Staudinger ligation of peptides at non-glycyl residues

Article abstract of DOI:10.1021/jo0620056

The Staudinger ligation provides a means to form an amide bond between a phosphinothioester and azide. This reaction holds promise for the ligation of peptides en route to the total chemical synthesis of proteins. (Diphenylphosphino)methanethiol is the mo

Full text of DOI:10.1021/jo0620056

Staudinger Ligation of Peptides at Non-Glycyl Residues
Matthew B. Soellner,†,‡ Annie Tam,† and Ronald T. Raines*,†,§
Departments of Chemistry and Biochemistry, UniVersity of Wisconsin-Madison, Madison, Wisconsin 53706
raines@biochem.wisc.edu
ReceiVed September 28, 2006
The Staudinger ligation provides a means to form an amide bond between a phosphinothioester and
azide. This reaction holds promise for the ligation of peptides en route to the total chemical synthesis of
proteins. (Diphenylphosphino)methanethiol is the most efficacious of known reagents for mediating the
Staudinger ligation of peptides, providing high (>90%) isolated yields for equimolar couplings in which
a glycine residue is at the nascent junction. Surprisingly, the yields are lower (<50%) for non-glycyl
couplings due to an aza-Wittig reaction that diverts the reaction toward a phosphonamide byproduct.
Here, the partitioning of the reaction toward Staudinger ligation (and away from the aza-Wittig reaction)
is shown to increase with increasing electron density on phosphorus. This electron density can be tuned
either by installing functional groups on the phenyl substituents of (diphenylphosphino)methanethiol or
by changing the polarity of the solvent. Installing p-methoxy groups and using a solvent of low polarity
(such as toluene or dioxane) provide especially high (>80%) isolated yields for the ligation of two non-
glycyl residues. These conditions retain the high chemoselectivity of the reaction and do not lead to a
substantial change in reaction rate. The traceless Staudinger ligation is now poised to enable the iterative
ligation of peptides with little regard for their sequence, as well as the synthesis of amide bonds for other
purposes.
Introduction
A limitation of native chemical ligation is its intrinsic reliance
on having a cysteine residue at the ligation junction. Cysteine
is uncommon, comprising only 1.7% of all residues in pro-
teins.7 Modern peptide synthesis is typically limited to peptides
of e40 residues,8 whereas most proteins contain hundreds of
residues. Hence, most proteins cannot be prepared by any
method that allows for peptides to be coupled only at cysteine
residues.1
Total chemical synthesis is beginning to provide ready access
to natural proteins, as well as enable the creation of non-natural
ones.1 Many proteins have already been assembled from
synthetic peptides. “Native chemical ligation”, the coupling of
a peptide (or protein) containing a C-terminal thioester with
another peptide containing an N-terminal cysteine residue, has
been especially powerful.2-4 “Expressed protein ligation” is a
method by which C-terminal thioesters for native chemical
ligation can be accessed with recombinant DNA techniques.5,6
An alternative strategy is to employ auxiliaries that act as
cysteine surrogates to mediate the chemical ligation of peptide
fragments.1,9,10 After the ligation, the auxiliary is removed to
† Department of Chemistry.
‡ Present address: Department of Chemistry, University of California,
Berkeley, CA 94720.
(5) Muir, T. W. Annu. ReV. Biochem. 2003, 72, 249-289.
(6) Evans, T. C., Jr.; Xu, M.-Q. Chem. ReV. 2002, 102, 4869-4883.
(7) McCaldon, P.; Argos, P. Proteins: Struct., Funct., Genet. 1988, 4,
99-122.
§ Department of Biochemistry.
(1) Nilsson, B. L.; Soellner, M. B.; Raines, R. T. Annu. ReV. Biophys.
Biomol. Struct. 2005, 34, 91-118.
(2) Dawson, P. E.; Kent, S. B. H. Annu. ReV. Biochem. 2000, 69, 923-
960.
(8) Bray, B. L. Nat. ReV. Drug DiscoVery 2003, 2, 587-593.
(9) Canne, L. E.; Bark, S. J.; Kent, S. B. H. J. Am. Chem. Soc. 1996,
118, 5891-5896.
(3) Kent, S. J. Pept. Sci. 2003, 9, 574-593.
(4) Tam, J. P.; Xu, J.; Eom, K. D. Biopolymers 2001, 60, 194-205.
(10) Offer, J.; Dawson, P. E. Org. Lett. 2000, 2, 23-26.
10.1021/jo0620056 CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/01/2006
9824
J. Org. Chem. 2006, 71, 9824-9830

Staudinger Ligation of Peptides at Non-Glycyl Residues
SCHEME 1. Putative Mechanism for the Traceless
Staudinger Ligation Mediated by
(Diphenylphosphino)methanethiol
(Diphenylphosphino)methanethiol (HSCH2PPh2) is the most
efficacious of known reagents for effecting the traceless
Staudinger ligation.17 This reagent has been used in the
orthogonal assembly of a protein,18 site-specific immobilization
of peptides and proteins to a surface,19,20 and synthesis of
glycopeptides.21 The reaction is rapid, displaying t1/2 ) 7 min
for the coupling of AcGlySCH2PPh2 and 2-azido-N-benzylac-
etamide (175 mM each) in a wet organic solvent.17 This value
is limited by the encounter of the reactants to form the initial
phosphazide intermediate. Another phosphine, o-(diphenylphos-
phino)benzoic acid, has been used to mediate the non-traceless
Staudinger ligation for biomolecular labeling experiments in
vitro22 and in vivo,23,24 and drug delivery.25 This reaction leaves
a phosphine oxide in the ligation product, making it inappropri-
ate for synthetic applications.
