Novel acylation catalysts in peptide synthesis: Derivatives of N-hydroxytriazoles and N-hydroxytetrazoles

Article abstract of DOI:10.1039/a707313e

Six new derivatives of 1-hydroxy-1,2,3-triazole (1-hydroxytriazole in the following) and N-hydroxytetrazole have been evaluated in a direct competition assay to investigate their efficiency as catalysts in the formation of peptide bonds. Furthermore, thre

Full text of DOI:10.1039/a707313e

View Article Online / Journal Homepage / Table of Contents for this issue
Novel acylation catalysts in peptide synthesis: derivatives of
N-hydroxytriazoles and N-hydroxytetrazoles
Jane C. Spetzler,a Morten Meldal,*,a Jakob Felding,b Per Vedsøb and
Mikael Begtrupb
a Carlsberg Laboratory, Department of Chemistry, Gamle Carlsberg Vej 10, DK-2500 Valby,
Denmark
b Department of Medicinal Chemistry, The Royal Danish School of Pharmacy,
Universitetsparken 2, DK-2100 Copenhagen, Denmark
Six new derivatives of 1-hydroxy-1,2,3-triazole (1-hydroxytriazole in the following) and N-hydroxy-
tetrazole have been evaluated in a direct competition assay to investigate their efficiency as catalysts in
the formation of peptide bonds. Furthermore, three well known catalysts, 1-hydroxy-7-azabenzotriazole
(HOAt), 3,4-dihydro-3-hydroxy-4-oxo-1,2,3-benzotriazine (Dhbt-OH) and 1-hydroxybenzotriazole
(HOBt) have been compared. All nine compounds have been used for activation in combination with
N,NЈ-diisopropylcarbodiimide (DIPCDI) in solid-phase synthesis using the fluoren-9-ylmethoxycarbonyl
(Fmoc) strategy. The capability of the catalysts to suppress racemization has also been analysed. The
results demonstrate that three of the new compounds are competitive with HOAt and HOBt in the
suppression of racemization. 5-Chloro-1-hydroxytriazole is found to be a highly efficient acylation
catalyst; however, it does not show sufficient suppression of racemization. Its catalytic effect in the
synthesis of Aib᎐Aib-containing peptides is superior to that of HOAt. Also, 2-hydroxytetrazole has a
catalytic efficiency superior to that of HOAt and suppressed racemization as efficiently as HOBt. The
hydroxytetrazoles are explosive in a hammer test whereas the triazoles are stable compounds.
tetrazole 10. The preparation of compounds 7–10 has been
Introduction
described elsewhere.11–14 The core of these structures is a five-
membered ring containing at least three vicinal nitrogens in the
ring and the first nitrogen is hydroxylated. Additionally the
neighboring carbon is substituted to provide a lone pair in the
vicinity of the hydroxy group, thereby mimicking the struc-
ture and catalytic mechanism of HOAt and Dhbt-OH. These
compounds have not previously been employed as acylation
catalysts in peptide synthesis. HOBt4 11, HOAt5 12 and
Dhbt-OH6,7 13 are commercially available. The investigation of
catalysis of peptide-bond formation was performed using
poly(ethylene glycol)-cross-linked polyamide (PEGA) as a
resin.15 PEGA swells 10–20-fold in both organic solvents and
aqueous buffer solution and has been successfully employed in
synthesis and biochemical assaying of peptide libraries.16,17
The coupling additives were evaluated under conditions
similar to those used in solid-phase Fmoc-based peptide syn-
thesis.18,19 Thus all reactions were performed with a three-molar
excess of reagents. In the assay Fmoc-Ile-OH was coupled to
Val-Gly-PEGA resin by adding the catalyst in the presence of
diisopropylcarbodiimide (DIPCDI).2 The coupling times were
kept relatively short (exactly 5 min, followed by quenching of
the reaction) in order to determine the reaction rates with essen-
tially constant concentration of amino groups on the resin. The
relative ratio of Ile to Val was determined with good accuracy
by amino acid analysis. In studies of the racemization, the
tripeptide benzyloxycarbonyl-Phe-Val-Pro-NH2 (Z-Phe-Val-
Pro-NH2) was selected, since the two diasteroisomers Z-Phe-
Val-Pro-NH2 and Z-Phe-val-Pro-NH2 (val indicates -Val
according to the recommendation by UIPAC) can easily be
separated by analytical reversed-phased HPLC.8
The addition of coupling additives to in situ coupling reagents
such as carbodiimides1,2 can favorably influence reactivity and
decomposition of intermediates in the formation of amide
bonds during the stepwise assembly of peptides on a solid
phase.3 Thus the rate of acylation can be enhanced between
sterically hindered amino acids and the racemization can be
suppressed. HOBt is a commonly used catalyst, which in
carbodiimide-mediated coupling prevents racemization and
other side-reactions such as dehydration of the side chains of
Asn and Gln.4 Recently, a new racemization-suppressing
catalyst for solid-phase peptide synthesis, 1-hydroxy-7-
azabenzotriazole (HOAt), has been described.5 The very favor-
able properties of 3,4-dihydro-3-hydroxy-4-oxo-1,2,3-benzo-
triazine (Dhbt-OH) as a catalyst have also been reported.6,7
Dhbt-OH furthermore indicates progress of the reaction
visually by a color change when the acylation is complete.
