Synthesis and utility of β-selenol-phenylalanine for native chemical ligation-deselenization chemistry

Article abstract of DOI:10.1021/ol3012265

An efficient synthetic route to a suitably protected β-selenol- phenylalanine derivative from commercially available Garner's aldehyde is described. The incorporation of this building block into peptides and its application in native chemical ligation reactions with peptide thioesters are demonstrated. Ligation products were chemoselectively deselenized (including in the presence of unprotected cysteine residues) to provide native peptides.

Full text of DOI:10.1021/ol3012265

ORGANIC
LETTERS
2012
Vol. 14, No. 12
3142–3145
Synthesis and Utility of β-Selenol-
Phenylalanine for Native Chemical
LigationꢀDeselenization Chemistry
Lara R. Malins and Richard J. Payne*
School of Chemistry, The University of Sydney, NSW 2006, Australia
richard.payne@sydney.edu.au
Received May 5, 2012
ABSTRACT
An efficient synthetic route to a suitably protected β-selenol-phenylalanine derivative from commercially available Garner’s aldehyde is
described. The incorporation of this building block into peptides and its application in native chemical ligation reactions with peptide thioesters
are demonstrated. Ligation products were chemoselectively deselenized (including in the presence of unprotected cysteine residues) to provide
native peptides.
Native chemical ligation is the most widely used method
for the chemoselective assembly of unprotected peptide
fragments to afford polypeptides and proteins.1 The meth-
od relies on the reaction of a peptide bearing an N-terminal
cysteine (Cys) residue with a peptide possessing a C-term-
inal thioester (or more recently a selenoester2) moiety. The
reaction involves a reversible thioesterification step fol-
lowed by a rapid intramolecular S to N acyl shift to
generate the native peptide bond in high yield and under
mild reactionconditions. Thepowerofthismethodologyis
exemplified by its application in the synthesis of hundreds
of proteins to date.1b,3
provide an alanine (Ala) residue at the ligation junction.4
In the same report, the authors proposed the concept of
further expanding native chemical ligation to any thiol-
derivatized proteinogenic amino acid at the N-terminus of
peptide fragments through the use of ligationꢀdesulfurization
chemistry.4,5
This concept has now been expanded through the use of
β-, γ-, or δ-thiol amino acids.6ꢀ12 Although much progress
has been made to maximize the scope of native chemical
(5) (a) Rohde, H.; Seitz, O. Biopolymers 2010, 94, 551. (b) Dawson,
P. E. Isr. J. Chem. 2011, 51, 862.
The requirement for a Cys residue at the N-terminus of a
peptide fragment has prompted significant research effort
directedtowardextending the repertoire of nativechemical
ligation-based transformations to other amino acids. This
concept was fuelled by the work of Dawson and co-work-
ers who demonstrated that peptides and proteins produced
via native chemical ligation could be desulfurized to
(6) (a) Crich, D.; Banerjee, A. J. Am. Chem. Soc. 2007, 129, 10064. (b)
Botti, P.; Tchertchian, S. WO/2006/133962.
(7) (a) Chen, J.; Wan, Q.; Yuan, Y.; Zhu, J.; Danishefsky, S. J.
Angew. Chem., Int. Ed. 2008, 47, 8521. (b) Haase, C.; Rohde, H.; Seitz,
O. Angew. Chem., Int. Ed. 2008, 47, 6807.
(8) (a) Yang, R. L.; Pasunooti, K. K.; Li, F. P.; Liu, X. W.; Liu, C. F.
J. Am. Chem. Soc. 2009, 131, 13592. (b) Kumar, K. S. A.; Haj-Yahya,
M.; Olschewski, D.; Lashuel, H. A.; Brik, A. Angew. Chem., Int. Ed.
2009, 48, 8090.
(9) (a) Harpaz, Z.; Siman, P.; Kumar, K. S. A.; Brik, A. ChemBio-
Chem 2010, 11, 1232. (b) Tan, Z. P.; Shang, S. Y.; Danishefsky, S. J.
Angew. Chem., Int. Ed. 2010, 49, 9500.
(1) (a) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H.
Science 1994, 266, 776. (b) Dawson, P. E.; Kent, S. B. H. Annu. Rev.
