Solid phase synthesis of peptide-selenoesters

Article abstract of DOI:10.1016/j.bmc.2013.03.076

The synthesis of proteins by native chemical ligation greatly enhances the application of chemistry to complex molecules such as proteins. The essential building blocks for this approach traditionally have been peptide-thioester segments that are linked chemoselectively in consecutive reactions. By using peptide selenoesters instead of thioesters, the ligation rate can be significantly accelerated permitting couplings at difficult sites and potentially enabling new ligation strategies. To facilitate the routine synthesis of selenoester peptides, a general and straightforward procedure has been developed that generates a suitably functionalized resin from which the desired selenoester peptide can be readily synthesized. This simple approach utilizes readily available and cheap chemical agents and enables production of peptide selenoesters of excellent quality in short time and with high recovery. In addition, the stability of peptide selenoesters was examined under different native chemical ligation conditions and compared to thioesters. Selenoesters are slightly more reactive and more susceptible to hydrolysis and aminolysis than thioesters but sufficiently stable under mildly acidic conditions (pH 6.5). Under these conditions, rapid selenoester-mediated ligation is kinetically favoured.

Full text of DOI:10.1016/j.bmc.2013.03.076

Bioorganic & Medicinal Chemistry 21 (2013) 3473–3478
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Solid phase synthesis of peptide-selenoesters
à
⇑
Artin Ghassemian , Xavier Vila-Farrés , Paul F. Alewood, Thomas Durek
Division of Chemistry and Structural Biology, Institute for Molecular Bioscience, The University of Queensland, St Lucia, Brisbane, Queensland 4072, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 6 April 2013
The synthesis of proteins by native chemical ligation greatly enhances the application of chemistry to
complex molecules such as proteins. The essential building blocks for this approach traditionally have
been peptide-thioester segments that are linked chemoselectively in consecutive reactions. By using pep-
tide selenoesters instead of thioesters, the ligation rate can be significantly accelerated permitting cou-
plings at difficult sites and potentially enabling new ligation strategies. To facilitate the routine
synthesis of selenoester peptides, a general and straightforward procedure has been developed that gen-
erates a suitably functionalized resin from which the desired selenoester peptide can be readily synthe-
sized. This simple approach utilizes readily available and cheap chemical agents and enables production
of peptide selenoesters of excellent quality in short time and with high recovery. In addition, the stability
of peptide selenoesters was examined under different native chemical ligation conditions and compared
to thioesters. Selenoesters are slightly more reactive and more susceptible to hydrolysis and aminolysis
than thioesters but sufficiently stable under mildly acidic conditions (pH 6.5). Under these conditions,
rapid selenoester-mediated ligation is kinetically favoured.
Keywords:
Native chemical ligation
Thioester
Selenoester
Hydrolysis
Solid phase peptide synthesis
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
Kent in 1994.2 The former allows linking of amino acid building
blocks in rapid and precise fashion yielding peptides of up to 50
Proteins are the essential ingredients of life, being involved in
virtually all known biological processes. Chemists were the first
to realize their potential and early on sought of ways to manipulate
and study these molecules with the ultimate aim to chemically
synthesize these highly complex biopolymers. Today, in the post-
genome era, the discovery of novel bioactive peptides and proteins
has assumed a breathtaking pace. Many of these peptides and
medium-sized proteins have therapeutic or diagnostic potential,
which necessitates reliable techniques to facilitate the validation
of their structure, detailed structure–function studies and their
production on large scale in a timely fashion. Many of these
requirements can only satisfactorily be achieved through chemical
means and our need for robust technologies that permit total
chemical access to large peptides and proteins is higher than ever
before. Two major developments in the last 50 years have signifi-
cantly contributed bringing us closer to these goals: the invention
of solid phase peptide synthesis by Merrifield in 19631 and the
conceptual development of so called orthogonal chemical ligation
techniques in the 80s and 90s of the last century which culminated
in the discovery of native chemical ligation (NCL) by Dawson and
amino acids in length. The latter discovery provides a practical
means to chemoselectively join these unprotected peptides to form
much longer polypeptides (>200 amino acids) thereby extending
chemical access from relatively small polypeptides to the realm
of full-fledged proteins and enzymes.
Of the many chemical ligation approaches developed, NCL is by
far the most widely used due to its robustness, versatility and
ability to yield a native peptide bond at the site of ligation.3,4
NCL is the reaction of an unprotected peptide carrying a C-terminal
a-thioester with another unprotected peptide equipped with an N-
terminal cysteine residue (Scheme 1). The nucleophilic sulfhydryl
group of the cysteine residue attacks the mildly activated carbonyl
carbon of the thioester moiety and replaces the thiol leaving group
effectively ‘capturing’ both peptide segments (transesterification).
