Catalytic staudinger-vilarrasa reaction for the direct ligation of carboxylic acids and azides

Article abstract of DOI:10.1021/jo802825e

2,2 -Dipyridyl diselenide (PySeSePy) is the catalyst or activator of choice for the direct reaction of carboxylic acids with azides and trimethylphosphine at room temperature. The mechanism of the process, which is not an aza-Wittig reaction, has been elucidated.

Full text of DOI:10.1021/jo802825e

SCHEME 1. Direct Formation of Amide/Peptide Bonds
from Carboxylic Acids, Organic Azides, and Et3P
Catalytic StaudingersVilarrasa Reaction for the
Direct Ligation of Carboxylic Acids and Azides
Jordi Bure´s, Manuel Mart´ın, Fe`lix Urp´ı, and
Jaume Vilarrasa*
Departament de Qu´ımica Orga`nica, Facultat de Qu´ımica,
UniVersitat de Barcelona, 08028 Barcelona, Catalonia, Spain
jVilarrasa@ub.edu
ReceiVed January 1, 2009
al.8 examined and optimized the reactions of Staudinger
phosphazenes with carboxyl derivatives (cyclic anhydrides,
mixed anhydrides, thioesters,9 among others) and applied their
findings to the synthesis of peptides4k and macrolactam-like
natural products. The reactions with other carboxyl and carbon-
ate derivatives (RCOCl,10 simple esters intramolecularly,6,11
CH3COOCHO,12 Boc2O,13 selenoesters,14 and activated esters15)
have several “fathers” apart from our own research group. While
some researchers16 name the general process RCO-LG + R3P
+ N3R the Staudinger-Vilarrasa reaction (henceforward, S-V
reaction or S-V peptide ligation),17 other authors refer to these
reactions as particular cases of “aza-Wittig” processes.4c-f
We report the first catalytic version of the S-V reaction. It
may be related to the classical Mukaiyama reaction of 2-pyridyl
2,2′-Dipyridyl diselenide (PySeSePy) is the catalyst or
activator of choice for the direct reaction of carboxylic acids
with azides and trimethylphosphine at room temperature. The
mechanism of the process, which is not an aza-Wittig
reaction, has been elucidated.
(4) This is one of the reactions discovered by Staudinger: (a) Staudinger,
H.; Meyer, J. HelV. Chim. Acta 1919, 2, 635 (Ph3P + N3Ar to give Ph3PdNAr,
their hydrolyses, and reaction with CO2 and CS2). (b) Staudinger, H.; Hauser,
E. HelV. Chim. Acta 1921, 4, 861 (Ph3P, PhEt2P, and Et3P). For recent reviews
about the Staudinger reaction and iminophosphorane chemistry, see: (c) Palacios,
F.; Alonso, C.; Aparicio, D.; Rubiales, G.; Santos, J. M. Tetrahedron 2007, 63,
523. (d) Eguchi, S. ArkiVoc 2005, 98. (e) Ko¨hn, M.; Breinbauer, R. Angew.
Chem., Int. Ed. 2004, 43, 3106. (f) Fresneda, P. M.; Molina, P. Synlett 2004, 1.
For ab initio calculations of the phosphatriazene to phosphazene conversion,
see: (g) Alajar´ın, M.; Conesa, C.; Rzepa, H. S. J. Chem. Soc., Perkin Trans. 2
1999, 1811. (h) Widauer, C.; Grutzmacher, H.; Shevchenko, I.; Gramlich, V.
Eur. J. Inorg. Chem. 1999, 1659. (i) Tian, W. Q.; Wang, Y. A. J. Org. Chem.
2004, 69, 4299, and references therein. For the participation of phosphatriazenes
in some reactions, see: (j) Inazu, T.; Kobayashi, K. Synlett 1993, 34, 4671 (gly-
copeptides). (k) Bosch, I.; Urp´ı, F.; Vilarrasa, J. Chem. Commun. 1995,
91 (peptides). (l) Velasco, M. D.; Molina, P.; Fresneda, P. M.; Sanz, M. A.
Tetrahedron 2000, 56, 4079 (trapping of a Z-phosphazide).