Unfortunately a ring-opening side-reaction leading to 2-azido-
benzoic acid Dhbt ester is associated with the use of Dhbt-OH
as an additive to DCC reactions. However, fully protected
amino acid esters of Dhbt-OH are stable and crystalline and
have increased activity compared with other active esters.6
Coupling efficiency and rates have been examined for several
coupling procedures.8–10 The coupling-efficiency studies have
been performed in either solution or on a solid phase using a
one-to-one mixture of the two components containing free
amino and carboxy groups.8–10
The present work compares nine different compounds,
including modified derivatives of 1-hydroxytriazoles and
N-hydroxytetrazoles and commercially available peptide-
coupling additives. Their efficiency as catalysts in the formation
of peptide bonds and the suppression of racemization was
determined. The novel catalysts were hydroxy-triazole and
-tetrazole derivatives: 1-hydroxy-5-(methoxymethyl)-1,2,3-
triazole 4, 5-acetyl-1-hydroxy-1,2,3-triazole 6, 5-chloro-1-
hydroxy-1,2,3-triazole 7 (either free or as the hydrochloride), 1-
hydroxy-1,2,3-triazole 8, 2-hydroxytetrazole 9 and 1-hydroxy-
Results and discussion
Nine compounds 4, 6–13 have been evaluated for their ability to
catalyze formation of peptide bonds by analysis of coupling
efficiency and suppression of racemization in model reactions
J. Chem. Soc., Perkin Trans. 1, 1998
1727

View Article Online
Table 1 Evaluation of analogs of N-hydroxytriazoles and N-hydroxy-
tetrazoles, Dhbt-OH, HOBt and HOAt as catalysts in peptide synthesis
using a comparison study
O
H
N
CH2OCH3
OCH2Ph
CH2OH
OCH2Ph
OCH2Ph
i
ii
N
N
97%
85%
Coupling
catalyst
Yield
Aib᎐Aib bond
formation (%)
N
N
N
N
N
N
(% Ile)a
LDL (%)
1
2
3
4
6.0
3.7
9.0
4.1
7.0
3.7
5.4
6.6
4.0
4.1
34.3b
6
20.3b
iii 99%
7
35.6b, 47.2c, 28.7d
18.3b, 24.9c
14.1b, 18.1d
4.5b
43
O
CH3
O
CH3
8
CH2OCH3
9
iii
10
OCH2Ph
OH
OH
N
N
N
10.9b
93%
11
6
38
12
0.95b
N
N
N
N
N
N
13
26.0b
5
6
4
no catalyst
50.0b
2
Scheme 1 Reagents and conditions: i, NaBH4; ii, CH3l, NaH; iii, H2,
Pd/C
a The amino acid Fmoc-Ile-OH and catalyst were added to Val-Gly-
PEGA resin. After 2 min, DIPCDI was added and the reaction was
allowed to proceed for 5 min. The yield of Ile was calculated based on
the ratio of Ile to Val found after hydrolysis of the peptide-resin.
The couplings with the 3 best catalysts (7, 9 and 12) were repeated
and standard deviations were < 0.5. The low conversions secure that
concentrations of amine component are essentially constant for all
catalysts investigated. b The dipeptide Z-Phe-Val-OH and catalyst were
added to Pro-Rink linker-PEGA resin. After 1 min, DIPCDI was
added. The reaction was carried out overnight and the peptide was
cleaved with 95% aq. TFA. The degree of racemization was calculated
based on the areas between the two diastereoisomers obtained from
HPLC. The experiments with compounds 7, 9, 11 and 12 with a
deviation < 2. c Catalyst and DIPCDI were dissolved in DMF at 0 ЊC
and the solution was left for 5 min. Then, Z-Phe-Val-OH was added to
the above solution at 0 ЊC. After 2 min, the mixture was added to the
Pro-Rink linker-PEGA resin. The coupling reaction was carried out
overnight followed by cleavage of the peptide with 95% aq. TFA. The
optical purity was analyzed as described above.b d Same procedure as
described abovec except that three mole equivalents of catalyst were
used.
on a solid phase. The compounds analyzed are N-hydroxy-
triazole and N-hydroxytetrazole derivatives designed to mimic
the catalytic mechanism of HOAt and Dhbt-OH.
1-Hydroxy-5-(methoxymethyl)-1,2,3-triazole 4 and 5-acetyl-
1-hydroxy-1,2,3-triazole 6 (see Scheme 1) were first synthesized.
Thus 1-benzyloxy-5-formyl-1,2,3-triazole14 1 was reduced with
NaBH4 to afford 1-benzyloxy-5-(hydroxymethyl)-1,2,3-triazole
2 in 97% yield. The intermediate 1-benzyloxy-5-(methoxy-
methyl)-1,2,3-triazole 3 was formed in 85% yield by treatment
of compound 2 with MeI in the presence of NaH, and com-
pound 3 was further converted into 1-hydroxy-5-(methoxy-
methyl)-1,2,3-triazole 4 in 99% yield by hydrogenation using
palladium/charcoal (10%). Similarly, 5-acetyl-1-hydroxy-1,2,3-
triazole 6 was obtained in 93% yield by hydrogenolytic debenz-
ylation of 5-acetyl-1-benzyloxy-1,2,3-triazole13 5. The first
series of experiments was designed to evaluate the relative
efficiency in forming peptide bonds on a solid phase (Fig. 1).
