Stability studies of N-acylimidazoles

Article abstract of DOI:10.1002/1099-0690(200208)2002:15<2633::AID-EJOC2633>3.0.CO;2-U

Studies of the stabilities of a series of N-acylimidazoles towards acidic and basic conditions of potential usefulness for the removal of common temporary protection in peptide and oligonucleotide synthesis are presented. N-Acylimidazoles with a variety of substituents in the acyl component were prepared and treated with 3% trifluoroacetic acid (TFA) in chloroform and with 2% 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in N,N-dimethylformamide (DMF), the extent of their degradation being determined by proton NMR. N-(2,4,6-Trimethylbenzoyl)imidazole (1) and N-(2,6-dimethoxybenzoyl)-imidazole (2) remained unaffected under the above acidic and basic conditions after 4 d and 2 d, respectively. In addition, 1 and 2 were resistant to treatment with a solution of 2% piperidine/2% DBU in DMF for 24 h. Under ammonolytic conditions, 2 was rapidly cleaved (less than 1 h), whereas 1 was 64% degraded after 48 h, as ascertained by reversed-phase HPLC. Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002.

Full text of DOI:10.1002/1099-0690(200208)2002:15<2633::AID-EJOC2633>3.0.CO;2-U

FULL PAPER
Stability Studies of N-Acylimidazoles[‡]
Simone Zaramella,[a] Roger Strömberg,[a] and Esther Yeheskiely*[a]
Keywords: Nitrogen heterocycles / Cleavage / Protecting groups
Studies of the stabilities of a series of N-acylimidazoles to-
wards acidic and basic conditions of potential usefulness for
the removal of common temporary protection in peptide and
oligonucleotide synthesis are presented. N-Acylimidazoles
with a variety of substituents in the acyl component were
prepared and treated with 3% trifluoroacetic acid (TFA) in
chloroform and with 2% 1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU) in N,N-dimethylformamide (DMF), the extent of their
imidazole (2) remained unaffected under the above acidic
and basic conditions after 4 d and 2 d, respectively. In addi-
tion, 1 and 2 were resistant to treatment with a solution of
2% piperidine/2% DBU in DMF for 24 h. Under ammonolytic
conditions, 2 was rapidly cleaved (less than 1 h), whereas 1
was 64% degraded after 48 h, as ascertained by reversed-
phase HPLC.
degradation being determined by proton NMR. N-(2,4,6-Tri- ( Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany,
methylbenzoyl)imidazole (1) and N-(2,6-dimethoxybenzoyl)- 2002)
Introduction
rate imidazole-containing components in conjugates to oli-
gonucleotides both prior to and subsequent to oligonucleo-
tide synthesis, provided that the side-chain protection be
adapted to that technique. A standard oligonucleotide pro-
cedure entails several repeated steps, such as the removal of
mono- or dimethoxytrityl (MMTr and DMTr) groups from
the 5Ј-hydroxy groups of RNA and DNA, respectively.
Moreover, deprotection of the nucleobases by ammonolysis
is also included. It would thus be preferable if an imidazole
protecting group for the synthesis of nucleopeptides and oli-
gonucleotide-imidazole conjugates were stable to repeated
detritylation, but cleaved by ammonolysis. It would also be
an advantage, and would make it more versatile, if it were
stable to Fmoc removal conditions.
Imidazole groups are found not only in peptides/proteins,
but also in various modified biomolecules, such as peptido-
mimetics,[1] nucleopeptides,[2] or oligonucleotide conjugates
in which the conjugate may be a peptide[3] or another imid-
azole-containing structure.[4] Since imidazoles are known to
play an essential role in many enzyme mechanisms, low-
molecular-weight model compounds containing imidazole
moieties are important tools in studies aimed at more de-
tailed knowledge of enzyme catalysis.[5] Derivatives of imid-
azole have also been used in artificial enzymes.[6] As part of
an ongoing program, we are interested in the synthesis of
several of the above classes of molecules.
Most of the commonly used protecting groups for imida-
zole functions are of the alkyl,[10Ϫ14] urethane,[15,16] or sul-
fonyl types.[17] Protecting groups of the first type are usually
removed under strongly acidic conditions, which are incom-
patible with the presence of acid-sensitive structures. Those
of the second and third classes may be unstable to some
reagents used in standard peptide and oligonucleotide
chemistry.[15,17] We were curious to find out whether acyl-
ation could provide sufficient protection for imidazole func-
tions. N-Acylimidazoles are generally considered to be mild
acylating agents,[18] but their reactivity, like that of other
acylating reagents, depends on steric and electronic factors.
It is known that sterically hindered aliphatic N-acylimid-
azoles show better stability towards nucleophiles than un-
hindered ones do.[19] In addition, it has also been reported
that electron-donating groups in substituted aromatic N-
acylimidazoles exhibit enhanced stability towards hydro-
lysis.[20,21] Several stability studies of various N-acylimid-
azoles have already been published.[20Ϫ22] However, these
Oligonucleotide conjugates are both acid- and base-sensi-
tive, due to the presence of the nucleic acid component. It
is well established that native RNA and, especially, DNA
are labile in strongly acidic media.[7] In addition, RNA
strands are prone to degradation upon prolonged exposure
to base. Since nucleic acids are rather sensitive to acids, the
use of tert-butyl- or benzyl-type protecting groups in the
synthesis of these conjugates is precluded. It has neverthe-
less been reported[8,9] that nucleopeptides can be prepared
by employment of the 9-fluorenylmethyloxycarbonyl
(Fmoc) system as an α-amino protecting group for the pep-
tide component, yet maintaining standard techniques for
the nucleic acid segment. It should be possible to incorpo-
[‡]
Acyl Groups as Prospective Protection for Imidazole, I
[a]
Organic and Bioorganic Chemistry, MBB, Scheele Laboratory,
Karolinska Institutet
17177 Stockholm, Sweden
Fax: (internat.) ϩ 46-8/311052
E-mail: esther.kuyl-yeheskiely@mbb.ki.se
Eur. J. Org. Chem. 2002, 2633Ϫ2639  WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 1434Ϫ193X/02/0815Ϫ2633 $ 20.00ϩ.50/0
2633

S. Zaramella, R. Strömberg, E. Yeheskiely
FULL PAPER
were limited, and the conditions employed were generally
not those most suitable for the synthesis of acid- and base-
sensitive compounds. As a first step towards the develop-
ment of acyl-type protecting groups for the imidazole func-
tions of oligonucleotide-imidazole-containing conjugates,
we subjected a set of (a few) aliphatic and (mostly) aromatic
derivatives to stability tests. They were exposed to acidic
and basic conditions that can be used for the removal of
MMTr and Fmoc, respectively, and to ammonolysis. This
paper reports the results of these investigations, performed
in order to provide a basis for the development of acyl-
based imidazole protection.
