New and highly efficient immonium type peptide coupling reagents: Synthesis, mechanism and application

Article abstract of DOI:10.1016/S0040-4020(00)00365-3

A series of novel immonium type coupling reagents BOMI, BDMP, BPMP, DOMP, SOMP, FOMP and AOMP were designed and synthesized. It was shown that most of these reagents were more efficient in peptide synthesis than conventional methods due to the advantages

Full text of DOI:10.1016/S0040-4020(00)00365-3

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 4437±4445
New and Highly Ef®cient Immonium Type Peptide Coupling
Reagents: Synthesis, Mechanism and Application
*
Peng Li and Jie-Cheng Xu
Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, 354 Feng Lin Road, Shanghai 200032, People's Republic of China
Received 11 January 2000; revised 3 April 2000; accepted 20 April 2000
AbstractÐA series of novel immonium type coupling reagents BOMI, BDMP, BPMP, DOMP, SOMP, FOMP and AOMP were designed
and synthesized. It was shown that most of these reagents were more ef®cient in peptide synthesis than conventional methods due to the
1
advantages of high reactivity, low racemization and good yields. The reaction mechanism was proposed and studied by H NMR, IR, UV and
HPLC. q 2000 Elsevier Science Ltd. All rights reserved.
Introduction
ef®ciency, the HOAt-derived phosphonium and uronium
reagents, such as HATU, HAPyU, AOP, PyAOP, were
9
In recent years a large number of the HOBt-based uronium1
and phosphonium type coupling reagents, such as PyBOP,
suggested with good virtues because of the anchimeric
assistance effect of HOAt. Here we report an alternative
10
2
3
4
5
HBTU, HBPyU, HBPipU and HBMDU, have been
developed and widely employed in peptide synthesis both
in solution and solid phase since the ®rst HOBt-derived
phosphonium coupling reagent BOP was designed by
pathway to enhance the coupling ef®ciency by modifying
the carbon skeleton of uronium salts by replacing one of the
substituted amino groups with a hydrogen, alkyl or aryl
group. Thus the novel immonium type coupling reagents
were exploited and we observed that the reactivity of
these reagents was much higher than that of the correspond-
ing uronium salts. Using these immonium reagents to
synthesize peptides not only enhanced the coupling
ef®ciency but also substantially suppressed the extent of
racemization under relatively mild reaction conditions.
6
Castro et al. (Fig. 1). Although the predominance of carbo-
diimides and active ester methods are being gradually
replaced with these onium salts based upon 1-hydroxy-
benzotriazole (HOBt), the problem of racemization is still
7
3
a,8
a challenge during the use of the above reagents. To
suppress the level of racemization and enhance the coupling
Figure 1. General structures of HOBt- and HOAt-based uronium/aminium and phosphonium type coupling reagents.
Keywords: peptide coupling reagents; reaction mechanism; immonium salts.
*
Corresponding author. Tel.: 186-21-6416-3300 ext.1235; fax: 186-21-6416-6128; e-mail: xjcheng@pub.sioc.ac.cn
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00365-3

4
438
P. Li, J.-C. Xu / Tetrahedron 56 (2000) 4437±4445
Scheme 1. Synthesis of immonium type coupling reagents.
Results and Discussion
two nitrogen atom are delocalized within the bonds N±C±N,
which causes the carbocation to share a relatively high elec-
tron density; consequently the nucleophilic attack of the
carboxylate anion is impeded during the activation step.
To overcome this drawback, a series of novel immonium
salts were designed and synthesized by replacing one of the
substituted amino groups of the uronium reagents with
hydrogen, alkyl or aryl. Thus the electron density of the
central carbon of immonium salts was much lower than
that of corresponding uronium salts, and resulted in
improvement of the reactivity of these immonium type
reagents due to the electronic effect.
HOBt- or HOAt-based onium salt-mediated amide forma-
tion reaction generally involves two steps, activation and
1
1
coupling. The activation step, in which coupling reagent
reacts with N-protected amino acid to form the HOBt or
HOAt active ester of the carboxylic acid, was mainly in¯u-
enced by the nature of the carbon skeleton structure besides
the in¯uence of the carboxylic component. The coupling
step, in which the active ester of the N-protected amino
acid reacts with the amino component to form the peptide,
was mainly determined by the nature of the active ester
moiety of the coupling reagent and, of course, amino
component. Therefore, one effective way to enhance the
coupling ef®ciency is to modify the active ester moiety of
the molecules, which has been proved by the successful
design and application of HOAt-based onium salts to replace
the HOBt, HOPfp, HOSu or HODhbt-derived onium salts.
While our approach to enhance the coupling ef®ciency is
based on the modi®cation of the structures of the carbocation
skeleton moiety of uronium/aminium type reagents which
display relatively higher reactivity and lower racemization
These immonium type reagents can be readily prepared by
treating N-disubstituted amides with bis(trichloromethyl)
carbonate to yield the corresponding immonium chlorides,
which were stabilized with SbCl and subsequently reacted
5
with the potassium salts of active hydroxyl compounds or
hydroxyl compounds themselves in the presence of tertiary
amine, such as NEt , to afford the desired compounds as
3
crystalline and shelf-stable solids (Scheme 1).
1
2
than the corresponding phosphonium analogues.
Concerning the stability, activity, simplicity and easy avail-
ability of the amides, N,N-dimethyl formamide (DMF),
N-methyl pyrrolidone (NMP) and N-benzoyl pyrrolidine
were selected as substrates, thus a series of immonium
type coupling reagents were synthesized. Some represent-
ative immonium salts, such as BOMI, BDMP, BPMP,
Considering the structural feature of uronium salts shown in
Fig. 1, it is obvious that the two substituted amino groups of
these molecules provide two equal resonance structures to
stabilize these uronium salts, the lone electron pairs of the
Figure 2. Structures of novel immonium type coupling reagents and their intermediates.

P. Li, J.-C. Xu / Tetrahedron 56 (2000) 4437±4445
4439
both in reactivity and in racemization, HOBt-based immo-
nium salts are superior to the uronium salt HBPyU which
was very reactive among HOBt-based phosphonium and
3
a,12
uronium salts,
salt HAPyU, one of the most ef®cient reagents up to
even better than HOAt-derived uronium
1
1,19
now.
9.6% for BDMP, 81.7% for BPMP, meantime only 3.3%
for HBPyU and 11.2% for HAPyU after 10 min reaction at
108C with 2,6-lutidine as base in THF (Table 1, entries 3,
, 9, 13, 14). The extremely high reactivity of these reagents
For example, the yield is up to 20.1% for BOMI,
8
2
7
can also be seen by the result of entry 2. Adopting strong
base DIEA or NMM, the coupling reaction was completed
within 5 min using BOMI, even at 2708C. The ef®ciency of
HOBt-base immonium salts, such as BOMI, can also be
further enhanced by the addition of HOAt (entry 6).