The couplings were mediated by the in situ reagent DIPCDI2
and formation of the peptide bond was performed on the
PEGA-resin.15 In this assay Fmoc-Ile-OH was coupled to
Val-Gly-PEGA resin. The resin was equally divided into 20
columns of a library generator.20 Amino acid, catalyst and
DIPCDI were dissolved in DMF and used in three-fold excess
in order to perform the experiment under the usual solid-phase
peptide synthesis conditions. First, the amino acid and catalyst
were added to the resin. After 2 min pre-incubation, DIPCDI
was added. A relatively short coupling time of 5 min was
selected in order to obtain a low degree of conversion, while
still allowing reagents to diffuse into the resin. The result
ranked the compounds as potential acylation catalysts and
peptide-coupling additives. The ratio of Ile to Val, which is a
value for the relative efficiency of formation of the peptide
bond was calculated from the relative Ile and Val content found
by amino acid analysis. The results are presented in Table 1.
The experiment showed that triazole derivative 7 was the
best catalyst and considerably more efficient than HOAt. The
2-hydroxytetrazole derivative 9 was also slightly more efficient
as compared with HOAt in forming the peptide bond. Com-
pound 4 had a catalytic activity between those of HOBt and
HOAt. The remaining compounds 6, 8, 10 and 13 were poor
catalysts with relative coupling efficiencies comparable to that
of in situ activation with no addition of catalyst.
OH
X
N
N
N
N
N
N
N
N
OH
OH
N
N
N
4 X = CH2OCH3
6 X = COCH3
7 X = Cl
9
10
8 X = H
O
OH
N
N
N
N
N
X
N
OH
11 X = CH
12 X = N
13
Fmoc-Aib-Lys(Boc)-Ser(tBu)-Ser(tBu)-Tyr(tBu)-Lys(Boc)-Rink linker-PEGA
14
Structures of the derivatives of triazole (4, 6–8) and tetrazole (9, 10),
HOBt 11, HOAt 12, Dhbt-OH 13 and 14
LLL and LDL were first obtained in a 6:4 ratio by reaction of
dipeptide with Pro-linker-PEGA resin and 2-(1H-benzotriazol-
1-yl)-1,1,3,3-tetramethyluronium tetrafluoroborate (TBTU) in
the presence of N-ethylmorpholine (NEM) and pre-activation
for 8 min.22 Cleavage of the peptide from the resin was carried
out by treatment with 95% aq. (TFA) and HPLC separation of
the two diasteroisomers is presented in Fig. 2. Throughout this
assay, the coupling procedure was the same as that described
above for the coupling efficiency assay except that four-fold
excess of reagents was used, preactivation was 1 min and the
acylation reaction mixture was left overnight. Subsequently, the
peptides were cleaved from the linker by 95% aq. TFA and
analyzed by HPLC (Fig. 1). This model reaction proved to be
extremely sensitive in testing carbodiimide-mediated couplings
since the racemization ranged between 0.95% and 50.0% for the
In the second series of experiments, the ability to suppress
racemization was studied for the nine compounds mentioned
above. Previously, a peptide model reaction in which the mix-
ture of Z-Phe-Val-Pro-NH-Resin and Z-Phe-val-Pro-NH-Resin
is formed by coupling Z-Phe-Val to a Pro residue on the solid
phase has been employed in racemization studies.13 The two
LLL and LDL diastereoisomers are easily separated by
analytical HPLC. This model was selected to study the ability
of the new catalysts to suppress racemization. The acid-labile
p-[(α-Fmoc-amino)-2,4-dimethoxybenzyl]phenoxyacetic acid
(Rink amide linker)21 was employed. The two diastereoisomers
1728
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
O
OH
O
+
NH
O
O
NH
H2N
NH
O
X2
N
X1
1)
X4
N
2) Wash,
deprotect,
wash
X3
N
N C N
O
NH
H2N
NH
O
O
NH
NH
H2N
NH
O
O
X1
Cl
X2 X3 X4
H
H
H
H
H
H
OH
OH
OH
H
C
C
C
N
N
C
CH3OCH2
Ac
OH
H
Fig. 2 Separation of the two diastereoisomers LLL (Rf 36 min) and
LDL (Rf 38 min) by analytical HPLC and obtained from DIPCDI-
mediated coupling of Z-Phe-Val-OH with Pro linked to PEGA-resin.