Results and Discussion
It is well established that steric hindrance at the carbonyl
function and the presence of electron-releasing groups
render acylating agents less prone to nucleophilic
attack.[23Ϫ26] The most stable compounds are not the best
choice for all purposes, and so we decided to investigate a
series of derivatives to provide a basis for choosing the most
suitable for each application. For this purpose, N-benzoyl-
imidazoles with methyl or methoxy groups in their ortho
and/or para positions (compounds 1Ϫ7) were synthesized,
together with the naphthoyl (12 and 13), pivaloyl (14), and
adamantoyl (15) derivatives. Bromo- and iodo-substituted
benzoylimidazoles were also prepared (compounds 8, 9,
and 11). We reasoned that the bulkiness of a halogen atom
in the ortho position might overcome the inductive effect
and that, as a consequence, derivatives 8 and 9 could be
more stable towards degradation than the parent N-
benzoylimidazole (10). All the above-mentioned molecules
were obtained in good yields by treatment of imidazole in
twofold excess with the individual acyl chlorides at room
temperature, except for compounds 1, 2, and 3, which
were synthesized from the corresponding acid and
1,1Ј-carbonylbis(1H-imidazole) in equimolar amounts.
We chose to follow the cleavage of N-acylimidazoles by
proton NMR. The stability studies were performed in 3%
(v/v) trifluoroacetic acid (TFA) in chloroform, conditions
resembling those used for the removal of MMTr from the
5Ј-hydroxy moieties of nucleotides. Solutions such as 1%
TFA or 2Ϫ3% dichloroacetic acid (advisable with DNA) in
a chlorinated solvent are normally employed, and complete
detritylation occurs within 2 min.[27,28] However, in order to
provide a large safety margin and to stretch the limits of
applicability to other groups that may require stronger acid
for removal, we decided to use 3% TFA solution. It should
water in order to evaluate the effect of water on the stability
of the substrate. Derivative 14 was partially degraded (10%)
within 30 min in deuterated chloroform containing 200 ppm
water, compared with 2% degradation of the same substrate
in 15 ppm of water. To circumvent possible partial degrada-
tion of the N-acylimidazoles, the water level in all further
TFA experiments was monitored, and did not exceed
15 ppm.
The results of the experiments under anhydrous condi-
tions (water content Ͻ 15 ppm) are reported in Table 1. De-
gradation of the aromatic derivatives was almost undetect-
able in the first 5Ϫ6 h and limited in most cases to a few
percent after 4 d. Compounds 1 and 2 showed no cleavage
at all within the detection limits of the instrument. As ex-
pected, substitution of the aromatic ring in N-benzoylimid-
azoles with electron-donating groups such as methoxy and
methyl resulted in substantial stabilization (see compounds
1Ϫ7) in comparison with the parent unsubstituted N-
benzoylimidazole (10). The data presented in Table 1 also
indicate that compounds bearing bulky substituents in their
ortho positions (see compounds 1Ϫ5, 7Ϫ9, and 12) were
significantly more stable than N-benzoylimidazole and sub-
stantially more stable than the aliphatic N-acylimidazoles.
The results of the stability tests on the substituted N-
be mentioned that concentrations of TFA much higher than acylimidazoles in diluted TFA prompted us to investigate
this would be precluded due to rapid degradation of the whether they would also withstand basic conditions. In
oligonucleotide component of the conjugate. As a con- view of possible applications to peptide-containing mole-
sequence, conditions commonly used for removal of Boc cules, such as nucleopeptides, we chose conditions that can
or tert-butyl ethers and esters are incompatible with most be used for removal of the Fmoc group from α-amino func-
nucleic acids.
tions of amino acids. In most cases, piperidine is employed
Initially, N-pivaloylimidazole (compound 14) was treated to promote cleavage of Fmoc by β-elimination. Unfortu-
with 3% TFA in chloroform containing 15 or 200 ppm nately, treatment with the nucleophilic piperidine resulted
2634
Eur. J. Org. Chem. 2002, 2633Ϫ2639

Stability Studies of N-Acylimidazoles
FULL PAPER
biomolecules, in particular oligonucleotides. To ascertain
how quickly the acyl group could be removed from the im-
idazole residue, the reactions were monitored by reversed-
phase HPLC. The results are summarized in Table 3 and
indicate that removal of the acyl groups from the imidazole
residue was particularly fast, except for the 2,4,6-trimethyl-
benzoyl group, which exhibited a substantial resistance to
cleavage.
Table 1. Stability of N-acylimidazoles in 3% (v/v) TFA in CDCl3,
at 25 °C; [substrate] ϭ 1.3·10Ϫ2 ; [TFA] ϭ 4·10Ϫ1

Substrate[a]
Substrate cleavage [%][b]
δ(1H)
23 h
48 h
70 h
96 h
[ppm][c]
2,4,6-Me3BzIm (1) n.d.[d]
2,6-MeO2BzIm (2) n.d.
n.d.
n.d.
Ͻ 1
Ͻ 1
Ͻ 1
1
1
2
2
2
4
6
20
60
Ϫ
n.d.
n.d.
Ͻ 1
Ͻ 1
1
2
2
3
3
3
7
n.d.
n.d.