Among these immonium salts, HOBt-based immonium salts
were shown to be the most promising. The rationality of the
molecular design based upon the modi®cation of HOBt-
based uronium salts was also con®rmed by the high reac-
tivity of the HOBt-immonium salts. Although HOAt-
derived immonium salt AOMP demonstrated high reactivity
and low racemization, the coupling yield was not satis-
factory yet. It was most likely that the extremely high reac-
tivity of AOMP led to decomposition before activation of
the N-protected amino acid was achieved in the presence of
base, such as 2,6-lutidine. As for the HODhbt-derived
reagent DOMP, it also showed low racemization, but the
yield was relatively low. HOPfp- and HOSu-derived immo-
nium salts give poor results due to the low reactivity of the
penta¯uorophenyl esters or succinimidyl esters formed in
situ during coupling. It has been reported that benzotriazolyl
Figure 3. Comparison of reactivity of HOBt-based immonium salts vs
HBPyU and HAPyU (reaction conditions: T, 108C; base, 2,6-lutidine;
solvent, THF. Substrate ratio: N-protected amino acid/amino acid ester
hydrochloride/coupling reagent/baseˆ1:1.1:1.1:3 (mol)).
DOMP, FOMP, SOMP, AOMP and their intermediates are
listed in Fig. 2. X-Ray analysis indicated that the reagent
BOMI presents as the N-substituted form shown in Fig. 2,
1
3
1
4,15
rather than as the isomeric O-substituted form. This kind
of isomerization was also observed in the HOBt-based
1
6
esters, and can be explained by molecular orbital calcu-
3
esters undergo aminolysis up to 10 times faster than suc-
20
1
7
lations using the semi-empirical method PM3.
cinimidyl and related esters. Among the developed HOBt-
based immonium salts, the reactivity of BDMP was the
highest which was probably due to the tension of the
intra-annular imide bond of the molecule. The angles of
The ef®ciency of these immonium reagents was evaluated
by HPLC monitoring the segment coupling reaction: Z-Gly-
Phe-OH1Val-OCH ´HCl!Z-Gly-Phe-Val-OCH , which
sp2
sp3
sp2
sp3
C
±N±C2 and N±C ±C4 in the molecule of BDMP
are equal to 105.7 and 113.78, respectively, by calculating
3
3
1
8
was ®rst adopted by Van der Auwera et al. From the
results shown in Fig. 3 and Table 1, it is obviously that,
2
with PCMODEL software. This tension or distortion may
1
Table 1. Comparison of reactivity and racemization of immonium type reagents with uronium type reagents and active esters using HPLC (model reaction:
Z-Gly-Phe-OH1Val-OCH
3
´HCl1coupling reagent1base!Z-Gly-Phe-Val-OCH
3
)
1
Reactivity t (min)
/2
a
Entry
Coupling reagent
Reaction conditions
Yield (%)
10 min
Racemization dl%
Base
Temp. (8C)
Solvent
2 min
120 min
1
2
3
4
5
6
7
8
9
BOMI
BOMI
BOMI
BOMI
BOMI
BOMI1HOAt
BDMP
AOMP
BPMP
BPMP
DOMP
FOMP
NMM
NMM
2,6-Lutidine
DIEA
230
270
210
210
210
210
210
210
210
210
210
210
210
210
210
210
CH
CH
THF
CH CN
3
CN
Cl
89.9
87.6
6.43
12.7
69.1
41.1
51.0
64.1
37.6
49.6
±
9.2
2.0
2.4
42.0
90.9
90.3
89.1
20.1
49.2
80.4
73.5
89.6
64.2
81.7
67.9
±
13.8
3.29
11.2
90.8
91.5
90.4
89.3
93.4
72.7
80.5
86.0
95.3
64.2
96.2
69.1
5.57
54.3
16.1
78.3
94.8
91.7
,1
,1
20
10.1
,1
4.5
1.9
,1
2.8
3.1
3.0
3.1
4.2
3.3
1.7
2.2
1.6
2.3
2.3
19.1
12.2
14.6
3.3
4.4
2
2
b
b
3
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
2,6-Lutidine
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
10
11
12
13
14
15
16
2.0
.120
110
.120
49
2.7
,1
HBPyU
HAPyU
Z-G-F-OBt
Z-G-F-OAt
a
dl% equal to d-isomer% multiplied by two.
Preactivated 10 min before the amino component Val-OCH ´HCl was added.
3
b

4
440
P. Li, J.-C. Xu / Tetrahedron 56 (2000) 4437±4445
Figure 4. The two steps of the HOBt-based reagents mediated coupling reactions.
disturb the delocation of the lone electron pair of the
nitrogen atom over the CvN bond, consequently, lead
the carbonation to possess relatively low electron density,
which will be in favor of the nucleophilic attack of
carboxylate anion.
the amino component Val-OCH ´HCl at 2108C in the
3
1
presence of 2,6-lutidine to give over 90% yield after only
10 min (Table 1, entry 15). This indicated that the relatively
higher reactivity of HOBt-based immonium salts than
uronium salts HBPyU and HAPyU was mainly because
the immonium carbon skeleton was more sensitive towards
the attack of carboxylate anion than the uronium/aminium
To explain the high ef®ciency of HOBt-based immonium
salts, the mechanism of the immonium salts mediated
coupling reaction was proposed and studied. Since the
amino component of the adopted model reaction was a
primary amine without obvious hindrance and the benzo-
triazolyl esters elicited very high reactivity towards amino-
lysis in the presence of base, the rate of the amide formation
reaction was substantially determined by the activation step,
the formation of the benzotriazolyl esters, which determined
0
carbon skeleton, that is, k .k (Fig. 5). The acyloxyimmo-
1
1
nium salt intermediate I was also more reactive than the
corresponding acyloxyuronium salt intermediate I , that is,
0
k .k , but this may not be the rate-determining step and not
0
2
2
in¯uence the overall reaction speed due to their extremely
high reactivity.
Unexpectedly the reaction speed was slow and the racemi-
zation was increased during the model reaction carried out
by the 10 min preactivation of the carboxylic component via
BOMI mediation using DIEA as base (Table 1, entry 4).