(5) Horner, L.; Gross, A. Liebigs Ann. Chem. 1955, 591, 117. (“Tertiary
phosphines. IV. Use of phosphine imines in causing the introduction of primary
amino groups”). Such a reaction cannot be located via SciFinder; even currently,
with >190 entries via Ph3PdNPh, none is associated with this paper.
(6) Hickey, D. M. B.; MacKenzie, A. R.; Moody, C. J.; Rees, C. W. Chem.
Commun. 1984, 776.
(7) Zaloom, J.; Calandra, M.; Roberts, D. C. J. Org. Chem. 1985, 50, 2601.
(8) Relevant papers: (a) Garcia, J.; Vilarrasa, J.; Bordas, X.; Banaszek, A.
Tetrahedron Lett. 1986, 27, 639 (phthalic anhydride). (b) Bosch, I.; Romea, P.;
Urp´ı, F.; Vilarrasa, J. Tetrahedron Lett. 1993, 34, 4671 (mixed anhydrides and
thioesters, macrolactamization). (c) Reference 4k (peptides). (d) Bosch, I.;
Gonzalez, A.; Urp´ı, F.; Vilarrasa, J. J. Org. Chem. 1996, 61, 5638 (acyl chlorides
and mixed anhydrides, mechanistic studies). (e) Ariza, X.; Urp´ı, F.; Viladomat,
C.; Vilarrasa, J. Tetrahedron Lett. 1998, 39, 9101 (Boc-ON). (f) Ariza, X.; Urp´ı,
F.; Vilarrasa, J. Tetrahedron Lett. 1999, 40, 7515 (ClCOOR). (g) Ariza, X.;
Pineda, O.; Urp´ı, F.; Vilarrasa, J. Tetrahedron Lett. 2001, 42, 4995 (vicinal azido
alcohols with Boc2O or CO2).
In our search for the direct macrolactamization of ω-azido
carboxylic acids, we discovered years ago1 that carboxylic acids,
organic azides, and tertiary phosphines reacted slowly in benzene
or toluene to give carboxamides, N2, and phosphine oxide.
Triphenylphosphine was soon replaced by Bu3P and by Et3P
(Scheme 1),1b which is even more reactive and practical because
the water-soluble Et3PdO is easier to remove.2 Since organic
azides and phosphines give phosphatriazenes,3 which at room
temperature (rt) are converted to phosphazenes (Staudinger
phosphazenes)4 and N2, the rate-limiting step of our reaction
must be the last one, that is, the collapse of the aminophos-
phonium carboxylates (RCOO- Et3P+NHR′) to RCONHR′ and
Et3PdO.1
We were aware later that Horner and Gross had carried out
a related experiment: among series of reactions to prepare azides,
phosphimines, isothiocyanates, and thioureas, they indicated that
heating of Ph3PdNPh with benzoic acid in xylene gave
PhCONHPh in 30% yield.5 A few weeks before publication of
our first paper,1a Moody et al.6 reported an intramolecular
reaction of a (EtO)3P-generated phosphazene with an o-COOH
group, in the context of aza-Wittig-type cyclization reactions,
whereas a few weeks afterward Roberts et al.7 reported similar
results to our own with simple peptides. These authors are
independent codiscoverers of the reaction. Later, Vilarrasa et
(9) The “traceless ligation” developed by Raines et al. involving RCOSC6H4-
o-PR2 and later RCOSCH2PAr2 is an exciting application to the peptide field
and molecular biology. See: (a) Nilsson, B. L.; Kiessling, L. L.; Raines, R. T.
Org. Lett. 2000, 2, 1939. (b) Soellner, M. B.; Nilsson, B. L.; Raines, R. T. J.
Org. Chem. 2002, 67, 4993. (c) Soellner, M. B.; Nilsson, B. L.; Raines, R. T.
J. Am. Chem. Soc. 2006, 128, 8820, and references therein. Also see: (d) David,
O.; Meester, W. J. N.; Biera¨ugel, H.; Schoemaker, H. E.; Hiemstra, H.; van
Maarseveen, J. H. Angew. Chem., Int. Ed. 2003, 42, 4373 (medium-sized lactams).