Ͻ 1
1
1
2
3
3
3
5
10
10
40
Ϫ
Ϫ
8.83 (2)
7.87 (5)
7.84 (5)
7.86 (5)
7.85 (5)
9.03 (2)
9.19 (2)
2,4-Me2BzIm (3)
2-MeBzIm (4)
2-IBzIm (8)
2-BrBzIm (9)
4-MeOBzIm (6)
1-NaphtIm (12)
2,4-MeO2BzIm (5)
2-MeOBzIm (7)
BzIm (10)
2-NaphtIm (13)
4-BrBzIm (11)
AdCOIm (15)
PivIm (14)
Ͻ 1
Ͻ 1
Ͻ 1
1
Ͻ 1
1
1
1
2
2
The data presented in Tables 1Ϫ3 show the expected sta-
bilizing effect of an ortho substituent, which it is particu-
larly evident in the bis(ortho)-substituted derivatives (com-
9.14 (2) pounds 1 and 2). This effect of steric hindrance on nucleo-
7.82 (5)
7.86 (5)
9.23 (2)
9.30 (2)
9.25 (2)
7.92 (5)
7.91 (5)
philic attack at the carbonyl group is best illustrated by
comparison of the stabilities of compounds 1 vs. 3, 2 vs. 5,
9 vs. 11, and 12 vs. 13. In the ortho-iodo and -bromo deriv-
atives (8 and 9) the steric effect is clearly larger than the
electron-withdrawing influence, rendering these compounds
as stable as their methyl and methoxy counterparts (4 and
7).
9
7
30
70
20
80
Ϫ
[a]
Substrate concentration for compounds 1, 6, 9, 10, and 11 was
1.4·10Ϫ2 .
Experiments are based on a single run, except for
[b]
The stability exhibited by compounds 1Ϫ9 and 12 in di-
substrate 3 which was shown to be reproducible with a maximum luted TFA solution can be considered satisfactory, a fact
[c]
deviation of 20%. Chemical shifts of the substrate protons used
that should make these acyl groups suitable for imidazole
for quantification; cleavage is referred to the 5-H of free imidazole
protection in association with the acid-labile MMTr and
at δ ϭ 7.48 ppm, except for compounds 4 and 8, which referred to
DMTr groups commonly used in oligonucleotide chemistry.
the o-CH3 signal at δ ϭ 2.66 ppm and to the 2-H signal at δ ϭ
8.73 ppm, respectively. n.d. means not detected.
[d]
The stability of these acyl groups should be sufficient to
allow prolonged exposure to 3% TFA with negligible loss
of imidazole protection.
In association with the Fmoc group, only the use of acyl
in rapid degradation of substrate 5 (20% cleavage after groups in N-acylimidazoles 1 and 2 is recommended. The
60 min with 20% piperidine in DMF) even under anhydrous Fmoc group can be removed with DBU in less than 10 min;
conditions. As a consequence, we turned our attention to however, some procedures require repetition of the treat-
the nonnucleophilic base 1,8-diazabicyclo[5.4.0]undec-7-ene ment and the overall time of contact with the cleaving solu-
(DBU), which has also been used for removal of Fmoc.[29] tion can be rather long.[31] Because of their lability in basic
Rapid degradation of substrate 5 was observed in a solution solutions, compounds 3Ϫ15 should be excluded from ap-
of 2% DBU in DMF containing 350 ppm of water (70% plication in procedures requiring repeated removal of the
cleavage in 60 min). In order to obtain a relatively anhy- Fmoc group.
drous solvent, molecular sieves were added prior to use, re-
The fact that the 2,6-dimethoxybenzoyl group was com-
ducing the water content to about 30 ppm. Table 2 summa- pletely removed in aqueous ammonia within 1 h should
rizes the results of the stability tests towards DBU in anhy- make it more suitable for general use in synthesis. The 2,4,6-
drous DMF. The order of stability was similar to that ob- trimethylbenzoyl residue required prolonged ammonolysis,
served under acidic conditions; however, the relative conditions that may not be ideal for base-sensitive molec-
stabilities of the halogenated compounds 8, 9, and 11 were ules.
not as high in this basic environment as under dilute TFA
solutions. In most cases, a significant amount of cleavage
had already occurred in the early stages of the experiments,
Conclusion
due to the unavoidable presence of traces of water in the
From the results of the acid stability tests it is evident
samples. Compounds 1 and 2, however, were stable in this that N-acylimidazoles 3Ϫ15, and in particular derivatives 1
environment. In fact, no detectable cleavage was observed and 2, can survive prolonged exposure to anhydrous 3%
even after 2 d. In addition to the above tests, compounds 1 TFA in chloroform. They should thus remain intact during
and 2 were subjected to treatment with a mixture of 2:2:96 the repeated removal of, for example, MMTr and DMTr.
(v/v/v) DBU/piperidine/DMF at room temperature. These On the other hand, compounds 3Ϫ15 are not very stable in
conditions are applied in batch peptide synthesis, the addi- basic environments, a fact that precludes the application of
tional piperidine being used to scavenge undesired dibenzo- their acyl groups in procedures that require repeated re-
fulvene.[30] In these experiments the substrates were unaf- moval of, for example, Fmoc groups. However, compounds
fected after 24 h.
1 and 2 are quite stable and their acyl groups could there-
Ammonolysis tests were also conducted on compounds fore be applied for imidazole protection in the synthesis of
1Ϫ3 and 5. These tests were performed because ammono- compounds by procedures that require mild basic condi-
lysis is often used to effect final deprotection of different tions.
Eur. J. Org. Chem. 2002, 2633Ϫ2639
2635

S. Zaramella, R. Strömberg, E. Yeheskiely
Table 2. Cleavage of N-acylimidazoles in 2% (v/v) DBU in [D7]DMF, at 25 °C; [substrate] ϭ 1.3·10Ϫ2 ; [DBU] ϭ 1.3·10Ϫ1
FULL PAPER

Substrate[a]
Substrate cleavage [%][b]
Relative error
[%]
δ(1H)
15 min
30 min
60 min
120 min
[ppm][c]
2,4,6-Me3BzIm (1)
2,6-MeO2BzIm (2)
2,4-Me2BzIm (3)
2-MeBzIm (4)
2,4-MeO2BzIm (5)
4-MeOBzIm (6)
2-MeOBzIm[f] (7)
2-IBzIm (8)
n.d.
n.d.