The putative mechanism for the traceless Staudinger ligation
appears to be indifferent to amino acid substitution at the ligation
sites (Scheme 1). Yet, as with auxiliary-mediated ligations,11,12
all reported traceless Staudinger ligations of peptides that
proceed with a high (>90%) yield have had a glycine residue
at the ligation junction.26 Reactions of more encumbered residues
proceed with a lower (<50%) yield.17,27 Likewise, yields for
the synthesis of glycopeptides are diminished by steric strain.21
Somehow, steric effects are being manifested during the
reaction.
reveal the native amide bond. Efforts with extant auxiliaries
have, however, revealed a requirement for a glycine residue to
be present at the ligation junction.11,12 Although these methods
are not yet sequence-independent, they do represent an advance
beyond native chemical ligation, as glycine is among the most
common amino acids in proteins (7.2% of all residues).7
An elegant new strategy for peptide coupling entails the
decarboxylative condensation of R-ketoacids and N-alkylhy-
droxylamines.13 This reaction suffers, however, from the ac-
cumulation of numerous intermediates, leading to low overall
rates. Modifications to the reactants or reaction conditions that
overcome this intrinsic deficiency are not apparent. Moreover,
many of the R-ketoacids that correspond to the proteinogenic
amino acids are not readily accessible.
The traceless Staudinger ligation was designed to overcome
such limitations and provide a general means to synthesize
proteins from component peptides (Scheme 1).14,15 In this
reaction, a peptide with a C-terminal phosphinothioester (1)
is coupled with a second peptide having an N-terminal azido
acid (2) through the intermediacy of an iminophosphorane (3).
The iminophosphorane can form a tetrahedral intermediate
(4), which collapses to give an amidophosphonium salt (5)
that is hydrolyzed in aqueous solution. The final ligation prod-
uct is a peptide (6) without either residual atoms14,15 or
racemization.16 In nonaqueous solvents, amidophosphonium
salt 5 can fragment to form peptide 6 and a thiaphosphiranium
salt.17
(17) Soellner, M. B.; Nilsson, B. L.; Raines, R. T. J. Am. Chem. Soc.
2006, 128, 8820-8828.
(18) Nilsson, B. L.; Hondal, R. J.; Soellner, M. B.; Raines, R. T. J. Am.
Chem. Soc. 2003, 125, 5268-5269.
(19) Soellner, M. B.; Dickson, K. A.; Nilsson, B. L.; Raines, R. T. J.
Am. Chem. Soc. 2003, 125, 11790-11791.
(20) Gauchet, C.; Labadie, G. R.; Poulter, C. D. J. Am. Chem. Soc. 2006,
128, 9274-9275.
(21) For examples, see: (a) He, Y.; Hinklin, R. J.; Chang, J. Y.; Kiessling,
L. L. Org. Lett. 2004, 6, 4479-4482. (b) Bianchi, A.; Bernardi, A.
Tetrahedron Lett. 2004, 45, 2231-2234. (c) Bianchi, A.; Russo, A.;
Bernardi, A. Tetrahedron: Asymmetry 2005, 16, 381-386. (d) Liu, L.;
Hong, Z.-Y.; Wong, C.-H. ChemBioChem 2006, 2006, 429-432.
(22) For examples, see: (a) Saxon, E.; Bertozzi, C. R. Science 2000,
287, 2007-2010. (b) Saxon, E.; Luchansky, S. J.; Hang, H. C.; Yu, C.;
Lee, S. C.; Bertozzi, C. R. J. Am. Chem. Soc. 2002, 124, 14893-14902.
(c) Kiick, K. L.; Saxon, E.; Tirrell, D. A.; Bertozzi, C. R. Proc. Natl. Acad.
Sci. U.S.A. 2002, 99, 19-24. (d) Ovaa, H.; van Swieten, P. F.; Kessler, B.
M.; Leeuwenburgh, M. A.; Fiebiger, E.; van den Nieuwendijk, A. M. C.
H.; Galardy, P. J.; van der Marel, G. A.; Ploegh, H. L.; Overkleeft, H. S.
Angew. Chem., Int. Ed. 2003, 42, 3626-3629. (e) Wang, C. C.-Y.; Seo, T.
S.; Li, Z.; Ruparel, H.; Ju, J. Bioconjugate Chem. 2003, 14, 697-701. (f)
Restituyo, J. A.; Comstock, L. R.; Petersen, S. G.; Stringfellow, T.; Rajski,
S. R. Org. Lett. 2003, 5, 4357-4360. (g) Vocadlo, D. J.; Hang, H. C.;
Kim, E.-J.; Hanover, J. A.; Bertozzi, C. R. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 9116-9121. (h) Hang, H. C.; Yu, C.; Kato, D. L.; Bertozzi, C.
R. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 14846-14851. (i) Luchansky,
S. J.; Argade, S.; Hayes, B. K.; Bertozzi, C. R. Biochemistry 2004, 43,
12358-12366. (j) Hang, H. C.; Yu, C.; Pratt, M. R.; Bertozzi, C. R. J. Am.
Chem. Soc. 2004, 126, 6-7. (k) Hosoya, T.; Hiramatsu, T.; Ikemoto, T.;
Nakanishi, M.; Aoyama, H.; Hosoya, A.; Iwata, T.; Maruyama, K.; Endo,
M.; Suzuki, M. Org. Biomol. Chem. 2004, 2, 637-641. (l) Kho, Y.; Kim,
S. C.; Jiang, C.; Barma, D.; Kwon, S. W.; Cheng, J.; Jaunbergs, J.;
Weinbaum, C.; Tamanoi, F.; Falck, J.; Zhao, Y. Proc. Natl. Acad. Sci. U.S.A.
