cleared, and after 75 min
reaction was complete. The solution was diluted with 250 mL
of water and extracted twice with 100 mL of ethyl acetate to
remove unreacted di-tert-butyl pyrocarbonate The aqueous
layer was carefully acidified to pH 1 with 10 mL of 6 M HCl
and extracted twice with 100 mL of ethyl acetate. The
combined organic layers were washed twice with 25 mL of
water and twice with a saturated NaCl solution. The organic
phase was dried with MgSO4, and the ethyl acetate was
removed in-vacuo leaving an oil, which was dissolved in ethyl
acetate (30 mL) and then triturated with petroleum ether (100
mL) and cooled to 0 °C. After standing for 30 min the resultant
precipitate was collected by filtration and washed with 40 mL
of ethyl acetate:petroleum ether (1:3). The solid was dried in-
vacuo to yield 1.97 g of 4-[N-Boc-2-aminoethylmercapto]-
methylphenoxyacetic acid (AMPA linker 1, 5.8 mmol, 58%
(55) Sarin, V.; Kent, S. B. H.; Tam, J . P.; Merrifield, R. B. Anal.
Biochem. 1981, 117, 147-157.
(56) In principle, the ligation chemistries selected could involve
differential regioselective chemistry for each cassette.19,52 Further, more
general versions of this strategy can be readily envisaged. These could
include N- to C-terminal protein synthesis where each cassette is
incorporated at the C-terminus of the growing peptide chain and
supported solid-phase cassette ligation where the C- or N-terminal
cassette is attached to a support with subsequent addition of each
cassette in a fashion similar to stepwise SPPS.
(57) Boc SPPS refers to use of tert-butoxylcarbonyl and benzyl
temporary and permanent protecting groups, respectively. Abbrevia-
tions use are the following: CH3CN, acetonitrile; DCM, dichloro-
methane; DIEA, diisopropylethylamine; DMF, N,N-dimethylforma-
mide; Da, Dalton; ES-MS, electrospray mass spectrometry; HBTU,
2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophos-
phate; HF, anhydrous hydrogen fluoride; LC-MS, liquid chromatog-
raphy mass spectrometry; NMR, nuclear magnetic resonance; Pam,
phenylacetamidomethyl polystyrene resin, RP-RP-HPLC, reversed-
phase high performance liquid chromatography; SPPS, solid-phase
peptide synthesis; TFA, trifluoroacetic acid. Standard IUPAC single
and triple letter codes for amino acids are used throughout.
(58) Ter m in ology. In this report we refer to the N-terminally
functionalized R-chloro-, bromo-, and iodo-acetylated peptides by
the respective prefixes ClAc, BrAc, and IAc. Thus NR-chloroacetyl-Ile-
Gly-Gly-Phe is denoted as ClAc-IGGF. The C-terminal 2-mercapto
ethylamido functional group is enclosed in brackets; thus Leu-Pro-Gly-
Lys-Trp-Lys-Pro-Lys-Met-Ile-NH(CH2)2-SH is represented as LPGK-
WKPKMI-[NH(CH2)2-SH]. The thioether surrogate for Gly-Gly di-
peptides, -[NH-CH2-CH2-S-CH2-CO]-, is represented using IUPAC
notation where the amide bond (CONH) which is replaced by the
thioether surrogate is denoted -ψ(CH2-S)-; thus the product of the
ligation of LPGKWKPKMI-[NH(CH2)2-SH] and BrAc-IGGF is shown
as LPGKWKPKMIG-ψ(CH2-S)-GIGGF.
(59) Note that the pKa of the thiol group in both cysteine and
cystamine is approximately 8.2.
J . Org. Chem, Vol. 67, No. 17, 2002 5889

Englebretsen et al.
(-5 °C, 90 min). HF was removed in-vacuo, and the peptides
were precipitated with 10 mL of ether, collected by filtration
on a sintered glass funnel under a flow of nitrogen, and washed
twice with 25 mL of ether. Following dissolution in 50% acetic
acid, peptide solutions were diluted with water and lyophilized.