However the reaction speed was remarkably increased by
the use of the weak base 2,6-lutidine (entry 5). This
the net overall rate k<k (Fig. 4). Thus the reaction speed
a
probably depends on the nature of the carbon skeleton of
the coupling reagents when the amino component was an
unhindered coded amino acid. This was con®rmed by treat-
ing the preformed active ester intermediate Z-G-F-OBt with
Figure 5. Proposed mechanism of the HOBt-derived immonium and uronium salts mediated coupling reaction.

P. Li, J.-C. Xu / Tetrahedron 56 (2000) 4437±4445
Table 2. The spectra of the model active ester Z-G-F-OBt
4441
1
Solvent
UV
IR
n(CvO) (cm
O-form
H NMR (300 MHz, CDCl
TMS, ppm)
3
,
21
l
max (nm)
Absorption (A)
)
None (KBr)
±
±
1808 (m)
722 (vs)
1660 (s)
1
N-form (overlapped)
CH
3
CN
7.05±8.38 (m, 14H, aryl)
2
54
2.2114
1821 (m)
O-form
P6 .h9 e5) , 6.76 (2d, Jˆ7Hz,1H,NH-
5.74, 4.83 (2m, 1H, a-CH-Phe)
5.56, 5.61 (2br, 1H, NH-Gly)
282
328
342
O-form
N-form
1.0088
0.8148
0.7819
1729 (vs)
1715 (sh)
1685 (m)
N-form (partly overlapped)
5.08±5.20 (m,2H,CH
3.76±4.09 (m, 2H, a-CH
3.18, 3.39 (2m, 2H, b-CH
2
2
-CBZ)
-Gly)
-Phe)
1635 (s)
2
CHCl
3
256
0.5692
0.3337
0.0517
0.0453
1818 (m)
1732 (vs)
1689 (vs)
O-form
N-form (partly overlapped)
2
3
3
86
28
42
O-form
N-form
2
1
phenomenon can be attributed to the fact that the preformed
active intermediate benzotriazolyl ester of N-protected
amino acid was gradually transformed into the less reactive
and racemization prone species 5-(4H) oxazolone in the
1805±1825 cm , Z-G-F-OBt did exist, to some extent, in
the N-acylated form revealed by the characteristic absorp-
tion of N-oxide at 328 and 342 nm, and it appeared to a
greater extent in CH CN than in CHCl due to the higher
3
3
2
2
1
presence of base, especially the strong base DIEA.
k ,k ,k ).
polarity of CH CN stabilizing the N-oxide (Table 2). The H
3
(
NMR also revealed numerous signals corresponding to
benzotriazole residues from the two forms of Z-G-F-OBt
in the dˆ7.1±8.4 ppm region. Similarly, during the prepa-
ration of Fmoc-MeVal-Sar-OBzl by treating Fmoc-MeVal-
OH and Sar-OBzl-HCl with reagent BDMP in the presence
of 2,6-lutidine at 2108C, the active intermediate Fmoc-
MeVal-OBt was also isolated besides the dipeptide. This
benzotriazolyl ester also exists in two forms, the less polar
O-acylated isomer and the more polar N-acylated isomer,
which can be separated by preparative thin layer chroma-
tography. Either of the separated isomers gradually
isomerises into two forms when dissolved in solvent.
7
4
5
As shown in Table 1, the rate of the coupling reagent BDMP
mediated reaction is equal to, or even faster than, that of
Z-G-F-OBt (Table 1, entries 7 and 15). It is suggested that
the structural feature of BDMP considerably accelerates the
speed of the activation step and leads to k nearly equal to k .
a
c
This result may provide an insight into the mechanism of
immonium salt-mediated coupling reactions and support the
rationality of the molecular design of BDMP. The reason for
the slightly higher reactivity of BDMP than the preformed
active ester Z-G-F-OBt was most probably that the O-form
isomer of the benzotriazolyl ester intermediate II is more
2
3
reactive than the corresponding N-oxide isomer III
k .k , Fig. 5) and the initially generated active ester
from the BDMP mediation exists as the O-acylated isomer
As examined above, we evaluated the excellent perfor-
mance of HOBt-based immonium type coupling reagents
and discussed the mechanism and kinetics of coupling reac-
tions mediated by these reagents using the model reaction:
Z-Gly-Phe-OH1Val-OCH ´HCl!Z-Gly-Phe-Val-OCH ,
(
4
5
1
6b
II under mild reaction conditions, and it reacts rapidly
with the amino component to afford product (k .k ); while
4
3
3
3
a certain amount of the less reactive N-acylated isomer III
seems to be involved in the case of the preformed ester Z-G-
F-OBt. To verify the above speculation we studied and
characterized the isomerization of the active ester by UV,
in which the amino component is an unhindered primary
amine. When an N-alkyl or C , C -disubstituted amino
a
acid, such as Sar-OBzl´HCl or Aib-OCH ´HCl was used as
3
amino component, the coupling step becomes the key step
of the peptide formation reaction (k .k ). There is still a
certain amount of active ester intermediate unreacted, and it
a
1
IR and H NMR. The UV spectra indicated that, besides the
existence of the O-isomer characterized by the absorption at
a
c
Table 3. Syntheses of amides and esters using HOBt-based immonium salts as coupling reagents (reaction conditions: base, 2,6-lutidine; temp., 2108C!r.t.;
substrates ratio, carboxylic acid/amine/coupling reagent/baseˆ1:1.1:1.1:3 (mol))
a
Entry
Substrate
Main product
Reagent
Reaction time
Yield (%)
Carboxylic acid
Z-Gly-Phe-OH
Z-Gly-Phe-OH´DCHA
PhCOOH
PhCOOH
PhCOOH
PhCOOH
Z-Aib-OH
Z-Aib-OH
Fmoc-MeVal-OH
Amine
Val-OCH
Val-OCH
1
2
3
4
5
6
7
8
9
3
´HCl
´HCl
Z-Gly-Phe-Val-OCH
Z-Gly-Phe-Val-OCH
PhCOOBt
3
BPMP
BOMI
BOMI
BOMI
BOMI
BOMI
BDMP
BOMI
BDMP
2 h
2 h
2 h
2 h
2 h
2 h
1 h
1 h
3 h
96.2
64.5
87.5
90.0
79.4
75.8
88.7
36.2
51.6
3
3
Not added
PhNH
CHA
2
PhCONHPh
PhCONHC
PhCOOBt
Z-Aib-OBt
Z-Aib-OBt
6 11
H
DCHA
Not added
Aib-OCH
3
´HCl
Sar-OBzl-OCH
3
´HCl
Fmoc-MeVal-OBt
a
Isolated yield based on carboxylic acid component except entry 1 which was the yield determined by HPLC.