The new thioester is short-lived and rapidly re-arranges through
an S?N acyl shift presumably involving a favorable five-mem-
bered transition state to form a native amide bond. In mechanistic
terms it is important to point out that the key to the exquisite
chemoselectivity lies in the reversibility of the first transesterifica-
tion step and the (near)5 irreversibility of the second acyl transfer
step. The former ensures that unproductive thioesters (e.g.,
thioesters involving non-ligation site cysteines) are short-lived
and in conjunction with the latter guarantees near quantitative
(typically >95%) reactions in short time (typically 2 h). NCL is
normally carried out in aqueous guanidine HCl under mild condi-
tions (neutral pH and room temperature) using thiol catalysts
⇑
Corresponding author. Tel.: +61 7 3346 2985; fax: +61 7 3346 2090.
E-mail address: thomas.durek@gmail.com (T. Durek).
Present address: McMaster University, Hamilton, ON, Canada L8S 4L8.
Present address: Barcelona Centre for International Health Research (CRESIB,
à
Hospital Clínic—Universitat de Barcelona), Barcelona, Spain.
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.03.076

3474
A. Ghassemian et al. / Bioorg. Med. Chem. 21 (2013) 3473–3478
advantages and limitations. We also compare the stability of sele-
no- and thioesters towards hydrolysis and aminolysis allowing a
critical evaluation of these compounds under typical NCL reaction
conditions.
2. Results and discussion
2.1. Synthesis of peptide selenoesters by solution phase
approaches
There are surprisingly few chemical approaches available for
the incorporation and intermediate protection of selenium into or-
ganic molecules, when compared to its chalcogen peers sulfur and
oxygen.15 Most approaches have focused on selenocysteine (Sec,
the ‘21st amino acid’), which plays an important role in enzymatic
redox pathways16 and also has some utility in Sec-mediated
NCL.9,17–19 Selenols have significantly lower pKa values than thiols
of similar structure, which makes them considerably more nucleo-
philic at neutral and acidic pH.20 In addition, their reduction poten-
tials are substantially lower (more negative) and are more rapidly
oxidized to diselenides than thiols to disulfides.21
In our initial attempt to generate peptide selenoesters,13 we
developed a solution phase approach in which unprotected peptide
thioesters were converted to selenoacids by treatment with NaHSe
at pH 7 followed by alkylation with alkyl halides at pH 4 to yield
the corresponding selenoesters (Scheme 2). This sequence has to
be carried out in one-pot under inert conditions as the intermedi-
ate selenoacids are highly unstable when exposed to air.22 While
this approach is compatible with a variety of functionalized amino
acids and long polypeptides,13 our continuing studies suggest that
a few amino acids (most likely Trp) and certain side chain protect-
ing groups (e.g., Cys(Fm)) can undergo substantial side reactions.
These side reactions primarily occur during the first reaction step
in the presence of the highly nucleophilic and reducing hydrogen-
selenide agent rather than during the alkylation step. While these
major side products were not fully characterized, it is evident that
novel approaches for the synthesis of selenoesters are needed.
We initially resorted to two well-established strategies that
have proven useful for the Fmoc SPPS of peptide thioesters: Daw-
son’s N-acyl-benzimidazolinone (Nbz) approach and Liu’s hydra-
zide method.23,24 Both approaches rely on relatively unreactive
precursors during chain assembly (o-aminoanilides and hydra-
zides, respectively) that can be activated under mild conditions
(to give Nbz and azides, respectively) and thiolysed with mercap-
tans yielding the corresponding peptide thioesters. We anticipated
that these precursors might be sufficiently activated towards dis-
placement with selenols. LYRAF-Nbz and LYRAF-azide were pre-
pared as described and treated with 30 mM diphenyldiselenide,
100 mM TCEP in 6 M GdmHCl, 0.1 M sodium phosphate buffer,
pH 7.0 under argon atmosphere. However, analysis of the reaction
mixture after 24 h indicated that only trace amounts of the ex-
pected LYRAF-[COSe]-phenyl ester were formed and the majority
of peptide was still in the precursor form. This suggests that selen-
ophenol is a comparatively poor nucleophile under these condi-
tions, in agreement with our earlier observations.13 This problem
is further compounded by the poor solubility of organic diselenides
in aqueous solution and the strong tendency of selenols to oxidize
rapidly and quantitatively, which make it difficult to achieve and
maintain high concentrations of free selenol nucleophile. Hence,
we next turned our attention to solid phase strategies to avoid
most of the limitations of the solution phase methods.