P ep tid e Isola tion a n d Ch a r a cter iza tion . Crude peptides
(up to 70 mg) were purified by preparative RP-HPLC using
either a Vydac C-4 (10 µm, 2.2 cm × 25 cm) or Waters Delta-
pak C-18 (10 µm, 1.9 cm × 30) column with a linear gradient
from 0 to 60% B over 60 min at 8 mL/min. Fractions were
collected based on monitoring of the column effluent at 230
nm. Those fractions shown by ES-MS analysis to contain solely
the target peptide were combined and lyophilized. Purified
peptides were homogeneous by HPLC analysis and showed
the expected molecular ion masses by ES-MS: ClAc-IGGS 6
Mr 409.1, calcd 409.8; ClAc-IGGF 3 Mr 469.3, calcd 470.0;
BrAc-IGGF 4 Mr 514.0, calcd 514.4; IAc-IGGS 8 Mr 500.8,
calcd 501.3; LPGKWKPKMI-[NH(CH2)2-SH] 2 Mr 1256.9,
samples were taken for RP-HPLC and ES-MS analysis. The
experiment was repeated using 0.1% TFA solution saturated
with KI (pH 2). The exchange was also studied at a peptide
concentration of 22 mM in saturated KI, 100 mM phosphate,
pH 7.0.
Syn th esis of EP F 19 by Sequ en tia l Liga tion : Liga tion
of ClAc-EP F (58-75)-[NH(CH2)2SH] 13 a n d Br Ac-EP F
(78-101) 12. 13 (13.4 µmol) and 12 (11.0 µmol) were dissolved
in 5 mL of 8 M urea, 100 mM Tris‚HCl, pH 8.5. A sample taken
after 40 min for RP-HPLC analysis showed the ligation to be
almost complete. After 2 h the reaction was quenched with
500 µL of acetic acid and purified by RP-HPLC. Fractions were
analyzed by ES-MS, and those containing the target peptide
were pooled and lyophilized. The ligated peptide ClAc-EPF-
(58-75)-G-ψ(CH2S)-G-(78-101) 16 (Mr 5025.0, calcd 5025.3,
5.0 µmol) was dissolved in 5 mL of 8 M urea saturated with
NaI (pH 3.5) to effect halide exchange. After 30 min RP-HPLC
and ES-MS analysis showed complete exchange, and the IAc-
EPF(58-75)-G-ψ(CH2S)-G-(78-101) 17 was isolated by RP-
HPLC to give 4.9 µmol of peptide (Mr 5117.1, calcd For
calcd 1257.6; ClAc-LPGKWKPKMI-[NH(CH2)2-SH]
9 Mr
1332.4, calcd 1334.1; BrAc-TKVVLDDKDYFLFRDGDILGKYVD
{BrAc-EPF(78-101)} 12 Mr 2956.4, calcd For C133H202N30O41
-
C
230H360N52O69SI 5116.7).
Br, 2957.2; ClAc-EIQPVSVKVGDKVLLPEY-[NH(CH2)2-SH]
Liga tion of ClAc-EP F (30-55)-[NH(CH2)2SH] 14 a n d
{ClAc-EPF(58-75)-[NH(CH2)2-SH]} 13 Mr 2149.5, calcd for
IAc-EP F (58-75)-G-ψ(CH2S)-G(78-101) 17. 17 (4.9 µmol)
and 14 (8.5 µmol) were dissolved in 4 mL of 8 M urea, pH 8.5.
The reaction was quenched with 100 µL of acetic acid after 3
h. RP-HPLC purification gave 4.8 µmol of ClAc-EPF(30-55)-
G-ψ(CH2S)-G-(58-75)-G-ψ(CH2S)-G-(78-101) 18 (Mr 7736.5,
calcd for C348H567N85O104S3Cl, 7737.5).