2004, 101, 12479-12484. (m) Tsao, M.-L.; Tian, F.; Schultz, P. G. Chem
BioChem 2005, 6, 2147-2149. (n) Sprung, R.; Nandi, A.; Chen, Y.; Kim,
S. C.; Barma, D.; Falck, J. R.; Zhao, Y. J. Proteome Res. 2005, 4, 950-
957. (o) Comstock, L. R.; Rajski, S. R. Nucleic Acids Res. 2005, 33, 1644-
1652. (p) Rose, M. W.; Xu, J.; Kale, T. A.; O’Doherty, G.; Barany, G.;
Distefano, M. D. Pept. Sci. 2005, 80, 164-171. (q) Xu, J.; DeGraw, A. J.;
Duckworth, B. P.; Lenevich, S.; Tann, C.-M.; Jenson, E. C.; Gruber, S. J.;
Barany, G.; Distefano, M. D. Chem. Biol. Drug Des. 2006, 68, 85-96.
(23) Prescher, J. A.; Dube, D. H.; Bertozzi, C. R. Nature 2004, 430,
873-877.
(11) Offer, J.; Boddy, C. N. C.; Dawson, P. E. J. Am. Chem. Soc. 2002,
124, 4642-4646.
(12) Marinzi, C.; Bark, S. J.; Offer, J.; Dawson, P. E. Bioorg. Med. Chem.
2001, 9, 2323-2328.
(13) Bode, J. W.; Fox, R. M.; Baucom, K. D. Angew. Chem., Int. Ed.
2006, 45, 1248-1252.
(14) Nilsson, B. L.; Kiessling, L. L.; Raines, R. T. Org. Lett. 2001, 3,
9-12.
(15) Nilsson, B. L.; Kiessling, L. L.; Raines, R. T. Org. Lett. 2000, 2,
1939-1941.
(24) Dube, D. H.; Prescher, J. A.; Quang, C. N.; Bertozzi, C. R. Proc.
Natl. Acad. Sci. U.S.A. 2006, 103, 4819-4824.
(25) Azoulay, M.; Tuffin, G.; Sallem, W.; Florent, J.-C. Bioorg. Med.
Chem. Lett. 2006, 16, 3147-3149.
(16) Soellner, M. B.; Nilsson, B. L.; Raines, R. T. J. Org. Chem. 2002,
67, 4993-4996.
J. Org. Chem, Vol. 71, No. 26, 2006 9825

Soellner et al.
SCHEME 2. Reaction of Alanyl Phosphinothioester 10 with
Alanyl Azide 11
SCHEME 3. Synthesis of Alanyl Phosphinothioesters 10, 14,
and 15
Herein, we overcome the limitation of having a glycine
residue at the ligation junction during a traceless Staudinger
ligation. To do so, we discern the basis for the inefficiency of
non-glycyl couplings. We then discover that the rational tuning
of the electronic structure of tetrahedral intermediate 4 (Scheme
1) by both substitution on the phenyl groups of (diphenylphos-
phino)methanethiol and solvent effects can enable the traceless
Staudinger ligation between non-glycyl residues to proceed in
high yield. This discovery expands the synthetic utility of the
traceless Staudinger ligation for the synthesis of proteins as well
as other molecules.
Results and Discussion
NMR-Based Assay To Monitor the Staudinger Ligation.
We recently reported on a means to monitor the traceless
Staudinger ligation by 13C NMR spectroscopy using an or-
ganic azide enriched with 13C at its R-carbon.17 This NMR-
based assay offers the ability to observe the reaction continu-
ously and without its perturbation, and (most importantly) it
reveals a product-distribution profile that is unperturbed by the
vagaries of chemical purification. This NMR-based assay was
used for determining all product yields herein, unless noted
otherwise.
Effect of Phenyl Substitutents. Ala + Ala couplings were
examined as a model ligation involving two non-glycyl resi-
dues.28 As reported previously, the coupling of alanyl phosphi-
nothioester 10 with 13C-labeled alanyl azide 11 in DMF results
in a yield of only 36% (Scheme 2).17 This moderate yield is
largely due to the formation of a byproduct, phosphonamide
13, which results from the aza-Wittig reaction of the imino-
phosphorane.
that oxygen forms a covalent bond with the oxophilic phos-
phorus atom, resulting in oxazaphosphetane 8. Why is Path A
disfavored relative to Path B in an Ala + Ala coupling? In such
a non-glycyl coupling, both the thioester group of 1 (RC(O)-
SCH2PPh2) and the azido group of 2 (R′N3) are secondary. In
tetrahedral intermediate 4, the bulky R and R′ groups of a non-
glycyl coupling are brought into close proximity. The ensuing
steric strain could be relieved upon formation of the O-P bond
in oxazaphosphetane 8. We reasoned that tuning the electron
density on the phosphorus atom could discourage the formation
of this bond and thereby alter the partitioning of tetrahedral
intermediate 4. The result would be a higher yield for the
coupling of non-glycyl residues.
To test our hypothesis, we synthesized alanyl phosphi-
nothioesters 14 and 15 and determined their abilities relative to
alanyl phosphinothioester 10 to undergo a traceless Staudinger
ligation with an alanyl azide. The p-chloro groups in phosphi-
nothioester 14 should decrease the electron density on phos-
phorus in tetrahedral intermediate 4. According to our hypoth-
esis, this decrease in electron density should lead to greater
partitioning toward the aza-Wittig reaction (Scheme 1, Path B).