Scheme 1. Mechanism of native chemical ligation. P1 and P2 designate unpro-
tected polypeptide segments.
and strong reducing agents such as tris(2-carboxyethyl)phosphine
(TCEP) in order to prevent formation of unproductive thioesters
and disulfides, respectively. Many of the original limitations of
NCL, such as the requirement for a cysteine at the site of ligation
or chemical access to the essential peptide thioester segments by
more user-friendly and milder Fmoc SPPS protocols (rather than
the traditional Boc SPPS strategies)6 have been resolved in the past
and are covered by recent excellent reviews.7,8
However, of the few remaining challenges, the issue relating to
the dependence of the NCL reaction rate on the identity of the ami-
no acid adjacent to the thioester moiety has been poorly addressed
(Xaa in a peptidyl-Xaa-[COSR] + Cys-peptide ligation). Hackeng
et al. observed early, that b-branched amino acids such as Thr,
Val and Ile in this position react extremely slowly under standard
NCL reaction conditions (>48 h).6 Long NCL reaction times are gen-
erally discouraged, due to potential side-reactions (thioester
hydrolysis, desulfurization of cysteine and methionine oxidation)9
under the conditions employed, hence, ligations at these sites tra-
ditionally have been avoided. Furthermore, proline in this position
has been found to react even more sluggishly with, at best, conver-
sions of around 20% after 48 h.6,10,11 As demonstrated by Danishef-
sky and coworkers, the problem can be partially solved by
employing p-nitrophenyl oxo-esters (50% conversion after 15 h),
however, hydrolysis of the highly activated esters even at slightly
acidic pH values largely prevents quantitative conversions.12
By extending this line of thought of replacing sulfur with its
chalcogen cousins, we recently investigated peptidyl-selenoesters
as acyl donors in NCL reactions.13 We reasoned that a selenoate
would be a superior leaving group than the corresponding thiolate
and alkoxide due to the better polarizibility of selenium versus sul-
fur and oxygen.14 Detailed kinetic studies confirmed that selenoes-
ters are indeed superior acyl donors in NCL reactions with
observed rate enhancements of up to 350 times for ligations at pro-
line. Stereochemistry and chemoselectivity were maintained under
these conditions. Hence, this poorly studied compound class
appears to have great potential not only as a practical way to en-
able NCL at difficult sites, but also in the optimization of existing
chemical ligation strategies or the development of new ligation
chemistries.
2.2. Synthesis of peptide selenoesters by solid phase approaches
To enable the synthesis of selenoester peptides, a general sele-
noester-generating resin linker for Boc SPPS was developed
Here we outline two approaches for the solid phase chemical
synthesis of peptide selenoester building blocks and discuss their

A. Ghassemian et al. / Bioorg. Med. Chem. 21 (2013) 3473–3478
3475
Scheme 2. Synthetic strategies for the solution and solid-phase synthesis of peptide selenoesters.
(Scheme 2). This linker (HSe-CH2-CH2-CONH-CH(R)-COO-CH2-
PAM-Polystyrene) is a selenol variant of the thioester-generating
resin described by Hackeng et al.6 and employs the HF-labile
PAM linker.25 Starting from pre-loaded Ile-OCH2-PAM-resin,
3-bromo- or 3-iodopropionic acid was coupled by using symmetric
anhydride conditions and dicyclohexylcarbodiimide (DCC). Intro-
duction of the selenol functionality was achieved by two alterna-
tive procedures as shown in Scheme 2.
In our initial attempt lithium diselenide (Li2Se2) was generated
from the reduction of stoichiometric amounts of elemental sele-
nium with lithium triethylborohydride.26 The resulting mixture
was used to selenate the halogenated resin under inert conditions.
Diselenides were reduced with 1,4-dithiothreitol (DTT, in DMF)
and the selenol acylated with the first Boc-protected amino acid
building block using standard SPPS protocols with an extended
coupling time (1 h).27 Manual peptide synthesis by standard proto-
cols was then continued and the completed peptide selenoester
was cleaved from the resin with concomitant side chain deprotec-
tion using HF/p-cresol (9:1 (v/v)) at ꢀ5 °C. The crude peptide (LYR-
AF-[COSe-CH2-CH2-CO]-Ile) was recovered in good yield (70%
based on resin loading) and purity (Fig. 1a). The described ap-
proach facilitates generation of peptide selenoesters though it does
require the handling of toxic and malodorous selenide solution un-
der inert conditions and demands safety precautions for the poten-
tially hazardous hydrogen. In addition, the Li2Se2 solution has to be
freshly prepared every time, which may limit the attractiveness of
this approach.
During our studies we came across potassium selenocyanate as
a suitable alternative to Li2Se2/NaHSe for incorporation of the
selenol functionality.28,29 KSeCN is a stable salt, less odorous than
selenides, easily handled, cheap and the cyano group effectively
functions as an intermediate protecting group to prevent
undesired oxidation or substitution reactions.15,28,29 Substitution
of the halogenated resin was achieved by a modest excess of
KSeCN (5 equiv) in under 24 h at ambient temperature (45 °C)
and the cyano group removed by treatment with sodium borohy-
Fig. 1. HPLC and MS analysis of crude peptide selenoester LYRAF-[COSe-CH2-CH2-
CO]-Ile obtained by the Li2Se2 (A) or KSeCN (B) SPPS approaches. The insert shows
an ESI-MS of the crude peptide (calculated monoisotopic mass (most abundant
isotope composition): 918.4 Da) and the desired product is indicated by an asterisk.
