Ring-deactivated hydroxyalkylpyrrole-based inhibitors of α-chymotrypsin: Synthesis and mechanism of action

Article abstract of DOI:10.1039/b302411c

¹³C NMR and mass spectrometry studies have been used to demonstrate that the inhibition of α-chymotrypsin by N-sulfonylhydroxymethylpyrrole inhibitors (10) is non-covalent. Hydroxyalkylpyrroles in which an electron-withdrawing group (acyl substituent) is introduced at the alternative C2 position have been synthesised and also shown to inactivate α-chymotrypsin. SAR studies on this class suggests that the incorporation of phenylalanine at C2 is favoured, however, there is little gain in introducing a hydrophobic substituent at C5.

Full text of DOI:10.1039/b302411c

View Article Online / Journal Homepage / Table of Contents for this issue
Ring-deactivated hydroxyalkylpyrrole-based inhibitors of
ꢀ-chymotrypsin: synthesis and mechanism of action
Derek C. Martyn, Andrea J. Vernall, Bruce M. Clark and Andrew D. Abell*
Department of Chemistry, University of Canterbury, Christchurch, New Zealand.
E-mail: a.abell@chem.canterbury.ac.nz; Fax: 64-3-3642110; Tel: 64-3-3642818
Received 5th March 2003, Accepted 12th May 2003
First published as an Advance Article on the web 27th May 2003
13C NMR and mass spectrometry studies have been used to demonstrate that the inhibition of α-chymotrypsin
by N-sulfonylhydroxymethylpyrrole inhibitors (10) is non-covalent. Hydroxyalkylpyrroles in which an electron-
withdrawing group (acyl substituent) is introduced at the alternative C2 position have been synthesised and also
shown to inactivate α-chymotrypsin. SAR studies on this class suggests that the incorporation of phenylalanine at
C2 is favoured, however, there is little gain in introducing a hydrophobic substituent at C5.
Introduction
In previous work1 we reported a series of hydroxymethyl-
pyrroles substituted on nitrogen with an electron withdrawing
group (EWG) (see 1 in Scheme 1) as a new class of inhibitor of
α-chymotrypsin, a representative example of a class of enzyme
known as serine proteases.2 These compounds were postulated
to act as mechanism-based inhibitors based upon model
hydrolysis studies3 and extensive chemical precedence.4,5 In par-
ticular, hydrolytic removal of the EWG from 1, in a process that
mimics normal protease action on a substrate peptide, allows
for the liberation of a highly electrophilic azafulvene (3) that is
free to react with an available nucleophile (Scheme 1).3 We pro-
posed that this process could result in enzyme inactivation if it
occurred in the vicinity of a catalytic group in the enzyme active
site.1
Fig. 1 Schematic representation of two conformations of a natural
substrate (middle) bound to the active site of α-chymotrypsin, and how
these might correlate with the proposed N-acylhydroxymethylpyrrole
5 (top) and the C2-acylhydroxyalkylpyrrole 6 (bottom) inhibitors.
The point of cleavage of the natural substrate is indicated by the bold
arrow. Note the use of Schechter–Berger nomenclature.†
alkylpyrroles extended in the C-direction with an amino acid or
dipeptide that also inhibit α-chymotrypsin by a non-covalent
mechanism. These compounds have the EWG at C2 and as such
mimic an alternative conformation of a natural substrate
(see Fig. 1).