DIPCDI in combination with the evaluated compounds derivatives
of triazole, tetrazole and commercially available peptide coupling addi-
tives: A, no additive; B, 4; C, 6; D, 7; E, 8; F, 9; G, 10; H, 11 (HOBt);
I, 12 (HOAt) and J, 13 (Dhbt-OH)
OH
OH
H
Dhbt-OH, HOBt or HOAt
Fig. 1 Evaluation of the relative efficiency in forming peptide bonds
on a solid phase. Racemization in the solid-phase reaction was investi-
gated in a similar system where Z-Phe-Val-OH was coupled to solid-
phase-linked Pro using different catalysts (see Fig. 2).
evaluated compounds (Table 1). The excellent racemization-
suppressing property of HOAt afforded the product with only
0.95% racemization, whereas for HOBt, and the two tetrazoles
9 and 10, the amount of the LDL diastereoisomer was 10.9,
14.1 and 4.5%, respectively. The modified triazole 7, which in
the efficiency assay was found to be the most effective catalyst,
gave the highest degree of racemization (35.6%) of the evalu-
ated compounds. However, this is still a large improvement as
compared with the reaction without catalyst (50%) and the
catalyst could still be valuable in peptide synthesis. It may be
speculated that a different mechanism of catalysis is in oper-
ation with this catalyst. For the coupling with DIPCDI and no
addition of catalyst, a ratio of one to one of the two diastereo-
isomers was found. Racemization studies with the most efficient
additives 7 and 9–12 were repeated without significant devi-
ation of results.
In an effort to suppress the racemization and avoid the usual
formation of O-acylisourea, a reversed-addition protocol was
performed.23 DIPCDI and a small excess of catalyst were mixed
for 5 min at 0 ЊC to generate the intermediate adduct. The
protected dipeptide was added to the formed intermediate RO-
(RЈO)C-(NHR)2 which probably reacted to give the protected
active dipeptide ester of the catalyst and N,NЈ-diisopropylurea.
This coupling procedure was employed with the modified
triazole 7 as well as for N-hydroxytriazole (Table 1), but
unexpectedly it resulted in a small increase in the racemization.
However, using the protocol with a 3 molar excess of catalyst
decreased the amount of the LDL form from 47.2 to 28.7% for
the modified triazole 7 (Table 1).
Fig. 3 Determination of the efficiency of catalyst 11 (B), 12 (C) and 7
(D) in Aib᎐Aib bond formation on a solid phase. In trace A no catalyst
was added. After deprotection of the peptides and cleavage from the
resin HPLC separation showed the relative formation of Aib-Aib-Lys-
Ser-Ser-Tyr-Lys-NH2 (Rf 39.5) from Aib-Lys-Ser-Ser-Tyr-Lys-NH2 (Rf
36.9) (Table 1).
The efficiency of compounds 7, 11 and 12 as catalysts in the
synthesis of peptides containing sterically hindered bonds, such
as Aib᎐Aib for which racemization is not problematic, was
analyzed. The Fmoc-protected peptide Fmoc-Aib-Lys(Boc)-
Ser(tBu)-Ser(tBu)-Tyr(tBu)-Lys(Boc)- was synthesized on a
Rink amide linker on a PEGA resin (compound 14). The
peptide resin was first deprotected with piperidine and then it
was mixed with 3 equiv. of catalyst and amino acid and, after
equilibrium had been established, DIPCDI was added. The
reactions were terminated after 30 min by filtration and
washing, and after cleavage with 95% aq. TFA the amount of
J. Chem. Soc., Perkin Trans. 1, 1998
1729

View Article Online
coupling was monitored by HPLC. The results are presented in
Fig. 3. The reaction without addition of catalyst was a control
reaction, affording only 3% formation of an Aib᎐Aib bond.
Addition of HOBt or HOAt gave 6 and 38% conversion to
product respectively, whereas compound 7 afforded 43%
conversion to the peptide containing the Aib᎐Aib bond.
Compound 7 is therefore superior to HOAt in the catalysis of
αα-dialkylglycine-coupling reactions.
min. 1H and 13C NMR spectra were recorded on a 200 MHz or
300 MHz Bruker instrument. J-Values are given in Hz.
Hammer tests were performed on compounds 7–10 by placing a
2–4 mg sample of a compound on a block of iron and dropping
a hammer-head on the compound from height of 50 cm.
Compounds 7 and 8 were not explosive whereas compounds 9
and 10 both exploded.
The synthesis of fluorine-substituted analogs of compound 7
was attempted by a range of different strategies with both elec-
trophilic and nucleophilic substitution reactions; however, none
were successful.
1-Hydroxy-5-(methoxymethyl)-1,2,3-triazole 4
A solution of 1-benzyloxy-5-formyl-1,2,3-triazole 114 (163 mg,
0.80 mmol) and MeOH (10 cm3) was cooled to 0 ЊC and was
then treated with NaBH4 (41 mg, 1.08 mmol). After 40 min, the
solvent was removed in vacuo. The obtained solid was dissolved
in CH2Cl2 (15 cm3) and washed with saturated aq. NaCl (3 × 10
cm3). The organic phase was dried (MgSO4) and the solvent was
removed in vacuo to obtain 160 mg (97%) of 1-benzyloxy-5-
(hydroxymethyl)-1,2,3-triazole 2, Rf (EtOAc–heptane, 1:1)
0.05; mp 71–72 ЊC; δH (CDCl3) 7.49 (1 H, s), 7.42–7.31 (5 H, m),
5.47 (2 H, s), 4.33 (2 H, d, J 4.9) and 3.06 (1 H, br t, J ~5); δC
(CDCl3) 132.5 (s), 131.6 (s), 131.2 (d), 130.1 (d), 130.0 (d), 128.9
(d), 82.6 (t) and 52,1 (t) (Calc. for C10H11N3O2: C, 58.53; H,
5.40; N, 20.48%. Found: C, 58.60; H, 5.59; N, 20.24).