Biochem. 2000, 69, 923. (c) Hackeng, T. M.; Griffin, J. H.; Dawson, P. E.
Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 10068.
(2) Durek, T.; Alewood, P. F. Angew. Chem., Int. Ed. 2011, 50, 12042.
(3) Kent, S. B. H. Chem. Soc. Rev. 2009, 38, 338.
(10) Chen, J.; Wang, P.; Zhu, J. L.; Wan, Q.; Danishefsky, S. J.
Tetrahedron 2010, 66, 2277.
(11) (a) Shang, S. Y.; Tan, Z. P.; Dong, S. W.; Danishefsky, S. J.
J. Am. Chem. Soc. 2011, 133, 10784. (b) Townsend, S. D.; Tan, Z.; Dong,
S.; Shang, S.; Brailsford, J. A.; Danishefsky, S. J. J. Am. Chem. Soc.
2012, 134, 3912.
(4) Yan, L. Z.; Dawson, P. E. J. Am. Chem. Soc. 2001, 123, 526.
(12) Siman, P.; Karthikeyan, S. V.; Brik, A. Org. Lett. 2012, 14, 1520.
r
10.1021/ol3012265
Published on Web 05/29/2012
2012 American Chemical Society

ligationꢀdesulfurization chemistry, the various conditions
(reductive or radical) used to effect desulfurization are not
selective in the presenceofotherCys residues inthe ligation
product which are concomitantly converted to Ala. This
unwanted side reaction can be prevented by global protec-
tion of Cys side chains in the sequence;13 however, this
prevents the use of biologically expressed peptide and
protein fragments and the use of expressed protein ligation
(EPL) methodologies.14 A recent report from Dawson and
co-workers elegantly circumventsthisissue by demonstrat-
ing that peptides obtainable from selenocysteine (Sec)-
mediated native chemical ligation reactions15 can be chemo-
selectively deselenized with tris(2-carboxyethyl)phosphine
(TCEP) and dithiothreitol (DTT) to afford an Ala residue
without affecting unprotected Cys residues.16
reaction enabled resolution of the two diastereoisomers, as
only the anti-mesylate proved to be competent in the SN2
process. This exclusively provided the desired syn-seleno-
cyanate, which, based on prior work on β-thiol leucine
ligations,9b was predicted to facilitate more facile peptide
ligations than the corresponding β-epimer. It is worth
noting, however, that an oxidation/reduction protocol
could be used to obtain anti-enriched 3 (1:8 syn/anti),
thereby improving the overall yield of the inversion to
provide syn-diastereoisomer 4 in 55% yield over two steps
(see Supporting Information (SI) for details). Acidic cleav-
age of the hemiaminal protecting group followed by
oxidation with pyridinium dichromate (PDC), yielded
carboxylic acid 6 in 57% yield. Importantly, under these
reaction conditions, no selenium oxidation byproduct was
observed. Reduction of the selenocyanate functionality with
sodium borohydride followed by protection afforded the
corresponding p-methoxybenzyl (PMB) protected β-selenol
building block 1, ready for incorporation into peptides.
Scheme 1. β-Selenol-Mediated LigationꢀDeselenization
Scheme 2. Synthesis of β-Selenol-Phe Building Block 1 from
Garner’s Aldehyde 2
Given the potential utility of this chemoselective desele-
nization methodology, we envisaged the use of unnatural,
β-selenol amino acid derivatives (in a similar manner to
β-thiol amino acids) so as to expand the number of
available ligation sites by taking advantage of a ligation-
deselenization theme (Scheme 1). Recently, this type of
methodology was shown to be achievable at 4-selenopro-
line residues at the N-terminus of peptide fragments.11b
Herein, we report the preparation of the first β-selenol-
derived amino acid, namely the suitably protected
β-selenol-Phe building block 1. In addition, we demon-
strate the utility of this building block in selenol-mediated
native chemical ligationꢀdeselenization reactions. Parti-
cular emphasis is placed on the compatibility of this
methodology with unprotected Cys residues.