Scheme 1
Peptidomimetics 1 can be considered to be chemically stable
analogues of a dipeptide substrate of α-chymotrypsin where the
EWG mimics an amino acid on the N-terminus of the substrate
scissile bond as depicted in 5, Fig. 1. In support of this notion
we found1 that the best inhibitors of type 1 contained a hydro-
phobic EWG that mimics the aromatic P1 group (see Fig. 1)† of
a natural substrate of α-chymotrypsin, i.e. the point of natural
cleavage of a peptide by α-chymotrypsin.6 We now report
results of 13C NMR and mass spectrometric studies, using
α-chymotrypsin and labelled inhibitor, that suggest that the
mechanism of inhibition is in fact non-covalent in nature. We
also report a new but related series of C2-acylated hydroxy-
Results and discussion
We chose to use N-phenylsulfonylpyrrole 10a (see Scheme 2) for
the mechanistic studies since it is the most potent inhibitor of
this type identified to date and also because it is readily avail-
able.1 Initial 1H NMR studies on 10a, whereby it was incubated
in CDCl3 at room temperature for 7 days without degradation,
illustrated that this compound is relatively stable and hence
suitable for mechanistic studies. In addition, we have demon-
strated that an N-sulfonyl group, of the type found in 10a, is the
strongest ring-deactivating group of those studied, and as such
it is the best at suppressing azafulvene formation.5
In the first series of experiments we decided to incorporate
† Note the use of Schechter–Berger nomenclature22 – the residues on
the N-terminal side of the peptide bond that is cleaved are denoted (in
order) P1–Pn, and those on the C-terminus are denoted P1Ј–PnЈ. In turn,
the corresponding subsites on the enzyme are denoted Sn–SnЈ
a
13C-label into 10a as a marker for monitoring changes to
the electronic environment of the hydroxymethyl carbon on
exposure to α-chymotrypsin. The target 13C labelled compound
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 0 3 – 2 1 1 0
2103

View Article Online
Scheme 2 Reagents and conditions: (i) POCl3, 13C-DMF, 1,2-DCE,
reflux, 15 min, then CH3CO2Naؒ3H2O(aq), reflux, 15 min (59%);
(ii) PhSO2Cl, 30% NaOH(aq)
, Bu4NI, CH2Cl2, rt, 30 h (91%);
(iii) Zn(BH4)2, Et2O, 0 ЊC, 30 min (90%).
10b was prepared using the method previously reported for
10a1 (Scheme 2) and a modification reported by Hinz et al.7 In
particular, pyrrole was formylated with 13C–N,N-dimethyl-
formamide under Vilismeier–Haack conditions8 to give 8. This
was reacted with phenylsulfonyl chloride under phase transfer
catalysis7 to give 9, which was subsequently reduced with zinc
borohydride5 to give the desired labelled derivative 10b.
Scheme 3 Reagents and conditions: (i) POCl3, DMF, 1,2-DCE, reflux,
15 min, then CH3CO2Naؒ3H2O, reflux, 15 min (61%); (ii) KOH, H2O,
40–50 ЊC, 2 h (60%); (iii) EDCI, HOBT, DIEA, CH2Cl2, rt, 16 h and
either –PheOMeؒHCl (13a, 71%), –LeuOMeؒHCl (13b, 61%), or
–ProOMeؒHCl (13c, 74%); (vi) Zn(BH4)2, ether, 0 ЊC, 1 h (14a, 57%),
(14b, 42%), (14c, 24%); (v) LiBH4, ether, Ϫ78 ЊC, 1 h (14b, 66%).
An analysis of the 13C NMR spectrum of a sample of 10b
incubated with α-chymotrypsin, using conditions previously
reported for a different class of inhibitor,9 revealed an enriched
resonance at 56 ppm, the same chemical shift as was observed
for 10b in the absence of enzyme.‡ In addition, an electrospray
mass spectrum of the product from the incubation of 10a with
α-chymotrypsin revealed signals at 25,236 g molϪ1 (corre-
sponding to the native enzyme), and 25,473 g molϪ1, a differ-
ence of 237 g molϪ1 which corresponds to the molecular mass
of intact 10a. These results strongly suggest that inhibitors 10
are not covalently attached to α-chymotrypsin as previously
postulated.1
ate 12b in 60% yield.14 The free acid of 12b was then separately
coupled with -phenylalanine methyl ester hydrochloride,
-leucine methyl ester hydrochloride and -proline methyl ester
hydrochloride, using standard EDCI conditions,15 to give 13a,b
and c respectively. Reduction of the 5-formyl moiety of 13a–c
with zinc borohydride gave the desired dipeptide analogues
14a–c in reasonable yields. The use of lithium borohydride,
instead of zinc borohydride, for the reduction of 13b gave an
improvement in yield from 42% to 66%.
