Studies related to the relative thermodynamic stability of C-terminal peptidyl esters of O-hydroxy thiophenol: Emergence of a doable strategy for non-cysteine ligation applicable to the chemical synthesis of glycopeptides

Article abstract of DOI:10.1021/ja061588y

A pathway has been devised, wherein a phenolic ester of a C-terminal peptide is ligated with an N-terminal peptide through two consecutive acyl migrations. In the first transacylation, the C-terminus is transferred from a phenol to a newly liberated ortho

Full text of DOI:10.1021/ja061588y

Published on Web 05/19/2006
Studies Related to the Relative Thermodynamic Stability of C-Terminal
Peptidyl Esters of O-Hydroxy Thiophenol: Emergence of a Doable Strategy
for Non-Cysteine Ligation Applicable to the Chemical Synthesis of
Glycopeptides
Gong Chen,† J. David Warren,† Jiehao Chen,† Bin Wu,† Qian Wan,† and Samuel J. Danishefsky*,†,‡
Laboratory for Bioorganic Chemistry, Sloan-Kettering Institute for Cancer Research, 1275 York AVenue, New York,
New York 10021, and Department of Chemistry, Columbia UniVersity, HaVemeyer Hall, 3000 Broadway,
New York, New York 10027
Received March 7, 2006; E-mail: s-danishefsky@ski.mskcc.org
Scheme 1. (a) Previously Disclosed Cysteine-Based Ligation
Our laboratory has been studying methods for assembling
polypeptides bearing complex oligosaccharide domains at defined
sites.1 Toward this end, we had previously disclosed a method for
ligating two glycopeptide fragments, each possessing a fully
functionalized carbohydrate domain.2 A C-terminal glycopeptide
fragment is equipped with a phenolic ester bearing a protected
ortho-thiol moiety (1). Disulfide reduction, followed by exposure
of the acyl anion system to an N-terminal cysteine-bearing
glycopeptide (2), leads to formation of a homogeneous glycopeptide
(3), displaying two different carbohydrate domains (Scheme 1a).
In our initial conception, we had envisioned a possible O f S
acyl migration with the formation of a transient thioester intermedi-
ate (A f B). This event would then set the stage for a cysteine-
based native chemical ligation (NCL).3 We had also recognized
the possibility of the intervention of more subtle factors. For
example, the free ortho-thiol group in A might well enhance the
acylating ability of the O-ester through intramolecular H-bonding.4
In this interpretation, there need not be a thioester involved in the
ligation. Aside from intrinsically interesting mechanistic issues,
which drove the study described below, we hoped that a fact-based
understanding of the actual pathway could be helpful in broadening
the scope and utility of this ligation logic.5 Accordingly, we set
out to examine these and related questions in some detail. Below,
we report what were to us some surprising findings. These
discoveries led us to explore a promising strategy toward the
realization of a cysteine-free NCL that could become broadly
applicable to glycopeptide synthesis.
Protocol.2a (b) Possible Reaction Intermediates
reaction mixture, at pH 7, rapidly consumed cysteine to generate
the amide product 5, with a chemical shift of 172.4 ppm (Figure
1f).
It is clear from the 13C NMR spectra that the thioester
intermediate (B) is thermodynamically predominant at low pH (2-
4), perhaps due to stabilization imparted by an extra intramolecular
H-bonding interaction. When the pH is increased to 6-8, apparently
the equilibrium exists in the form of intermediates C and D, which
are exchanging very rapidly. Of these, we would intuitively suppose
that D is the more stable, based on its “lower energy” acyl linkage
and its less basic anionic center. However, this matter has certainly
not been established.