In conclusion, efficiency of the formation of a peptide bond
and measurement of the racemization during peptide coupling
have been determined using 6 new hydroxy-triazole and
-tetrazole derivatives and 3 commonly used compounds as
coupling catalysts in DIPCDI mediated couplings on a solid
phase. The compounds were modified 1-hydroxytriazoles and
N-hydroxytetrazoles.11–14 The synthesis of compounds 4 and 6
was achieved and afforded between 85% and 99% yield in the
four reaction steps reported. The results showed that three of
the new triazole and tetrazole derivatives could compete with
HOAt and HOBt. One of the catalysts was even found to be
superior to HOAt with respect to coupling efficiency. These
compounds contain only one five-membered ring and thus are
less sterically crowded. In particular, compound 7 could have an
advantage over HOAt and HOBt in coupling reactions, espe-
cially between sterically hindered amino acids such as Aib
residues. Furthermore, 5-chloro-1-hydroxy-1,2,3-triazole 7 was
a superior catalyst, giving faster coupling reactions than HOAt
but with less suppression of racemization. The results showed
that 2-hydroxytetrazole was as effective as HOAt in forming
peptide bonds. 2-Hydroxytetrazole afforded optical purity
comparable to that obtained with HOBt. 1-Hydroxytetrazole
had no significant catalytic activity but suppressed racemiz-
ation quite efficiently. CAUTION: The N-hydroxytetrazoles
were both explosive in a hammer test.
1-Benzyloxy-5-(hydroxymethyl)-1,2,3-triazole
2 (160 mg,
0.78 mmol) was dissolved in dry THF (5 cm3) and MeI (166 mg,
1.17 mmol) was added. The reaction mixture was cooled to
Ϫ78 ЊC and then NaH (24 mg, 1.0 mmol) was added. The cool-
ing bath was removed and the solution was stirred at room
temperature for 1 h. Water (2 cm3) was added and the solution
was extracted with CH2Cl2 (3 × 10 cm3). The organic phase was
dried (MgSO4) and the solvent was removed in vacuo. Purifica-
tion by column chromatography (EtOAc–heptane, 1:1) gave
145 mg (85%) of 1-benzyloxy-5-(methoxymethyl)-1,2,3-triazole
3, Rf (EtOAc–heptane, 1:1) 0.43; δH (CDCl3) 7.58 (1 H, s),
7.46–7.22 (5 H, m), 5.45 (2 H, s), 4.17 (2 H, s) and 3.25 (3 H, s);
δC (CDCl3) 132.2 (s), 131.7 (s), 129.7 (d), 129.5 (d), 128.5 (d),
128.3 (d), 82.4 (t), 60.8 (q) and 57.8 (t).
1-Benzyloxy-5-(methoxymethyl)-1,2,3-triazole 3 (315 mg,
1.43 mmol) was dissolved in EtOH (20 cm3), the solution was
cooled to 0 ЊC, 10% palladium on charcoal (50 mg) was added,
and the reaction mixture was stirred under hydrogen for 30 min.
After filtration through Celite and evaporation in vacuo, 182 mg
(99%) of 1-hydroxy-5-(methoxymethyl)-1,2,3-triazole 4 was
obtained, mp 108–110 ЊC; δH (D2O) 8.04 (1 H, s), 4.53 (2 H, s)
and 3.36 (3 H, s); δC (D2O) 128.7 (s), 127.4 (d), 61.4 (q) and 58.6
(t) (Calc. for C4H7N3O2: C, 37.21; H, 5.46; N, 32.54%. Found:
C, 37.52; H, 5.25; N, 32.30).
Experimental
General
Analytical TLC was performed on Merck silica gel 60F254
alumina sheets with detection of the protected triazoles by
charring with mostain [Ce(SO4)2ؒH2O (0.4 g) and
(NH4)6Mo7O24 (20 g) dissolved in 10% H2SO4 (400 ml)]. All
organic solvents were purchased from Labscan Ltd. (Dublin,
Ireland). Concentrations were performed under reduced
pressure at temperatures <40 ЊC. Suitable protected Nα-Fmoc-
amino acids, Z-Phe-Val and the Rink-amide-linker were pur-
chased from Nova Biochem (Switzerland); TBTU, NaBH4,
DIPCDI, HOBt and Dhbt-OH from Fluka (Switzerland);
NEM from Merck (Germany); NaCl and MgSO4 from Aldrich;
and HOAt from Millipore. The optical purity of Z-Phe-Val-
Pro-NH2 was analyzed by analytical HPLC. The catalysts
1-hydroxy-1,2,3-triazole 812,14 5-chloro-1-hydroxy-1,2,3-triazole
711,14 1-hydroxytetrazole 10,12 and 2-hydroxytetrazole 912 were
prepared as previously described. The relative molecular masses
of the peptide compounds were determined, using matrix-
assisted laser desorption time-of-flight mass spectroscopy
(MALDITOF-MS), recorded on a Lasermat 2000 (Finnigan
Mat) using a matrix of α-cyano-4-hydroxycinnamic acid. Quan-
titative amino acid analyses were performed on a Pharmacia
LKB Alpha Plus amino acid analyser following hydrolysis with
6  HCl at 110 ЊC for 36 h. Analytical HPLC was performed
using a Waters RCM 8 × 10 module and with a Deltapak C-18
column (19 × 300 mm). The solvent system for the analytical
HPLC was buffer A; 0.1% TFA in water and buffer B; 0.1%
TFA in 90% acetonitrile–10% water and UV detection was at
215 and 280 nm. The gradient for analytical HPLC (1 cm3
minϪ1); a linear gradient of 0–100% buffer B over a period of 50
5-Acetyl-1-hydroxy-1,2,3-triazole 6
5-Acetyl-1-benzyloxy-1,2,3-triazole 5 (138 mg, 0.64 mmol)13
was dissolved in MeOH (10 cm3) and the solution was cooled to
0 ЊC. Then, palladium on charcoal (30 mg) was added and the
reaction mixture was stirred under hydrogen for 30 min. After
filtration through Celite and evaporation in vacuo 76 mg (93%)
of 5-acetyl-1-hydroxy-1,2,3-triazole 6 was obtained, mp 143–
145 ЊC (decomp.); δH (CD3OD) 8.40 (1 H, s) and 2.65 (3 H, s);
δC (CD3OD) 186.2 (s), 132.3 (d), 128.7 (s) and 27.22 (q) (Calc.
