The mercaptomethyl group facilitates an efficient one-pot ligation at Xaa-Ser/Thr for (Glyco)peptide synthesis

Article abstract of DOI:10.1002/anie.201000384

Going native: A mercaptomethyl group on the side-chain hydroxy group of serine and threonine residues facilitates a native chemical ligation reaction at the Xaa-Ser/Thr site (see scheme; R=H or Me). The intermediate thioester is treated to achieve an S- t

Full text of DOI:10.1002/anie.201000384

Communications
DOI: 10.1002/anie.201000384
Glycopeptide Synthesis
The Mercaptomethyl Group Facilitates an Efficient One-Pot Ligation
at Xaa-Ser/Thr for (Glyco)peptide Synthesis**
Hironobu Hojo,* Chinatsu Ozawa, Hidekazu Katayama, Akiharu Ueki, Yuko Nakahara, and
Yoshiaki Nakahara*
The native chemical ligation (NCL) method is generally used
for the synthesis of (glyco)proteins.[1] It efficiently condenses
peptide segments without the protection of functional groups,
except for the terminal carboxyl group of the carboxyl
component by the thioester group. Thus, the segments can
be easily prepared by the solid-phase method or recombinant
DNA technology. Recently, the preparation of peptide
thioesters by the recombinant DNA technique has also
been realized by intein technology.[2] The major problem in
NCL is that a cysteine residue is required at the ligation site to
achieve chemoselective coupling. Since the natural abun-
dance of the cysteine residue in proteins is low (1.5%), the
ligation site is severely limited by this problem. Thus, much
effort has been made to develop thiol auxiliary groups to
overcome this limitation.[3]
especially sulfur-containing groups. Recently, Wan and Dan-
ishefsky developed the free-radical reduction of thiol groups,
which will be useful for (glyco)peptide synthesis by the NCL
method.[3o]
Herein, we focused on the use of a mercaptomethyl group
on the side-chain hydroxy groups of serine and threonine,
which is a thiohemiacetal that is generally regarded as too
labile to be used as an auxiliary group. In return, an additional
deprotection step for this auxiliary group would not be
required, as it is spontaneously hydrolyzed after the ligation.
The use of the hydroxy groups of serine and threonine
residues provides far more candidates for the ligation site,
since these amino acids appear in proteins more frequently
(ca. 12% in total) than cysteine. A general route for the new
method is shown in Scheme 1. The mercaptomethyl group is
In initial studies,[3a–d,f–i] the a-amino group was used to
anchor thiol auxiliary groups, which were cleaved after the
ligation. However, this strategy might lead to incomplete
ligation as a result of steric hindrance, and thus has been
mainly used where the ligation site contains at least one
glycine residue. The other strategies used thiol groups at the
side-chain functional groups, which were subsequently con-
verted to proteinogenic amino acid forms. In these strategies
the ligation seems to proceed efficiently, as the acyl group in
the thioester intermediate is transferred to the primary amino
group through five-, six-, or even larger-membered rings using
the thiol group on carbohydrate moieties.[3i,m] These strategies
also require additional reaction for the removal of auxiliary
groups, which is mainly carried out by metal-based reduction.
However, the reaction might not be compatible when applied
to the synthesis of peptides having various functional groups,
Scheme 1. Novel ligation reaction at the Xaa-Ser/Thr site facilitated by
the mercaptomethyl group. Xaa=any amino acid. R=H or CH3,
R1 =C6H5 or C6H4CH2COOH.
[*] Prof. Dr. H. Hojo, C. Ozawa, Dr. H. Katayama, Dr. A. Ueki,
Dr. Y. Nakahara, Prof. Dr. Y. Nakahara
Department of Applied Biochemistry
generated in situ by the reduction of a disulfide-protected
parent peptide. Then the ligation with N-terminal thioester is
performed, followed by the S- to N-acyl shift via a seven-
membered ring. After ligation, the mercaptomethyl group is
hydrolyzed without any extra deprotection steps. If this
concept functions well, all reactions can be performed in one
pot, thus making this method efficient and practical. To
examine the usefulness of the new strategy, contulakin-G
(pGlu-Ser-Glu-Glu-Gly-Gly-Ser-Asn-Ala-Thr(GalNAc)-Lys-
Lys-Pro-Tyr-Ile-Leu-OH, 1a),[4] a glycopeptide toxin isolated
from Conus geographus venom, and human calcitonin were
synthesized.