HPLC analysis was done using an Agilent Zorbax 300SB-C18, 4.6 ꢁ 250 mm Column
and a gradient of 10–50% buffer B in buffer A over 35 min.
dride at 0 °C for 1 h. Acylation of the free selenol with the first
Boc-protected amino acid was then carried out as described above
and included a DTT reduction step in order to reverse any potential

3476
A. Ghassemian et al. / Bioorg. Med. Chem. 21 (2013) 3473–3478
diselenide formation during the washing steps. Test peptide LYR-
AF-[COSe-CH2-CH2-CO]-Ile was obtained in similar yields and
slightly higher purity when compared to the same peptide pre-
pared by the Li2Se2 approach (Fig. 1b).
difficult ligation sites as well as the genuine structure of the syn-
thetic selenoesters obtained by the described direct SPPS approach.
2.3. Evaluation of the stability of selenoesters versus thioesters
under NCL conditions
To demonstrate the applicability of this approach to the synthe-
sis of longer peptides, two medium-sized peptide selenoesters
corresponding to segments 1–13 and 14–30 of the spider toxin
PnTx2-6 from Phoneutria nigriventer were synthesized.30 This 48
amino acid toxin (also known as ‘spider-viagra’) contains ten cys-
teine residues, which should facilitate its synthesis by a number
of ligation strategies. However, the most practicable ligation sites
are exclusively ‘slow ligation sites’ (i.e., either T/I/V/P-C, six in to-
tal) making it an ideal target for selenoester-mediated NCL. Both
peptides were recovered in high yield (1–13: 83%; 14–30: 71% of
the theoretical yield based on resin loading) and purity following
HF cleavage (see Fig. 2). Thus, the described procedure is straight-
forward, simple, utilizes readily available and cheap chemical
agents and facilitates production of any desired peptide selenoest-
er in short time, high recovery and excellent quality.
Selenoesters have a higher reactivity than comparable thioes-
ters towards thiol nucleophiles in the first transesterification step
of NCL.13 One could reasonably argue that this higher reactivity
of selenoesters also extends towards other nucleophiles, for exam-
ple, amines and hydroxyl ions. This would have direct conse-
quences for their applicability in NCL, because such (undesired)
reactivity would affect chemoselectivity and/or stability towards
alkaline hydrolysis at even slightly elevated pH. To compare the
stability of selenoesters and thioesters under typical NCL condi-
tions, peptides LYRAF-[COS-CH2-CH2-CO]-Gly and LYRAF-[COSe-
CH2-CH2-CO]-Ile were incubated at a concentration of 1 mg/mL
(approximately 1.1 mM) in 6 M guanidine HCl, 200 mM Na2HPO4
which was adjusted to different pH values. A thiol or selenol cata-
lyst was not included, as this would give rise to transesterification
reactions that would complicate data analysis. Hence, the data
should be interpreted with caution as NCL is usually carried out
in the presence of aryl thiol catalysts that activate the poorly reac-
tive alkyl thioester. The resulting aryl thioester is typically short-
lived but also likely more prone to hydrolysis than the alkyl thio-
esters studied here. The mixtures were analyzed by HPLC and MS
and the fraction of remaining ester expressed as a function of time
(Fig. 3). Our data indicate that selenoesters at pH 7.0 and 7.5
decompose substantially with half-lives of 12 h and 7 h, respec-
tively. The hydrolyzed ester was the dominant product in both
cases, however small amounts of cyclic LYRAF and a guanidino
adduct were also detected, suggesting that direct aminolysis of
the selenoester may also play a role.31 In contrast, thioesters are
reasonably stable at pH 7.5. The minor hydrolysis observed, how-
ever, could be of significance if ligations are carried out over longer
periods (>24 h), suggesting ligations may best be carried out at
lower pH values. At a pH of 6.5 the selenoester was essentially as
stable as a thioester of similar structure at pH 8.0 (t1/2 in both cases
approximately 20 h).
PnTx2-6[14-30]-selenoester was ligated to peptides CFRANK
and PnTx2-6[31-48] (CRQGYFWIAWYKLANC(Acm)KK) using con-
ditions described previously.13 More than 80% conversion was
achieved within 2 h, thus demonstrating the utility of selenoest-
er-mediated NCL for the synthesis of longer polypeptides at
It has long been noted that oxoesters and analogous thioesters
have very similar reactivities towards hydroxide ion nucleophiles
Fig. 2. HPLC and MALDI-MS analysis of crude peptide selenoesters obtained using
the KSeCN approach. (A) crude Thz-DC(Acm)C(Acm)GERGEC(Acm)VC(Acm)GGPC
(Acm)I-[COSe-CH2-CH2-CO]-Ile after HF cleavage. The asterisk indicates the desired
product and the insert depicts a high-resolution MALDI-MS (calculated monoiso-
topic mass (most abundant isotope composition): 2319.78 Da). (B) ATC(Ac-
m)AGQDQPC(Acm)KET-[COSe-CH2-CH2-CO]-Ile after HF cleavage. The asterisk
indicates the desired product and the insert depicts a high-resolution MALDI-MS
(calculated monoisotopic mass (most abundant isotope composition): 1742.66 Da).