(e) Kleineweischede, R.; Hackenberger, C. P. R. Angew. Chem., Int. Ed. 2008,
47, 5984 (peptide cyclization).
(1) (a) Garcia, J.; Urp´ı, F.; Vilarrasa, J. Tetrahedron Lett. 1984, 25, 4841 (12
simple amides, pyrrolidin-2-one, Z-Ile-Val-NHR, Ac-Sar-Gly-OEt, and Z-Ile-
Ile-Gly-OEt were prepared). (b) Urp´ı, F.; Vilarrasa, J. Tetrahedron Lett. 1986,
27, 4623 (advantges of Et3P). (c) Garcia, J. Ph.D. Thesis, Universitat de
Barcelona, 1986. (d) Urp´ı, F. Ph.D. Thesis, Universitat de Barcelona, 1987.
(2) Me3PdO is even more soluble in water, but Me3P (pyrophoric, low-
boiling liquid) was not commercially available at that time.
(3) Phosphatriazenes are also called phosphazides. λ5-Phosphazenes, or
iminophosphoranes, were also called phosphinimines or phosphine imines.
10.1021/jo802825e CCC: $40.75
Published on Web 02/09/2009
2009 American Chemical Society
J. Org. Chem. 2009, 74, 2203–2206 2203

thiol esters (RCOSPy) with aliphatic amines,18 which is a
standard in the peptide field and has been used in macrolac-
tonization reactions;19 C-terminal thioesters are key in many
approaches to the synthesis of proteins and glyco-conjugates,20
so we shall not discuss this type of carboxyl activation. The
advantages of operating with the N3 group (as an amino
protecting group) are clear4k,8,9,11,16 and will not be discussed
either.
Our purpose was to identify appropriate catalysts and solvents
for reactions such as those depicted in Tables 1 and 2. Me3P (1
equiv; commercially available in toluene or THF solutions) was
added to convert the N3 group to a Me3PdN group, while an
additional 0.3-1.4 equiv of Me3P and an additive were required
to activate the carboxyl group. With hindered carboxylic acids
we used 2.4 equiv of Me3P overall.
TABLE 1. Catalyst Screening: Yields of Boc-Phe-Gly-OEt
entry additive (catalyst or activator) yield, 24 h (%) yield, 48 h (%)
1
2
3
4
5
6
none
PhSSPh
PhSeSePh
4,4′-PySSPy
PySSPy (2,2′-PySSPy)
PySeSePy (2,2′-PySeSePy)
35
37
39
42
74
91
44
46
51
55
87
100
a Boc-Phe-OH (1.0 mmol), ethyl 2-azidoacetate (1.0 mmol), and
various additives (0.2 mmol) were mixed with 2.4 mL of 1 M toluene
solution of Me3P at 0 °C under N2. A few minutes later, the solutions
1
were stirred at rt for 24 or 48 h, quenched, and analyzed by H NMR.
The rate of the direct reaction of Boc-Phe-OH, Me3P, and
N3CH2COOEt (entry 1 of Table 1) increased significantly when
0.2 equiv of 2,2′-dipyridyl disulfide (PySSPy, entry 5) or 2,2′-
dipyridyl diselenide (PySeSePy, entry 6) was added. PySeSePy
appeared to be the catalyst or activator of choice also with
isobutyric acid instead of Boc-Phe-OH or when ethyl azidoac-
etate was replaced by benzyl azide (results not included for the
sake of simplification). Among the solvents (Table 2), we
confirmed1 that toluene (entry 6) was the best solvent. The 1:1
TABLE 2. Solvent Screeninga
entry
solvent
amide yield (%)
1
2
3
4
5
6
CH3CN-toluene (1:1)
pyridine-toluene (1:1)
CH2Cl2-toluene (1:1)
THF
DMF-toluene (1:1)
toluene
17
35
43
47
97
(10) The reaction of phosphazenes with acid chlorides had precedents: (a)
Masriera, M. An. Soc. Esp. F´ıs. Qu´ım. 1923, 21, 418 (Ph3PdNPh + AcCl, AcBr,
or PhCOCl). (b) Zbiral, E.; Stroh, J. Liebigs Ann. Chem. 1969, 29 (imidoyl
chlorides, to obtain tetrazoles). (c) Bachi, M. D.; Vaya, J. J. Org. Chem. 1979,
44, 4393 (reaction of PhOCH2COCl with a monobactam triphenylphosphazene).