4
442
P. Li, J.-C. Xu / Tetrahedron 56 (2000) 4437±4445
can be isolated from the reaction mixture during reaction
with hindered amines, such as Z-Aib-OBt and Fmoc-
MeVal-OBt (Table 3, entries 8, 9). In the case of more
hindered dicyclohexylamine (DCHA), the reaction stops at
the HOBt ester stage and the corresponding amide was not
observed (entry 6), it appears that the DCHA salt of
carboxylic acid component can be used directly in peptide
synthesis (entry 2). In the absence of amine components, the
corresponding active esters are obtained as the ®nal product
Varian-SY-5000 with Kromasil RP-18 (0.5£25 cm)
column. Flash column chromatography was performed
with 300±400 meshes silica gel, and analytical thin layer
chromatography was performed on precoated silica gel
plates (60F-254) with the systems (v/v) indicated. Solvents
and reagents were puri®ed by standard methods as neces-
sary. Amino acids were l-con®guration if not otherwise
stated.
(entries 3, 7). These results also give an insight into the
mechanism discussed above.
Preparation of immonium type coupling reagents
Dimethylchloromethaniminium hexachloroantimonate
(MCMI, general procedure I). A solution of N,N-dimethyl
formamide (0.77 mL, 10 mmol) in CH Cl was added drop-
To verify the applicability of these novel immonium salts,
we synthesized a series of oligopeptides, biological peptides
both in solution and solid phase with good yield, and low
2
2
wise to a solution of bis(trichloromethyl)carbonate (0.989 g,
3.33 mmol) in CH Cl (5 mL) at 08C under nitrogen atmos-
1
4
racemization. We also synthesized an immunosuppressive
cyclic undecapeptide cyclosporin O using these reagents
2
2
phere. After being stirred approximately 1 h, when the
evolution of carbon dioxide has ceased, a 0.887 M solution
of SbCl in CHCl (10.7 mL) was added dropwise at 2308C
2
4
combined with other reagents designed by us. These
reagents can also be used in ester formation with satisfac-
tory yield, especially the preparation of active esters such as
benzotriazolyl ester, penta¯uorophenyl ester and succini-
midyl ester which are widely used in the syntheses of
lactones and lactams.
5
3
under vigorous stirring. The reaction mixture was stirred at
08C for 2 h, the resultant suspension was ®ltered under nitro-
gen atmosphere, the ®lter cake was washed with cold
CH Cl2 and dried in vacuo. Recrystallization from
2
CH CN±CH Cl gave pure product as colorless crystals
3
2
2
In conclusion, several novel immonium coupling reagents
BOMI, BDMP, BPMP, SOMP, DOMP, FOMP and AOMP,
which can be easily prepared from the corresponding amide,
were designed and synthesized by modifying the structure
of the uronium/aminium salts. HPLC monitoring of these
immonium salts mediated coupling reaction demonstrated
that these novel reagents, especially the HOBt and HOAt-
derived immonium salts, displayed extremely high reac-
tivity and low racemization. Among the HOBt-based immo-
nium salts, BDMP was the most ef®cient in terms of high
reactivity, low racemization and good yields. In the case of
the coupling of normal coded amino acids and peptide
segments, HOBt-derived immonium salts, such as BOMI,
BDMP and BPMP were even more ef®cient than the HOAt-
derived uronium/aminium reagent HAPyU, the high reac-
tivity of these reagents may be attributed to their structural
features. However, in the case of the coupling of N-methyl
or C ,C -disubstituted amino acids, the HOAt-based
(4.07 g, 86.9%), mp 220±2228C; [Found: C, 8.59; H, 1.72;
N, 3.25. C H Cl NSb requires C, 8.44; H, 1.65; N, 3.28%];
nmax (KBr) 3066w, 1658s, 1436m, 1407sh, 1322m, 1152m,
3
7
7
1
1033m, 800w, 710w; H NMR ([D ]acetone): dˆ8.25 (s,
6
1H, a-H), 3.20 (s, 3H, CH ), 3.02 (s, 3H, CH ). EI-MS m/z:
3
3
1
92 [M2SbCl ] , 94 [M2SbCl 12] .
1
6
6
N-(1H-Benzotriazol-1-ylmethylene)-N-methylmethan-
aminium hexachloroantimonate N-oxide (BOMI,
general procedure II). KOBt (0.173 g, 1 mmol) was
added to a solution of MCMI (0.427 g, 1 mmol) in dry
CH CN (3 mL) at 2308C with stirring under nitrogen atmo-
3
sphere. After the reaction mixture was stirred at room
temperature for 2 h, it was ®ltered and the ®ltrate was
concentrated under reduced pressure, the residue was
recrystallized from CH CN±Et O to give the yellowish
3
2
crystalline product BOMI (0.460 g, 87.4%), mp 152±
1538C (dec.); [Found: C, 20.78; H, 2.12; N 10.63.
C H Cl N Osb requires C, 20.56, H, 2.11, N, 10.66%];
a
a
reagents are more ef®cient since the reaction rate is mainly
determined by the reactivity of the active ester inter-
mediates.
9
11
6
4
nmax (KBr) 3050w, 1702s, 1496m, 1461sh, 1446m,
1388m, 1340m, 1043m, 755w, 732w; H NMR ([D ]ace-
1
6
tone): dˆ7.99 (1H, s, a-H), 7.95±7.45 (4H, m, aryl), 2.97
(
3H, s, CH ), 2.81 (3H, s, CH ).
3 3
Experimental
5-Chloro-3,4-dihydro-1-methyl 2H-pyrrolium hexachloro-
Amino acid derivatives were obtained from commercial
sources or synthesized according to the literature. Melting
points were determined in capillary tubes and are uncor-
rected. IR spectra were measured with an IR-440 spectro-
meter. H NMR spectra were recorded on Bruker AM
300 MHz and are reported as parts-per-million (ppm)
antimonate (CDMP). CDMP was synthesized from
N-methyl pyrrolidone according to the general procedure
I. Yield 90.7%. Mp 191±1938C (dec). [Found: C, 13.47;
H, 2.06; N, 3.03; Cl, 54.99. C H Cl NSb requires C,
5
9
7
1
13.25; H, 2.00; N, 3.09; Cl, 54.78.]; nmax (KBr) 2922w,
1661vs, 1462w, 1446w, 1410m, 1320s, 1121m, 1095w. H
1
down®eld from a tetramethylsilane internal standard. The
following abbreviations are used: singlet (s), doublet (d),
triplet (t), quartet (q), multiplet (m), broad (br). Mass spectra
were taken with HP5890A, and VG QUATTRO mass
spectrometers. Elemental analyses for carbon, hydrogen
and nitrogen were determined on a MOD-1106 elemental
analyzer. Optical rotations were determined using a Perkin±
Elmer 241 MC polarimeter. HPLC were carried out on
NMR ([D ]acetone): dˆ3.60 (t, Jˆ7 Hz, 2H, a-CH ), 2.91
6
2
(s, 3H, CH ), 2.59 (t, Jˆ8 Hz, 2H, g-CH ), 2.04±2.16 (m,
3
2
1
2
2H, b-CH ). FABMS: 118 [M2SbCl ] .
