S-(2-Pyrimidinyl)- and S-(2-(4,6-dimethylpyrimidinyl))-1,1,3,3- tetramethylthiouronium hexafluorophosphates: Novel reagents for in situ peptide coupling

Article abstract of DOI:10.1016/j.tetlet.2005.11.067

Two new reagents for in situ peptide coupling based on the 2-mercaptopyrimidine core have been developed. The readily prepared thiouronium salts were found to promote both peptide and segment coupling efficiently with low racemization/epimerization levels

Full text of DOI:10.1016/j.tetlet.2005.11.067

Tetrahedron
Letters
Tetrahedron Letters 47 (2006) 483–486
S-(2-Pyrimidinyl)- and S-(2-(4,6-dimethylpyrimidinyl))-1,1,3,3-
tetramethylthiouronium hexafluorophosphates: novel
reagents for in situ peptide coupling
*
¨
¨
Philip Garner, Ozge Sßeßseno g˘ lu and H. Umit Kaniskan
Department of Chemistry, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, OH 44106-7078, USA
Received 14 October 2005; revised 4 November 2005; accepted 11 November 2005
Available online 1 December 2005
This paper is dedicated to Professor Miklos Bodanszky on the occasion of his 90th birthday and in
appreciation of his lasting contributions to the field of peptide chemistry
Abstract—Two new reagents for in situ peptide coupling based on the 2-mercaptopyrimidine core have been developed. The readily
prepared thiouronium salts were found to promote both peptide and segment coupling efficiently with low racemization/epimeriza-
tion levels.
ꢀ
2005 Elsevier Ltd. All rights reserved.
Because of the importance of both natural and unnatu-
ral peptides as structurally defined molecular scaffolds,
there is continued interest in new and effective ways to
NMe2
O
NMe2
PF
6
N
N
N
O
Me N
N
N
N
N
NMe2
N
P
N
2
O
N
N
1
construct the peptide bond. One line of research has
N
PF6
Y
PF
6
Y
been concerned with the development of in situ peptide
coupling reagents that both activate the carboxylic acid
component and facilitate amide bond formation in one
pot. A working mechanism for this transformation
Y
6
8
(Y = CH)
(Y = N)
5
a
(Y = CH)
(Y = N)
5b
7b
7a
(
Eq. 1) starting from carboxylic acid 1 involves the for-
Figure 1. HOAt and HOBt-based peptide coupling reagents.
mation of an active ester 2 (X = leaving group) followed
by nucleophilic acyl substitution with amine 3 to give
the amide 4. Two general classes of in situ peptide cou-
pling reagents have emerged from these studies (Fig. 1).
5
stereochemical integrity of the a-carbon. An analogous
phosphonium reagent 8 (PyAOP, Y = N) was reported
6
the next year. While originally formulated as uronium
salts, the structures of HBTU and HATU were later
reformulated as guanidinium salts 5b (N-HBTU) and
7b (N-HATU) based on X-ray crystallographic data.
A deliberate synthesis of O-HATU (7a) was recently
reported. Based on the limited number of direct com-
parisons (Ref. 8), O-HATU appears to be a more effec-
tive peptide coupling reagent than N-HATU but it is not
commercially available.
The reagents are exemplified by the uronium salt 5
2
(
HBTU, Y = CH ) and phosphonium salt 6 (PyBOP,
2
3
Y = CH ), both of which were based on the peptide
coupling additive HOBt.
7
2
4
8
r
+
eagent
base
R' NH
R"
O
O
O
3
R'
R
N
ð1Þ
R
OH
R
X
R"
1
2
4
The design of N-HATU was based on the hypothesis
that the nitrogen atom at C-7 would accelerate peptide
coupling by facilitating intramolecular acylation of the
amine component via the hydrogen bonded ensemble 9
A major advance came in 1993, when Carpino reported
the effectiveness of the in situ peptide coupling reagent 7
(
HATU, Y = N) in terms of both chemical yield and
(
Fig. 2). It was argued that the seven-membered ring
*
Corresponding author. Tel.: +1 216 368 3696; fax: +1 216 368
006; e-mail: ppg@case.edu
transition state (TS) would be favored over the possible
alternative six-membered TS (H-bonding to N-2)
3
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.11.067

484
P. Garner et al. / Tetrahedron Letters 47 (2006) 483–486
O
O
iol active esters for peptide coupling was subsequently
O
N
O
N
reported in a patent by Japanese workers but this work
N
N
S
H
S
R
R
R
20
was apparently never followed up on.