1.3[d]
1.7[e]
5.1
9.3
10
6.3
6.1[g]
11
14
16
17
28
n.d.
n.d.
2.9
3.3
8.8
9.9
12
8.8
11
15
17
19
23
37
n.d.
n.d.
4.6
5.4
11
n.d.
n.d
7.5
8.2
13
12
14
Ϫ
Ϫ
7.20 (4)
7.15 (4)
7.19 (4)
7.20 (4)
6.84 (3Ј)
8.23 (2)
7.34 (3Ј)
7.65 (5)
8.12 (2)
7.22 (4)
7.21 (4)
7.25 (4)
8.53 (2)
7.21 (4)
Ϯ15
Ϯ15
Ϯ5
Ϯ10
Ϯ5
Ϯ15
Ϯ5
Ϯ5
Ϯ5
Ϯ5
Ϯ15
Ϯ15
11
13
12[e]
14
16
2-BrBzIm (9)
20[g]
23
1-NaphtIm (12)
BzIm (10)
2-NaphtIm (13)
PivIm (14)
4-BrBzIm (11)
20
23
37[h]
30[g]
40
26[g]
30
48
59
[a]
[b]
[c]
Substrate concentration for compounds 3, 4, 10, and 11 is 1.4·10Ϫ2 . Experiments are based on duplicate runs. Chemical shifts
of the substrate protons used for quantification; cleavage relates to the 5-H of free imidazole at δ ϭ 7.09 ppm, except for compounds 5
[d]
[e]
[f]
and 14, in which it relates to the 2-H signal at δ ϭ 7.72 ppm. Relative error 55%. Relative error 25%. Cleavage of substrate 7 is
[g]
[h]
somewhat overestimated due to overlapping of proton signals. This value is based on a single measurement.
Value corresponding
to the cleavage after 180 min.
peptides, oligonucleotides, and oligonucleotide conjugates.
On the other hand, the increased ammonolytic stability of
1 might prove advantageous in term of storage, handling,
and workup, and so this may be the protection of choice
for more base-stable compounds. Further evaluation of 2,6-
dimethoxy- and 2,4,6-trimethylbenzoyl groups for imid-
azole protection in the synthesis of nucleopeptides, oligonu-
cleotide conjugates, and other biomolecules is in progress.
These investigations are also addressing the selective acyl-
ation of one of the two distinct nitrogen atoms of a substi-
tuted imidazole ring in different imidazole derivatives, in-
cluding histidine.
Table 3. Cleavage of N-acylimidazoles in 3:1 (v/v) aq. NH3/EtOH
at room temperature; [substrate] ϭ 4·10Ϫ2

Substrate
Substrate cleavage [%][a]
Peak tR
[min]
0.5 h
1 h
4 h
21 h
2,4,6-Me3BzIm (1) 16
2,6-MeO2BzIm (2) 97
20
100
Ϫ
31
Ϫ
Ϫ
Ϫ
46
Ϫ
Ϫ
Ϫ
28.1
21.0
2,4-Me2BzIm (3)
100
25.7
2,4-MeO2BzIm (5) 100
Ϫ
24.1[b]
[a] Experiments are based on duplicate runs. [b] Gradient: 16% hold
for 5 min, then 16Ϫ80% over 40 min (see for mobile phases in the
General Remarks section).
Experimental Section
The fact that N-acylimidazoles 3Ϫ15 are fairly stable to
3% TFA in chloroform but not very stable under basic con-
ditions may be a useful combination, which could result in
applications of such acyl groups as temporary or transient
protecting groups in organic chemical procedures involving
imidazole rings.
General Remarks: Imidazole (Lancaster) was recrystallized from
chloroform. 1,1Ј-Carbonylbis(1H-imidazole) (Fluka) and all acyl
chlorides (Fluka, Aldrich, and Lancaster) were of synthesis grade
and used without further purification. THF was distilled at atmo-
spheric pressure from a K/Na mixture. All other solvents used dur-
ing syntheses and kinetic measurements were stored over 4 A mo-
lecular sieves and kept under nitrogen. Solvents were evaporated
under reduced pressure at a water bath temperature not exceeding
˚
As mentioned, N-(2,4,6-trimethylbenzoyl)- and N-(2,6-di-
methoxybenzoyl)imidazoles both have excellent stabilities in
dilute TFA and DBU solutions. They exhibit different labil- 40 °C. TFA and piperidine were distilled at atmospheric pressure,
and DBU was distilled in vacuo. All glassware, syringes, and
needles used during syntheses and kinetic measurements were dried
in an oven prior to use. The glassware was assembled while hot and
cooled under a flow of dry nitrogen. N-Acylimidazoles were stored
in the presence of a desiccant and kept cold. Sample preparation
for kinetics was performed under a flow of dry nitrogen. NMR
spectra were recorded with a Bruker Avance DRX 400 spectro-
meter at 400 and 100 MHz for proton and carbon-13, respectively.