2
6
5-(1H-Benzotriazol-1-yloxy)-3,4-dihydro-1-methyl 2H-
pyrrolium hexachloroantimonate (BDMP). BDMP was
synthesized from CDMP and KOBt according to the general
procedure II. Yield 88.2%. Mp 165±1668C (dec.); [Found:

P. Li, J.-C. Xu / Tetrahedron 56 (2000) 4437±4445
4443
1
C, 23.83; H, 2.13; N, 10.24. C H Cl N Osb requires C,
1
638w; H NMR ([D ]acetone): dˆ1.95 (m, 4H, 2b-CH ),
1
13
6
4
6
2
2
1
6
3.95; H, 2.37; N, 10.15%]; nmax (KBr) 3127w, 1655vs,
496s, 1479m, 1445m, 1314m, 1165m, 1066m, 763m,
3.57 (m, 4H, 2a-CH ), 7.34±7.97 (m, 9H, aryl).
2
1
40w. H NMR ([D ]acetone): dˆ7.34±7.95 (m, 4H,
5-(1H-7-Azabenzotriazol-1-yloxy)-3,4-dihydro-1-methyl
2H-pyrrolium hexachloroantimonate (AOMP). The solu-
tion of HOAt (0.136 g, 1 mmol) and NEt (139 mL, 1 mol)
6
aryl), 3.48 (t, Jˆ7 Hz, 2H, a-CH ), 2.83 (s, 3H, CH ),
2
3
2.39 (t, Jˆ8 Hz, 2H, g-CH ), 1.98±2.09 (m, 2H, b-CH ).
2
2
3
2
1
FABMS: 217 [M2SbCl ] .
in CH Cl (5 mL) was added dropwise to a suspension of
2
6
2
CDMP (0.453 g, 1 mmol) in dry CH Cl (6 mL) at 08C with
2
2
5
-(N-Succinimidyloxyl)-3,4-dihydro-1-methyl 2H-pyrro-
stirring under argon atmosphere. After the reaction mixture
was stirred at room temperature for 2 h, it was ®ltered and
the ®lter cake was washed with cold CH Cl and dried in
lium hexachloroantimonate (SOMP). SOMP was synthe-
sized from CDMP and KOSu according to the general
procedure II. Yield 89.3%. Mp 166±1688C (dec.); [Found:
C, 20.22; H, 2.45; N, 5.23.C H Cl N O Sb requires C,
2
2
vacuo. Recrystallization from CH COCH ±Et O gave
3
3
2
0.47 g pure product as yellowish crystalline solid. Yield
85.1%. Mp 109±1108C (dec.); [Found: C, 21.60; H, 2.11;
N, 12.74. C H Cl N Osb requires C, 21.73; H, 2.19; N,
9
13
6
2
3
2
1
1
0.33; H, 2.46; N, 5.27%]; nmax (KBr) 2978w, 1754vs,
710vs, 1454s, 1420m, 1400m, 1335m, 1216s, 1162s,
067m, 801m, 633w; H NMR ([D ]acetone): dˆ3.70 (t,
1
0
12
6
5
1
12.67%]; nmax (KBr) 3111w, 1667vs, 1496s, 1475s, 1459s,
1408m, 1314m, 1070s, 793m, 760w, 638w; H NMR
6
1
Jˆ7.2 Hz, 2H, a-CH ), 3.00 (s, 3H, N-CH ), 2.71 (t,
2
3
3
4
Jˆ7.5 Hz, 2H, g-CH ), 2.68 (s, 4H, 2CH ±OSu), 2.12±
([D ]acetone): dˆ8.76 (dd, Jˆ4.4 Hz, Jˆ1.4 Hz, 1H, a-
2
2
6
2
6
1
.24 (m, 2H, b-CH ). FABMS: 197 [M2SbCl ] .
3
4
2
H-aryl), 8.45 (dd, Jˆ8.4 Hz, Jˆ1.4 Hz, 1H, g-H-aryl),
2
3
3
7
.53 (dd, Jˆ4.4 Hz, Jˆ8.4 Hz, 1H, b-H-aryl), 3.62 (t,
5-(3,4-Dihydro-4-oxo-1,2,3-benzotriazin-3yloxy)-1H-ben-
Jˆ7.2 Hz, 2H, a-CH ), 2.93 (s, 3H, N-CH ), 2.59 (t,
2
3
zotriazol-1-yloxy)-3,4-dihydro-1-methyl pyrrolium hexa-
chloroantimonate (DOMP). DOMP was synthesized from
CDMP and KODhbt according to the general procedure II.
Yield 83.6%. Mp 120±1218C (dec.); [Found: C, 24.71; H,
2
2
1
Jˆ8.0 Hz, 2H, g-CH ), 2.13 (m, 2H, b-CH ).
2
2
N-(Benzyloxycarbonyl)-glycyl-phenylalanine 1H-benzo-
triazolyl ester (Z-G-F-OBt, general procedure III). To
a solution of Z-Gly-Phe-OH (0.713 g, 2 mmol) and HOBt
(0.270 g, 2 mmol) in THF (6 mL), DCC (0.413 g, 2 mmol)
was added at room temperature. After the reaction mixture
was stirred for 1 h, it was cooled to 2308C for 2 h to make
DCU precipitate completely. The cold mixture was ®ltered,
the ®ltrate was concentrated and dried in vacuo. Recrystal-
lization from THF±AcOEt±Pe gave 0.766 g pure product as
colorless crystalline solid. Yield 80.9%. Mp 124±1258C.
.24; N, 9.72. C H Cl N O Sb requires C, 24.86; H,
12 13 6 4 2
.26; N, 9.66%]; nmax (KBr) 1725vs, 1708sh, 1681sh,
460m, 1395s, 1174w, 943s, 768m, 676m; H NMR
1
(
8
[D ]acetone): dˆ8.29 (d, Jˆ8.3 Hz, 1H, 8-CH-aryl),
6
.18 (d, Jˆ8.1 Hz, 1H, 5-H-aryl), 8.08 (m, 1H, 7-CH-
aryl), 7.92 (m, 1H, 6-CH-aryl), 3.66 (t, Jˆ7.4 Hz, 2H,
a-CH ), 2.96 (s, 3H, CH ), 2.66 (t, Jˆ7.4 Hz, 2H,
2
3
g-CH ), 2.15 (m, 2H, b-CH ).