N
N
N
N
R'
H
N
N
R'
H
R'
R''
R''
R''
Two new coupling reagents, S-(2-pyrimidinyl)-1,1,3,3-
tetramethylthiouronium hexafluorophosphate (14) and
S-(4,6-dimethyl-2-pyrimidinyl)-1,1,3,3-tetramethylthio-
uronium hexafluorophosphate (16) were prepared by
routes analogous to the one we had developed for HPTT
9
10
11
Figure 2. Proposed activated complexes.
2
1
because the former better accommodated a more linear
H-transfer. The enhanced rate could also be attributed
to the increased acidity of 7-amino-1-hydroxybenzo-
triazole (HOAt, pK = 8.7 in DMSO). However, sub-
(Eq. 2). Thus, the known chloroformamidinium salt 12
9
was combined with 2-mercaptopyrimidine 13 and 4,6-
dimethyl-2-mercaptopyrimidine 15 in the presence of
triethylamine to give the thiouronium hexafluorophos-
phate salt 14 (mp 84–87 ꢁC) in 65% yield and the corre-
sponding dimethyl analog 16 (mp 100–102 ꢁC) in 94%
1
0
a
sequent studies with N-HATU isomers underscored
the special influence of the 7-amino group.11 Because
of its effectiveness, N-HATU has become the reagent
of choice for difficult peptide couplings. Unfortunately,
N-HATU is an expensive reagent and this has limited its
use to some extent. The cost of N-HATU can be traced
to its 7-aza-1-hydroxybenzotriazole (HOAt) core, which
is prepared by a multistep synthesis. With this back-
ground in mind, we explored a new class of in situ
peptide coupling reagents based on the inexpensive
yield, respectively. The molecular structures of these
1
compounds and their purity were established by
H
1
3
and C NMR spectroscopy as well as combustion analy-
sis. Both of these compounds are air stable, non-hygro-
2
2
scopic crystalline solids.
1
2
R
R
R
PF
6
NMe2
PF
6
N
N
N
N
NMe2
Cl
Et N/DCM
3
2
-mercaptopyrimidine core.
+
HS
S
Me N
2
Me N
ð2Þ
2
1
2
R
This study actually began during our development of S-
2-pyridinyl)-1,1,3,3-tetramethylthiouronium hexafluoro
phosphate (HPTT) for the synthesis of 2-pyridinethiol
esters. At the time, the possibility that this reagent
might also be used for in situ peptide coupling was con-
sidered. The use of preformed 2-pyridinethiol active
esters for peptide synthesis had been reported by Loyd
and Young in 1968. Shortly thereafter, Mukaiyama
described an in situ variant of this reaction in conjunction
with his oxidation–reduction condensation technology
1
1
3 (R = H)
5 (R = Me)
14 (R = H)
16 (R = Me)
(
1
3
To evaluate the potential of 14 as a peptide coupling
reagent, we examined its use for the synthesis of three
2
3
dipeptides:
Cbz-Phe-Val-OMe, Cbz-Gly-Phe-OMe,
1
4
24
and Cbz-Phg-Pro-NH2. We also used these reagents
for the assembly of four tripeptides by segment cou-
pling: Cbz-Phe-Val-Ala-OMe, Cbz-Phe-Val-Pro-NH2,
Cbz-Gly-Phe-Val-OMe, and Cbz-Gly-Phe-Pro-NH2.