HPLC analysis was done using an Agilent Zorbax 300SB-C18, 4.6x250 mm column
and a gradient of 10-50% buffer B in buffer A over 40 min. The hash symbol
indicates a side product corresponding to the loss of one Acm group.
Fig. 3. Stability of seleno- and thioester-peptides under NCL conditions. Peptides
LYRAF-[COS-CH2-CH2-CO]-Gly (thioester) or LYRAF-[COSe-CH2-CH2-CO]-Ile (sele-
noester) were incubated in 6 M guanidine HCl, 200 mM Na2HPO4, adjusted to
different pH values and the mixtures analyzed by HPLC at the indicated times. Data
were fitted to a single exponential equation. Legend and calculated half-lifes (t1/2):
(d) thioester, pH 7.5, t1/2 ꢂ 20,000 h; (j) thioester, pH 8.0, t1/2 = 19 h; (4)
selenoester, pH 6.5, t1/2 = 22 h; (}) selenoester, pH 7.0, t1/2=12 h; (s) selenoester,
pH 7.5, t1/2=7 h.

A. Ghassemian et al. / Bioorg. Med. Chem. 21 (2013) 3473–3478
3477
during alkaline hydrolysis.32 Our results indicate that selenoesters
are more susceptible than thioesters towards base-mediated
hydrolysis under NCL conditions. This is in agreement with results
from Chu and Mautner, who found that selenoesters are hydro-
lyzed approximately 3–10 times faster than corresponding thioes-
ters across a wide pH range.33 Interestingly, the same report found
that the rate of hydrolysis is significantly accelerated by phosphate
ions, suggesting that under NCL conditions, selenoesters may be
stabilized by substituting the widely used phosphate buffer. More
importantly, however, these changes are relatively small on the
grand kinetic scale and similar to rate changes observed when
comparing thioesters to oxoesters.
In contrast, thioesters are much more reactive than oxoesters of
similar structure towards ‘softer’ nucleophiles such as amines and
thiols.32,34 This trend appears to continue in the chalcogen series
with selenoesters being substantially more reactive towards thiol13
and amine nucleophiles33 than thioesters. For example it was ob-
served that aminolysis of selenoesters is about 100 times faster
than the same reaction of thioesters under identical conditions
(oxoesters did not react).33 Hence, the greater susceptibility of
selenoesters to hydrolysis when compared to thioesters does not
diminish their value in NCL as long as a safe optimum pH, where
the rate of thiolysis far exceeds the rate of hydrolysis, is
maintained.35 Depending on the exact reaction conditions,
selenoester-mediated NCLs are typically quantitative within
minutes to a few hours, even at slightly acidic pH values and in
the worst-case scenario (Pro-Cys ligations).13 Thus, under most
conditions selenoester-mediated NCL will be kinetically favoured.
taneous side chain removal and cleavage of peptides from the dried
resins were carried out with 9 mL of HF and 1 mL p-cresol as scav-
enger for 1.5 hour at ꢀ5 °C. Cleaved peptides were precipitated and
washed twice with cold diethyl ether. The peptides were then dis-
solved in 50% acetonitrile, 0.1% TFA (v/v) in water and lyophilized.
Peptides were purified by preparative reverse-phase (RP) HPLC and
fractions containing the desired peptide were pooled, lyophilized
and stored at ꢀ20 °C.
4.2. Characterization of peptides
Peptides were characterized by analytical HPLC on a Shimadzu
Prominence systems using a solvent system of 0.05% TFA in water
(buffer A) and 90% acetonitrile, 0.043% TFA in water (buffer B) and
an Agilent Zorbax 300SB-C18, 4.6 ꢁ 250 mm column. Peptide pur-
ity and identity were assessed by ESI-MS on a API-2000 mass spec-
trometer (Applied Biosystems) and by MALDI-MS on a Applied
Biosystems 4700 TOF/TOF.
LYRAF-[COS-CH2-CH2-CO]-Gly Mfound: 814.3 Da; calculated
monoisotopic mass (most abundant isotope composition): 814.4
Da
CFRANK Mfound: 737.4 Da; calculated monoisotopic mass (most
abundant isotope composition): 737.3 Da.
LYRAF-[COSe-CH2-CH2-CO]-Ile Mfound: 918.3 Da; calculated
monoisotopic mass (most abundant isotope composition):
918.4 Da.
ATC(Acm)AGQDQPC(Acm)KET-[COSe-CH2-CH2-CO]-Ile Mfound
:
1742.79 Da; calculated monoisotopic mass (most abundant isotope
composition): 1742.66 Da.
Thz-DC(Acm)C(Acm)GERGEC(Acm)VC(Acm)GGP-C(Acm)I-[COSe-
3. Conclusions
CH2-CH2-CO]-Ile Mfound: 2319.95 Da; calculated monoisotopic
mass (most abundant isotope composition): 2319.78 Da.