2
2
1
1
1
FABMS: 474 [M1H] , 339 [M2OBt] , 91 [PhCH ] .
2
5-Penta¯uorophenyloxy-3,4-dihydro-1-methyl 2H-pyrro-
Other spectra are shown in Table 2.
lium hexachloroantimonate (FOMP). FOMP was synthe-
sized from CDMP and KOPfp according to the general
procedure II. Yield 85.7%. Mp 182±1838C (dec.); [Found:
C, 21.50; H, 1.34; N, 2.16. C H Cl F NOSb requires C,
N-(Benzyloxycarbonyl)-glycyl-phenylalanine 1H-7-aza-
benzotriazolyl ester (Z-G-F-OAt). This compound was
synthesized from Z-Gly-Phe-OH and HOAt according to
the general procedure III. Yield 86.9%, mp 157±1598C.
nmax (KBr) 3069w, 2929w, 1815m, 1732vs, 1627vs,
1
1
9
6 5
2
1.99; H, 1.50; N, 2.33%]; nmax(KBr) 1705s, 1530sh,
1 1
2
520s, 1398m, 1034w, 1001m, 954w cm . H NMR
1
(
3
[D ]acetone): dˆ3.46 (t, Jˆ7 Hz, 2H, a-CH ), 2.81 (s,
1539s, 1393m, 1237s, 800m, 698m, 516w; UV (CHCl )
3
6
2
1
H, CH ), 2.35 (t, Jˆ8 Hz, 2H, g-CH ), 2.02 (m, 2H,
lmaxˆ258, 288 nm; H NMR ([D ]-DMSO) dˆ8.85±7.11
3
2
6
1
9
b-CH ). F NMR ([D ]acetone, CF COOH) dˆ276.8 to
(m, 13H, aryl), 7.05 (br, 1H, NH-Phe), 5.62 (d, 1H, NH-
Gly), 5.04 (s, 2H, CH2 ), 4.47 (m, 1H, a-CH-Phe), 3.75±
2
6
3
CBZ
2
1
77.1 (m, 2F), 280.2 to 80.3 (m, 2F), 291.1 to 291.4 (m,
F). FABMS: 266 [M2SbCl ] .
2
1
3.52 (m, 2H, a-CH -Gly), 3.12±2.83 (m, 2H, b-CH -Phe).
2
6
2
Chlorophenylmethylene pyrrolidinium hexachloroanti-
monate (CPPH). CPPH was synthesized from N-benzoyl
pyrrolidine according to the general procedure I. Yield
Reaction speed and racemization test by HPLC using the
model reaction: Z-Gly-Phe-OH1Val-OMe´HCl!Z-
Gly-D/L-Phe-Val-OMe. Z-Gly-Phe-OH (50 mg, 0.14
mmol) and Val-OMe´HCl (26 mg, 0.154 mmol) were
coupled with tested immonium type reagent (0.154 mmol)
or the same equivalent of other coupling reagent. Boc-Phe-
Val-OMe (66 mg, 0.175 mmol) was added as the internal
reference. Test of reaction were performed on total 1.5 mL
scale. Aliquots (10 mL) from the reaction mixture was
quenched and dissolved in 100 mL buffer solution
(CH OH/H O/TFA: 50/50/1). The resultant samples were
8
1
2.6%. Mp 176±1778C; nmax (KBr) 1621vs, 1438s,
343m, 1267m, 769m, 692s, 647w; H NMR ([D ]acetone):
1
6
dˆ1.98 (m, 4H, 2b-CH ), 3.62 (m, 4H, 2a-CH ), 7.34±7.63
2
2
2
1
(
[
m, 5H, Ph). FABMS: 194 [M2SbCl ] , 196
6
2
M2SbCl 12] .
1
6
(
1H-Benzotriazol-1-yloxy)phenylmethylene pyrrolidi-
nium hexachloroantimonate (BPMP). BPMP was synthe-
sized from CPPH and KOBt according to the general
procedure II. Yield 91.1%. Mp 93±948C (dec.); [Found:
C, 32.23; H, 2.61; N, 8.84.C H Cl N Osb requires C,
3
2
analyzed by HPLC to give the following result: Z-
Gly-Phe-OH
(t ˆ4.04 min);
Z-Gly-L-Phe-Val-OMe
(t ˆ 9.24 min); Z-Gly-D-Phe-Val-OMe (t ˆ10.28 min);
R
1
7
17
6
4
R
R
32.52; H, 2.73; N, 8.92%]; nmax (KBr) 3100w, 1616vs,
Boc-Phe-Val-OMe (t ˆ15.82 min). Peak areas were
R
1494s, 1467s, 1417m, 1331s, 1151m, 1065m, 752s, 705m,
compared in order to obtain the chemical yields

4
444
P. Li, J.-C. Xu / Tetrahedron 56 (2000) 4437±4445
ꢀ
yieldꢀ%† ˆ ‰ꢀLL=X 1 DL=X †=a´SŠ £ 100%†: Percentage of
procedure and puri®ed by ¯ash chromatography on silica
gel column eluted with AcOEt/Petroleum ether (1:2) to
give 64 mg of Z-Aib-OBt (36%), accompanied with the
1
2
epimers was calculated according to the equation: D% ˆ
‰
DL=X =ꢀLL=X 1 DL=X †Š £ 100%; where LL refers to the
2
1
2
area of Z-Gly-L-Phe-Val-OMe, DL refers to that of
Z-Gly-D-Phe-Val-OMe, S refers to that of Boc-Phe-Val-
OMe, aˆ0.778, which is the molar ratio between Z-Gly-
OH and Boc-Phe-Val-OMe; X ˆ1.269 and X ˆ1.254
dipeptide Z-Aib-Aib-OCH (76 mg, yield:45%). [Found:
3
C, 61.26; H, 5.28; N, 15.53.C H N O requires C, 61.01;
1
8
18
4
4
1
H, 5.12; N, 15.81%]; R (AcOEt/Peˆ1:2) 0.51; H NMR
f
(CDCl ):dˆ7.26±8.04 (m, 9H, aryl), 5.57 (s, 1H, NH),
1
2
3
which are the determined correction factors for absorption
difference (220 nm) between the references.