Given the distinctive chemical shift difference between the
possible isomeric ester intermediates (e.g., S-ester ∼ 200, O-ester
∼ 170, cf. A and B), we were hopeful that 13C NMR analysis with
a
13C-labeled substrate would provide meaningful insight into the
reaction dynamics. Thus, 13C-labeled phenylalanine was incorpo-
rated into peptide substrate 4. Upon addition of tris(2-carboxyethyl)-
phosphine hydrochloride (TCEP‚HCl, 3 equiv) to 4 in water,
disulfide reduction was completed in less than 2 min, and the
resulting 13C NMR spectrum was subsequently obtained (Figure
1b). Surprisingly, only a single peak at 201.5 ppm was obserVed,
corresponding to the thioester (cf. B). It is worthy of note that the
pH of the reaction medium had dropped to 2-3 upon addition of
TCEP‚HCl. Increasing the pH to 4-5 did not change the spectrum
significantly (Figure 1c). However, when the pH was increased
above 6 (Figure 1d), all of the 13C NMR signals disappeared,
indicating a dynamic chemical exchange between the O- and S-acyl
species (Scheme 1b).6 Interestingly, the thioester peak reappeared
when the pH of the solution was readjusted to 3 (Figure 1e). The
† Sloan-Kettering Institute for Cancer Research
‡ Columbia University.
Figure 1. 13C NMR spectra of reaction (125 MHz).
9
7460
J. AM. CHEM. SOC. 2006, 128, 7460-7462
10.1021/ja061588y CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2. Design of a Cysteine-Free Ligation Protocol
Scheme 3. Synthesis of Auxiliary 17 and Peptide Conjugate 19a
a Key: (a) NaH, MEMCl, DMF, 0 °C to rt, 58%; (b) KOH, ClCSNMe2,
THF/H2O, 0 °C, 92%; (c) reflux, triethyleneglycol dimethyl ether, 74%;
(d) 4% TFA/CH2Cl2, 95%; (e) NaOH, MeOH/H2O, 93%; (f) TmobOH,
NCS, PPh3, CH2Cl2, 0 °C; (g) K2CO3, DMF, two steps, 58%; (h) 20, EDC,
HOBt, DIEA, DMF, 83%; (i) 21, NaBH3CN, AcOH, MeOH, 77%.
Given the evidence for the emergence of a thioester intermediate,
via stoichiometric O f S acyl transfer, the possibility of a modified
system, wherein the thioester, generated in situ, would be intercepted
by an intramolecular nucleophilic amine group, presented itself.7
In this way, one could envision developing a useful new peptide
ligation method, independent of an N-terminal cysteine, wherein
the organizing matrix emerged as a protecting group marking each
ligation event.8
Scheme 4. Peptide Ligation through Consecutive Acyl Migrationa
As is apparent in browsing the generalized Scheme 2, the
realization of the proposed method is contingent on two factors.
First, it must be possible to construct the two-domain peptide
conjugate (cf. 6) in a facile manner.9 In addition, both acyl transfers
(cf. 7 f 8, 8 f 9) must proceed efficiently. With this goal in mind,
a novel 1,2,3-trisubstituted aromatic auxiliary unit (12) was
designed. A convergent two-step procedure was adopted for
assembling the conjugate (6), wherein the auxiliary unit would first
be installed onto the CR-carboxyl terminus of peptide 1 (11) through
phenol ester coupling. Next, peptide 2 (13) would be introduced
via reductive amination.
a Key: (a) 50% TFA/0.5% TIPS, CH2Cl2, 30 min; (b) evaporate solvent,
redissolve in MeOH, 82%; (c) TCEP‚HCl.
Table 1. Ligation Yields of Selected Peptide Substrates
The acid-labile 2,4,6-trimethoxybenzyl group (2,4,6-Tmob) was
selected for protection of the thiol group because of its ready
removal by TFA10 and its tolerance of reductive amination
conditions. Two appropriately protected peptide fragments, 20 and
21, were prepared using Fmoc solid-phase peptide synthesis. The
auxiliary 17sitself prepared from 14swas appended to the CR-
carboxyl group of 20 to produce ester 18 (Scheme 3). Although
the coupling proceeded in good yield, racemization of the C-
terminal phenylalanine presented a significant problem, presumably
from oxazolone formation during activation of the CR-carboxyl
group (vide infra).11 Unable to identify conditions by which to
suppress racemization, we nonetheless explored the subsequent
reductive amination reaction. Thus, aldehyde 18 and peptide 21
were treated with NaCNBH3 and AcOH in MeOH to afford peptide
substrate 19 in good yield (>75%). Although compromised by
racemization in formation of the O-ester, the overall sequence did
accomplish the gross merge we sought.