(d) Barluenga, J.; Ferrero, M.; Palacios, F. J. Chem. Soc. Perkin Trans. 1 1990,
2193 (N-acylated vinylphosphazenes). (e) Molina, P.; Alajar´ın, M.; Lo´pez-
Leonardo, C. Tetrahedron Lett. 1991, 32, 4041 (rearrangement to C-phosphonium
salts, see below).
100
a To 1.0 mmol of isobutyric acid, 1.0 mmol of benzyl azide, and 0.2
mmol of PySeSePy in 2.4 mL of the “first” solvent were added 2.4 mL
of 1 M toluene (THF in entry 4) solutions of Me3P under N2.
(11) Named by Bertozzi et al. “the Staudinger ligation” in 2000: (a) Saxon,
E.; Bertozzi, C. R. Science 2000, 287, 2007 More than 20 papers since then
have been dedicated by this group to applications of this methodology to the
biology field. For other leading references, see: (b) Lin, F. L.; Hoyt, H. M.; van
Halbeek, H.; Bergman, R. G.; Bertozzi, C. R. J. Am. Chem. Soc. 2005, 127,
2686. (c) Baskin, J. M.; Bertozzi, C. R. QSAR Comb. Sci. 2007, 26, 1211.
(12) Hiebl, J.; Zbiral, E. Liebigs Ann. Chem. 1988, 765.
DMF-toluene mixture (entry 5) was as efficient as toluene for
substrates such as those of Table 2, but for more hindered cases
toluene alone was better.
When linear carboxylic acids were used instead of the
R-branched examples of Tables 1 and 2, the reactions were
always much more rapid but the final yields of the desired
amides dropped significantly with PhSSPh or PhSeSePh and
moderately when the additive was PySSPy; this was due to the
appearance of byproducts (see below). Again, the highest amide
yields were obtained with PySeSePy.
On the basis of these results, we carried out the reactions of
Table 3, with 240 mol % of Me3P and 20 mol % of PySeSePy.
The simplest amides (entries 1-5) were obtained in 90-100%
yields within 2 h; indeed, in these cases the amide yields were
practically quantitative even with only 130-150 mol % of Me3P
and 20 mol % of PySeSePy, in 3-4 h at rt.
When one of the reaction partners contained R-branched
chains (entries 6-9 of Table 3, substrates of Tables 1 and 2),
longer reaction times were required to complete the conversion.
As usual, reaction times could be shortened by (i) increasing
the amount of PySeSePy (footnote b of Table 3) or (ii) raising
the temperature (footnote d). No racemization took place
with the lactic acid derivative (entry 8 of Table 3) or in the
reaction of Boc-Phe-OH with methyl 2-azidoacetate (entry 6
of Table 1), as checked by chiral HPLC. No racemization
occurred in the preparation of Boc-Gly-Ala-OEt (entry 9 of
Table 3), but this was expected as enantiopure phosphazenes
of general formula R3PdNCHRCOOR do not racemize.4k
However, no method is perfect. The reaction of N-protected
amino acids (such as Boc-Phe-OH) with sterically hindered
(13) Afonso, C. Tetrahedron Lett. 1995, 36, 8857.
(14) The reaction of RCOOH, N3R′, PhSeSePh, and Bu3P was reported under
conditionssnearly equimolar amounts of substrates and reagentssin which
PhSeH was assumed to reduce the azide to amine (phosphazenes were not
involved). See: (a) Ghosh, S. K.; Singh, U.; Mamdapur, V. R. Tetrahedron Lett.
1992, 33, 805. (b) Ghosh, S. K.; Verma, R.; Ghosh, U.; Mamdapur, V. R. Bull.
Chem. Soc. Jpn. 1996, 69, 1705.