Chemical shifts (δ) are reported in ppm downfield from tetra-
methylsilane, with the residual nondeuterated solvent signal as an
ities when exposed to ammonolysis, upon which compound
1 is more stable than compound 2. This feature implies that
removal of the acyl group of 1 would require prolonged
and/or higher temperatures during ammonolysis. This may
compromise the structural integrity of base-labile mole-
cules. To circumvent such degradation, the duration of am-
monolysis should be limited as much as possible, thus mak-
ing the 2,6-dimethoxybenzoyl group the protecting group
of choice for the general synthesis of biologically active
compounds of various biomolecules, in particular nucleo- internal standard. The substrate concentration was set to be
2636 Eur. J. Org. Chem. 2002, 2633Ϫ2639

Stability Studies of N-Acylimidazoles
FULL PAPER
1.3·10Ϫ2 , so that a distinguishable signal could be obtained
129.5, 131.4, 132.6, 138.2, 138.5, 142.9, 167.0 ppm. HRMS (ES-
within a spectrum acquisition time of 2 min. The extent of degrada- TOF): m/z calcd. for C12H13N2O [M ϩ H]ϩ: 201.1028; found
tion was quantified by integration of the proton signals of the imid-
azole ring (normally the 2-H or the 5-H signal in the acidic tests,
and the 2-H or the 4-H signal in the basic tests) in the substrate
and in the free imidazole, at different time intervals. The signals
used for quantification were selected because they did not overlap
with other signals. Chemical shifts of these signals are reported in
Tables 1 and 2. Mass analysis of N-acylimidazoles was performed
with a Micromass LCT (ES-TOF) mass spectrometer in positive
ion mode. Reversed-phase HPLC analysis was carried out with a
Jasco HPLC system equipped with a Jones Genesis C18 analytical
column (250 ϫ 4.6 mm). A gradient (16% hold for 5 min, then
16Ϫ80% over 30 min) of CH3CN in 0.1  triethylammonium acet-
ate (pH ϭ 6.5) was used with a flow rate of 0.8 mL/min, at 30 °C.
Chromatograms were recorded at 254 nm.
201.1023.
N-(2-Methylbenzoyl)imidazole (4): 1H NMR (400 MHz, CDCl3):
δ ϭ 2.39 (s, 3 H), 7.13 (s, 1 H), 7.32Ϫ7.37 (m, 2 H), 7.42 (d, J ϭ
7.1 Hz, 1 H), 7.46 (s, 1 H), 7.47 (td, J ϭ 7.7, J ϭ 1.3, 1 H), 7.89
(s, 1 H) ppm. 13C NMR (100 MHz, CDCl3): δ ϭ 19.9, 117.7, 126.2,
128.6, 131.5, 131.8, 132.3, 132.5, 137.8, 138.4, 166.9 ppm. HRMS
(ES-TOF): m/z calcd. for C11H11N2O [M ϩ H]ϩ: 187.0871; found
187.0869.
N-(2,4-Dimethoxybenzoyl)imidazole (5): 1H NMR (400 MHz,
CDCl3): δ ϭ 3.80 (s, 3 H), 3.90 (s, 3 H), 6.55 (d, J ϭ 1.9 Hz, 1 H),
6.60 (dd, J ϭ 8.5, 2.0 Hz, 1 H), 7.08 (s, 1 H), 7.46Ϫ7.48 (m, 2 H),
7.90 (s, 1 H) ppm. 13C NMR (100 MHz, CDCl3): δ ϭ 56.1, 56.2,
99.3, 105.6, 115.1, 117.6, 130.8, 132.7, 138.7, 159.5, 164.9, 165.0
ppm. HRMS (ES-TOF): m/z calcd. for C12H13N2O3 [M ϩ H]ϩ:
233.0926; found 233.0927.
Synthesis of the N-Acylimidazole Derivatives: Imidazole (2 mmol)
was dissolved in dry CH2Cl2 (1 mL), and a solution of acyl chloride
(1 mmol) in CH2Cl2 (1 mL) was slowly added to the stirred solu-
tion, resulting in the precipitation of imidazolium chloride. The
mixture was stirred for 2Ϫ3 h at room temperature. The mixture
was filtered, and the solution was rapidly washed with cold water
(2 ϫ 10 mL). The organic phase was dried with MgSO4 and fil-
tered, and the solvent was evaporated under reduced pressure,
yielding the crude N-acylimidazole. Some of the crude products
were successfully recrystallized from dichloromethane/n-hexane or
cyclohexane mixtures. In an alternative method, the parent car-
boxylic acid (1 mmol) and 1,1Ј-carbonylbis(1H-imidazole)
(1 mmol) were dissolved in anhydrous THF (2 mL) and stirred for
2Ϫ3 h. The THF was then evaporated, and the residue was dis-
solved in CH2Cl2 (10 mL), washed with water (2 ϫ 10 mL), and
dried with MgSO4, and the solvents were evaporated to dryness.
Compounds 1, 2, 5, 6, and 11Ϫ15 were isolated in crystalline form,
whereas compounds 3, 4, and 7Ϫ10 were recovered as viscous oils
and used without further purification. NMR analysis showed that
the only organic impurities were traces of hydrolysis products (im-
idazole and parent carboxylic acids, some of which may have
formed during NMR analysis) and the purity was Ͼ 99Ϫ99.5%
(down to 95%, 97.5%, and 98% for 7, 12, and 14, respectively).
Furthermore, the synthetic method used precludes the formation
of inorganic impurities. All N-acylimidazoles employed in this work
have previously been reported in the literature,[18,32Ϫ37] except for
compounds 3 and 12. Compounds were characterized by high-
1
N-(4-Methoxybenzoyl)imidazole (6): H NMR (400 MHz, CDCl3):
δ ϭ 3.92 (s, 3 H), 7.05 (d, J ϭ 8.9 Hz, 2 H), 7.18 (s, 1 H), 7.55 (t,
J ϭ 1.3 Hz, 1 H), 7.82 (d, J ϭ 8.9 Hz, 2 H), 8.10 (s, 1 H) ppm. 13
C
NMR (100 MHz, CDCl3): δ ϭ 56.1, 114.7, 118.7, 124.3, 131.1,
132.8, 138.6, 164.5, 165.9 ppm. HRMS (ES-TOF): m/z calcd. for
C11H11N2O2 [M ϩ H]ϩ: 203.0821; found 203.0827.