The stage was now set to choreograph the O f S f N acyl
transfer (Scheme 4). The actual operation was first triggered by
acidic removal of the Tmob blocking group, followed by transfer
of the reagents to a proper solvent system in order to promote the
desired amide formation. In the event, upon treatment of 19 in 50%
TFA in CH2Cl2 with 0.5% TIPS as a cation scavenger, the Tmob
was completely cleaved within 30 min (along with the other acid-
labile protecting groups of the peptide side chains). Next, the solvent
was evaporated, and the resultant reaction residue was redissolved
in MeOH for LC/MS monitoring. Surprisingly, a notable amount
of amide-ligated product 24 had rapidly formed (less than 5 min)
in the MeOH solution, which was still fairly acidic due to residual
TFA from the prior step. The reaction was complete in under an
hour, and the free thiol 24 was obtained along with its corresponding
reaction
isolated
entry
substrates: peptide1 (a) peptide2
conditionsa
yield (%)
1
2
3
4
5
6
7
8
9
19 BocK(Boc)GF (a) GFFNH2
19 BocK(Boc)GF (a) GFFNH2
26 BocK(Boc)FG (a) GFFNH2
27 BocK(Boc)GP (a) GFFNH2
28 BocK(Boc)FG (a) AFFNH2
29 BocK(Boc)FG (a) K(ivDde)FFNH2
30 BocK(Boc)GP (a) AFFNH2
31 BocK(Boc)FG (a) GFS(Tn)GNH2
32 BocK(Boc)GP (a) GFS(Tn)GNH2
a (1 h)
b (1 h)
b (30 min)
c (8 h)
d (4 h)
d (12 h)
d (24 h)
b (30 min)
c (8 h)
67b
82
87
83
85
81
53
79
74
72
10 33 BocK(Boc)S(STn)AGP (a) GFS(Tn)GNH2 c (10 h)
a Key: (a) MeOH/H2O (1:1); (b) MeOH; (c) MeOH, NaH2PO4; (d) DMF,
NaH2PO4. b 67% product + 13% hydrolysis side product.
oxidized form, disulfide 25; the disulfide was easily reduced to 24
upon exposure to TCEP. Although ligation was feasible in aqueous
buffer media, hydrolysis of the ester intermediate emerged as a
competing reaction under these conditions (Table 1, entry 1).12 This
issue was solved by the use of MeOH as a solvent (entry 2). Overall,
this two-step protocol was essentially performed as a one-flask
reaction, cleanly providing the desired ligation adduct.
Although an excellent yield was achieved with the Phe-Gly
junction, the racemization problem (vide supra) would, of course,
limit the practical utility of this ligation method. However, peptide
fragments bearing C-terminal glycine or proline residues obviously
do not suffer from this type of issue.13 Indeed, the sterically less
demanding Gly-Gly ligation of substrate 26 occurred very rapidly
in acidic MeOH (Table 1). In the sterically and electronically more
demanding Pro-Gly ligation of substrate 27, S f N acyl transfer
was much slower; however, adding solid NaH2PO4 to the MeOH
solution significantly increased the acyl transfer rate without leading
9
J. AM. CHEM. SOC. VOL. 128, NO. 23, 2006 7461

C O M M U N I C A T I O N S
Scheme 5. Synthesis of Bidomainal Glycopeptide 34a
acyl transfer to deliver the intact amide. While issues of substrate
racemization and auxiliary removal remain to be solved, we view
this new strategy as a promising platform for future initiatives. The
full range of applicability of chemistry-centered glycopeptide
ligation “devices” is a matter of continuing interest in our laboratory.
Acknowledgment. This work was supported by the NIH
(CA28824). We thank Dr. George Sukenick (NMR Core Facility,
CA02848) for help with NMR analyses, and Ms. Sylvi Rusli and
Ms. Hui Fang for help with mass spectral analyses and LC/MS
separations. Postdoctoral fellowship support is gratefully acknowl-
edged by J.D.W. (NIH, CA62948).
Supporting Information Available: Experimental procedures and
compound characterization data, including LC/MS and NMR data
(PDF). This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) (a) Miller, J. S.; Dudkin, V. Y.; Lyon, G. J.; Muir, T. W.; Danishefsky,
S. J. Angew. Chem., Int. Ed. 2003, 42, 431-434. (b) Mandal, M.; Dudkin,
V. Y.; Geng, X.; Danishefsky, S. J. Angew. Chem., Int. Ed. 2004, 43,
2557-2561. (c) Geng, X.; Dudkin, V. Y.; Mandal, M.; Danishefsky, S.