These particular test cases were chosen primarily be-
cause they had been previously synthesized using N-
HATU. In this way, a direct comparison between these
reagents could be made in terms of reaction rate and
racemization/epimerization levels. With the exception
of reagent used, we followed the general reaction condi-
tions and workup procedure of Carpino. The initial cou-
plings were effected by combining equimolar quantities
of the carboxylic acid, coupling reagent, and the free
amine (or its HCl salt), in the presence of 2 (or 3) equiv-
alents of diisopropylethylamine (DIEA) in DMF. As
indicated by the data in Table 1, reagent 14 is an effec-
tive peptide and segment coupling reagent. When com-
pared to N-HATU, the only significant differences
occurred with two of the tripeptides (entries 2 and 6)
and with the very stringent test case involving the stereo-
chemically labile amino acid Cbz-Phg-OH (entry 7),
0
15
using 2,2 -dipyridyldisulfide and triphenylphosphine.
In a significant application, Kurokawa and Ohfune
successfully used 2-pyridinethiol esters for peptide cou-
pling to silylated a-aminoacids in their total synthesis
of echinocandin D. It can be argued that, in these reac-
tions the peptide bond forming step proceeded via a
pre-organized complex akin to 10.17 While HPPT did
promote in situ peptide coupling, it was clearly inferior
to N-HATU in terms of the reaction rate and racemiza-
tion level when applied to a stringent test case.
1
6
Mechanistic considerations suggested to us that the
overall rate of peptide coupling might be enhanced if
the 2-pyridinethiol moiety (pK = 9.81 in H O) of
1
8
a
2
HPTT were replaced with a 2-pyrimidinethiol moiety
1
9
(
pK = 7.01 in H O). Such a replacement would not
a 2
only make the active ester carbonyl more electrophilic
but also cause the tetrahedral intermediate to collapse
faster. This substitution would also preclude the possi-
bility of an unproductive thiol ester conformation (plac-
ing the carbonyl anti to the heterocyclic nitrogen), thus
enhancing the rate of coupling on entropic grounds.
Finally, 2-mercaptopyrimidine is commercially available
at a cost that is less than 1/10 of HOAt. The first hint
of the benefit of 2-pyrimidinethiol esters for peptide cou-
pling can actually be found (as a single example) in the
Lloyd and Young full paper cited above. A more exten-
sive description of the use of preformed 2-pyrimidineth-
2
5
where the degree of racemization was higher.
We next set out to optimize the reaction conditions to
reduce the level of racemization observed during the
coupling of Cbz-Phg-OH to H-Pro-NH (Table 2). Since
2
longer reaction times did not lead to increased levels of
racemization, it was hypothesized that a-deprotonation
was occurring at the active ester stage and was related
2
6
to the base strength of DIEA (pK = 10.1). Reducing
a
the amount of DIEA to one equivalent did not help

P. Garner et al. / Tetrahedron Letters 47 (2006) 483–486
Table 1. Coupling reactions with reagent 14
485
23
Entry
Peptide
Base (equiv)
Crude yield (%)
DL or LDL (%)
1
2
3
4
5
6
7
Cbz-Phe- -Val-OMe
Cbz-Phe-Val- -Ala-OMe
DIEA (3)
DIEA (3)
DIEA (2)
DIEA (3)
DIEA (3)
DIEA (2)
DIEA (2)
93
2.7 ± 0.1
6.2 ± 0.1
4.1 ± 0.15
79
Cbz-Phe-Val- -Pro-NH
2
86
Cbz-Gly- -Phe-OMe
quant.
69
—
Cbz-Gly-Phe- -Val-OMe
Cbz-Gly-Phe- -Pro-NH
Cbz-Phg- -Pro-NH
2.7 ± 0.1
4.3 ± 0.1
22.0 ± 1
2
61
2
95
Table 2. Optimization studies with Cbz-Phg-Pro-NH
2
Entry
Reagent
Base (equiv)
Additive (equiv)
Crude yield (%)
DL (%)
1
2
3
4
5
6
14
14
14
14
16
16
DIEA (1)
TMP (1)
—
DIEA (1)
DIEA (1)
TMP (1)
—
—
—
84
87
51
85
86
82
21.0
11.7
9.6
31.0
8.8
2-Mercaptopyrimidine (1)
—
—
9.1
(
entry 1). However, switching to the weaker base 2,4,6-
of the American Chemical Society Petroleum Research
Fund.
collidine (TMP) (pK = 7.43) cut racemization in half
a
(
entry 2). A similar level of racemization was observed
without any added base but the yield of product was
lower (entry 3). The fact that racemization actually
increased in the presence of an extra equivalent of 2-mer-
captopyrimidine (entry 4) suggested that this by-product
actually contributed to the racemization problem. This
reasoning led us to try the dimethylated reagent 16 since
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tetlet.