Conversely, the p-methoxy groups of phosphinothioester 15
should increase the electron density on phosphorus in tetrahedral
intermediate 4 and thereby favor the Staudinger ligation.
Phosphinothioesters 10, 14, and 15 were synthesized by a route
analogous to that used to synthesize related phosphinothioesters
(Scheme 3).14,29
The key species in the partitioning toward Staudinger ligation
product 6 (Scheme 1, Path A) and aza-Wittig byproduct 9 (Path
B) is tetrahedral intermediate 4. In the Staudinger ligation (Path
A), an electron pair on the oxygen of tetrahedral intermediate
4 displaces a thiol to form amidophosphonium salt 5. In the
alternative aza-Wittig reaction (Path B), an electron pair from
(26) Lee and co-workers have reported that 14 different non-glycyl
couplings mediated by (diphenylphosphino)methanethiol proceed in “quan-
titative yield” (Kim, H.; Cho, J. K.; Aimoto, S.; Lee, Y.-S. Org. Lett. 2006,
8, 1149-1151). These results cannot be reproduced in our laboratory (ref
17) and are inconsistent with reports from other laboratories (Merkx, R.;
Rijkers, D. T. S.; Kemmink, J.; Liskamp, R. M. J. Tetrahedron Lett. 2003,
44, 4515-4518. He, Y.; Hinklin, R. J.; Chang, J.; Kiessling, L. L. Org.
Lett. 2004, 6, 4479-4482.).
(27) Merkx, R.; Rijkers, D. T. S.; Demink, J.; Liskamp, R. M. J.
Tetrahedron Lett. 2003, 44, 4515-4518.
(28) We chose an Ala + Ala coupling because alanine is especially
abundant in proteins (8.0% of all residues; ref 7).
The reaction of alanyl azide 11 with each alanyl phosphi-
nothioester (10, 14, and 15) was performed in DMF, and yields
(29) Grandjean, C.; Boutonnier, A.; Guerreiro, C.; Fournier, J. M.;
Mulard, L. A. J. Org. Chem. 2005, 70, 7123-7132.
9826 J. Org. Chem., Vol. 71, No. 26, 2006

Staudinger Ligation of Peptides at Non-Glycyl Residues
FIGURE 1. Plot of the yield of amide 12 product versus Hammett σp
value for the reaction of alanyl phosphinothioesters 10, 14, and 15 with
13C-labeled alanyl azide 11 in DMF and dioxane. The lines depict a
linear least-squares fit of the data.
TABLE 1. Effect of para Substituents on the Calculated
Gas-Phase Electron Density on Phosphorus in Tetrahedral
Intermediate 4
were obtained with the NMR-based assay.17 Under these
conditions, alanyl phosphinothioester 10 provided the desired
amide 12 product in 47% yield. In agreement with our
hypothesis, the p-chloro compound 14 afforded a diminished,
34%, yield of amide 12. The p-methoxy compound 15 also
followed the expected trend, conferring the amide product in
61% yield. This yield represents a substantial increase from any
reported previously for a non-glycyl Staudinger ligation.17,27 The
yield of amide product in this Ala + Ala coupling correlates
inversely with the Hammett σp value of the phenyl substituent
in the phosphinothioester (Figure 1).
Density functional theory (DFT) calculations30 were per-
formed with the Gaussian 98 package31 to quantify the effect
of the phenyl substituents on the electron density on phosphorus
in tetrahedral intermediate 4 (R ) R′ ) CH3). The DFT
calculations indicated that this electron density increases
considerably as the substituent became more electron-donating
(Table 1). (By comparison, the electron density on nitrogen
remains relatively constant.) Accordingly, both experiment and
theory are consistent with the weakening of the O‚‚‚P interaction
in tetrahedral intermediate 4 as being responsible for the
observed increase in the yield of amide product.
X
parameter
Cl
H
OMe
σp
+0.23
+0.131
-0.342
0.00
+0.062
-0.326
-0.27
+0.044
-0.322
P, charge
N, charge
determining step: phosphazide formation.17 To determine the
effect of solvent on the partitioning of iminophosphorane 3
between Staudinger ligation product 6 (Path A in Scheme 1)
and aza-Wittig byproduct 9 (Path B), the reaction of p-methoxy
phosphinothioester 15 and 13C-labeled alanyl azide 11 was
performed in nine solvents of varying polarity (acetone, aceto-
nitrile, dimethylformamide, dioxane, methanol, methylene chlo-
ride, nitromethane, tetrahydrofuran, and toluene), and the yields
were obtained by using the NMR-based assay.17
The yield of amide product exhibited a strong inverse
correlation with solvent polarity polarizability (SPP) value32
(Figure 2). Thus, the highest yields (in contrast to the highest
rates17) were obtained with nonpolar solvents. The yield of
AcAlaAlaNHBn product was high (83%) in dioxane and nearly
quantitative (99%) in toluene. These yields represent a sub-
stantial increase from those published previously for peptide
ligation at two non-glycyl residues.17,26 Each of the previously
described alanyl phosphinothioesters (p-chlorophenyl 14, p-
methoxyphenyl 15, and phenyl 10) was reacted with alanyl azide
11 in dioxane. The yields of AcAlaAlaNHBn in dioxane again
correlated well with the Hammett σp constant for the substituent
but were uniformly greater than the yields in the more polar
solvent, DMF (Figure 1).
Effect of Solvent Polarity. Previous efforts aimed at
optimizing the Staudinger ligation showed that the reaction
proceeds more rapidly in polar solvents, presumably due to
favorable stabilization of a polar transition state in the rate-
(30) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density
Functional Theory, 2nd ed.; Wiley-VCH: Weinheim, Germany, 2001.