for C4H5N3O2: C, 37.80; H, 3.97; N, 33.06%. Found: C, 37.91;
H, 3.81; N, 32.71).
Solid-phase synthesis and analysis, general procedure
The synthesis of the peptides using the PEGA resin15 were per-
formed by the plastic syringe technique24 or a 20-column
library generator.25 The protected Nα-Fmoc--amino acid
pentafluorophenyl (OPfp) esters (3 equiv.)22 were coupled in
DMF with the addition of Dhbt-OH (1 equiv.) as an acylation
catalyst and an indicator of the end-point of the acylation reac-
tion. The dipeptide Z-Phe-Val was coupled as its free acid using
TBTU in situ activation22 to prepare the two diastereoisomers
Z-Phe-Val-Pro-NH2 and Z-Phe-val-Pro-NH2.7 The Nα-Fmoc
group was removed by 20% piperidine in DMF. The resin was
1730
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
then washed with DMF (×5) and CH2Cl2 (×5) and dried. The
peptides which were used for racemization studies were cleaved
simultaneous from the Rink-linker20 by treatment with 95% aq.
TFA for 2 h. The resin was then rinsed with 95% aq. acetic acid
(× 4). Both TFA and acetic acid were removed under reduced
pressure and, after precipitation with diethyl ether, the crude
product was analyzed by analytical HPLC. However, determin-
ation and comparison of the ratio Ile:Val was performed by
quantitative amino acid analysis.
mg × 9, 0.135 mmol/g) was deprotected with 20% piperidine in
DMF and, after washing with DMF, transferred to a 20 column
Teflon multiple synthesizer. The dipeptide Z-Phe-Val-OH (23.9
mg, 0.06 mmol, 4 equiv.) was dissolved in DMF (0.1 cm3) and
catalyst (4 equiv.) in DMF (0.5 cm3) were added to the resin.
After 1 min, DIPCDI (8 mm3, 0.06 mmol, 4 equiv.) was simul-
taneously added to all columns and the coupling reaction was
left overnight. All the nine compounds 4, 6–13 were examined.
For the hydrochloride of 5-chloro-1-hydroxy-1,2,3-triazole 7,
DIEA (10 mm3, 0.06 mmol) was added. In one experiment, the
acylation was performed as described above but without
catalyst. The peptide was cleaved from the resin by 95% aq.
TFA and the degree of racemization was analyzed by HPLC
using the conditions as described above. For those compounds
showing catalytic activity the experiment was repeated.
Anchoring of the Rink linker
Rink-amide-linker (539.6 mg, 1 mmol) was dissolved in DMF.
The solution was cooled to 0 ЊC and TBTU (308.3 mg, 0.96
mmol) and NEM (0.239 cm3, 1.6 mmol) were added. After
8 min, the solution was added to the PEGA resin (1 g, 0.4
mmol) which had been swelling in DMF (1 h). The flask was
gently shaken for 2 h and the resin was then filtered off, and
rinsed with DMF (× 6) and CH2Cl2 (× 5). The remaining amino
groups were capped by addition of 10% acetic anhydride in
DMF (20 min).
The racemization assay B (see Table 1c)
In these experiments, DIPCDI (0.008 cm3, 0.06 mmol, 4 equiv.)
and catalyst (4.1 equiv.) were dissolved in DMF (0.5 cm3) at
0 ЊC. For hydrochloride of 5-chloro-1-hydroxy-1,2,3-triazole 7,
DIEA (0.011 cm3, 0.066 mmol) was added. The solution was
left for 5 min. Then, Z-Phe-Val-OH (23.9 mg, 0.06 mmol, 4
equiv.) as a solution in DMF (0.2 cm3) was added at 0 ЊC. After
2 min, the mixture was transferred to the Pro-linker(Rink)-resin
(50 mg, 0.015 mmol). The coupling reaction was carried out
overnight. The peptide was cleaved and analyzed as described
above (see Table 1b).