Thz-DC(Acm)C(Acm)GERGEC(Acm)VC(Acm)GGP-
C(Acm)ICRQGYFWIAWYKLANC(Acm)KK Mfound: 4402.5 Da; calcu-
lated mass (average isotope composition): 4402.7 Da.
Thz-DC(Acm)C(Acm)GERGEC(Acm)VC(Acm)GGP-C(Acm)ICFRANK
In summary, herein we describe two high-yield strategies for
the efficient Boc-based synthesis of peptidyl selenoesters directly
on the solid phase. The use of KSeCN and Li2Se2 nucleophiles for
on-resin selenoester synthesis provides an alternative means to ac-
cess medium-length peptides through NCL reactions at difficult
sites that are otherwise impractical via traditional thioester liga-
tion chemistry. The solid phase approaches described overcome
several obstacles encountered in our solution-state synthesis, such
as considerable side reactions with protecting groups and Trp ami-
no acids. In addition, these protocols are quick, user friendly, and
result in excellent yields with high purity. Though they are more
prone to hydrolysis and aminolysis than thioesters, selenoesters
may be used for NCL in slightly acidic buffers for relatively short
ligation periods with minimal losses. The utility of selenoesters
in peptide ligation chemistry is a subfield in its infancy and wide-
spread utility awaits further development such as the development
of Fmoc-compatible techniques.
Mfound: 2790.6 Da; calculated mass (average isotope composition):
2790.8 Da.
4.3. Solid phase synthesis of peptide-selenoester: KSeCN
protocol
1.43 g of Boc-Ile-PAM resin (0.70 mmol/g, 1 mmol) was swollen
in DMF for 1 hour, after which the terminal Boc group was re-
moved with 2 ꢁ 1 min treatment with neat TFA. The resin was
flow-washed with DMF for 30 s, and neutralized with 2 ꢁ 1 min
treatments with 10% DIEA in DMF (v/v). The neutralized resin
was then flow washed for 30 s with DMF, then 30 s with DCM.
10 equiv of DCC (2.063 g, 10 mmol) were dissolved in 5 mL of
DCM and 20 equiv of 3-iodopropionic acid (4.0 g, 20 mmol) were
dissolved in ꢂ 6 mL of DCM. Both solutions were cooled to 0 °C
for 5 min before being combined and allowed to react at 0 °C for
30 min. The resulting solution containing 3-iodopropionic anhy-
dride was gravity filtered to remove the dicyclohexylurea precipi-
tate. The residue was then washed with 2 mL of DCM, filtered, and
pooled with the 3-iodopropionic anhydride filtrate. The pooled fil-
trates were then added to the deprotected neutralized Ile-PAM re-
sin and allowed to couple for 45 min. The resin was drained and
washed with DCM for 30 s. Quantitative ninhydrin test indicated
>99% coupling. The resin was then dried under vacuum. Yield:
1.49 g (new loading: 0.661 mmol/g, 0.984 mmol) yellowish resin.
550 mg (0.661 mmol/g, 0.363 mmol) of the dried acylated resin
was then swollen in 10 mL of anhydrous THF for 1 h. 5 equiv of
KSeCN (262 mg, 1.82 mmol) were added to the swollen resin in
THF. The mixture was then incubated under argon atmosphere at
45 °C and gently agitated for 22 h (magnetic stirring should be
4. Materials and methods
4.1. Peptide synthesis
All peptides were manually synthesized by step-wise Boc SPPS
using highly optimized in-situ neutralization protocols.27 Peptide
thioesters were synthesized on Trt-S-CH2-CH2-CONH-Xaa-O-CH2-
PAM resins as described previously.6 Peptide selenoesters were syn-
thesized on HSe-CH2-CH2-CONH-Xaa-O-CH2-PAM resins as de-
scribed below. Coupling efficiencies were measured using the
quantitative ninhydrin test.36 Boc-protected amino acids were used
with the following side chain protecting groups: D(OcHx); E(OcHX);
H(DNP); K(Cl-Z); N(Xan); Q(Xan); R(Tos); S(Bzl); T(Bzl); Y(BrZ). All
cysteine side-chains were protected with acetamidomethyl (Acm).
After chain assembly, the resin was washed with DMF and 1:1
DCM/MeOH and dried under vacuum prior to HF cleavage. Simul-

3478
A. Ghassemian et al. / Bioorg. Med. Chem. 21 (2013) 3473–3478
avoided as the stir bar will crush the resin which will later clog the
SPPS synthesis vessel frits). The resin was drained and washed
extensively with THF and DCM and dried under vacuum. Yield:
452 mg (new loading: 0.67 mmol/g, 0.303 mmol) of greyish-white
resin.
washed with DMF for 30 s, and reduced and coupled as described
for a second time. After double coupling of the first amino acid,
subsequent amino acids were coupled according to Boc in-situ
neutralization protocols.