(31) Frisch, M. J.; Trucks, G. W.; Schlegel, G. W.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.;
Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-
Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P.
M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.;
Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.9;
Gaussian, Inc.: Pittsburgh, PA, 1998.
To obtain insight on the ability of nonpolar solvents to provide
higher yields of the desired peptide product, we performed DFT
calculations on the effect of solvent on the electron density on
phosphorus in tetrahedral intermediate 4 (R ) R′ ) CH3). The
(32) Catalan, J.; Lopez, V.; Perez, P.; Martinvillamil, R.; Rodriguez, J.
G. Liebigs Ann. 1995, 241-252.
J. Org. Chem, Vol. 71, No. 26, 2006 9827

Soellner et al.
TABLE 3. Isolated Yields of Amide Product from the Reactions of
Phosphinothioesters 15 and 25 with 13C-Labeled Alanyl Azide 11
FIGURE 2. Plot of the yield of amide 12 product versus SPP value
for the reaction of alanyl phosphinothioester 15 with 13C-labeled alanyl
azide 11. The line depicts a linear least-squares fit of the data.
TABLE 2. Effect of Solvent Polarity on the Calculated Electron
Density on Phosphorus and Nitrogen in Tetrahedral Intermediate 4
solvent
amide product. This finding can be depicted by the dissection
parameter
H2O
THF
+
of tetrahedral intermediate 4 into resonance forms 4P (which
P, charge
N, charge
+0.082
-0.348
+0.028
-0.335
δ+
has less charge dispersal) and 4P . To emphasize the importance
+
δ+
of 4P and 4P to the product distribution, we have refined the
mechanism of the Staudinger ligation to be that in Scheme 4.
Coupling of Non-Glycyl Residues. Using those conditions
most favorable for non-glycyl ligations, we performed a series
of model ligations to determine the scope of these reactions.
Dioxane was chosen as the solvent, because many starting
materials and products were insoluble in the most nonpolar
solvent tested, toluene. 13C-Labeled alanyl azide 11 was reacted
(dioxane, 12 h) with alanyl phosphinothioester 15 and phenyl-
alanyl phosphinothioester 25 to form AcAlaAlaNHBn (12) and
AcPheAlaNHBn (26), respectively. After chromatography on
silica gel, each of these couplings afforded the desired dipeptide
in >80% isolated yield (Table 3). (Yields from the NMR-based
assay17 were indistinguishable from those in Table 3.) These
yields, which are for the coupling of equimolar reactants,
represent a substantial improvement upon those reported previ-
ously for traceless Staudinger ligations at two non-glycyl amino
acids.17,26,33 Unlike with (diphenylphosphino)methanethiol, in-
creasing temperature was not found to increase yield of ligation
product (data not shown).
SCHEME 4. Refined Mechanism of the Staudinger Ligation
Including the Resonance Forms of Tetrahedral Intermediate
4
The Staudinger ligation is highly chemoselective when
(diphenylphosphino)methanethiol is the coupling reagent.17,19
To determine the effect of the p-methoxy groups on chemose-
lectivity, the coupling of 13C-labeled alanyl azide 11 and alanyl
phosphinothioester 15 was performed in the presence of glycyl
amine 2717 (Scheme 5). Gratifyingly, no decrease in the
AcAlaAlaNHBn 12 product was observed, indicating that this
new phosphinothioester provides the high chemoselectivity of
reagents reported previously.17
Finally, Staudinger ligations involving a glycyl residue are
rapid.17,19 The NMR-based assay17 was used to monitor the rate
at which 13C-labeled alanyl azide 11 reacted with alanyl
phosphinothioester 15 (Table 3). In dioxane, this reaction was
calculations indicated that this electron density increases in
nonpolar solvents (Table 2), presumably due to the dispersal
of charge into the phenyl groups. (Again, the electron density
on nitrogen remains relatively constant.)
All of the experimental and theoretical results from reactions
with different phenyl substituents and in solvents of different
polarity are consistent. These results indicate that increasing
the electron density on phosphorus leads to a higher yield of
(33) Merkx, R.; Rijkers, D. T. S.; Kemmink, J.; Liskamp, R. M. J.
Tetrahedron Lett. 2003, 44, 4515-4518.
9828 J. Org. Chem., Vol. 71, No. 26, 2006

Staudinger Ligation of Peptides at Non-Glycyl Residues
SCHEME 5. Chemoselectivity of the Reaction of Alanyl
Phosphinothioester 15 and 13C-Labeled Alanyl Azide 11
chromatography (3% v/v MeOH in CH2Cl2). Phosphine oxide 21
was isolated as a clear, colorless oil in 87% yield.
1
Spectral Data. H NMR (CDCl3, 400 MHz) δ 7.70-7.65 (m,
4H), 6.99-6.97 (m, 4H), 3.85 (s, 6H), 3.70 (d, J ) 8.7 Hz, 2H),
2.28 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 162.9, 133.2,
123.4, 123.3, 114.4, 55.6, 30.4, 28.0 (d, J ) 70.6 Hz) ppm; 31P
NMR (CDCl3, 161 MHz) δ 29.19 ppm; MS (ESI) m/z 373.0627
(MNa+ [C17H19O4PSNa+] ) 373.0634).
AcSCH2P(C6H4-p-Cl)2 (22) and AcSCH2P(C6H5)2 (23). Phos-
phines 2229 and 2314 were synthesized from phosphine oxides 19
and 20 according to reports published previously. Spectral Data.