Synthesis of 1 began with a Grignard addition of
bromobenzene to readily available Garner’s aldehyde 2
which provided 3 as an inseparable mixture of diastereo-
isomers (2:3 syn/anti) in 80% yield (Scheme 2). Anticipating
the potential for significantly different relative reactivities
of the anti- and syn-diastereoisomers, we chose to progress
this mixture without further attempts to optimize the
selectivity of the Grignard addition. As such, activation
of alcohol 3 as the corresponding mesylate, followed by
introduction of the crucial selenium moiety via SN2displace-
ment with potassium selenocyanate, afforded compound 4
in 29% yield over the two steps. Importantly, this inversion
Compound 1 was subsequently incorporated at the
N-terminus of model hexapeptides using Fmoc-strategy
solid-phase peptide synthesis (SPPS) starting from Rink
amide resin (Scheme 3). Following Fmoc-SPPS, cleavage
from the resin and purification, peptide 7 was isolated in
52% yield (see SI for synthetic details). The protected
seleno-peptide was then treated with 2,20-dithiobis(5-
nitropyridine) (DTNP)17 in trifluoroacetic acid (TFA) to
afford the symmetrical diselenide 8 in 88% yield. A variety
of C-terminal peptide thioesters (Ac-LYRANX-S(CH2)2-
CO2Et, X = Gly, Ala, Met, Phe, Val) were also prepared,
as previously described,18 thus enabling the scope of the
proposed ligation reaction to be investigated.
With β-selenol peptide 8 and an array of peptide thio-
esters now in hand, we next turned our attention to ligation
reactions which we investigated under native chemical
ligation conditions. After screening a variety of conditions
and thiol additives, optimal results were obtained using
(13) (a) Pentelute, B. L.; Kent, S. B. H. Org. Lett. 2007, 9, 687. (b)
Yang, Y.-Y.; Ficht, S.; Brik, A.; Wong, C.-H. J. Am. Chem. Soc. 2007,
129, 7690.
(14) (a) Muir, T. W.; Sondhi, D.; Cole, P. A. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 6705. (b) Muir, T. W. Annu. Rev. Biochem. 2003, 72, 249.
(15) (a) Hondal, R. J.; Nilsson, B. L.; Raines, R. T. J. Am. Chem. Soc.
2001, 123, 5140. (b) Quaderer, R.; Hilvert, D. Chem. Commun. 2002,
2620. (c) Gieselman, M. D.; Xie, L.; van Der Donk, W. A. Org. Lett.
2001, 3, 1331.
ligation buffer [6 M Gn HCl, 100 mM Na2HPO4, 5 mM
3
(17) Harris, K. M.; Flemer, S.; Hondal, R. J. J. Pept. Sci. 2007, 13, 81.
(18) (a) Ficht, S.; Payne, R. J.; Guy, R. T.; Wong, C.-H. Chem.;Eur.
J. 2008, 14, 3620. (b) Thomas, G. L.; Hsieh, Y. S. Y.; Chun, C. K. Y.;
Cai, Z. L.; Reimers, J. R.; Payne, R. J. Org. Lett. 2011, 13, 4770.
(16) Metanis, N.; Keinan, E.; Dawson, P. E. Angew. Chem., Int. Ed.
2010, 49, 7049.
Org. Lett., Vol. 14, No. 12, 2012
3143

Table 1. Scope of β-Selenol-Phe LigationꢀDeselenizations
Scheme 3. Synthesis of Model Diselenide Peptide 8
ligationꢀ
thioester
peptide
ligation
yielda
deselenization
yielda
entry
(X =)
(Z =)
1
2
3
4
5
6
7
Gly
Ala
Met
Phe
Val
8: Ser
8: Ser
8: Ser
8: Ser
8: Ser
9: Lys
9: Lys
56%b
53%b
51%b
53%b
21%c
52%b
54%d
35%e,g
38%e,g
47%f
43%f
21%f
N.D.