The synthesis of Fmoc-Ser/Thr (Fmoc = 9-fluorenylme-
thoxycarbonyl) units carrying a protected mercaptomethyl
group was easily accomplished in three steps by following the
procedure of Semenyuk et al.,[5] as shown in Scheme 2. The
commercially available Fmoc-Ser/Thr tBu esters (4 and 5)
Institute of Glycoscience, Tokai University
1117 Kitakaname, Hiratsuka, Kanagawa 259-1292 (Japan)
Fax: (+81)463-50-2075
E-mail: hojo@keyaki.cc.u-tokai.ac.jp
yonak@keyaki.cc.u-tokai.ac.jp
[**] This work was supported by a grant-in-aid for creative scientific
research from the Japan Society for the Promotion of Science
(17GS0420), and in part by a grant-in-aid for scientific research from
the Ministry of Education, Sports, Science, and Technology of Japan
(20380069). We thank Tokai University for their support with a
grant-in-aid for high technology research. We also thank Prof.
Hiromichi Nagasawa and Shinji Nagata of Tokyo University for
sequence analysis and Tsuyoshi Ohira of Kanagawa University for
providing Glu-C. Xaa=any amino acid.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201000384.
5318
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 5318 –5321

Angewandte
Chemie
5-sulfonyl),[6f] the peptide chain was elongated by adopting
the Fmoc strategy. The obtained resin was treated with
Reagent K,[7] and the crude peptide was dissolved in aqueous
CH3CN containing 6m urea in the presence of 4-mercapto-
phenylacetic acid (MPAA)[8] or thiophenol and AcOH. The
NAC moiety was converted to the corresponding thioester
within 24 h. The crude peptide was purified by reversed-phase
(RP) HPLC to give peptide thioesters 10a and 10b.
The ligation was first conducted using thioester 10a and
glycopeptide 11a in sodium phosphate (pH 7.0) containing
6m guanidine HCl and MPAA. Glycopeptide 11a was quickly
reduced by MPAA to yield 11b, which was ligated with
thioester 10a (Scheme 4). However, the product 1b was not
obtained. Under these conditions, only a small amount of the
intermediate 12 was detected by HPLC. This might be
Scheme 2. Synthetic route for the Ser/Thr units 2 and 3 carrying the
thiol auxiliary group.
were methylthiomethylated by the Pummerer rearrangement
using dimethyl sulfoxide (DMSO), acetic anhydride (Ac2O),
and acetic acid (AcOH). The products were chlorinated by
SO2Cl2 in diisopropylethylamine (DIEA), then treated with
potassium thiotosylate (TsSK) and tBuSH, followed by
trifluoroacetic acid (TFA) to give Fmoc-Ser/Thr carrying
the disulfide-protected auxiliary group (2 and 3).
The serine unit 2 was used for the synthesis of contulakin-
G through ligation at Gly6–Ser7. The C-terminal segment
(H-Ser(CH2SStBu)-Asn-Ala-Thr(GalNAcBn)-Lys-Lys-Pro-
Tyr-Ile-Leu-OH (Bn = benzyl), 11a) was prepared by the
conventional Fmoc strategy introducing compound 2 as the
final amino acid. It was essential to use a thiol-free TFA
cocktail for deprotection to minimize the decomposition of
the mercaptomethyl group to less than 15%.
N-Terminal thioesters 10a and 10b were prepared by the
Fmoc method using our N-alkylcysteine (NAC)-assisted
thioesterification method,[6] as shown in Scheme 3. Starting
from Fmoc-Gly-(Et)Cys(Trt)-Arg(Pbf)-Arg(Pbf)-NH-resin
(Trt = trityl; Pbf = 2,2,4,6,7-pentamethyldihydrobenzofuran-
Scheme 4. Pathways of the ligation between peptide thioesters and
glycopeptide carrying the thiol auxiliary group, and side reactions.
because of an equilibrium between the intermediate 12 and
the peptides 10a and 11b, which is shifted to the latter by the
excess MPAA. Since the S- to N-acyl shift (Scheme 4, path A)
proceeds through the seven-membered ring, the reaction was
too slow to be productive. As a result, peptides 10a and 11b
were gradually hydrolyzed to produce nonreactive peptides
13 and 14 (Scheme 4, path B).
To suppress the undesirable reverse reaction, the ligation
between 10a and 11a was conducted without the external
MPAA (Table 1, entry 1). Guanidine HCl was also excluded
to simplify the buffer composition. The reduction of the
disulfide bond in 11a was achieved with triscarboxyethyl-
phosphine (TCEP). Under these conditions, about half of the
generated glycopeptide 11b was converted to intermediate 12
within 10 min (Supporting Information, Figure S1a). How-
ever, the formation of 12 did not become quantitative even
after 30 min, which indicated that the intermediate 12 and the
liberated MPAA are still in equilibrium with peptides 10a and
11b. The relative content of the desired peptide 1b after
Scheme 3. Synthetic route for the N-terminal peptide thioesters.