(15) (a) Kurosawa, W.; Kan, T.; Fukuyama, T. J. Am. Chem. Soc. 2003,
125, 8112 (pentafluorophenyl esters). (b) Malkinson, J. P.; Falconer, R. A.; Toth,
I. J. Org. Chem. 2000, 65, 5249 (activation with DIC or DCC and 1-hydroxy-
benzotriazole, HOBt). (c) Chapuis, H.; Strazewski, P. Tetrahedron 2006, 62,
12108 (DIC, HOBt).
(16) (a) Charafeddine, A.; Chapuis, H.; Strazewski, P. Chem.sEur. J. 2007,
13, 5566. (b) Charafeddine, A.; Chapuis, H.; Strazewski, P. Org. Lett. 2007, 9,
2787. (c) Chapuis, H.; Bui, L.; Bestel, I.; Barthe´le´my, P. Tetrahedron Lett. 2008,
49, 6838. (d) Saneyoshi, H.; Michel, B. Y.; Choi, Y.; Strazewski, P.; Marquez,
V. E. J. Org. Chem. 2008, 73, 9435.
(17) It avoids confusion with the reaction of azides and phosphines (and the
reaction of phosphazenes with H2O, with CO2, with CS2, with RNCO, and with
R2CdCdO, the classical Staudinger reactions) and the [2 + 2]-cycloaddition
of ketenes and imines (another Staudinger reaction).
(18) (a) Mukaiyama, T.; Matsueda, R.; Marayama, H. Bull. Chem. Soc. Jpn.
1970, 43, 1271. (b) Mukaiyama, T.; Matsueda, R.; Suzuki, M. Tetrahedron Lett.
1970, 1901. Also see: (c) Lloyd, K.; Young, G. T. Chem. Commun. 1968, 1400.
(19) (a) Corey, E. J.; Nicolaou, K. J. Am. Chem. Soc. 1974, 96, 5614. For a
recent review, see: (b) Parenty, A.; Moreau, X.; Campagne, J.-M. Chem. ReV.
2006, 106, 911.
(20) Reviews on the chemical synthesis of proteins: (a) Nilsson, B. L.;
Soellner, M. B.; Raines, R. T. Annu. ReV. Biophys. Biomol. Struct. 2005, 34,
91. (b) Dawson, P. Chem. Biol. 2007, 2, 567. (c) Hackenberger, C. P. R.;
Schwarzer, D. Angew. Chem., Int. Ed. 2008, 47, 10030. For recent papers on
chemical ligation, cf. (d) Haase, C.; Rohde, H.; Seitz, O. Angew. Chem., Int.
Ed. 2008, 47, 6807. (e) Blanco-Canosa, J. B.; Dawson, P. E. Angew. Chem.,
Int. Ed. 2008, 47, 6851. Activation via 4-nitrophenyl ester: (f) Wan, Q.; Chen,
J.; Yuan, Y.; Danishefsky, S. J. J. Am. Chem. Soc. 2008, 130, 15814.
2204 J. Org. Chem. Vol. 74, No. 5, 2009

TABLE 3. Direct Ligation of Carboxylic Acids and Organic
Azides, Mediated by Me3P and Catalyzed or Promoted by
PySeSePya
SCHEME 2. Plausible Mechanism of the Reaction of
RCH2COOH with Azides, Me3P, and ArXXAra
a ArXXAr ) PhSSPh, PhSeSePh, PySSPy, or PySeSePy.
of Me3PdNCH2Ph with PhCOSePh, PhCH(OTBS)COSPh,22
PhCH2CH2COSePy, and CH3(CH2)16COSePy.23 All reactants
were prepared separately and mixed in NMR tubes under N2,
and the spectra were registered at rt every 10 min. The signals
of the initial intermediates, corresponding to the expected “N-
phosphonium species” (see N-P in Scheme 2 as an example),
were maintained. We did not detect other intermediates. In other
words, C(XAr)dNCH2Ph moieties coming from aza-Wittig
reactions between the COXAr groups and Me3PdNCH2Ph were
not observed; the direct loss of Me3PdO if moisture and acid
were avoided was not observed either. In the last two experi-
ments (RCH2COSePy cases), only a very slow appearance of
1
other phosphonium species was detected, characterized by H
and 31P NMR as “C-phosphonium species” (C-P, see the
Supporting Information).