J. Angew. Chem., Int. Ed. 2004, 43, 2562-2575. For the most recent
review, see: (d) Liu, L.; Bennett, C. S.; Wong, C.-H. Chem. Commun.
2006, 21-33. For other glycopeptide syntheses, see: (e) Shin, Y.; Winans,
K. A.; Backes, B. J.; Kent, S. B. H.; Ellman, J. A.; Bertozzi, C. R. J. Am.
Chem. Soc. 1999, 121, 11684-11689. (f) Hojo, H.; Hagnoya, E.;
Matsumoto, Y.; Nakahara, Y.; Nabeshima, K.; Toole, B. P.; Watanabe,
Y. Tetrahedron Lett. 2003, 44, 2961-2964. (g) Macmillan, D.; Bertozzi,
C. R. Angew. Chem., Int. Ed. 2004, 43, 1355-1359. (h) Mezzato, S.;
Schaffrath, M.; Unverzagt, C. Angew. Chem., Int. Ed. 2005, 44, 1650-
1654. (i) Hackenberger, C. P. R.; Friel, C. T.; Radford, S. E.; Imperiali,
B. J. Am. Chem. Soc. 2005, 127, 12882-12889. (j) Kaneshiro, C. M.;
Michael, K. Angew. Chem., Int. Ed. 2006, 45, 1077-1081.
(2) (a) Warren, J. D.; Miller, J. S.; Keding, S. J.; Danishefsky, S. J. J. Am.
Chem. Soc. 2004, 126, 6576-6578. (b) Chen, J.; Warren, J. D.; Wu, B.;
Chen, G.; Wan, Q.; Danishefsky, S. J. Tetrahedron Lett. 2006 47, 1969-
1972. For a conceptually similar approach, see: (c) Botti, P.; Villain, M.;
Manganiello, S.; Gaertner, H. Org. Lett. 2004, 6, 4861-4864.
(3) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H. Science 1994,
266, 776-779.
a Key: (a) 50% TFA/0.5% TIPS, CH2Cl2, 35 min; (b) MeOH, NaH2PO4,
10 h, 72% (two steps).
to problems of hydrolysis. As expected, excellent ligation yields
and racemization suppression were achieved. Interestingly, we found
a stable intermediate in the reaction mixture following acidic
treatment. Using a 13C-labeled analogue, the intermediate was
identified as a thioester. As the S f N acyl transfer proceeded,
LC/MS indicated a gradual decrease in the amount of thioester,
while the amide product was enhanced correspondingly.
We note that Gly-non-Gly linkages occur frequently in natural
peptide sequences. The efficiency of this type of amide ligation
was found to be dependent on the substrate and reaction conditions
employed. Thus, in substrate 28, the decreased nucleophilicity and
steric hindrance of the secondary amine in alanine resulted in a
slow S f N acyl transfer. In the same molecule, the glycine
thioester was found to be very prone to hydrolysis under various
aqueous conditions. Adding NaH2PO4 to the MeOH solution not
only noticeably accelerated the desired S f N acyl transfer but
also produced a very high level of hydrolytic and even methanolytic
side products. Various combinations of solvents and additives were
tested, yielding only moderate improvement. Fortunately, by simply
dissolving the reaction mixture in DMF after Tmob removal,
followed by the addition of a slight excess of NaH2PO4, the ligated
adducts could be obtained in very good yields with minimal
hydrolysis. Under these conditions, ligation of Gly-Lys(ivDde) in
substrate 29 and the even more sterically demanding Pro-Ala in
30 was successfully achieved.
Happily, the logic described above could be extended to
encompass the synthesis of bidomainal glycopeptides. As listed in
Table 1, two substrates, 31 and 32, were prepared, containing Gly-
Gly and Pro-Gly junctions, respectively, each bearing a protected
Tn antigen14 on the N-terminal peptide segment. These compounds
were excellent substrates for ligation, proceeding in very good yield.