1
N-(2-Methoxybenzoyl)imidazole (7): H NMR (400 MHz, CDCl3):
δ ϭ 3.83 (s, 3 H), 7.06 (d, J ϭ 8.4 Hz, 1 H), 7.09Ϫ7.14 (m, 2 H),
7.49Ϫ7.45 (m, 2 H), 7.58 (td, J ϭ 8.1, J ϭ 1.6, 1 H), 8.16 (s, 1 H)
ppm. 13C NMR (100 MHz, CDCl3): δ ϭ 56.2, 112.1, 117.4, 121.3,
122.6, 130.3, 131.1, 134.0, 138.6, 157.3, 165.4 ppm. HRMS (ES-
TOF): m/z calcd. for C11H11N2O2 [M ϩ H]ϩ: 203.0821; found
203.0828.
1
N-(2-Iodobenzoyl)imidazole (8): H NMR (400 MHz, CDCl3): δ ϭ
7.17 (d, J ϭ 0.9 Hz, 1 H), 7.31 (td, J ϭ 7.8, J ϭ 1.7, 1 H),
7.43Ϫ7.45 (m, 2 H), 7.54 (t, J ϭ 7.5 Hz, 1 H), 7.85 (s, 1 H), 7.99
(d, J ϭ 8.0 Hz, 1 H) ppm. 13C NMR (100 MHz, CDCl3): δ ϭ 92.7,
117.5, 128.7, 129.2, 132.0, 133.1, 138.4, 139.0, 140.5, 166.3 ppm.
HRMS (ES-TOF): m/z calcd. for C10H8N2OI [M ϩ H]ϩ: 298.9681;
found 298.9669.
N-(2-Bromobenzoyl)imidazole (9): 1H NMR (400 MHz, CDCl3):
δ ϭ 7.17 (s, 1 H), 7.46 (s, 1 H), 7.48Ϫ7.54 (m, 3 H), 7.74 (d, J ϭ
8.0 Hz, 1 H), 7.87 (s, 1 H) ppm. 13C NMR (100 MHz, CDCl3): δ ϭ
117.3, 120.2, 128.1, 129.6, 132.0, 133.3, 134.1, 135.0, 138.3, 165.0
ppm. HRMS (ES-TOF): m/z calcd. for C10H8N2OBr [M ϩ H]ϩ:
250.9820; found 250.9817.
1
resolution mass spectrometry and H and 13C NMR, and showed
the expected spectra.
N-(2,4,6-Trimethylbenzoyl)imidazole (1): 1H NMR (400 MHz,
CDCl3): δ ϭ 2.19 (s, 6 H), 2.36 (s, 3 H), 6.96 (s, 2 H), 7.14 (s, 1
H), 7.54 (br. s, 1 H), 7.72 (br. s, 1 H) ppm. 13C NMR (100 MHz,
CDCl3): δ ϭ 19.5, 21.7, 116.5, 129.1, 131.0, 131.9, 135.2, 137.5,
141.1, 168.1 ppm. HRMS (ES-TOF): m/z calcd. for C13H15N2O [M
ϩ H]ϩ: 215.1184; found 215.1178.
N-Benzoylimidazole (10): 1H NMR (400 MHz, CDCl3): δ ϭ 7.17
(s, 1 H), 7.54 (s, 1 H), 7.56 (t, J ϭ 7.8 Hz, 2 H), 7.69 (t, J ϭ 7.5 Hz,
1 H), 7.80 (d, J ϭ 8.6 Hz, 2 H), 8.07 (s, 1 H) ppm. 13C NMR
(100 MHz, CDCl3): δ ϭ 118.4, 129.3, 130.4, 131.4, 132.3, 134.0,
138.6, 166.5 ppm. HRMS (ES-TOF): m/z calcd. for C10H9N2O [M
ϩ H]ϩ: 173.0715; found 173.0718.
N-(2,6-Dimethoxybenzoyl)imidazole (2): 1H NMR (400 MHz,
CDCl3): δ ϭ 3.79 (s, 6 H), 6.65 (d, J ϭ 8.4 Hz, 2 H), 7.09 (s, 1 H),
7.44 (t, J ϭ 8.5 Hz, 1 H), 7.47 (s, 1 H), 7.83 (s, 1 H) ppm. 13C
NMR (100 MHz, CDCl3): δ ϭ 56.4, 104.4, 112.0, 116.9, 131.2,
133.3, 138.3, 158.1, 163.6 ppm. HRMS (ES-TOF): m/z calcd. for
C12H13N2O3 [M ϩ H]ϩ: 233.0926; found 233.0930.
N-(4-Bromobenzoyl)imidazole (11): 1H NMR (400 MHz, CDCl3):
δ ϭ 7.20 (s, 1 H), 7.54 (s, 1 H), 7.70 (d, J ϭ 8.6 Hz, 2 H), 7.75 (d,
J ϭ 8.6 Hz, 2 H), 8.08 (s, 1 H) ppm. 13C NMR (100 MHz, CDCl3):
δ ϭ 118.4, 129.3, 131.1, 131.6, 132.8, 138.4, 165.7 ppm. HRMS
(ES-TOF): m/z calcd. for C10H8N2OBr [M ϩ H]ϩ: 250.9820;
found 250.9827.
N-(2,4-Dimethylbenzoyl)imidazole (3): 1H NMR (400 MHz,
CDCl3): δ ϭ 2.38 (s, 3 H), 2.41 (s, 3 H), 7.11Ϫ7.13 (m, 2 H), 7.17
1
N-(1-Naphthoyl)imidazole (12): H NMR (400 MHz, CDCl3): δ ϭ
(s, 1 H), 7.31 (d, J ϭ 7.8 Hz, 1 H), 7.46 (s, 1 H), 7.90 (s, 1 H) ppm. 7.19 (s, 1 H), 7.57 (s, 1 H), 7.60Ϫ7.64 (m, 3 H), 7.73 (dd, J ϭ 7.0,
13C NMR (100 MHz, CDCl3): δ ϭ 19.9, 21.8, 117.8, 126.8, 129.0, J ϭ 0.7, 1 H), 7.97Ϫ8.03 (m, 3 H), 8.13 (d, J ϭ 8.2 Hz, 1 H) ppm.