a Standard conditions: carboxylic acid (1.0 mmol), azide (1.0 mmol),
PySeSePy (0.2 mmol), 2.4 mL of 1 M Me3P in toluene, 0 °C, under N2,
stirring at rt for 2-48 h depending on the case. b This reaction was
The formation of C-phosphonium salts PhCH(PPh3+Cl-)-
CONHCH2R in reactions of PhCH2COCl with Ph3PdNCH2R
was first reported by Molina, Alajar´ın, et al.10e These anomalous
reactions, which are sometimes very rapid and give relatively
stable C-P derivatives that are not cleaved readily to the desired
amides, were re-examined by us8d with PhCH2COCl or
PhCH2COOCOAr vs Bu3PdNCH2Ph, confirming the results of
our colleagues.10e,24
carried out using 100 mol
% ꢀ anomer
of PySeSePy. c The
predominated (R/ꢀ, 1/10); we think that the re-equilibration of anomers
occurs at the N-phosphonium intermediate stage (see below). d At 100
°C in a MW reactor.
phosphazenes (such as that arising from the azido derivative of
Val-OEt) was very slow under our standard conditions. In such
cases (when both partners were R-branched), we also observed
the racemization of the stereocenter R to the carboxylic acid
moiety. Likely, the reaction was so slow that it allowed the
corresponding phosphazene to act as a base. The racemization
of Boc-Phe-SePy may occur via standard enolization, by
oxazolone formation, or through a ketene intermediate.21
Therefore, a catalytic protocol cannot be applied to these cases.
For the formation of peptide bonds without epimerization from
two crowded substrates, we recommend our stoichiometric
procedure (Boc-NHCHRCO-LG + N3CHR′COOEt, toluene,
0-5 °C, addition of Me3P or Bu3P)4k or those of refs 15 and
16.
When the reaction of Me3PdNCH2Ph and CH3(CH2)16-
COSePy was repeated with other thio- and selenoesters of stearic
acid, the rearrangements from the N-P to C-P intermediates
took place more quickly. Spectra at different reaction times are
given as Supporting Information. The conversion of
RCH2CON(PMe3XAr)CH2Ph to RCH(PMe3XAr)CONHCH2Ph
followed an increasing order for XAr equal to
SePy , SPy < SePh , SPh
In other words, RCH2COSePy gave rise to lower amounts of
C-P byproducts than RCH2COSPy, RCH2COSePh, and
To gain insight into the key mechanistic steps, we followed
by H NMR spectroscopy, in toluene/C6D6 at rt, the reactions
(22) The reaction of PhCH(OTBS)COSPh with Me3PdNCH2Ph, both
prepared separately, gave rise to PhCH(OTBS)CON(PMe3SPh)CH2Ph (structure
confirmed by 1H and 13C NMR, as well as by a 31P-decoupled 1H NMR
experiment), which was kinetically stable at rt.
(23) We choose the stearic (octadecanoic) acid derivative because of the
hydrophobicity and higher solubility in toluene of the corresponding intermediates
(from its reaction with phosphazenes).
1
(21) Under the standard conditions of Table 3, the reaction of (S)-
PhCH(OTBS)COOH and Me3PdNCH2Ph also stopped quickly, and the yield
of the corresponding amide was poor. With stoichiometric amounts of PySeSePy,
a 52% yield of the amide was isolated after 2 h, but 15% of racemization had
already taken place. The acidity of the CH proton (now R to COXAr and Ph)
and the phosphazene basicity make easy this racemization.
(24) Molina, P.; Alajar´ın, M.; Lo´pez-Leonardo, C.; Alca´ntara, J. Tetrahedron
1993, 49, 5153.
J. Org. Chem. Vol. 74, No. 5, 2009 2205

RCH2COSPh. This explains why the final yields were excellent
in the RCH2COSePy cases (that is, with PySeSePy as the
activator), but often poor (with SPh and SePh derivatives) or
moderate to good (with SPy derivatives) even at longer reaction
times.