MBHA=4-methylbenzhydrylamine resin, Boc=tert-butoxycarbonyl.
Angew. Chem. Int. Ed. 2010, 49, 5318 –5321
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5319

Communications
Table 1: Yield of the product 1b by the ligation of peptide thioester with
11a.
Entry
Peptide thioester
Buffer[a]
Relative content [%][b]
1b
14
1
2
3
4
5
6
7
8
pESEEGG-MPAA
pESEEGG-SPh
pESEEGG-SPh
pESEEGG-SPh
pESEEGG-SPh
pESEEGG-SPh
pESEEGG-SPh
pESEEGL-SPh
10a
10b
10b
10b
10b
10b
10b
10c
A
A
B
C
D
E
27
42
47
37
80
60
77
69
73
58
53
15[c]
20
40
23
31
F
D
Figure 1. RP-HPLC profiles of the ligation to obtain glycopeptide 1b
under the conditions of Table 1, entry 7: a) 1 h ligation in buffer A
containing 6m guanidine HCl (pH 7.0) in the presence of ether;
b) overnight reaction after the addition of 10% AcOH in DMF,
c) 2 days reaction time. Elution conditions: column, Mightysil RP-18
GP (4.6ꢀ150 mm, Kanto, Japan); eluent: acetonitrile/water/TFA
[a] Buffer A: 0.1m sodium phosphate containing 15 mm TCEP (pH 7.0);
buffer B: 1:1 mixture of buffer A and CH3CN; buffers C and D: buffer A,
which after 10 min was diluted (ꢀ10) with DMF and 10% AcOH in DMF,
respectively; buffer E: buffer A containing 6m guanidine HCl (pH 7.0),
which was diluted (ꢀ10) with 10% AcOH in DMF after 2 h of ligation;
buffer F: buffer A containing 6m guanidine HCl (pH 7.0) in the presence
of ether, followed by the removal of ether and dilution with 10% AcOH in
DMF (ꢀ10) after 1 h of ligation. [b] The value shows the ratio of the area
of peptide 1b or 14 to that of the total area of peptides derived from
C-terminal peptides upon HPLC. [c] The remaining 48% is acyl shift
products to the side-chain amino group.
85:15:0.1!70:30:0.1 over 30 min, flow rate 1 mLminꢀ1
.
obtained by RP-HPLC purification was 45%. Thus, these
conditions can be generally used when segments retain low
solubility. Under the organic and acidic conditions, the thiol
auxiliary group was fairly stable: it took approximately 2 days
for complete removal. Since the S- to N-acyl shift itself is
completed within 6 h, the dilution of the mixture with neutral
buffer at this stage can accelerate the overall reaction time.
The native peptide bond between Gly6 and Ser7 in glycopep-
tide 1b was confirmed by sequence analysis of the Glu-C
digest of 1b (see the Supporting Information). Glycopeptide
1b was easily converted to the final product, contulakin-G
(1a), by low-acidity trifluoromethanesulfonic acid (TfOH)
treatment to remove a benzyl group.[9]
overnight reaction was 27% (Supporting Information, Fig-
ure S1b). Based on these results, the MPAA thioester seemed
to be unsuitable for this ligation. In contrast, the ligation with
the thiophenyl ester 10b gave the intermediate 12 in higher
yield within 10 min and the product 1b was obtained in an
increased yield after overnight reaction (Table 1, entry 2;
Supporting Information, Figure S1d). The addition of CH3CN
had little effect on the yield of the product 1b, as shown in
Table 1, entry 3.
To prevent the hydrolysis of the intermediate 12
(Scheme 4, path C), the ligation mixture derived from 10b
and 11a was diluted with DMF after 10 min, when a sufficient
amount of the intermediate 12 was formed. Although the
hydrolysis was reduced to less than 20%, about 50% of the
acyl shift products to the side-chain amino groups of lysine
was obtained, which might be because of the reduced
protonation state of the side-chain amino groups in DMF
(Table 1, entry 4; Supporting Information, Figure S1e). How-
ever, this side reaction was almost completely suppressed by
the addition of acetic acid to DMF, and the desired product
was successfully obtained after overnight reaction at a relative
content of 80% (Table 1, entry 5; Supporting Information,
Figure S1f).