Next we prepared compounds 18a (Scheme 4) in which R1
(see general structure 6, Fig. 1) more closely correlates to the
known specificity of the P1 group of a substrate of α–chymo-
trypsin (see Fig. 1). We also prepared compounds 18b in which
R1 (Me) is of intermediary size, cf. H and (CH2)2Ph of 14a–c
and 18a, respectively. To this end, the pyrrole 11 was acylated
with either hydrocinnamoyl chloride or acetyl chloride16 to give
15a and 15b, respectively. Hydrolysis of the ethyl esters of
15a and 15b with potassium hydroxide gave the acids 16a and
16b which were then separately coupled with -phenylalanine
methyl ester hydrochloride (as for 12b) to give 17a and 17b. A
final reduction of the 5-keto group of 17a and 17b with zinc
We next decided to investigate the effect of introducing an
EWG at the alternative C2-position of a hydroxymethylpyrrole
(see general structure 6 in Fig. 1). Unlike structures of type 1
and 5, these derivatives do not have the EWG directly attached
to the pyrrole nitrogen and as such they mimic an extended
conformation of a peptide backbone as depicted in Fig. 1. We
reasoned that, while a C2-acyl group as in 6 would still deactiv-
ate the pyrrole, it would do so to a lesser extent. As such azaful-
vene formation would be more favoured, which should in turn
promote inactivation of α-chymotrypsin by a covalent mechan-
ism. It is also important to note that compounds of type 6 allow
for chain extension in the C-direction and also the incorpor-
ation of a range of groups in the P1 position. A series of such
compounds was prepared and assayed against α-chymotrypsin.
Scheme 3 details the synthesis of compounds of type 6 that
bear a hydrogen at R1 (P1 group as defined for a substrate in
Fig. 1). Three derivatives (14a–c) were prepared to investigate
the effect of the nature of the appended amino acid on inhibi-
tory potency. Phenylalanine was used (14a) as it is known that
the S2Ј subsite of α–chymotrypsin preferentially accommodates
large, hydrophobic moieties of this type (see Fig. 1).10,11 Leucine
and proline were chosen in order to test this hypothesis in our
series and to gain some possible insight into the mode of
enzyme binding. The synthesis of compounds 14 began with
the formylation of the pyrrole ester 1112 under Vilismeier–
Haack conditions13 to give 12a, which would eventually give
rise to the hydroxymethyl side chain of 14a–c. Hydrolysis of the
ester of 12a with potassium hydroxide gave the key intermedi-
Scheme 4 Reagents and conditions: (i) ZnCl2, 1,2-DCE, 50 ЊC, 1 h and
Ph(CH2)2COCl (15a, 35%) or AcCl (15b, 37%); (ii) KOH, 2 : 1 H2O–
THF, 40–50 ЊC, 2 h (16a, 86%), (16b, 93%); (iii) –PheOMeؒHCl,
EDCI, HOBT, DIEA, CH2Cl2, rt, 16 h (17a, 70%), (17b, 91%);
(iv) Zn(BH4)2, ether, 0 ЊC, 1 h (18a, 21%), (18c); (v) LiBH4, ether,
Ϫ78 ЊC, 1 h, then rt, 1 h (18b, 46%).
‡ A difference spectrum was determined by subtracting the 13C NMR
spectrum of 10a (unlabelled) with α-chymotrypsin from that of 10b
(labelled) with α-chymotrypsin according to the published method.9
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 0 3 – 2 1 1 0
2104

View Article Online
Table 1 Inhibition of α-chymotrypsin by 14a–c, 18a and 22
% Inhibition (µg mLϪ1
)
Compound
R1
X
1250
125
12.5
14a
14b
14c
18a
22
H
H
H
NH––PheOMe
NH––LeuOMe
N––ProOMe
NH––PheOMe
NH––Phe––LeuOMe
75
10
45
75
100
35
5
25
35
20
45
20a
5
15
15
Ph(CH2)2
H
a This apparent anomalous result where greater inhibition is observed at lower concentration is under further investigation.
borohydride gave the extended peptide analogues 18a and 18b
as 1 : 1 mixtures of epimers by 1H NMR. An over reduction was
observed in the case of 17a, with the production of an almost
equal amount of by-product 18c. Compound 18b proved to be
somewhat unstable and its purification problematic. An altern-
ative preparation of 18b was also investigated as outlined
in Scheme 5. Here, the formyl group of pyrrole 12a was
reacted with methyllithium to give 19a. Subsequent ethyl ester
hydrolysis with sodium hydroxide gave 19b in low yield, which
was then coupled to -phenylalanine methyl ester hydrochloride
to give 18b, which again proved difficult to purify and was
ultimately unstable.
presence of BOP19 to give 21 in 52% yield over two steps.