5.21 (s, 2H, PHCH ), 1.78 (s, 6H, 2CH ); n
2 3 max
2
1
(Nujol)ˆ3322, 1817, 1713, 1521, 1270, 1047 cm .
1
FABMS m/z: 355 [M1H] , 220 [M2BtO] , 91 [PhCH ] .
1
1
2
General procedure for peptide synthesis using
immonium type coupling reagents
Fmoc-MeVal-OBt (Table 3, entry 9). 2,6-Lutidine (0.37
mL, 3.2 mmol) was added to a cold mixture (2108C) of
Fmoc-MeVal-OH (353 mg, 1.0 mmol), Sar-OBzl´HCl
(237 mg, 1.1 mmol), and BDMP (607 mg, 1.1 mmol), in
THF (5 mL), stirred for 1 min cold and 3 h at room tempera-
ture. The reaction mixture was directly puri®ed by ¯ash
chromatography on silica gel column eluted with AcOEt/
Petroleum ether (1:4) to give Fmoc-MeVal-OBt besides the
dipeptide Fmoc-MeVal-Sar-OBzl. Yield: 243 mg (51.6%),
mp 43±448C; [Found: C, 68.39; H, 5.57; N, 11.62.
2
,6-Lutidine (3 equiv.) was added to a cold mixture
2108C) of N-protected amino acid (1 equiv.), amino acid
ester hydrochloride (1.1 equiv.), and coupling reagent
1.1 equiv.) in THF (3±5 mL/mmol), stirred for 1 min
(
(
cold and 1 h at room temperature. For large scale and
coupling between hindered amino acids, the reaction time
should be moderately prolonged. After the completion of
reaction, the reaction mixture was diluted with THF, the
resultant suspension was ®ltered and the ®ltered cake was
washed with THF. The combined ®ltrate was concentrated
under reduced pressure to give the crude product which was
puri®ed by ¯ash chromatography on silica gel column to
afford the desired product.
C H N O ´0.25H O requires C, 68.27; H, 5.62; N,
27 26 4 4 2
11.79%]; Rf1 (O-form, AcOEt/Peˆ1:2) 0.67, Rf2 (N-form,
AcOEt/Peˆ1:2) 0.40. nmax (KBr) 3067w, 2966w, 1818m,
2
1
1702vs, 1464m, 1451sh, 1169m, 740s cm ; UV (CH Cl )
2 2
1
l
maxˆ262, 289, 329, 344 nm; H NMR (CDCl
): dˆ0.70,
3
0
(2m, 1H, b-CH-Val), 2.92, 2.98, 3.02, 3.04 (4s, 3H, N-CH
.85, 1.01, 1.23 (4d, Jˆ6.4 Hz, 6H, 2n-CH -Val), 2.16, 2.41
3
General procedure for esters synthesis using immonium
type coupling reagents
3
),
-Fmoc, 9-CH-Fluorenyl, a-CH-Val),
4.08±4.85 (m, 4H, CH
7
2
.08±8.12 (m, 12H, aryl).
2
,6-Lutidine (2 equiv.) was added to a cold mixture
2108C) of carboxylic acid (1 equiv.), alcohol (1.1
equiv.), and coupling reagent (1.1 equiv.) in THF
3±5 mL/mmol), stirred for 1 min cold, then raise to room
temperature, the reaction was monitored by TLC. For the
preparation of HOBt active ester, no alcohol need be added
and the quantity of base was one equivalent. The work-up
was the same as that of peptide synthesis.
(
Abbreviations: Aib, a-aminoisobutyric acid; AOMP,
5-(1H-7-azabenzotriazol-1-yloxy)-3,4-dihydro-1-methyl
2H-pyrrolium hexachloroantimonate; AOP, (1H-7-aza-
benzotriazol-1-yloxy)tris(dimethylamino)phosphonium
hexa¯uorophosphate; BOMI, N-(1H-benzotriazol-1-yl-
methylene)-N-methylmethanaminium hexachloroanti-
monate N-oxide; BDMP, 5-(1H-benzotriazol-1-yloxy)-
(
3
monate; BOP, (1H-benzotriazol-1-yloxy)tris(dimethyl-
,4-dihydro-1-methyl 2H-pyrrolium hexachloroanti-
Special examples for the mechanism studied of
immonium salts mediated coupling reaction
amino)phosphonium
1
hexa¯uorophosphate;
-(1H-benzotriazol-1-yloxy)phenylmethylene
BPMP,
pyrroli-
PhCOOBt (Table 3, entry 6). To a solution of PhCOOH
(
dinium hexachloroantimonate; CHA, cyclohexyl amine;
DCC, dicyclohexylcarbodiimide; DCHA, dicyclohexyl
amine; DIEA, N,N -diisopropylethylamine; DOMP;
61 mg, 0.5 mmol) in CH CN (2 mL), DCHA (101 mL) was
3
0
added, then BOMI (0.289 g, 0.55 mmol), 2,6-lutidine
116 mL, 1 mmol) were added at 2108C. The reaction
mixture was stirred 2 h, then it was ®ltered via a pad of
0
0
0
0
0
0
0
(
5-(3 ,4 -dihydro-4 -oxo-1 ,2 ,3 -benzotriazin-3 -yloxy)-
3,4-dihydro-1-methyl 2H-pyrrolium hexachloroanti-
monate; FOMP; 5-(penta¯uorophenyloxy)-3,4-dihydro-
1-methyl 2H-pyrrolium hexachloroantimonate; HAPyU,
1-(1-pyrrolidinyl-1H-1,2,3-triazolo[4,5-b]pyridin-1-
ylmethylene)pyrrolidinium hexa¯uorophosphate N-oxide;
HBMDU, (1H-benzotriazol-1-yloxy)-1,3-dimethyl-1,3-
dimethyleneuronium hexa¯uorophosphate; HBPipU,
silica gel, and washed with AcOEt/Peˆ1:4 (200 mL), the
®ltrate was concentrated, dried and recrystallized from
benzene±hexane to give 0.91 g PhCOOBt as main product.
Yield 75.8%, mp 80±818C; [Found: C, 65.30; H, 3.64; N,
17.92. C H N O requires C, 65.27; H, 3.79; N, 17.56%];
nmax (KBr) 3067w, 1808sh, 1779vs, 1597m, 1454m, 1230s,
13 9 3 2
1
0
0
1
085m, 987s, 705s; H NMR ([D ]acetone): dˆ7.48±8.46
(1H-benzo-triazol-1-yloxy)-N,N,N ,N -bis(pentamethylene)
uronium hexa¯uorophosphate; HBPyU, O-(1H-benzo-
triazol-1-yl)-N,N,N ,N -bis(tetramethylene)uronium hexa-
uorophosphate; HBTU, N-[(1H-benzotriazol-1-yl)-
6
1
m, 9H, aryl). EI-MS m/z: 239 M , 105 [M2BtO] , 77
1
(
[
1
C H ] .