Eur. J. Org. Chem. 2002, 2633Ϫ2639
2637

S. Zaramella, R. Strömberg, E. Yeheskiely
FULL PAPER
13C NMR (100 MHz, CDCl3): δ ϭ 118.1, 124.8, 125.0, 127.6,
were withdrawn at time intervals and diluted in 50% CH3CN in 0.1
128.3, 128.6, 129.1, 130.1, 130.8, 131.6, 133.3, 134.0, 138.7, 166.5  triethylammonium acetate at pH ϭ 6.5 (1 mL). The pH was ad-
ppm. HRMS (ES-TOF): m/z calcd. for C14H11N2O [M ϩ H]ϩ:
223.0871; found 223.0878.
justed to 6.5 by addition of diluted acetic acid. Aliquots were ana-
lyzed by reversed-phase HPLC. The extent of substrate cleavage
at a given time was determined from the ratio of products to N-
acylimidazole. Differences in the extinction coefficients were cor-
rected for by calibration with carboxylic acid/N-acylimidazole mix-
tures of known composition. Minor extinction coefficient differ-
ences between amide, ester, and acid were corrected by using the
values obtained from benzamide, ethyl benzoate, and benzoic acid.
1
N-(2-Naphthoyl)imidazole (13): H NMR (400 MHz, CDCl3): δ ϭ
7.23 (s, 1 H), 7.64 (s, 1 H), 7.65Ϫ7.72 (m, 2 H), 7.87 (dd, J ϭ 8.6,
J ϭ 1.6, 1 H), 7.97 (d, J ϭ 8.1 Hz, 1 H), 7.99 (d, J ϭ 8.1 Hz, 1 H),
8.04 (d, J ϭ 8.6 Hz, 1 H), 8.18 (s, 1 H), 8.34 (s, 1 H) ppm. 13C
NMR (100 MHz, CDCl3): δ ϭ 118.6, 125.5, 128.0, 128.4, 129.5,
129.6, 129.7, 131.4, 131.9, 132.5, 135.9, 138.8, 166.7 ppm. HRMS
(ES-TOF): m/z calcd. for C14H11N2O [M ϩ H]ϩ: 223.0871; found
223.0872.
Acknowledgments
We thank the Swedish Natural Science Research Council and the
Swedish Research Council for Engineering Sciences for the finan-
cial support.
N-Pivaloylimidazole (14): 1H NMR (400 MHz, CDCl3): δ ϭ 1.46
(s, 9 H), 7.06 (d, J ϭ 0.8 Hz, 1 H), 7.57 (t, J ϭ 1.4 Hz, 1 H), 8.29
(s, 1 H) ppm. 13C NMR (100 MHz, CDCl3): δ ϭ 28.4, 41.4, 117.9,
130.3, 137.7, 175.4 ppm. HRMS (ES-TOF): m/z calcd. for
C8H13N2O [M ϩ H]ϩ: 153.1028; found 153.1027.
N-(1-Adamantylcarbonyl)imidazole (15): 1H NMR (400 MHz,
CDCl3): δ ϭ 1.77Ϫ1.85 (m, 6 H), 2.14Ϫ2.16 (m, 9 H), 7.06 (s, 1
H), 7.64 (s, 1 H), 8.38 (s, 1 H) ppm. 13C NMR (100 MHz, CDCl3):
δ ϭ 28.3, 36.7, 39.6, 44.2, 117.9, 130.1, 137.6, 175.0 ppm. HRMS
(ES-TOF): m/z calcd. for C14H9N2O [M ϩ H]ϩ: 231.1497; found
231.1494.
[1]
[1a] G. J. Moore, Trends Pharmacol. Sci. 1994, 15, 124Ϫ129. [1b]
G. L. Olson, D. R. Bolin, M. P. Bonner, M. Bös, C. M. Cook,
D. C. Fry, B. J. Graves, M. Hatada, D. E. Hill, M. Kahn, V. S.
Madison, V. K. Rusiecki, R. Sarabu, J. Sepinwall, G. P. Vinc-
ent, M. E. Voss, J. Med. Chem. 1993, 36, 3039Ϫ3049.
[2]
´
M. Beltran, E. Pedroso, A. Grandas, Tetrahedron Lett. 1998,
39, 4115Ϫ4118.
General Procedures for Stability Studies, Demonstrated by Experi-
ments with Compound 1
[3]
[4]
[5]
[6]
C.-H. Tung, S. Stein, Bioconjugate Chem. 2000, 11, 605Ϫ618,
and references cited therein.
B. N. Trawick, A. T. Daniher, J. K. Bashkin, Chem. Rev. 1998,
98, 939Ϫ960, and references cited therein.
T. H. Fife, Acc. Chem. Res. 1993, 26, 325Ϫ331, and references
cited therein.
Studies in 3% (v/v) TFA in CDCl3: A stock solution of 1 was pre-
pared by dissolving the compound (14.3 mg) in CDCl3 (0.5 mL),
in a vial sealed with a septum. The stock solution was stored over
˚
4 A molecular sieves at 4 °C under nitrogen. An aliquot of the
´
R. Giege, B. Felden, M. A. Zenkova, V. N. Sil’nikov, V. V. Vlas-
stock solution (50 µL) was withdrawn and transferred to an NMR
tube, previously dried in an oven. CDCl3 (450 µL) was added, and
the first spectrum was recorded. Distilled TFA (15.5 µL) was
rapidly added, and the tube was tightly sealed with Parafilm.
When out of the spectrometer, the sample was kept in a water bath
at 25 °C.
sov, Methods Enzymol. 2000, 318, 147Ϫ165.
M. Septak, Nucleic Acids Res. 1996, 24, 3053Ϫ3058.
D. Filippov, E. Kuyl-Yeheskiely, G. A. van der Marel, G. I.
Tesser, J. H. van Boom, Tetrahedron Lett. 1998, 39, 3597Ϫ3600.
M. Antopolsky, E. Azhayeva, U. Tengvall, A. Azhayev, Tetra-
hedron Lett. 2002, 43, 527Ϫ530.
P. Sieber, B. Riniker, Tetrahedron Lett. 1987, 28, 6031Ϫ6034.
V. du Vigneaud, O. K. Behrens, J. Biol. Chem. 1937, 117,
27Ϫ36.