with respect to seleno-peptide 8] in the presence of 200 mM
4-mercaptophenylacetic acid (MPAA) at room tempera-
ture and a final pH of 7.0ꢀ7.3, a slight modification of
conditions used by Dawson and co-workers for Sec
ligations.16 Gratifyingly, the first ligation of peptide 8 with
a peptide thioester bearing a C-terminal Gly residue (entry 1,
Table 1) proceeded to completion in less than 24 h and in
56% isolatedyield, comparable topreviouslyreportedSec-
mediated ligations.15,16 Under these reaction conditions,
the ligation product was obtained as a mixture of the di-
selenidedimerand theselenyl-MPAAsulfideadductwhich
were separated using reversed-phase HPLC (see SI). Liga-
tions with peptide thioesters bearing C-terminal Ala, Met,
and Phe (entries 2ꢀ4, Table 1) also proceeded to comple-
tion within 24 h, and the desired products were isolated in
51ꢀ53% yields. Unfortunately, ligation of 8 with a peptide
thioester bearing a sterically encumbered Val residue led to
poor conversion to the desired ligation product after 96 h
(21%, see entry 5, Table 1). The prolonged ligation times
are likely a result of the low steady state concentration of
free selenol present in solution,15c owing to the stability
of the diselenide and use of a relatively mild reductant
(MPAA) in the ligation reaction.16 The chemoselectivity of
the ligation in the presence of an unprotected Lys residue
was also demonstrated by the ligation of a modified
N-terminal peptide (β-Se-FSPGYK-NH2 dimer, 9) with
two model thioesters (Entries 6 and 7, Table 1) which
provided the desired ligation products with no direct
aminolysis byproduct detected.19
After successfully demonstrating the feasibility and
scope of β-selenol-Phe mediated ligations, we subjected
our recombined mixture of purified ligation products to
deselenization using the optimized conditions reported
previously (9.4 equiv of DTT, 2.9 equiv of TCEP in
0.1 M phosphate buffer, pH = 5.1).16 In our hands, how-
ever, these conditions afforded the desired deselenization
product along with a significant quantity (∼50%) of two
diastereomeric hydroxylated byproducts, consistent with
the conversion of the enantiopure β-selenol-Phe residue
within the peptide ligation products into a mixture of
the corresponding β-hydroxylated compounds (see SI).
Dawson and co-workers have also reported a minor
byproduct upon deselenization in the presence of excess
Gly
Phe
N.D.
a Isolated yields; reaction conditions: ligation: buffer (6 M Gn HCl,
3
100 mM Na2HPO4, 5 mM with respect to seleno-peptide fragment 8 or 9),
200 mM MPAA, rt, pH 7.0ꢀ7.3. b t = 24 h. c t = 96 h. d t = 48 h; one-pot
deselenization: addition of TCEP (42 equiv) and DTT (4.9 equiv) to
ligation mixture. e t = 48 h. f t = 24 h. g Additional TCEP (42 equiv) and
DTT (4.9 equiv) were added after 24 h.
TCEP that is consistent with Sec to Ser conversion.16 The
relative prevalence of the hydroxylation pathway in our
system is likely due to the enhanced stability of the benzylic
radical formed in the course of the deselenization of
β-selenol-Phe over the alanyl radical formed in the corre-
sponding deselenization of Sec residues.16
Given the difficulties associated with purification of the
ligation product mixture and in an attempt to suppress
the unwanted hydroxylation pathway, we were interested
in exploring a one-pot ligationꢀdeselenization protocol
which would avoid any intermediarypurification steps and
hopefully provide further insight into the intricacies of the
deselenization process. To this end, ligation reactions with
N-terminal peptide 8 and a range of thioesters were carried
out using the conditions previously described and sub-
jected directly (without purification) to deselenization
with DTT (4 equiv) and excess TCEP (40 equiv) (Table 1).
The deselenization reactionswere relativelyslow, requiring
24ꢀ48 h and, in some cases, additional dosing with DTT
and TCEP to reach completion (see SI). As previously
suggested,16,20 thesluggishreactionratesmaybe attributed
to the large excess (200 mM) of MPAA present in the
ligation buffer, which acts as a competitive radical scaven-
ger. Nonetheless, we were pleased to find that the one-pot
ligationꢀdeselenization protocol afforded the desired pep-
tide products in an average of 60ꢀ70% yield per step
(35ꢀ47% over two steps). Most notably, the hydroxyla-
tion pathway was almost entirely suppressed (<5% hy-
droxylated byproduct observed) likely a positive effect of
the excess MPAA, which may also act as a suitable
hydrogen donor to rapidly trap the reactive benzylic
radical and prevent unwanted side reactions.21
(19) Payne, R. J.; Ficht, S.; Greenberg, W. A.; Wong, C.-H. Angew.
Chem., Int. Ed. 2008, 47, 4411.