Reduction with lithium borohydride then gave 22.
The peptidomimetics 14a–c, 18a (as a mixture of dia-
stereomers) and 22 were assayed against α–chymotrypsin at
three concentrations using a simple screening assay as described
previously1 and the results are presented in Table 1 (compound
18b proved too unstable to assay). From these results it is
apparent that there is not a significant difference in the inhibitor
potency of compounds 14a and 18a, compounds that differ
solely in the nature of R1 (see 6 in Fig. 1). We had postulated
that the hydrophobic group of 18a might be favoured (see
earlier for a discussion). However, it is important to remember
that 18a was assayed as a mixture of diastereomers and as such
one isomer may be more active than the other. A comparison of
the inhibitory assay results obtained for the four compounds
(14a–c and 22) that contain a hydroxymethylpyrrole moiety
suggest a slight preference for a phenylalanine in the appended
X group. The tetrapeptide analogue 22 proved the most potent
and an IC50 value of 108 µM was determined for this com-
pound. LCMS studies were also performed on α-chymotrypsin
incubated with 14a at 37 ЊC for 3 hours in 50 mM Tris–HCl. As
for 10, there was no evidence of covalent attachment of the
inhibitor to the enzyme. Finally, it should be noted that 2-acyl-
hydroxyalkylpyrroles of general structure 6 (see 14a–c, 18a and
22 in Table 1) are similarly active to the compounds in which
the EWG is substituted on nitrogen (see 1 and 5).1
Scheme 5 Reagents and conditions: (i) MeLi, THF, Ϫ23 ЊC, 5 h (95%);
(ii) NaOH, 1 : 1 THF–H2O, 40–50 ЊC, 24 h (17%); (iii) –PheOMeؒHCl,
EDCI, HOBT, DIEA, CH2Cl2, rt, 16 h (54%).
Finally, we prepared an extended tetrapeptide analogue 22
based upon the simple hydroxymethylpyrrole scaffold of com-
pounds 14a–c. We chose to extend the peptide in the C-direc-
tion using -phenylalanyl--leucine since this sequence is
known to favour binding to α-chymotrypsin.17,18 This sequence
also allows for a direct comparison with 14a, which simply has
-phenylalanine attached. The required starting dipeptide ester
20b was prepared from commercially available -phenylalanyl-
-leucine hydrate 20a on treatment with diazomethane (Scheme
6). This dipeptide methyl ester was then coupled to 12b in the
Conclusion
N-Substituted hydroxymethylpyrroles of general type 1 are
known to inhibit α-chymotrypsin. In this paper we have used a
combination of mass spectrometry and 13C NMR experiments
to show that compounds of type 1 (specifically 10) do not form
a covalent attachment to the enzyme. A modified series of com-
pounds was then prepared in which a peptidic side chain is
attached to the pyrrole nucleus at C2. Assay results for this
series reveal that, while compounds with a phenylalanine resi-
due as part of the C2 side chain (X in Table 1) are the most
potent inhibitors of α-chymotrypsin, there is little gain in intro-
ducing a more hydrophobic substituent at C5 (R1 in Table 1).
The IC50 value for the most potent compound (22) in this new
series of inhibitor of α-chymotrypsin was determined to be 108
µM. Again it appears that these compounds inhibit α-chymo-
trypsin via a non-covalent mechanism.
Experimental
All melting points were obtained on an Electrothermal appar-
atus and are uncorrected. Proton NMR spectra were obtained
on a Varian Inova spectrometer, operating at 500 MHz. Carbon
NMR were obtained on a Varian Unity 300 spectrometer, oper-
ating at 75 MHz. IR spectra were obtained using a Shimadzu
8201PC series FTIR, either in CHCl3 (solution phase), or in
Scheme 6 Reagents and conditions: (i) CH2N2, THF, rt, 15 h; (ii) 12b,
BOP, DIEA, CH2Cl2, rt, 16 h (52% yield over two steps); (iii) LiBH4,
THF, Ϫ78 ЊC, 1 h, then rt, 1 h (66%).