Procedure used for peptide formation comparison studies (see
Table 1a)
Synthesis of the Fmoc-Val-Gly-PEGA-resin was carried out as
described above in a plastic syringe using the PEGA-resin
(0.3–0.4 mmol gϪ1; 1 g). The resin was then dried under high
vacuum. The dried Fmoc-peptide resin (200 mg) was treated
with 20% piperidine in DMF for 20 min, washed with DMF
and then transferred to a 20-column library generator (10 mg
of Fmoc-peptide resin in each column), such that the resin was
equally divided into the 20 columns. Then, DMF (0.1 cm3),
Fmoc-Leu-OH (4.7 mg, 13.5 µmol) in DMF (50 mm3) and
coupling additive (3 equiv.) in DMF (0.1 cm3) were added in
each column. After 2 min, a solution of DIPCDI (2.2 mm3, 13.5
µmol) in DMF (50 mm3) was added and the mixture was
allowed to react for exactly 5 min. The evaluated catalysts were
1-hydroxy-5-(methoxymethyl)-1,2,3-triazole 4 (1.76 mg, 13.5
µmol), 5-acetyl-1-hydroxy-1,2,3-triazole 6 (1.70 mg, 13.5 µmol),
5-chloro-1-hydroxy-1,2,3-triazole hydrochloride 7ؒHCl (2.10
mg, 13.5 µmol), 1-hydroxy-1,2,3-triazole 8 (1.15 mg, 13.5 µmol),
2-hydroxytetrazole 9 (1.91 mg, 13.5 µmol), 1-hydroxytetrazole
10 (1.91 mg, 13.5 µmol), HOBt 11 (1.84 mg, 13.5 µmol), HOAt
12 (1.84 mg, 13.5 µmol) and Dhbt-OH 13 (2.20 mg, 13.5 µmol).
For 5-chloro-1-hydroxy-1,2,3-triazole hydrochloride 7ؒHCl
diisopropylethylamine (DIEA) (0.0023 cm3, 13.5 µmol) was
added for neutralization. One experiment was performed
without catalyst and the protocol was the same as described
above except that catalyst was not added. Then, resin was
drained, washed and the Fmoc-group was removed by treat-
ment with 20% piperidine in DMF followed by washing with
DMF. The peptides were hydrolyzed by 6  HCl at 110 ЊC
for 36 h and the Val:Ile ratio was determined by amino acid
analysis.
The racemization assay B (see Table 1d)
In these experiments, the racemization assay B protocol was
used, but with an increase to 12 equiv. of catalyst as compared
with resin-bound amino groups.
Preparation of Fmoc-Aib-Lys(Boc)-Ser(tBu)-Ser(tBu)-Tyr(tBu)-
Lys(Boc)-Rink linker-PEGA resin 14
The precursor Fmoc-Lys(Boc)-Ser(tBu)-Ser(tBu)-Tyr(tBu)-
Lys(Boc)-Rink linker-PEGA resin (0.12 mmol gϪ1) was syn-
thesized as described above using the PEGA resin. After Fmoc
deprotection with 20% piperidine in DMF, Fmoc-Aib-OH (54
mg, 0.166 mmol) was coupled to the protected peptide PEGA
resin (460 mg, 0.055 mmol) using TBTU (51.1 mg, 0.16 mmol)
and NEM (0.0033 cm3, 0.266 mmol). The coupling reaction was
carried out for 24 h. A small amount of the peptide resin was
then Fmoc-deprotected. The side-chain deprotection and cleav-
age from the resin was mediated by 95% aq. TFA. The crude
product was analyzed by analytical HPLC (tR 37.8 min) using a
isotropic gradient of 100% buffer A during 20 min followed by
a linear gradient of 0–100% buffer B during 70 min. Analysis by
MALDITOF-MS [m/z 696.3 (M ϩ H)ϩ and 718.2 (M ϩ Na)ϩ,
(C31H53N9O9 requires M, 695.8)] confirmed the expected
product.