Spectral data were as reported previously.14,29
AcSCH2P(C6H4-p-OMe)2 (24). Phosphine oxide 21 (1.06 g, 2.95
mmol) was dissolved in anhydrous chloroform (10 mL). Trichlo-
rosilane (8 mL, 79 mmol) was added, and the resulting solution
was stirred under Ar(g) for 72 h. The solvent was then removed
under reduced pressure. (CAUTION: Excess trichlorosilane in the
removed solvent was quenched by the slow addition of saturated
sodium bicarbonate in a well-ventilated hood.) The residue was
purified by flash chromatography (silica gel, 3% v/v MeOH in CH2-
Cl2). Phosphine 24 was isolated as a white solid in 92% yield.
found to proceed with a second-order rate constant of k2 ) 2.1
× 10-3 M-1 s-1, which is similar to that reported previously
for a Gly + Gly coupling mediated by (diphenylphosphino)-
methanethiol in DMF/D2O (6:1) (k2 ) 7.7 × 10-3 M-1 s-1).17
Conclusions
1
Spectral Data. H NMR (CDCl3, 400 MHz) δ 7.38-7.36 (m,
This work has illuminated factors critical for achieving the
traceless Staudinger ligation of peptides at non-glycyl sites using
novel phosphinothiols. Proper tuning of the electron density on
phosphorus enables non-glycyl amino acid residues to be
coupled rapidly and in high (>80%) yield, providing a
benchmark for the sequence-independent ligation of peptides
at amino acid residues. These findings enhance the utility of
the Staudinger ligation for the iterative ligation of peptides in
the convergent synthesis of proteins.
4H), 6.93-6.89 (m, 4H), 3.85 (s, 6 H), 3.47 (bs, 2H), 2.32 (s, 3H)
ppm; 13C NMR (CDCl3, 125 MHz) δ 194.4, 160.5, 134.2, 134.0,
127.9, 127.8, 114.2, 114.2, 55.2, 30.3, 26.4 (d, J ) 20.4 Hz) ppm;
31P NMR (CDCl3, 161 MHz) δ -19.71 ppm; MS (ESI) m/z
357.0679 (MNa+ [C17H19O3PSNa+] ) 357.0685).
AcAlaSCH2P(C6H4-p-Cl)2 (14). NaOH (44.8 mg, 1.12 mmol)
was added to a solution of phosphine 22 (384 mg, 1.12 mmol)
dissolved in degassed MeOH (12 mL). The resulting mixture was
allowed to stir under Ar(g) for 1.5 h. The solvent was then removed
under reduced pressure. The residue was dissolved in CH2Cl2, and
this solution was washed with 2 N HCl and brine. The organic
layer was dried over MgSO4(s) and filtered, and the solvent was
removed under reduced pressure. The resulting residue was used
below without further purification. In a separate round-bottom flask,
N-acetylalanine (157 mg, 1.2 mmol) was dissolved in anhydrous
DMF (8 mL). HOBT (148 mg, 1.1 mmol) was added to the resulting
solution, followed by DCC (138 mg, 1.1 mmol). Once precipitate
(DCU) was observed, freshly deprotected phosphinothiol from
above was added (330 mg, 1.09 mmol). The reaction mixture was
allowed to stir under Ar(g) for 4 h. The solvent was then removed
under reduced pressure. The residue was purified by automated
chromatography (50% v/v ethyl acetate in hexanes). Phosphi-
nothioester 14 was isolated as a white solid in 93% yield.
Experimental Procedures
[2-13C]-(2S)-2-Azido-N-benzyl-1-propionamide (11). 13C-
Labeled alanyl azide 11 was synthesized as described previously.17
ClCH2P(O)(C6H4-p-Cl)2 (16), ClCH2P(O)(C6H5)2 (17). Phos-
phine oxides 1629 and 1714 were prepared according to reports
published previously.
Spectral Data. Spectral data were as reported previously.14,29
ClCH2P(O)(C6H4-p-OMe)2 (18). Chloromethylphosphonic dichlo-
ride (20 g, 120 mmol) was dissolved in anyhdrous THF (120 mL).
A solution of 4-methoxyphenylmagnesium bromide (0.5 M) in THF
(480 mL, 240 mmol) was added dropwise over 1 h. The resulting
mixture was stirred at reflux for 24 h. The reaction was then
quenched by the addition of water (20 mL), and the solvent was
removed under reduced pressure. The residue was dissolved in CH2-
Cl2, and the resulting solution was washed once with water (50
mL) and once with brine (50 mL). The organic layer was dried
over anhydrous MgSO4(s) and filtered, and the solvent was removed
under reduced pressure. The residue was purified by automated
chromatography (3% v/v MeOH in CH2Cl2). Phosphine oxide 18
was isolated as a white solid in 63% yield.
1
Spectral Data. H NMR (CDCl3, 400 MHz) δ 7.36-7.32 (m,
8H), 5.28 (bt, 1H), 4.70 (bt, 1H), 3.46 (d, J ) 6.3 Hz, 2H), 2.03 (s,
3H), 1.62 (m, 1H), 1.32 (d, J ) 6.3 Hz, 3H) ppm; 13C NMR (CDCl3,
125 MHz) δ 199.6, 169.6, 135.8, 134.1, 133.9, 129.0, 128.9, 54.9,
25.2 (d, J ) 24.6 Hz), 23.12, 18.77 ppm; 31P NMR (CDCl3, 161
MHz) δ -16.94 ppm; MS (ESI) m/z 436.0059 (MNa+ [C18H18Cl2-
NO2PSNa+] ) 436.0065).