The formation of the intermediate 12 was also tested in
the presence of 6m guanidine HCl (Table 1, entry 6). How-
ever, the reaction became slower and incomplete. The relative
content of the product 1b after 2 h of ligation followed by S–N
shift for 2 days in AcOH/DMF was decreased to 60%. Then,
the ligation was conducted in the presence of ether with
vortexing to extract a part of the generated thiophenol and
make the equilibrium more productive (Table 1, entry 7). As
a result, the intermediate 12 was obtained in good yield within
1 h, as shown in Figure 1a. After 2 days, the product 1b was
obtained at a relative content of 77%, which is comparable to
that of Table 1, entry 5 (Figure 1c). The yield of isolated 1b
The peptide thioester 10c with a C-terminal Leu residue
was also ligated with peptide 11a. As shown in Table 1
(entry 8), the desired Leu6-contulakin-G (1c) was successfully
obtained at a 69% relative content and 31% yield of isolated
product, thus showing the applicability of this method to
ligation with a sterically demanding amino acid.
The stability of the mercaptomethyl group on the
glycopeptide 11b was examined in buffer A containing 6m
guanidine HCl at pH 6, 7, and 8 (see the Supporting
Information). The half-lives were about 4, 2.5, and 2 h,
respectively, which might be sufficient periods to perform the
initial transthioesterification step of the ligation at these pH
values.
The method was further applied to the synthesis of human
calcitonin using Thr derivative 3, as shown in Scheme 5.
Peptide thioester 15 and peptide 16 were prepared according
to the same procedure as that for contulakin-G. The ligation
was performed under the conditions of Table 1, entry 7.
Within 2 h at room temperature, the formation of the
intermediate 17 was maximum. After dilution with AcOH/
DMF, the solution was left undisturbed for 2 days. The desired
peptide 18 was obtained at a relative content of 66% and
40% yield of isolated product.[10] Considering that this is a
total yield of the ligation and the removal of the auxiliary
group, the value is acceptable compared with those of other
auxiliary-mediated ligation reactions.[3] The removal of the
5320
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 5318 –5321

Angewandte
Chemie
Tetrahedron Lett. 2001, 42, 1831 – 1833; e) L. Z. Yan, P. E.
Dawson, J. Am. Chem. Soc. 2001, 123, 526 – 533; f) J. Offer,
C. N. C. Boddy, P. E. Dawson, J. Am. Chem. Soc. 2002, 124,
4642 – 4646; g) T. Kawakami, S. Aimoto, Tetrahedron Lett. 2003,
44, 6059 – 6061; h) D. Macmillan, D. W. Anderson, Org. Lett.
2004, 6, 4659 – 4662; i) B. Wu, J. Chen, J. D. Warren, G. Chen, Z.
Hua, S. J. Danishefsky, Angew. Chem. 2006, 118, 4222 – 4231;
Angew. Chem. Int. Ed. 2006, 45, 4116 – 4125; j) A. Brik, Y. Y.
Yang, S. Ficht, C. H. Wong, J. Am. Chem. Soc. 2006, 128, 5626 –
5627; k) A. Saporito, D. Marasco, A. Chambery, P. Botti, S. M.
Monti, C. Pedone, M. Ruvo, Biopolymers 2006, 83, 508 – 518;
l) B. L. Pentelute, S. B. H. Kent, Org. Lett. 2007, 9, 687 – 690;
m) Y.-Y. Yang, S. Ficht, A. Brik, C.-H. Wong, J. Am. Chem. Soc.
2007, 129, 7690 – 7701; n) D. Crich, A. Banerjee,
J. Am. Chem. Soc. 2007, 129, 10064 – 10065; o) Q. Wan, S. J.
Danishefsky, Angew. Chem. 2007, 119, 9408 – 9412; Angew.
Chem. Int. Ed. 2007, 46, 9248 – 9252; p) C. Chatterjee, R. K.
McGinty, J.-P. Pellois, T. W. Muir, Angew. Chem. 2007, 119,
2872 – 2876; Angew. Chem. Int. Ed. 2007, 46, 2814 – 2818; q) R.
Okamoto, Y. Kajihara, Angew. Chem. 2008, 120, 5482 – 5486;
Angew. Chem. Int. Ed. 2008, 47, 5402 – 5406; r) C. Haase, H.
Rohde, O. Seitz, Angew. Chem. 2008, 120, 6912 – 6915; Angew.
Chem. Int. Ed. 2008, 47, 6807 – 6810; s) J. Chen, P. Wang, J. Zhu,
Q. Wan, S. J. Danishefsky, Tetrahedron 2010, 66, 2277 – 2283.
[4] A. G. Craig, T. Norberg, D. Griffin, C. Hoeger, M. Akhtar, K.