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 0 3 – 2 1 1 0
2105

View Article Online
solid KBr (diffuse reflectance). Molecular masses of products
were detected on a Kratos MS80 RFA mass spectrometer.
Inhibitor–enzyme samples were analysed initially by electro-
spray ionisation mass spectrometry (ESI) using a Micromass
LCT mass spectrometer scanning at 100–2500 atomic mass
units (scans every 0.9 s, 0.1 s delay). The source was at 80 ЊC
(ESI probe = 150 ЊC, nebuliser = 190 L hϪ1, desolvation = 350 L
General procedure A: peptide couplings using EDCI
To a stirred solution (∼0.1 M) of the pyrrole carboxylic acid
(1 equiv.) and the -amino acid ester hydrochloride (1.1 equiv.)
in dry dichloromethane (∼8 cm3), at rt under an inert atmos-
phere, were added 1-(3-dimethylaminopropyl)-3-ethylcarbo-
diimide hydrochloride (1.3 equiv.) and 1-hydroxybenzotriazole
hydrate (1.5 equiv.). N,N-Diisopropylethylamine (1.1 equiv.)
was added, and the reaction mixture was stirred at rt for 16 h.
The solution was then diluted with dichloromethane (10 cm3),
washed with 3 M aqueous hydrochloric acid (2 × 10 cm3), water
(2 × 10 cm3), dried (MgSO4) and the solvent was removed by
evaporation under reduced pressure. The resultant material
was purified by flash chromatography on silica. See individual
experiments for details.
h
Ϫ1). Liquid chromatography mass spectrometry (LCMS) was
carried out on a Waters 2790 LC coupled to the Micromass
LCT spectrometer using a Waters 996 PDA Zorbax C3 column
(150 mm × 2 mm × 5 µm, flow 0.2 cm3 minϪ1). The mobile
phase consisted of a gradient of 100% [water–0.5% formic acid]
to 40% [water–0.5% formic acid]–60% acetonitrile over 80 min,
then to 100% acetonitrile over 5 min, held at 100% acetonitrile
for 10 min, then back to 100% [water–0.5% formic acid] over 5
min. Data was processed with Maxent® software. α-Chymot-
rypsin assays were carried out in microtitre plates (twelve 1 × 8
well Titertek® microtitre strips) and read in a BMG Labtech-
nologies Fluostar Galaxy 96-well plate reader. Compounds
11,12 12a13 and 12b14 were prepared by literature methods.
Flash chromatography was performed using 230–400 mesh
Merck Silica Gel 60 under positive pressure. All solvents were
distilled prior to use using literature procedures.20
Modification to general procedure A
The reaction was performed in the manner as described in gen-
eral procedure A, except that the combined aqueous washings
were back-extracted with dichloromethane (2 × 10 cm3), and
the combined organic fractions were then dried (MgSO4). The
solvent was removed by evaporation under reduced pressure,
and the resultant material was purified by flash chromato-
graphy on silica. See individual experiments for details.