2005.11.067.
4
,6-dimethyl-2-mercaptopyrimidine was expected to be
a more sterically hindered, kinetically weaker base.
Indeed, the use of reagent 16 with 1 equivalent of either
DIEA or 2,4,6-collidine brought the level of racemiza-
tion down to 9% while retaining a high conversion
References and notes
. (a) Bodanszky, M. Principles of Peptide Synthesis, 2nd ed.;
Springer: New York, 1993; (b) Chemical Synthesis of
Peptides. In Bioorganic Chemistry: Peptides and Proteins;
Hecht, S. M., Ed.; Oxford University Press: New York,
1998; pp 27–64; (c) Han, S. Y.; Kim, Y. A. Tetrahedron
2004, 60, 2447.
1
(
entries 5 and 6). While this improvement was gratifying,
reagent 16 still did not outperform N-HATU in terms of
reaction rate and racemization level. The former differ-
ence was clearly seen with a very hindered dipeptide.
N-HATU mediated coupling of Cbz-Aib-OH and H-
Aib-OMe produced Cbz-Aib-Aib-OMe in 56% yield.
However, the reaction stalled at the active ester stage
2. Dourtoglou, V.; Ziegler, J.-C.; Gross, B. Tetrahedron Lett.
978, 15, 1269.
1
2
7
3. Castro, B.; Domoy, J. R.; Evin, G.; Selve, C. Tetrahedron
Lett. 1975, 1219.
. Reagent abbreviations used: HOAt = 7-aza-1-hydroxy-
when either 14 or 16 was used. These results may
underscore the necessity of a geometrically favored H-
bond between amine and active ester in TS.
4
benzotriazole; HOBt = N-hydroxybenzotriazole; PyAOP =
7
-azobenzotriazolyoxytris(pyrrolidino)phosphonium hexa-
In conclusion, we have developed a new type of reagent
for in situ peptide coupling based on the inexpensive
fluorophosphate; PyBOP = 1-benzotriazolyoxytris(pyrr-
olidino)phosphonium hexafluorophosphate; N-HBTU =
N-[(1H-benzotriazol-1-yl)-(dimethylamino)methylene]-N-
methylmethanaminium hexafluorophosphate N-oxide; N-
HATU = N-[(dimethylamino)-1H-1,2,3-triazolo-[4,5-b]pyr-
idin-1-yl-methylene]-N-methylmethanaminium hexafluoro-
phosphate N-oxide.
2
-mercaptopyrimidine core. The resulting thiouronium
salts 14 and 16 are easy to prepare and, in favorable
instances, may provide a cost effective alternative to
N-HATU and N-HBTU.
5
6
7
. Carpino, L. A. J. Am. Chem. Soc. 1993, 115,
4
397.
Acknowledgements
. Carpino, L. A.; El-Faham, A.; Minor, C. A.; Albericio, F.
J. Chem. Soc., Chem. Commun. 1994, 201.
. Abdelmoty, I.; Albericio, F.; Carpino, L. A.; Foxman, B.
M.; Kates, S. Lett. Pept. Sci. 1994, 1, 57.
We thank Hugh Burgoon and Nicholas Sizemore for
their technical assistance. This work was supported by
a grant from the National Science Foundation (CHE-
8. Carpino, L. A.; Imazumi, H.; El-Faham, A.; Ferrer, F. J.;
Zhang, C.; Lee, Y.; Foxman, B. M.; Henklein, P.;
0
111831). Acknowledgement is also made to the Donors

486
P. Garner et al. / Tetrahedron Letters 47 (2006) 483–486
Hanay, C.; M u¨ gge, C.; Wenschuh, H.; Klose, J.; Beyer-
11 19 6 4
C H F N PS: C, 34.38; H, 4.98; N, 14.58. Found: C,
mann, M.; Bienert, M. Angew. Chem., Int. Ed. 2002, 41,
34.39; H, 5.12; N, 14.40.
442.
23. General procedure for peptide couplings:
9
. Gandour, R. D. Tetrahedron Lett. 1974, 295.