Schmidt, W. Low, J. Dykert, E. Richelson, V. Navarro, J. Mazella,
M. Watkins, D. Hillyard, J. Imperial, L. J. Cruz, B. M. Olivera,
J. Biol. Chem. 1999, 274, 13752 – 13759.
Scheme 5. Synthetic route for human calcitonin. Acm=acetamido-
methyl.
Acm group and disulfide bond formation successfully gave
calcitonin 19 (see the Supporting Information).
In conclusion, novel ligation at the naturally abundant
Xaa-Ser/Thr site using the mercaptomethyl group as a thiol
auxiliary was demonstrated by the syntheses of contulakin-G
and human calcitonin. The ease of the synthesis of the Ser/Thr
unit carrying the thiol auxiliary group as well as its autocleav-
able property make this method versatile for the synthesis of
various (glyco)proteins. Currently, this method is being
applied to glycoprotein synthesis.
[5] A. Semenyuk, A. Foldesi, T. Johansson, C. Estmer-Nilsson, P.
Blomgren, M. Brannvall, L. A. Kirsebom, M. Kwiatkowski,
J. Am. Chem. Soc. 2006, 128, 12356 – 12357.
[6] a) H. Hojo, Y. Onuma, Y. Akimoto, Y. Nakahara, Y. Nakahara,
Tetrahedron Lett. 2007, 48, 25 – 28; b) C. Ozawa, H. Katayama,
H. Hojo, Y. Nakahara, Y. Nakahara, Org. Lett. 2008, 10, 3531 –
3533; c) H. Katayama, H. Hojo, T. Ohira, Y. Nakahara,
Tetrahedron Lett. 2008, 49, 5492 – 5494; d) H. Hojo, Y. Mur-
asawa, H. Katayama, T. Ohira, Y. Nakahara, Y. Nakahara, Org.
Biomol. Chem. 2008, 6, 1808 – 1813; e) Y. Nakahara, I. Matsuo,
Y. Ito, R. Ubagai, H. Hojo, Y. Nakahara, Tetrahedron Lett. 2010,
51, 407 – 410; f) H. Katayama, H. Hojo, I. Shimizu, Y. Nakahara,
Y. Nakahara, Org. Biomol. Chem. 2010, 8, 1966 – 1972.
[7] D. S. King, C. G. Fields, G. B. Fields, Int. J. Pept. Protein Res.
1990, 36, 255 – 266.
[8] D. Bang, B. L. Pentelute, Z. P. Gates, S. B. H. Kent, Org. Lett.
2006, 8, 1049 – 1052.
[9] a) J. P. Tam, W. F. Heath, R. B. Merrifield, J. Am. Chem. Soc.
1986, 108, 5242 – 5251; b) H. Hojo, E. Haginoya, Y. Matsumoto,
Y. Nakahara, K. Nabeshima, B. P. Toole, Y. Watanabe, Tetrahe-
dron Lett. 2003, 44, 2961 – 2964; c) H. Hojo, Y. Matsumoto, Y.
Nakahara, E. Ito, Y. Suzuki, M. Suzuki, A. Suzuki, Y. Nakahara,
J. Am. Chem. Soc. 2005, 127, 13720 – 13725.
[10] The ligation at Val–Thr, Leu–Thr, and Tyr–Thr sites was also
tested. The first one was unproductive, but the latter two were
useful for ligation (see the Supporting Information).
Received: January 22, 2010
Keywords: glycopeptides · ligation · peptide thioesters ·
.
synthetic methods
[1] a) P. E. Dawson, T. W. Muir, I. Clark-Lewis, S. B. Kent, Science
1994, 266, 776 – 779; b) P. E. Dawson, S. B. Kent, Annu. Rev.
Biochem. 2000, 69, 923 – 960.
[2] T. W. Muir, D. Sondhi, P. A. Cole, Proc. Natl. Acad. Sci. USA
1998, 95, 6705 – 6710; b) T. W. Muir, Annu. Rev. Biochem. 2003,
72, 249 – 289.
[3] a) L. E. Canne, S. J. Bark, S. B. H. Kent, J. Am. Chem. Soc. 1996,
118, 5891 – 5896; b) D. W. Low, M. G. Hill, M. R. Carrasco,
S. B. H. Kent, P. Botti, Proc. Natl. Acad. Sci. USA 2001, 98,
6554 – 6559; c) T. Kawakami, K. Akaji, S. Aimoto, Org. Lett.
2001, 3, 1403 – 1405; d) P. Botti, M. R. Carrasco, S. B. H. Kent,
Angew. Chem. Int. Ed. 2010, 49, 5318 –5321
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5321