Coupling of Fmoc-Aib-OH to Aib-Lys(Boc)-Ser(tBu)-Ser(tBu)-
Tyr(tBu)-Lys(Boc)-Rink linker-PEGA resin using catalyst-
mediated couplings
The racemization assay A (see Table 1b)
The two stereoisomers for racemization studies,8 Z-Phe-Val-
Pro-NH2 and Z-Phe-val-Pro-NH2, were synthesized as follows;
Z-Phe-Val-OH (87.6 mg, 0.22 mmol) was dissolved in DMF,
then NEM (0.044 cm3, 0.35 mmol) and TBTU (67.1 mg, 0.20
mmol) were added. The mixture was stirred at 0 ЊC for 8 min
and was then added to the Pro-Rink-linker PEGA resin (180
mg, 0.054 mmol). The mixture was allowed to react for 2 h. The
resin was then washed thoroughly with DMF. The resin was
dried and the two stereoisomers were cleaved from the resin by
95% aq. TFA. They were analyzed by analytical HPLC using a
linear gradient of 0–100% buffer B over a period of 50 min
[tR 36.7 min (LLL) and tR 38.1 min (LDL)] as well as by
MALDITOF-MS; m/z 517.5 (M ϩ Na)ϩ (LLL) and 517.3
(M ϩ Na)ϩ (LDL), respectively (C27H34N4NaO5 requires m/z,
517.2). The dried Fmoc-Pro-Rink linker-PEGA resin (50
The dried protected-peptide resin 14 (400 mg, 0.12 mmol gϪ1
)
was Fmoc-deprotected with 20% piperidine in DMF for 20
min, washed with DMF and then divided into eight plastic
syringes. The protected amino acid Fmoc-Aib-OH in DMF
(5.86 mg, 18 µmol in 0.050 cm3) and either HOBt (2.43 mg, 18
µmol), HOAt (2.44 mg, 18 µmol) or 5-chloro-1-hydroxy-1,2,3-
triazole hydrochloride 7ؒHCl (2.8 mg, 18 µmol) in DMF (0.05
cm3) were added to the syringe (for 2 sets of reactions). After
2 min, a solution of DIPCDI in DMF (0.0028 cm3, 18 µmol in
0.1 cm3) was added and the coupling reactions were performed
in 25 min and 4 h respectively. In one experiment, the reaction
was carried out in the absence of catalyst. For 5-chloro-1-
hydroxy-1,2,3-triazole hydrochloride 7ؒHCl, one mole equiv. of
DIEA (0.0031 cm3, 18 µmol) was added for neutralization. The
J. Chem. Soc., Perkin Trans. 1, 1998
1731

View Article Online
11 M. Begtrup and P. Vedsø, Acta Chem. Scand., 1996, 50, 549.
Fmoc group was removed and the peptide was side-chain-
deprotected and cleaved with 95% aq. TFA. The progress of the
reaction was monitored by analytical HPLC (tR 39.5 min).
The yield of the desired peptide Aib-Aib-Lys-Ser-Ser-Tyr-Lys-
NH2 was calculated based on the areas of the two obtained
HPLC peaks (see Fig. 3). The product was characterized by
MALDITOF-MS; m/z 782.5 (M ϩ H)ϩ, 806.5 (M ϩ Na)ϩ and
822.4 (M ϩ K)ϩ (C35H60N10O10 requires M, 780.9).
12 M. Begtrup and P. Vedsø, J. Chem. Soc., Perkin Trans. 1, 1995, 243.
13 J. Felding, P. Uhlmann, J. Kristensen, P. Vedsø and M. Begtrup,
Synthesis, in the press.
14 P. Uhlmann, J. Felding, P. Vedsø and M. Begtrup, J. Org. Chem.,
1997, 62, 9177.
15 M. Meldal, Tetrahedron Lett., 1992, 33, 3077.
16 M. Meldal, I. Svendsen, K. Breddam and F. I. Auzanneau, Proc.
Natl. Acad. Sci. USA, 1994, 91, 3314.
17 M. Meldal, F.-L. Auzanneau, O. Hindsgaul and M. M. Palcic,
J. Chem. Soc., Chem. Commun., 1994, 1849.
18 C.-D. Chang and J. Meienhofer, Int. J. Pept. Protein Res., 1978, 11,
Acknowledgements
We thank Dr Ib Svendsen for amino acid analysis.
246.
19 E. Atherton, H. Fox, D. Harkiss, C. J. Logan, R. C. Sheppard and
B. J. Williams, J. Chem. Soc., Chem. Commun., 1978, 537.
20 M. Meldal, Methods: A Companion to Methods in Enzymology,
Academic Press, 1994, vol. 6, p. 417.
21 H. Rink, Tetrahedron Lett., 1987, 28, 3787.
22 R. Knorr, A. Trzeciak, W. Bannwarth and D. Gillesen, Tetrahedron
Lett., 1989, 30, 1927.
23 G. Doleschall and K. Lempert, Tetrahedron Lett., 1963, 1195.
24 I. Christiansen-Brams, M. Meldal and K. Bock, J. Chem. Soc.,
Perkin Trans. 1, 1993, 1461.
25 E. Atherton and R. C. Sheppard, J. Chem. Soc., Chem. Commun.,
1985, 165.
References
1 J. C. Sheehan and G. P. Hess, J. Am. Chem. Soc., 1955, 77, 1067.
2 N. L. Benoiton and F. M. F. Chen, J. Chem. Soc., Chem. Commun.,
1981, 543.
3 B. R. Merrifield, J. Am. Chem. Soc., 1963, 85, 2149.
4 W. Konig and R. Geiger, Chem. Ber., 1970, 103, 788.
5 L. A. Carpino, J. Am. Chem. Soc., 1993, 115, 4397.
6 E. Atherton, L. Cameron, M. Meldal and R. C. Sheppard, J. Chem.
Soc., Chem. Commun., 1986, 1763.
7 W. Konig and R. Geiger, Chem. Ber., 1970, 103, 2024.
8 L. A. Carpino and A. El-Faham, J. Org. Chem., 1995, 60, 3561.
9 D. Hudson, J. Org. Chem., 1988, 53, 617.
Paper 7/07313E
Received 9th October 1997
Accepted 24th February 1998
10 D. Hudson, Pept. Res., 1990, 3, 51.
1732
J. Chem. Soc., Perkin Trans. 1, 1998