(20) Siman, P.; Blatt, O.; Moyal, T.; Danieli, T.; Lebendiker, M.;
Lashuel, H. A.; Friedler, A.; Brik, A. ChemBioChem 2011, 12, 1097.
3144
Org. Lett., Vol. 14, No. 12, 2012

Table 2. One-Pot LigationꢀDeselenization in the Presence
of Cys
ligationꢀ
deselenization
entry
peptide A
peptide B
yielda
1
2
3
SPGYC (10)
SPGYS (8)
SPGYS (8)
LYRANG
LYRCNG
LYRCNM
52%
61%
48%
a Isolated yields; reaction conditions: ligation: buffer (6 M Gn HCl,
3
100 mM Na2HPO4, 5 mM with respect to seleno-peptide fragments),
200 mM MPAA, rt, pH 7.0ꢀ7.3, 48 h; one-pot deselenization: addition
of TCEP (42 equiv), DTT (4.9 equiv) to ligation mixture, 48 h.
Figure 1. (A) Representative crude reaction mixture of ligation
between peptide 10 and Ac-LYRANG-S(CH2)2CO2Et; peak
a = unreacted 10 [M þ H]þ = 750.5; b = hydrolyzed thioester
[M þ H]þ = 735.7; c = MPAA-thioester [M þ H]þ = 885.9;
d = ligation product [M þ H]þ = 1467.5, [M þ 2H]2þ = 734.45.
(B) One-pot deselenization; e = SedP[(CH2)2CO2H]3 [M þ H]þ =
331.2 ; f = hydroxylated byproducts [M þ 2H]2þ = 703.5;
Finally, to evaluate whether the modified conditions
used to effect deselenization in our one-pot ligationꢀ
deselenization protocol were chemoselective in the
presence of unprotected Cys residues, we synthesized
an alternative peptide 10 bearing both a Cys residue
and the β-selenol-Phe building block, as well as two Cys-
containing thioester fragments (Ac-LYRCNX-S(CH2)2-
CO2Et, X = Gly, Met), and subjected these compounds
to the ligationꢀdeselenization cascade (Table 2). Ligation
reactions required slightly longer reaction times to reach
completion (∼48 h), owing to the formation of a stable
intramolecular selenyl-sulfide bond (entry 1, Table 2),
and unproductive thioester formation with the unpro-
tected Cys residues (entries 2 and 3, Table 2). None-
theless, we were pleased to find that, in all cases, the
two-step process afforded the desired peptide products
bearing a native Phe residue at the ligation junction in
good yields (48ꢀ61% over two steps) without affecting
the internal, unprotected Cys residues (Figure 1). This
represents the first study of β-selenol-mediated ligations
at an amino acid other than Sec and suggests that this
chemistry will be amenable to other β-selenol amino
acids. The chemoselectivity of the deselenization protocol
in the presence of free Cys residues has important im-
plications for protein synthesis via selenol-based ligation
techniques.
g = deselenization product [M þ H]þ = 1389.65, [M þ 2H]2þ
=
695.55.
In summary, we have developed a novel synthetic route
to β-selenol-Phe 1 that can be readily modified to enable
access to a range of β-selenol (and β-thiol) amino acid
building blocks for use in native chemical ligationꢀ
deselenization chemistry. We have successfully utilized this
building block in ligation reactions with a variety of pep-
tide thioesters and have demonstrated a one-pot ligationꢀ
deselenization cascade that is efficient and chemoselec-
tive in the presence of unprotected Cys residues, thereby
expanding the repertoire ofligationmethodologies. Future
work in our laboratory will focus on the synthesis of
additional β-selenol amino acids and their utility in peptide
ligation chemistry and protein synthesis.
Acknowledgment. We acknowledge the University of
Sydney and the International Postgraduate Research
Scholarship Scheme (L.R.M.).
Supporting Information Available. Detailed experimen-
tal procedures, analytical HPLC traces and characteriza-
tion data. This material is available free of charge via the
Internet at http://pubs.acs.org.
(21) Crich and coworkers previously reported that the inclusion
of benzeneselenol, a suitable hydrogen donor, suppressed unwanted
rearrangements of radical alkyl intermediates. Crich, D.; Yao, Q. W.
J. Org. Chem. 1995, 60, 84.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 12, 2012
3145