The dried resin (450 mg, 0.3 mmol) was re-swollen in 8 mL of
THF for 1 h, cooled to 0 °C on an ice/water bath, and sodium boro-
hydride (37.83 mg, 1 mmol) in 95% (v/v) EtOH/H2O (1 mL) was
added in one portion. The reaction was placed at 0 °C for 1 h with
occasional agitation, drained, and washed with THF, DCM, and
DMF.
Prior to coupling of the first amino acid, the resin was treated
with 3 mmol 1,4-dithiothreitol (DTT) in 6 mL DMF for 10 min.
The resin was then drained, and without washing, a 10 equiv
pre-activated mixture of Boc-Xaa-OH (3 mmol), HATU (1140 mg,
3 mmol), and 1.038 mL DIEA (6 mmol) in 6 mL of DMF was added
and allowed to couple for 1 h. The resin was then drained and
washed with DMF, and the reduction and coupling steps were per-
formed a second time. The resin was washed with DMF and step-
wise synthesis was subsequently carried out using standard Boc
in-situ neutralization protocols as described previously.
4.5. Stability assay
A 1 mg/mL solution of H-Leu-Tyr-Arg-Ala-Phe-[COS-CH2-CH2-
CO]-Gly-OH or H-Leu-Tyr-Arg-Ala-Phe-[COSe-CH2-CH2-CO]-Ile-
OH was prepared by dissolving 1 mg of peptide in 1 mL of ligation
buffer (6 M guanidine HCl, 200 mM Na2HPO4, adjusted with 6 M
HCl or NaOH to the indicated pH). The dissolved peptidyl esters
were incubated for a total of 13.4 h at 22 °C and 10 ll aliquots were
removed every 40 min and analyzed by RP-HPLC using a gradient
of 20–40% of buffer B (90% acetonitrile, 0.043% TFA in water) in
buffer A (0.05% TFA in water) over 40 min (Column: Agilent Zorbax
300SB-C18, 4.6 ꢁ 250 mm column). Peak areas were integrated
using the Shimadzu Labsolutions software package and normalized
to 100% at reaction time t = 0. Data were fitted to a single exponen-
tial decay curve.
Acknowledgments
4.4. Solid phase synthesis of peptide-selenoester: Li2Se2 protocol
This work was supported by a National Health and Medical Re-
search Council (NHMRC) Program Grant (569927 to PFA) and by
The University of Queensland.
1.43 g of Boc-Ile-PAM resin (0.70 mmol/g, 1 mmol) was swollen
in DMF for 1 h, after which Boc groups were cleaved with
2 ꢁ 1 min treatment with neat TFA, flow washed with DMF for
30 s, and neutralized with 2 ꢁ 1 min treatments with 10% DIEA
in DMF (v/v). The neutralized resin was then flow washed for
30 s with DMF, then 30 s with DCM. The resin was then stored in
DCM. 15 equiv of DCC (3.09 g, 15 mmol) and 30 equiv of 3-
bromopropionic acid (4.59 g, 30 mmol) were each dissolved in
10 mL DCM and cooled at 0 °C for 5 min before being combined.
The combined solution was then allowed to react at 0 °C for 1 h.
The resulting solution containing 3-bromopropionic anhydride
was then gravity filtered to remove the dicyclohexylurea precipi-
tate. The residue was then washed with 2 mL of DCM and the fil-
trates pooled. The combined filtrates were added to the
deprotected and neutralized Ile-PAM resin and allowed to couple
for 45 min. The resin was drained and washed with DCM for
30 s. Quantitative ninhydrin test indicated more than 98% coupling
efficiency and the resin was dried under vacuum.
References and notes
1. Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149.
2. Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. Science 1994, 266, 776.
3. Pattabiraman, V. R.; Bode, J. W. Nature 2011, 480, 471.
4. Macmillan, D. Angew. Chem., Int. Ed. 2006, 45, 7668.
5. Macmillan, D.; Adams, A.; Premdjee, B. Isr. J. Chem. 2011, 51, 885.
6. Hackeng, T. M.; Griffin, J. H.; Dawson, P. E. Proc. Natl. Acad. Sci. U.S.A. 1999, 96,
10068.
7. Dawson, P. E. Isr. J. Chem. 2011, 51, 862.
8. Mende, F.; Seitz, O. Angew. Chem., Int. Ed. 2011, 50, 1232.
9. Metanis, N.; Keinan, E.; Dawson, P. E. Angew. Chem., Int. Ed. 2010, 49, 7049.
10. Pollock, S. B.; Kent, S. B. Chem. Commun. 2011, 2342.
11. Townsend, S. D.; Tan, Z. P.; Dong, S. W.; Shang, S. Y.; Brailsford, J. A.;
Danishefsky, S. J. J. Am. Chem. Soc. 2012, 134, 3912.