To 0.25 mmol of the acid, 0.25 mmol of amine (or amine
salt), and 0.50 (or 0.75) mmol base in 2 (or 3) mL of DMF
was added 0.25 mmol of coupling reagent at 0 ꢁC. The
reaction mixture was stirred at 0 ꢁC for 1 h and at rt for
1.5 h for dipeptides and 3 h for tripeptides. The mixture
was diluted with 50 mL of EtOAc, and extracted with 1 N
1
0. Koppel, I.; Kopple, J.; Leito, I.; Pihl, V.; Grehn, L.;
Ragnarsson, U. J. Chem. Res., Synop. 1993, 446.
1. Carpino, L. A.; Imazumi, H.; Foxman, B. M.; Vela, M. J.;
Henklein, P.; El-Faham, A.; Klose, J.; Bienert, M. Org.
Lett. 2000, 2, 2253.
1
1
1
1
2. HOAt is commercially available ($2000/mol) or can be
synthesized from 3-hydroxy-2-nitropyridine.
3. Scardovi, N.; Garner, P. P.; Protasiewicz, J. D. Org. Lett.
HCl (2 · 20 mL), satd NaHCO (2 · 20 mL), brine (2 ·
3
4
20 mL), and dried over MgSO . The solvent was removed
under vacuum and crude peptide was directly analyzed by
HPLC unless otherwise indicated. Solid NaCl was added
to break up emulsions formed during the extractions.
According to NMR and HPLC analysis no more than 3%
(w/w) of 2-mercaptopyrimidine dimer was detected in
crude products prepared with reagent 14. All of the
peptides synthesized in this paper are known. The NMR
data obtained were consistent with that reported previ-
2003, 5, 1633.
4. (a) Lloyd, K.; Young, G. T. J. Chem. Soc., Chem.
Commun. 1968, 1400; (b) Lloyd, K.; Young, G. T. J.
Chem. Soc. (C) 1971, 2890.
1
5. (a) Mukaiyama, T.; Matsueda, R.; Suzuki, M. Tetra-
hedron Lett. 1970, 1901; (b) Matsueda, R.; Maruyama, H.;
Ueki, M.; Mukaiyama, T. Bull. Chem. Soc. Jpn. 1971, 44,
2
8–31
1
373.
6. Kurokawa, N.; Ohfune, Y. J. Am. Chem. Soc. 1986, 108,
043.
7. The Me
ously.
1
1
24. Abbreviations used: Cbz = (benzy1oxy)carbonyl; Phe =
(S)-phenylalanine; Val = (S)-valine; Ala = (S)-alanine;
Gly = glycine; Phg = (S)-phenylglycine; Pro = (S)-pro-
line; Aib = 2-aminoisobutyric acid.
25. Literature values: Cbz-Phe-Val-Ala-OMe: 82% yield, 1.9%
LDL (N-HATU + DIEA, Ref. 29a), 82% yield, 6.0% LDL
6
3
Al promoted condensation of 2-pyridinethiol
esters with amino esters has also been reported Kurosu,
M. Tetrahedron Lett. 2000, 41, 591.
8. Jones, R. A.; Katrizky, A. R. J. Chem. Soc. 1958, 3106.
9. Reddy, M. S.; Reddy, K. V. J. Ind. Chem. Soc. 1996, 73,
1
1
2
(N-HBTU + DIEA, Ref. 29a); Cbz-Phe-Val-Pro-NH :
3
45.
0. Nagasawa, T.; Kuroiwa, K.; Narita, K. U.S. Patent
,904,612, 1975.