0
0
6
5
¯
Z-Aib-OBt (Table 3, entry 8). 2,6-Lutidine (0.174 mL,
.5 mmol) was added to a cold mixture (2108C) of
Z-Aib-OH (119 mg, 0.5 mmol), Aib-OCH .HCl (84 mg,
(dimethylamino)methylene]-N-methylmethanaminium
hexa¯uorophosphate N-oxide; HOAt, 1-hydroxy-7-aza-
benzotriazole; HOBt, 1-hydroxybenzotriazole; HODhbt,
3-hydroxy-3,4-dihydro-4-oxo-1,2,3-benzotriazine; HOSu,
N-hydroxysuccinimide; PyBOP, (1H-benzotriazol-1-yloxy)-
1
3
0
.55 mmol), and BOMI (289 mg, 0.55 mmol) in THF
2 mL), stirred for 1 min cold and 1 h room temperature.
The reaction mixture was treated according to the general
(
tris(pyrrolidino)phosphonium
hexa¯uorophosphate;

P. Li, J.-C. Xu / Tetrahedron 56 (2000) 4437±4445
4445
SOMP, 5-(succinimiddyloxy)-3,4-dihydro-1-methyl
2H-pyrrolium hexachloroantimonate. Nomenclature and
11. Albericio, F.; Bo®ll, J. M.; El-Faham, A.; Kates, S. A. J. Org.
Chem. 1998, 63, 9678.
symbols of amino acids and peptides generally follow the
recommendations of the IUPAC±IUB Joint Commission
of Biochemical Nomenclature in: Pure Appl. Chem.
12. Carpino, L. A.; El-Faham, A.; Minor, C. A.; Albericio, F.
J. Chem. Soc., Chem. Commun. 1994, 201.
13. Crystal data for C H N OCl Sb BOMI: Mˆ525.68, triclinic,
9
11
4
6
Ê
1984, 56, 595±624.
aˆ9.058(2), bˆ15.729(4), cˆ6.598(1) A, aˆ94.21(2), bˆ
3
Ê
9
8.99(2), gˆ73.38(2)8, Vˆ889.4(4) A , Tˆ293 K, space group
2
1
23
P1(#2), Zˆ2, mˆ24.50 cm , D ˆ1.963 g cm , 2943 re¯ections
c
Acknowledgements
measured, 2727 were unique (R ˆ0.076). The ®nal R (R ) values
int
w
were 0.045 (0.057) for 2247 re¯ections [I.3.00s(I)) and 190
variables.
This work was ®nancially supported by the National Natural
Science Foundation of China (9772045).
14. Li, P.; Xu, J. C. Tetrahedron Lett. 1999, 50, 3605.
1
5. For analogous coupling reagents, such as BDMP and BPMP of
which X-ray data have not yet been obtained, the structural repre-
sentations have arbitrarily been assigned as the O-substituted forms
and their nomenclature also retained in this article.
References
1
2
2
1
3
. Coste, J.; Le-Nguyen, D.; Castro, B. Tetrahedron Lett. 1990, 31,
05.
1
6. (a) MeCarthy, D. G.; Hegarty, A. F.; Hathaway, B. J. J. Chem.
Soc., Perkin Trans. 1977, 224. (b) Singh, J.; Fox, R.; Wang, M.;
Kissick, T. P.; Moniot, T. L. J. Org. Chem. 1988, 53, 205. (c)
Nagarajan, S.; Wilson, S. R.; Rinehart, K. L. J. Org. Chem.
. Dourtoglou, V.; Ziegler, J. C.; Gross, B. Tetrahedron Lett.
978, 1269.
. (a) Chen, S. Q.; Xu, J. C. Tetrahedron Lett. 1992, 33, 647. (b)
1
985, 50, 2174.
7. Anders, E.; Katritzky, A. R.; Malhotra, N.; Stevens, J. J. Org.
Coste, J.; Frerot, E.; Jouin, P. Tetrahedron Lett. 1991, 32, 1967.
. Henklein, P.; Beyermann, M.; Bienert, M.; Knorr, R. In
Proceedings of the 21st European Peptide Symposium; Giralt, E.,
1
4
Chem. 1992, 3698.
8. Van der Auwera, C.; Van Damme, S.; Anteunis, M. J. O. Int. J.
Pept. Protein Res. 1987, 29, 464.
1
Andreu, D., Eds.; ESCOM, Science: Leiden, 1991; p 67.
5. Kiso, Y.; Fujiwara, Y.; Kimura, T.; Nishitani, A.; Akaji, K. Int.
J. Pept. Protein Res. 1992, 40, 308.
1
2
2
2
9. Humphrey, J. M.; Chamberlin, A. R. Chem. Rev. 1997, 97,
243.
6
1
7
8
. Castro, B.; Dromoy, J. R.; Evin, G.; Selve, C. Tetrahedron Lett.
975, 1219.
0. Konig, W.; Geiger, R. Chem. Ber. 1970, 103, 2034.
1. PCMODEL force ®eld calculation: They are based on MM2,
. Konig, W.; Geiger, R. Chem. Ber. 1970, 103, 788.
. Galpin, I. J.; Gordon, P. F.; Ramage, R.; Thorpe, W. D.
and the included VESCF routines of MMP1 were applied. The
program was distributed by Serena Software Corporation, U.S.A.
22. Carpino, L. A.; Chao, H. G.; Beyermann, M.; Bienert, M.
J. Org. Chem. 1991, 56, 2635.
Tetrahedron 1976, 32, 2417.
. (a) Albericio, F.; Cases, M.; Alsina, J.; Triolo, S. A.; Carpino,
9
L. A.; Kates, S. A. Tetrahedron Lett. 1997, 38, 4853. (b) Angell,
Y. M.; Thomas, T. L.; Rich, D. H. In Peptides: Chemistry,
Structure and Biology; Pravin, T. P., Kaumaya, Hodges, R. S.,
Eds.; May¯ower Scienti®c Ltd.; 1996; p 88.
23. Barlos, K.; Papaioannou, D.; Theodoropoulos, D. Int. J. Pept.
Protein Res. 1984, 23, 300.
24. Li, P.; Xu, J. C. J. Org. Chem. 2000, in press.
1
0. Carpino, L. A. J. Am. Chem. Soc. 1993, 115, 4397.