12. Wan, Q.; Chen, J.; Yuan, Y.; Danishefsky, S. J. J. Am. Chem. Soc. 2008, 130, 15814.
13. Durek, T.; Alewood, P. F. Angew. Chem., Int. Ed. 2011, 50, 12042.
14. Steinmann, D.; Nauser, T.; Koppenol, W. H. J. Org. Chem. 2010, 75, 6696.
15. Flemer, S. Molecules 2011, 16, 3232.
The acylated resin was then swollen in 30 mL dry THF under ar-
gon for 1 h, then cooled to ꢀ78 °C prior to selenization with Li2Se2.
Li2Se2 (6 mmol) was prepared by suspending elemental selenium
powder (480 mg, 6.1 mmol) in 60 mL of dry THF under argon and
slowly adding 1 equiv of LiEt3BH (1 M in dry THF, 6 mL, 6 mmol).
When hydrogen evolution stopped, the solution was refluxed un-
der argon atmosphere for 30 min. The solution was then allowed
to cool to room temperature and 5 equiv (55 mL, 5 mmol) were
added to the swollen resin in THF at ꢀ78 °C under argon. The mix-
ture was allowed to stir at ꢀ78 °C for 30 min, then allowed to stir
at room temperature for 2 h. The resin was then transferred to a
peptide synthesis reaction vessel, drained, washed once with
THF, then washed thoroughly with DMF and stored in DMF under
argon.
16. Muttenthaler, M.; Alewood, P. F. J. Pept. Sci. 2008, 14, 1223.
17. Gieselman, M. D.; Xie, L. L.; van der Donk, W. A. Org. Lett. 2001, 3, 1331.
18. Hondal, R. J.; Nilsson, B. L.; Raines, R. T. J. Am. Chem. Soc. 2001, 123, 5140.
19. McGrath, N. A.; Raines, R. T. Acc. Chem. Res. 2011, 44, 752.
20. Huber, R. E.; Criddle, R. S. Arch. Biochem. Biophys. 1967, 122, 164.
21. Beld, J.; Woycechowsky, K. J.; Hilvert, D. Biochemistry 2007, 46, 5382.
22. Castano, J. A. G.; Romano, R. M.; Beckers, H.; Willner, H.; Boese, R.; Della
Vedova, C. O. Angew. Chem., Int. Ed. 2008, 47, 10114.
23. Blanco-Canosa, J. B.; Dawson, P. E. Angew. Chem., Int. Ed. 2008, 47, 6851.
24. Fang, G. M.; Li, Y. M.; Shen, F.; Huang, Y. C.; Li, J. B.; Lin, Y.; Cui, H. K.; Liu, L.
Angew. Chem., Int. Ed. 2011, 50, 7645.
25. Mitchell,A.R.;Kent,S.B.H.;Engelhard,M.;Merrifield,R.B.J.Org.Chem.1978,43,2845.
26. Gladysz, J. A.; Hornby, J. L.; Garbe, J. E. J. Org. Chem. 1978, 43, 1204.
27. Schnölzer, M.; Alewood, P.; Jones, A.; Alewood, D.; Kent, S. B. Int. J. Pept. Protein
Res. 1992, 40, 180.
28. Vanende, D.; Krief, A. Tetrahedron Lett. 1975, 2709.
29. Ie, Y.; Hirose, T.; Yao, A.; Yamada, T.; Takagi, N.; Kawai, M.; Aso, Y. PCCP 2009,
11, 4949.
Approximately 0.1 mmol of selenized resin was transferred to
another reaction vessel in DMF and put under argon. Immediately
prior to coupling of the first Boc-amino acid, the selenized resin
was reduced with 5 equiv of 0.5 M DTT in DMF (1 mL, 0.5 mmol)
for 10 min. The resin was then drained, and without washing, a
10 equiv pre-activated mixture of Boc-Xaa-OH (1 mmol), HATU
30. Nunes, K. P.; Costa-Goncalves, A.; Lanza, L. F.; Cortes, S. F.; Cordeiro, M. N.;
Richardson, M.; Pimenta, A. M. C.; Webb, R. C.; Leite, R.; De Lima, M. E. Toxicon
2008, 51, 1197.
31. Gates, Z. P.; Stephan, J. R.; Lee, D. J.; Kent, S. B. Chem. Commun. 2012, 786.
32. Connors, K. A.; Bender, M. L. J. Org. Chem. 1961, 26, 2498.
33. Chu, S. H.; Mautner, H. G. J. Org. Chem. 1966, 31, 308.
34. Yang, W.; Drueckhammer, D. G. Org. Lett. 2000, 2, 4133.
35. Bracher, P. J.; Snyder, P. W.; Bohall, B. R.; Whitesides, G. M. Orig. Life Evol.
Biosph. 2011, 41, 399.
(380 mg, 1 mmol), and 260
lL DIEA in 2 mL DMF was added and
allowed to couple for 1 h. The resin was then drained and flow
36. Sarin, V. K.; Kent, S. B.; Tam, J. P.; Merrifield, R. B. Anal. Biochem. 1981, 117, 147.