ꢀ-Chymotrypsin inhibition experiments
General procedure B: zinc borohydride reductions
Assay21. Solutions of 14a–c, 18a and 22 were made to 1250
µg cm3, 125 µg cm3 and 12.5 µg cm3 in methanol. Tris–HCl (50
µL of a 0.4 M solution in water, pH 7.6), distilled water (50 µL),
inhibitor solution (50 µL) and α-chymotrypsin (50 µL, Sigma
ex-bovine pancreas, 9 units cm3 in 50 mM Tris–HCl, pH 7.6)
were added to each well of the microtitre plate. Incubation at
37 ЊC for 30 min was followed by the addition of N-succinyl--
phenylalanine-4-nitroanilide (100 µL, 1 mg per cm3 solution in
50 mM Tris–HCl buffer, pH 7.6). The absorbance was read at
405 nm at hourly intervals until the maximum measured
absorbance exceeded 1.00 absorbance units. Each inhibitor
concentration was assayed in triplicate and average absorbances
were used to calculate the percentage (%) inhibition. Sample
blanks to determine the absorbance for each inhibitor concen-
tration in the absence of enzyme (i.e. no substrate turnover)
were run concurrently, in which Tris–HCl (50 µL of a 50 mM
solution in water, pH 7.6), replaced α-chymotrypsin. Maximum
substrate turnover was determined by averaging the absorbance
readings from six wells in which methanol (50 µL) replaced the
inhibitor solution. All percentage (%) inhibition figures
reported are rounded to the nearest 5%. The IC50 for 22 was
determined using the same general method except that solu-
tions of 22 were made to 12.0 mM, 3.01 mM, 0.75 mM, 0.19
mM, 47.0 µM, 11.8 µM, 2.94 µM and 0.73 µM. Precipitation
occurred in all wells where 22 was present at a concentration of
12.0 mM, so no inhibition value could be calculated at this
concentration. Each sample was assayed in triplicate. The IC50
value was calculated by fitting the inhibitor concentration–
percent inhibition data into Microsoft Excel.
A stirred solution (∼0.1 M) of the pyrrole (1 equiv.), dissolved
in dry ether under an inert atmosphere, was cooled to 0 ЊC. Zinc
borohydride (2.2 equiv. of a 0.14 M solution in ether) was
added, and the resultant solution was stirred at 0 ЊC for 1 h.
Water (1 cm3), and then 10% aqueous acetic acid (1 cm3), were
carefully added to quench the reaction. The separated aqueous
phase was extracted with dichloromethane (2 × 10 cm3), and the
combined organic phases were washed with water (2 × 10 cm3),
saturated aqueous brine (10 cm3), dried (MgSO4) and the sol-
vent was removed by evaporation under reduced pressure. The
resultant material was purified by flash chromatography on
silica. See individual experiments for details.
[Formyl-13C]-1H-pyrrole-2-carboxaldehyde (8). Prepared
from 7 (210 mg, 3.13 mmol) using phosphorous oxychloride
(526 mg, 3.43 mmol) and N,N-dimethylformamide-carbonyl-
13C (255 mg, 3.43 mmol) as described for unlabelled 88 (177 mg,
59%); δH(500 MHz; CDCl3; Me4Si) 6.36 (1 H, m, pyrrole
4-H), 7.02 (1 H, m, pyrrole 3-H), 7.18 (1 H, m, pyrrole
5-H), 9.51 (1 H, m, J = 175.3 Hz, CHO), 10.39 (1 H, br s, NH );
δC(75 MHz; CDCl3; Me4Si) 111.3, 121.8, 126.9, 132.7, 179.4
(CHO enriched peak).
[Formyl-13C]-1-(phenylsulfonyl)pyrrole-2-carboxaldehyde (9).
Prepared from 8 (110 mg, 1.14 mmol) as described for
unlabelled 97 (212 mg, 91%); δH(500 MHz; CDCl3; Me4Si) 6.42
(1 H, t, J = 3.4 Hz, pyrrole 4-H), 7.16 (1 H, m, pyrrole 3-H),
7.54–7.64 (4 H, m, pyrrole 5-H and ArH ), 7.92 (2 H, m, ArH ),
9.95 (1 H, d, J = 182.6 Hz, CHO); δC(75 MHz; CDCl3; Me4Si)
112.5, 124.8, 127.4, 129.5, 133.5, 134.5, 138.1, 178.7 (CHO
enriched peak).
13C NMR and LCMS. Stock solution A was made up by
dissolving 10b (2 mg) in DMSO (225 µL), followed by addition
of D2O (2.775 cm3) to give a final concentration of 2.8 mM.
Stock solution B was made up by dissolving α-chymotrypsin
(71 mg, Sigma ex-bovine pancreas) in DMSO (75 µL) followed
by addition of D2O (925 µL) to also give a concentration of 2.8
mM. Stock solution C was made up as for A using 10a in place
of 10b. Solution A (300 µL) and separate mixtures of A and B
(300 µL, 1 : 1) and B and C (300 µL, 1 : 1) were then left at room
temperature for one hour, shaking occasionally. The samples
were transferred to NMR tubes and the 13C NMR spectra were
determined and analysed as described.9 LCMS experiments on
enzyme inhibitor complexes were carried out on 10a, 10b and
14a in solution in Tris–HCl buffer.