81% yield, 12.7% LDL (N-HATU + DIEA, Ref. 31b),
5.9% LDL (N-HATU + TMP-DIEA, Ref. 8), ꢀ10% LDL
(O-HATU + DIEA, Ref. 8), 3.4% LDL (O-HATU +
TMP-DIEA, Ref. 8), 90% yield, 27.4% LDL (N-
HBTU + DIEA, Ref. 31b), 20.6% LDL (N-HBTU +
TMP-DIEA, Ref. 8), 10.3% LDL (O-HBTU + TMP-
2
3
2
2
1. See Supporting Information of Ref. 13.
2. S-(2-Pyrimidinyl)-1,1,3,3-tetramethylthiouronium hexa-
fluorophosphate (14): Tetramethylchloroformamidinium
1
3
hexafluorophosphate (5.0 g, 0.018 mol) and 2-pyrimid-
inethiol (2.9 g, 0.018 mol) were dissolved in 10 mL of
freshly distilled DCM. Triethylamine (2.5 mL, 0.018 mol)
was added dropwise at rt (a white smoke evolved) and the
reaction mixture was stirred for 10 min. It was then
concentrated under reduced pressure and the resulting
crude orange oil was crystallized from a mixture of 3:1
MeOH–i-PrOH to give white crystals, which were washed
2
DIEA, Ref. 8); Cbz-Gly-Phe-Pro-NH : 86% yield, 0.8%
LDL (N-HATU + DIEA, Ref. 31b), 85% yield, 5.9% LDL
(N-HBTU + DIEA, Ref. 31b); Cbz-Gly-Pro-NH2 81%
yield, 6.3% LDL (N-HBTU + DIEA, Ref. 31b). In our
hands, N-HATU gave 96% yield, 2.6% LDL for Cbz-Gly-
Phe-Val-OMe and 94% yield, 3.1% DL for Cbz-Phg-Pro-
NH2.
26. Carpino, L. A.; Ionescu, D.; El-Faham, A. J. Org. Chem.
1996, 61, 2460.
with i-PrOH and then Et
4.17 g). 65% yield, mp = 84–87 ꢁC; H NMR (300 MHz,
DMSO-d ) d 3.24 (s, 12H), 7.53 (t, J = 4.9 Hz, 1H), 8.81
2
O to give 14 as a white solid
1
1
(
27. Diagnostic H NMR peaks for active ester with reagent
14: 8.80 ppm, doublet, 7.27 ppm, triplet (aromatic pro-
tons); with 16: 6.99 ppm, singlet (aromatic proton) and
2.53 ppm (2 · Me).
6
1
3
(
1
d, J = 4.9 Hz, 2H); C NMR (75 MHz, CDCl
20.0, 159.0, 164.8, 169.3; HRMS (FAB+) m/z calcd for
C H N S 211.1012, found 211.1003; Anal Calcd for
3
) d 44.1,
28. Saha, A. K.; Schultz, P.; Rapoport, H. J. Am. Chem. Soc.
1989, 111, 4856.
9
15
4
C H F N PS: C, 30.34; H, 4.24; N, 15.73. Found: C,
9
15
6
4
3
0.29; H, 4.43; N, 15.51.
29. (a) Carpino, L. A.; El-Faham, A.; Albericio, F. J. Org.
Chem. 1995, 60, 3561; (b) Shieh, W. C.; Carlson, J. A.;
Shore, M. E. Tetrahedron Lett. 1999, 40, 7167.
30. Albericio, F.; Bailen, M. A.; Chinchilla, R.; Dodsworth,
D. J.; Najera, C. Tetrahedron 2001, 57, 9607.
31. (a) Wenschuh, H.; Beyermann, M.; Haber, H.; Seydel, J.
K.; Krause, E.; Bienert, M.; Carpino, L. A.; El-Faham,
A.; Albericio, F. J. Org. Chem. 1995, 60, 405; (b) Carpino,
L. A.; Xia, J.; El-Faham, A. J. Org. Chem. 2004, 69, 54.
S-(4,6-Dimethyl-2-pyrimidinyl)-1,1,3,3-tetramethylthiouro-
nium hexafluorophosphate (16) A similar procedure using
4
,6-dimethyl-2-pyrimidinethiol gave 16 as a white solid
1
(
7.59 g). 94% yield, mp = 100–102 ꢁC;
H
NMR
(
600 MHz, CDCl ) d 2.45 (s, 6H), 3.33 (s, 12H), 6.98 (s,
3
1
3
1H); C NMR (75 MHz, CDCl ) d 23.5, 43.9, 118.9,
3
1
63.6, 169.2, 169.7; HRMS (FAB+) m/z calcd for
11 19 4
C H N S 239.1325, found 239.1312; Anal Calcd for