T. Brown, J. H. Jones, J. D. Richards, J. Chem. Soc., Perkin
Trans. 1 1982, 1553Ϫ1561.
R. Colombo, F. Colombo, J. H. Jones, J. Chem. Soc., Chem.
Commun. 1984, 292Ϫ293.
S. Shaltiel, M. Fridkin, Biochemistry 1970, 9, 5122Ϫ5127.
A. Patchornik, A. Berger, E. Katchalski, J. Am. Chem. Soc.
1957, 79, 6416Ϫ6420.
E. Schnabel, H. Herzog, P. Hoffmann, E. Klauke, I. Ugi, Justus
Liebigs Ann. Chem. 1968, 716, 175Ϫ185.
T. Fujii, T. Kimura, S. Sakakibara, Bull. Chem. Soc. Jpn. 1976,
49, 1595Ϫ1601.
H. A. Staab, Angew. Chem. Int. Ed. Engl. 1962, 1, 351Ϫ367,
and references cited therein.
H. A. Staab, Chem. Ber. 1956, 89, 2088Ϫ2093.
T. H. Fife, J. Am. Chem. Soc. 1965, 87, 4597Ϫ4600.
J. A. Fee, T. H. Fife, J. Org. Chem. 1966, 31, 2343Ϫ2346.
T. H. Fife, R. Natarajan, M. Werner, J. Org. Chem. 1987, 52,
740Ϫ746, and references cited therein.
A. R. Emery, V. Gold, J. Chem. Soc. 1950, 1455Ϫ1460.
C. A. Bunton, N. A. Fuller, S. G. Perry, V. J. Shiner, J. Chem.
Soc. 1963, 2918Ϫ2926.
F. Mueller, P. Speiser, Pharm. Acta Helv. 1965, 40, 449Ϫ466.
H. J. Mooij, J. B. F. N. Engberts, M. Charton, Recl. Trav. Chim.
Pays-Bas 1988, 107, 185Ϫ189.
[7]
[8]
[9]
[10]
[11]
Studies in 2% (v/v) DBU in [D7]DMF: A stock solution of 1 was
prepared by dissolving the substrate (14.3 mg) in deuterated DMF
(0.5 mL), in a vial provided with a septum. The solution was stored
[12]
[13]
˚
over 4 A molecular sieves at 4 °C under nitrogen. An aliquot of
the stock solution (50 µL) was withdrawn and transferred to an
NMR tube, previously dried in an oven, and deuterated DMF (450
µL) was then added. The first spectrum was recorded. DBU (10.2
µL) was then rapidly added, and the tube was tightly sealed with
Parafilm. When out of the spectrometer, the sample was kept in
a water bath at 25 °C.
[14]
[15]
[16]
[17]
[18]
Studies in 2% (v/v) DBU/2% (v/v) Piperidine in [D7]DMF: A stock
solution of 1 was prepared by dissolving the substrate (14.3 mg) in
deuterated DMF (0.5 mL), in a vial provided with a septum. The
[19]
[20]
[21]
[22]
˚
solution was stored over 4 A molecular sieves at 4 °C under nitro-
gen. An aliquot of the stock solution (50 µL) was withdrawn and
transferred to an NMR tube, previously dried in an oven. Deuter-
ated DMF (450 µL) was added, and the first spectrum was re-
corded. DBU (10.2 µL) and piperidine (10.2 µL) were rapidly ad-
ded, and the tube was tightly sealed with Parafilm. When out of
the spectrometer, the sample was kept in a water bath at 25 °C.
[23]
[24]
[25]
[26]
Studies in 3:1 (v/v) Aqueous Ammonia/Ethanol: Compound 1
(4.3 mg) was dissolved in a mixture of aq. ammonia and ethanol
(3:1, v/v; 0.5 mL total volume) in a tightly sealed vial, and left at
room temperature with occasional swirling. Aliquots of 20 µL each
[27]
E. Westman, S. Sigurdsson, J. Stawinski, R. Strömberg, Nucleic
Acids Symp. Ser. 1994, 31, 25Ϫ26.
2638
Eur. J. Org. Chem. 2002, 2633Ϫ2639

Stability Studies of N-Acylimidazoles
FULL PAPER
R. S. Alphin, W. L. Smith, H. R. Munson Jr., R. F. Boswell,
U. S. Patent 4717563, 1987.
S. Ohta, M. Okamoto, Tetrahedron Lett. 1981, 22, 3245Ϫ3246.
R. H. Seevers, S. W. Schwedner, S. L. Swayze, R. E. Counsell,
J. Med. Chem. 1982, 25, 618Ϫ621.
[28]
[33]
T. Atkinson, M. Smith in Oligonucleotide Synthesis: a Practical
Approach (Ed.: M. J. Gait), IRL Press, Oxford, UK, 1984,
[34]
[35]
pp. 35Ϫ81.
[29]
J. D. Wade, J. Bedford, R. C. Sheppard, G. W. Tregear, Peptide
Res. 1991, 4, 194Ϫ199.
[30]
[36]
[37]
´
S. A. Kates, N. A. Sole, M. Beyermann, G. Barany, F. Alber-
N. Ikemoto, L.-C. Lo, K. Nakanishi, Angew. Chem. 1992, 104,
918Ϫ919; Angew. Chem. Int. Ed. Engl. 1992, 31, 890Ϫ891.
T. Kitagawa, H. Kuroda, H. Sasaki, K. Kawasaki, Chem.
Pharm. Bull. 1987, 35, 4294Ϫ4301.
icio, Peptide Res. 1996, 9, 106Ϫ113.
Model 433A Peptide Synthesizer User Manual, Applied Biosy-
stems, 1993, vol. 1.
H. A. Staab, W. Rohr, F. Graf, Chem. Ber. 1965, 98,
1122Ϫ1127.
[31]
[32]
Received December 18, 2001
[O01581]
Eur. J. Org. Chem. 2002, 2633Ϫ2639
2639