[Methylene-13C]-2-hydroxymethyl-1-(phenylsulfonyl)pyrrole
(10b). Prepared from 9 (62 mg mg, 0.30 mmol) as described for
10a1 (56 mg, 90%); δH(500 MHz; CDCl3; Me4Si) 2.78 (1 H, br s,
OH ), 4.60 (2 H, d, J = 145.0 Hz, CH2OH), 6.24–6.27 (2 H, m,
pyrrole 4-H and pyrrole 3-H), 7.27 (1 H, m, pyrrole 5-H), 7.50–
7.62 (3 H, m, ArH ), 7.82 (2 H, m, ArH ); δC(75 MHz; CDCl3;
Me4Si) 57.7 (CH2OH enriched peak), 112.0, 115.4, 123.6,
126.6, 129.5, 134.0, 134.6, 139.0.
N-(5-Formyl-1H-pyrrole-2-carbonyl)-L-phenylalanine methyl
ester (13a). The pyrrole acid 12b14 (100 mg, 0.72 mmol, 1
equiv.) was coupled with -phenylalanine methyl ester hydro-
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 0 3 – 2 1 1 0
2106

View Article Online
chloride (171 mg, 0.79 mmol, 1.1 equiv.) according to general
procedure A. The resultant material was purified by flash
chromatography on silica (ethyl acetate–petroleum ether, 2 : 1)
to give 13a (152 mg, 71%) as a cream solid; mp 48–51Њ;
νmax(CHCl3)/cmϪ1 1529.4, 1550.7, 1676.0, 1739.7, 3020.3,
3423.4; δH(500 MHz; CDCl3; Me4Si) 3.22 (2 H, m, CH2Ph),
3.76 (3H, s, CO2CH3), 5.06 (1 H, m, NHCH ), 6.51 (1 H, d,
J = 7.3 Hz, NHCH), 6.61 (1 H, dd, J = 2.7, 4.2 Hz, pyrrole 3-H),
6.91 (1 H, dd, J = 2.7, 4.2 Hz, pyrrole 4-H), 7.10 (2 H, m, ArH ),
7.27 (3 H, m, ArH ), 9.61 (1 H, s, CHO), 10.11 (1 H, br s, pyrrole
NH ); 13C NMR (75 MHz; CDCl3; Me4Si) δ 38.0, 52.6, 53.2,
111.2, 120.0, 127.2, 128.6, 129.3, 131.2, 134.0, 135.5, 159.3,
171.9, 180.3; m/z (ES) 301.1188 (Mϩ ϩ H) C16H17N2O4 requires
301.1188, 301 (100%).
(pyrrole 2-C), 127.1, 128.5, 129.1, 135.7 (ArC), 136.8 (pyrrole
5-H), 161.3 (CONH), 172.4 (CO2); m/z (EI) 302.1264 (Mϩ.
C16H18N2O4 requires 302.1267), 284 (11%), 162 (20), 140 (77),
124 (100), 122 (21), 120 (21), 108 (30), 107 (24), 106 (88), 91 (70)
and 78 (38).