AcAlaSCH2P(C6H4)2 (10). NaOH (148 mg, 3.71 mmol) was
added to a solution of phosphine 23 (1.02 g, 3.71 mmol) dissolved
in degassed MeOH (35 mL). The resulting mixture was allowed to
stir under Ar(g) for 1.5 h. The solvent was then removed under
reduced pressure. The residue was dissolved in CH2Cl2, and this
solution was washed with 2 N HCl and brine. The organic layer
was dried over MgSO4(s) and filtered, and the solvent was removed
under reduced pressure. The resulting residue was used without
further purification below. In a separate round-bottom flask,
N-acetylalanine (557 mg, 4.25 mmol) was dissolved in anhydrous
DMF (40 mL). HOBT (527 mg, 3.90 mmol) was added to the
resulting solution, followed by DCC (805 mg, 3.90 mmol). Once
precipitate (DCU) was observed, freshly deprotected phosphinothiol
from above was added (900 mg, 3.87 mmol). The reaction mixture
was allowed to stir under Ar(g) for 4 h. The solvent was then
removed under reduced pressure. The residue was purified by flash
1
Spectral Data. H NMR (CDCl3, 400 MHz) δ 7.77-7.72 (m,
4H), 7.04-7.00 (m, 4H), 4.00 (d, J ) 7.0 Hz, 2H), 3.88 (s, 6H)
ppm; 13C NMR (CDCl3, 125 MHz) δ 162.9, 133.5, 133.4, 121.5,
120.37, 114.4, 114.2, 55.4, 38.1 (d, J ) 73.2 Hz) ppm; 31P NMR
(CDCl3, 161 MHz) δ 27.38 ppm; MS (ESI) m/z 310.0530 (MNa+
[C15H16ClO3PNa+] ) 310.0526).
AcSCH2P(O)(C6H4-p-Cl)2 (19) and AcSCH2P(O)(C6H5)2 (20).
Phosphine oxides 1929 and 2014 were synthesized from phosphine
oxides 16 and 17 according to reports published previously.
Spectral Data. Spectral data were as reported previously.14,29
AcSCH2P(O)(C6H4-p-OMe)2 (21). Phosphine oxide 18 (5.8 g,
18.2 mmol) was dissolved in DMF (50 mL). Potassium thioacetate
(2.5 g, 21.8 mmol) was then added, and the reaction mixture was
stirred under Ar(g) for 18 h. The solvent was then removed under
reduced pressure. The resulting oil was purified by automated
J. Org. Chem, Vol. 71, No. 26, 2006 9829

Soellner et al.
chromatography (silica gel, 70% v/v ethyl acetate in hexanes).
removed under reduced pressure. The residue was purified by flash
chromatography (silica gel, 50% v/v ethyl acetate in hexanes).
Phosphinoester 25 was isolated as an off-white, crusty solid in 90%
yield.
Phosphinothioester 10 was isolated as a white solid in 83% yield.
1
Spectral Data. H NMR (CDCl3, 400 MHz) δ 7.44-7.37 (m,
10H), 5.89 (bt, 1H), 4.71 (bt, 1H), 3.53 (s, 2H), 2.02 (s, 3H), 1.64
(m, 1H), 1.32 (d, J ) 5.8 Hz, 3H) ppm; 13C NMR (CDCl3, 125
MHz) δ 199.8, 169.7, 136.5, 132.7, 132.6, 132.6, 129.1, 128.5,
128.5, 54.8, 25.3 (d, J ) 24.3 Hz), 22.99, 18.66 ppm; 31P NMR
(CDCl3, 161 MHz) δ -17.97 ppm; MS (ESI) m/z 368.0853 (MNa+
[C18H20NO2PSNa+] ) 368.0845).
1
Spectral Data. H NMR (CDCl3, 400 MHz) δ 7.37-7.33 (m,
4H), 7.26-7.25 (m, 4H), 7.09-7.07 (m, 2H), 6.91-6.89 (m, 3H),
5.71 (m, 1H), 4.96 (m, 1H), 3.81 (s, 6H), 3.50-3.34 (m, 2H), 3.12-
2.96 (m, 2H), 1.93 (s, 3H) ppm; 13C NMR (CDCl3, 125 MHz) δ
199.2, 169.9, 160.8, 135.6, 134.4, 129.5, 128.9, 127.8, 127.4, 114.5,
55.4, 38.5, 26.6 (d, J ) 23.1 Hz), 23.34, 21.30 ppm; 31P NMR
(CDCl3, 161 MHz) δ -18.49 ppm; MS (ESI) m/z 504.1393 (MNa+
[C26H28NO4PSNa+] ) 504.1374).
AcAlaSCH2P(C6H4-p-OMe)2 (15). NaOH (330 mg, 8.26 mmol)
was added to a solution of phosphine 24 (2.76 g, 8.26 mmol)
dissolved in degassed MeOH (80 mL). The resulting mixture was
allowed to stir under Ar(g) for 1.5 h. The solvent was then removed
under reduced pressure. The residue was dissolved in CH2Cl2, and
this solution was washed with 2 N HCl and brine. The organic
layer was dried over MgSO4(s) and filtered, and the solvent was
removed under reduced pressure. The resulting residue was used
below without further purification. In a separate round-bottom flask,
N-acetylalanine (1.13 g, 8.70 mmol) was dissolved in anhydrous
DMF (40 mL). HOBT (1.05 g, 7.80 mmol) was added to the
resulting solution, followed by DCC (984 mg, 7.80 mmol). Once
precipitate (DCU) was observed, freshly deprotected phosphinothiol
from above was added (2.30 g, 7.86 mmol). The reaction mixture
was allowed to stir under Ar(g) for 4 h. The solvent was then
removed under reduced pressure. The residue was purified by
automated chromatography (70% v/v ethyl acetate in hexanes).