C
97H160N22O28SCl, 2149.9; ClAc-IMLPEKSQGKVLQATVVAV-
GSGSKGK-[NH(CH2)2-SH] {ClAc-EPF(30-55)-[NH(CH2)2-
SH]} 14 Mr 2749.4, calcd for C118H209N33O35S2Cl, 2749.6;
Ac-AGQAFRKFLPLFDRVLVERSAAETVTK-[NH(CH2)2-SH],
{Ac-EPF(1-27)-[NH(CH2)2-SH]} 15 Mr 3151.9, calcd for
C
143H233N40O38S, 3152.6 (all calcd weights are average iso-
Sim u lta n eou s Exch a n ge a n d Liga tion of Ac-EP F (1-
27)-[NH(CH2)2SH] 15 w ith ClAc-EP F (30-55)-G-ψ(CH2S)-
G-(58-75)-G-ψ(CH2S)-G-(78-101) 18. 15 (3.4 µmol) and 18
(2.3 µmol) were dissolved in 3 mL of 8 M urea, 100 mM Tris‚
HCl, 1 M KI, pH 9.5. RP-HPLC analysis after 30 min indicated
little reaction had occurred. KI, 2.0 g, was therefore added 60
min after the start of the reaction, and after a further 2 h RP-
HPLC analysis indicated that the reaction was almost com-
plete. After a total of 4 h the mixture was diluted with 10 mL
of 8 M urea, pH 8.5, and then quenched with 100 µL of acetic
acid. RP-HPLC purification yielded 1.0 µmol of ligated EPF:
Ac-EPF(1-27)-G-ψ(CH2S)-G-(30-55)-G-ψ(CH2S)-G-(58-75)-G-
ψ(CH2S)-G-(78-101) 19. The ligated protein was characterized
by RP-HPLC and ES-MS analysis (Mr 10853 ( 2, calcd for
tope masses).
Rea ction of LP GKWKP KMI-[NH(CH2)2SH] 2 a n d X-Ac-
IGGF . BrAc-IGGF 4 (0.72 µmol) was dissolved in 350 µL of 6
M Gu‚HCl, 100 mM Tris pH 8.5. The solution was added to
0.93 mg of 2 (0.74 µmol). Samples (20 µL) were withdrawn at
intervals, quenched with 80 µL of 20% TFA in water, and
analyzed by RP-HPLC. The experiment was repeated at pH
7.9. This experiment was also done using ClAc-IGGF 3 instead
of 4 at pH values of 7.9, 8.5, and 9.5.
Com p etition betw een (ClAc-IGGS 6 a n d Br Ac-IGGF
4) or (Br Ac-IGGF 4 a n d IAc-IGGS 8) for LP GKWKP KMI-
[NH(CH2)2SH] 2. 6 (1.95 µmol), 4 (1.84 µmol), and 2 (1.62
µmol) were dissolved in 900 µL of 6 M Gu‚HBr, 100 mM Tris
pH 8.5 to give an initial peptide concentration of ∼2 µM. At
intervals 200 µL samples were quenched in a mixture of 200
µL of 50% RP-HPLC solvent B and 200 µL of 50% TFA. The
quenched samples were analyzed by RP-HPLC (Vydac C18,
0.46 × 15 cm, 0-40% B over 40 min at 1 mL/min). Peaks from
RP-HPLC analyses were trapped and identified by ion spray
mass spectrometry. In similar experiments the competition
between 4 and 8 for 2 was studied at pH values of 8.5, 7.0,
and 6.0.
Rea ction of ClAc-LP GKWKP KMI-[NH(CH2)2SH] 9 w ith
Br Ac-IGGF 4. 4 (1.46 µmol) and 10 (1.39 µmol) were dissolved
in 700 µL of 6 M GuHBr, 100 mM Tris pH 8.5. Samples were
quenched after 1 and 2 h for RP-HPLC and ES-MS analysis
as described above.
Con ver sion of ClAc-IGGF 3 to IAc-IGGF 11. 3 (0.72
µmol) was dissolved in 340 µL of freshly prepared saturated
KI, 100 mM phosphate pH 7.0 (peptide concentration ap-
proximately 2 mM). During the course of the exchange reaction
C
491H798N125O142S4 10852.3).
Ack n ow led gm en t. This work was supported in part
by the Australian Research Grants commission. We
thank Drs. J ohn Wade and Les Miranda for helpful
comments in preparing the manuscript and Dianne
Alewood for preparing some of the schemes and figures.
We also thank Dr. J oel D. A. Tyndal and Dr. Nathan P.
Cowieson, Institute for Molecular Bioscience, for their
assistance with the cover art.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails. This material is available free of charge via the Internet
at http://pubs.acs.org.
J O016121E
5890 J . Org. Chem., Vol. 67, No. 17, 2002