RCH2COOH, the use of PySeSePy has the additional advantage
that a side reaction, the rearrangement of the N-phosphonium
species [RCH2CON(PMe3XAr)R′, prone to turnover] to the
relatively more stable C-phosphonium species [RCH(PMe3X)-
CONHR′, not prone to turnover] hardly takes place, in contrast
with the other additives. The background for improved ligation
procedures has been established.
Thus, these N-P to C-P rearrangements appear to be more
general than we had previously believed.8d The PR3+ or PR3XAr
groups migrate from the N atoms to the RCH2CO methylene
carbon atomssthe more acidic these methylene protons, the
faster the rearrangementsdepending on the temperature, me-
dium polarity, and mainly presence of bases (for example, of a
relative excess of phosphazene). On the other hand, there was
no migration of PR3XAr from the N atom to the more crowded
RR′CHCO methine carbon atoms in the cases we examined.
Such a migration is impossible in the PhCOOH derivatives.
Under our catalytic conditions, the carboxylic acids (relatively
in excess save at the end) can cleave the initial N-PMe3SePy
species as soon as they are formed. We noted that in the model
reaction, followed by NMR, of CH3(CH2)16COSePy and
Me3PdNCH2Ph, addition of the starting carboxylic acid had a
positive effect (much more significant than the addition of
2-selenopyridone), as the amide appeared immediately. Taking
into account the data currently available, a plausible mechanism
for the reaction of azides, Me3P, and ArXXAr with
RCH2COOHsthe “simplest” acids but the “most complex”
since, as indicated, N-P to C-P rearrangements may take
placesis summarized in Scheme 2. ArXXAr behave as activa-
tors rather than true catalysts.
In conclusion, a catalytic variant of the coupling reaction of
Staudinger phosphazenes with carboxyl groups is described here
for the first time. It is based on mild in situ activation of COOH
as COSePy groups (which overcome COSPy, COSePh, and
COSPh). Mechanistically, it is not an aza-Wittig process, a term
that should be confined to the reactions of phosphazenes with
aldehydes and ketones unless otherwise is demonstrated. The
Masriera reaction,10a for historical reasons, or the S-V
reaction16,17 are more appropriate names for the ligation of
azides and carboxylic acid derivatives. The quick reaction of
the intermediate N-PMe3SePy species with the remaining
carboxylic acid, which generates the amide and more sele-
noester, is key for the success of this catalytic version. For
Experimental Section
Typical Procedure. A commercially available solution of Me3P
(1.0 M in toluene, 2.40 mL, 2.4 mmol) was added to a mixture of
carboxylic acid (1.0 mmol), azide (1.0 mmol), and 2,2′-dipyridyl
diselenide (PySeSePy, 63 mg, 0.20 mmol) at 0 °C under nitrogen.
A few minutes later, when no more nitrogen bubbles were observed,
the ice bath was removed and the reaction was stirred at room
temperature for 2 h. Then, water (1 mL) was added and the mixture
was stirred for 10 min. The mixture was diluted with an excess of
CH2Cl2 and rinsed with aqueous NaHCO3, cold 1 M HCl (to remove
yellow 2-selenopyridone and 2,2′-dipyridyl diselenide), and finally
water. The organic phase was dried (MgSO4), filtered, and
concentrated to dryness. The crude amide or peptide was purified
by flash chromatography on silica gel; the elution conditions are
indicated in each case (Supporting Information). With hindered
substrates it was necessary to increase the reaction times, the catalyst
loading, or the temperature, as indicated in the main text.
Acknowledgment. The Ministerio de Educacio´n y Ciencia
contributed through Grant Nos. SAF2002-02728 and CTQ-2006-
15393 and a studentship to J.B. and the Generalitat de Catalunya
through Grant No. 2001SGR065. We acknowledge the work
of authors who, over the last 15 years, have used the methods
mentioned in the introduction, and we apologize for not citing
them (cf. refs 4c-f). This study is dedicated to Prof. Teruaki
Mukaiyama.
Supporting Information Available: General procedures and
experimental details for the preparation of the carboxamides.
1
NMR monitoring of different reactions. Copies of H and 13C
NMR spectra. This material is available free of charge via the
Internet at http://pubs.acs.org.
JO802825E
2206 J. Org. Chem. Vol. 74, No. 5, 2009