Phosphinothioester 15 was isolated as a white solid in 94% yield.
Ligation of 15 and 11. Alanyl phosphinothioester 15 (85 mg,
0.21 mmol) and alanyl azide 11 (43 mg, 0.21 mmol) were added
to a reaction vessel and dissolved in anhydrous dioxane (2 mL).
The resulting reaction mixture was allowed to stir for 12 h. Amide
12 was isolated as a white solid in 81% yield (silica gel, 3% v/v
MeOH in CH2Cl2).
Spectral Data. Spectra data were as reported previously.17
Ligation of 25 and 11. Alanyl phosphinothioester 25 (100 mg,
0.21 mmol) and alanyl azide 11 (43 mg, 0.21 mmol) were added
to a reaction vessel and dissolved in anhydrous dioxane (2 mL).
The resulting reaction mixture was allowed to stir for 12 h. Amide
26 was isolated as a white solid in 84% yield (silica gel, 3% v/v
MeOH in CH2Cl2).
Spectral Data. 1H NMR (CDCl3/CD3OD, 1:1, 400 MHz) δ
7.35-7.17 (m, 10H), 4.62-4.59 (m, 1H), 4.39-4.34 (m, 1H), 4.34
(bs, 2H), 3.10-2.90 (m, 2H), 1.94 (s, 3H), 1.35 (d, J ) 7.0 Hz,
3H) ppm; 13C NMR (CDCl3, 125 MHz) δ 172.7, 171.7, 137.7,
136.3, 128.9, 128.0, 49.7, 42.6, 38.6, 22.1, 20.9, 17.6, 16.3; MS
(ESI) m/z 391.1799 (MNa+ [C21H25N3O3Na+] ) 391.1794).
Computational Methods. A global minimum-energy structure
for the tetrahedral intermediate 4 was determined by Monte Carlo
calculations using the program Macromodel and the Merck mo-
lecular force field. The minimum-energy structure thus obtained
was imported into Gaussian 9831 and minimized further with DFT
calculations30 at the B3LYP/6-31+g(d,p) level of theory. (If a
negative frequency was obtained, then the structure was optimized
again at the same level of theory.) Single-point energy calculations
were then performed in a variety of solvents by using a conductor-
like polarizable continuum model.34 Mulliken charges were obtained
from the single-point energy calculations.
1
Spectral Data. H NMR (CDCl3, 400 MHz) δ 7.39-7.35 (m,
4H), 6.92-6.90 (m, 4H), 5.96 (bt, 1H), 5.70 (bt, 1H), 3.83 (s, 6H),
3.45 (m, 1H), 2.02 (s, 3H), 1.32 (d, J ) 6.4 Hz, 2H), 1.15 (d, J )
6.7 Hz, 3H) ppm; 13C NMR (CDCl3, 125 MHz, numbers in
parentheses indicate doubling due to rotational isomerism) δ 200.0,
170.2, 160.5, 134.26, 127.6, 126.6, 125.9, 117.5, 114.2, 114.2,
111.0, 55.2 (55.0), 42.2, 26.0 (d, J ) 23.2 Hz), 23.30 (23.04), 18.76
ppm; 31P NMR (CDCl3, 161 MHz) δ -19.02 ppm; MS (ESI) m/z
428.1065 (MNa+ [C20H24NO4PSNa+] ) 428.1056).
AcPheSCH2P(C6H4-p-OMe)2 (25). NaOH (180 mg, 4.49 mmol)
was added to a solution of phosphine 24 (1.50 g, 4.49 mmol)
dissolved in degassed MeOH (45 mL). The resulting mixture was
allowed to stir under Ar(g) for 1.5 h. The solvent was then removed
under reduced pressure. The residue was dissolved in CH2Cl2, and
this solution was washed with 2 N HCl and brine. The organic
layer was dried over MgSO4(s) and filtered, and the solvent was
removed under reduced pressure. The resulting residue was used
below without further purification. In a separate round-bottom flask,
N-acetylphenylalanine (920 mg, 4.44 mmol) was dissolved in
anhydrous DMF (20 mL). HOBT (546 mg, 4.04 mmol) was added
to the resulting solution, followed by DIC (510 mg, 4.04 mmol).
After being stirred for 20 min, freshly deprotected phosphinothiol
(1.18 g, 4.04 mmol) from above was added. The reaction mixture
was allowed to stir under Ar(g) for 4 h. The solvent was then
Acknowledgment. We are grateful to Dr. J. A. Hodges and
L. D. Lavis for contributive discussions. This work was
supported by Grant GM44783 (NIH). M.B.S. was supported
by an ACS Division of Organic Chemistry Fellowship, spon-
sored by Abbott Laboratories.
Supporting Information Available: General experimental
procedures, NMR spectra of compounds 10, 14, 15, 18, 21, 24,
25, and 26, and atom coordinates and absolute energies from the
theoretical calculations. This material is available free of charge
via the Internet at http://pubs.acs.org.
(34) (a) Klamt, A.; Schu¨u¨rmann, G. J. Chem. Soc., Perkin Trans. 1993,
2, 799. (b) Andzelm, J.; Ko¨lmel, C.; Klamt, A. J. Chem. Phys. 1995, 103,
9312-9320. (c) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995-
2001. (d) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem.
2003, 24, 669-681.
JO0620056
9830 J. Org. Chem., Vol. 71, No. 26, 2006