Notably, the differentially diglycosylated substrate 33 (displaying
Tn and STn15) readily underwent ligation in buffered MeOH to
provide the bifunctional glycopeptide 34 in high yield (Scheme 5).
In summary, we have determined the mechanistic basis of our
previously described ligation strategy,2a through the unexpected
finding that ligation progresses via quantitative O f S acyl transfer
of the N-terminal coupling partner. On the basis of this finding,
we have laid the groundwork for the development of a novel
cysteine-free glycopeptide ligation strategy. The method involves
a well-orchestrated sequential intramolecular O f S then S f N
(4) Verhaeghe, J.; Lacassie, E.; Bertrand, M.; Trudelle, Y. Tetrahedron Lett.
1993, 34, 461-464 and references therein.
(5) For an early study of O f S acyl transfer in o-hydroxybenzenethiol, see:
McKinnon, D. M. Can. J. Chem. 1980, 58, 2761-2764.
(6) Oki, M., Ed. Applications of Dynamic NMR Spectroscopy to Organic
Chemistry; VCH Publishers: Deerfield Beach, FL, 1985.
(7) Intramolecular acyl transfer for peptide ligation: (a) Kemp, D. S.
Biopolymers 1981, 20, 1793-1804. (b) Kemp D. S.; Carey, R. I. J. Org.
Chem. 1993, 58, 2216-2222. (c) Johnson, T.; Quibell, M. Tetrahedron
Lett. 1994, 35, 463-466. (d) Canne, L. E.; Bark, S. J.; Kent, S. B. H. J.
Am. Chem. Soc. 1996, 118, 5891-5896. (e) Miranda, L. P.; Meutermans,
W. D. F.; Smythe, M. L.; Alewood, P. F. J. Org. Chem. 2000, 65, 5460-
5468. (f) Offer, J.; Dawson, P. E. Org. Lett. 2000, 2, 23-26. (g) Botti,
P.; Carrasco, M. R.; Kent, S. B. H. Tetrahedron Lett. 2001, 42, 1831-
1833. For some helpful reviews, see: (h) Coltart, D. M. Tetrahedron 2000,
3449-3491. (i) Nilsson, B. L.; Soellner, M. B.; Raines, R. T. Annu. ReV.
Biophys. Biomol. Struct. 2005, 34, 91-118.
(8) For other cysteine-free peptide ligation strategies, see: (a) Low, D. W.;
Hill, M. G.; Carrasco, M. R.; Kent, S. B. H.; Botti, P. Proc. Natl. Acad.
Sci. U.S.A. 2001, 98, 6554-6559. (b) Offer, J.; Boddy, C. N. C.; Dawson,
P. E. J. Am. Chem. Soc. 2002, 124, 4642-4646. (c) Nilsson, B. L.; Hondal,
R. J.; Soellner, M. B.; Raines, R. T. J. Am. Chem. Soc. 2003, 125, 5268-
5269.
(9) (a) Liu, C. F.; Tam, J. P. J. Am. Chem. Soc. 1994, 116, 4149-4153. (b)
Tam, J. P.; Lu, Y. A.; Liu, C. F.; Shao, J. Proc. Natl. Acad. Sci. U.S.A.
1995, 92, 12485-12489.
(10) Munson, M. C.; Garcia-Echeverria, C.; Albericio, F.; Barany, G. J. Org.
Chem. 1992, 57, 3013-3018.
(11) Kemp, D. S. In The Peptides: Analysis, Synthesis, Biology; Gross, E.,
Meienhofer, J., Eds.; Academic Press: New York, 1979; Vol. 1, Chapter
7.
(12) In substrates where the N-terminal residue of peptide 2 was not glycine,
the level of ester hydrolysis under aqueous conditions was prohibitively
high.
(13) Quibell, M.; Packman, L. C.; Johnson, T. J. Am. Chem. Soc. 1995, 117,
11656-11668.
(14) Springer, G. F. Science 1984, 224, 1198-1206.
(15) Itzkowitz, S. H.; Bloom, E. J.; Kokal, W. A.; Modin, G.; Hakomori, S.;
Kim, Y. S. Cancer 1990, 66, 1960-1966.
JA061588Y
9
7462 J. AM. CHEM. SOC. VOL. 128, NO. 23, 2006