N-(5-Hydroxymethyl-1H-pyrrole-2-carbonyl)-L-leucine
methyl ester (14b). Method 1. The pyrrole 13b (116 mg, 0.44
mmol, 1 equiv.) was reduced with zinc borohydride according
to general procedure B. The resultant material was purified by
flash chromatography on silica (ethyl acetate–petroleum ether,
4 : 1) to give 14b (49 mg, 42%) as a cream solid; mp 92–96 ЊC
(Found: C, 58.35; H, 7.37; N, 10.36. C13H20N2O4 requires C,
58.19; H, 7.51; N, 10.44%); νmax(CHCl3)/cmϪ1 1525.6, 1641.3,
1739.7, 2958.6, 3438.8; δH(500 MHz; CDCl3; Me4Si) 0.94 (6 H,
m, CH(CH3)2), 1.63–1.75 (3 H, m, CH2CH ), 3.76 (3 H, s,
CO2CH3), 4.60 (2 H, d, J = 2.9 Hz, CH2OH), 4.76 (1 H, m,
NHCH ), 6.04 (1 H, dd, J = 2.4, 3.4 Hz, pyrrole 4-H), 6.57 (1 H,
dd, J = 2.4, 3.4 Hz, pyrrole 3-H), 6.62 (1 H, d, J = 8.8 Hz,
NHCH), 10.51 (1 H, br s, pyrrole NH ); δC(75 MHz; CDCl3;
Me4Si) 21.6, 22.8 (CH(CH3)2), 24.8 (CH2CH), 41.0 (CH2CH),
50.7 (NHCH), 52.4 (CO2CH3), 57.3 (CH2OH), 107.9 (pyrrole
4-C), 110.9 (pyrrole 3-C), 124.6 (pyrrole 2-C), 136.8 (pyrrole
5-C), 161.7 (CONH), 174.5 (CO2); m/z (EI) 268.1418 (Mϩ.
C13H20N2O4 requires 268.1423), 212 (27%), 209 (12), 191 (19),
180 (11), 175 (13), 174 (96), 130 (15), 126 (18), 125 (12), 124
(100), 108 (21), 106 (70), 86 (48) and 78 (31).
Method 2. The pyrrole 13b (35 mg, 0.13 mmol, 1 equiv.),
dissolved in dry ether (5 cm3) under an inert atmosphere, was
cooled to Ϫ78 ЊC (dry ice–acetone). Lithium borohydride
(6 mg, 0.26 mmol, 2 equiv.) was added, and the resultant solu-
tion was stirred at Ϫ78 ЊC for 1 h. Water (10 cm3) was carefully
added to quench the reaction, and the solution was extracted
with ethyl acetate (3 × 10 cm3). The combined organic phases
were washed with saturated aqueous brine (10 cm3), dried
(MgSO4), and the solvent was removed by evaporation under
reduced pressure. The resultant material was purified by flash
chromatography on silica (ethyl acetate–petroleum ether, 4 : 1)
to give 14b (23 mg, 66%) as an off-white solid; mp 94–97 ЊC.
Data as above.
N-(5-Formyl-1H-pyrrole-2-carbonyl)-L-leucine methyl ester
(13b). The pyrrole acid 12b14 (100 mg, 0.72 mmol, 1 equiv.)
was coupled with -leucine methyl ester hydrochloride (144 mg,
0.79 mmol, 1.1 equiv.) according to general procedure A. The
resultant material was purified by flash chromatography on
silica (ethyl acetate–petroleum ether, 2 : 1) to give 13b (116 mg,
61%) as a cream solid; mp 139–142 ЊC; νmax(CHCl3)/cmϪ1
1529.4, 1550.7, 1676.0, 1739.7, 2960.5, 3018.4, 3425.3; δH(500
MHz; CDCl3; Me4Si) 0.96 (6 H, m, CH(CH3)2), 1.61–1.77 (3 H,
m, CH2CH ), 3.77 (3 H, s, CO2CH3), 4.83 (1 H, m, NHCH ),
6.62 (1 H, d, J = 8.3 Hz, NHCH), 6.68 (1 H, d, J = 2.4 Hz,
pyrrole 3-H), 6.93 (1 H, d, J = 2.9 Hz, pyrrole 4-H), 9.62 (1 H, s,
CHO), 10.18 (1 H, br s, pyrrole NH ); δC(75 MHz; CDCl3;
Me4Si) 21.9, 22.8, 24.9, 41.7, 50.9, 52.6, 111.0, 120.0, 131.2,
133.9, 159.6, 173.7, 180.2; m/z (EI) 266.1266 (Mϩ. C13H18N2O4
requires 266.1267), 210 (30%), 207 (38), 138 (12), 123 (26),
122 (75), 86 (100), 66 (20) and 44 (14).
N-(5-Formyl-1H-pyrrole-2-carbonyl)-L-proline methyl ester
(13c). The pyrrole acid 12b14 (100 mg, 0.72 mmol, 1 equiv.) was
coupled with -proline methyl ester hydrochloride (131 mg, 0.79
mmol, 1.1 equiv.) according to modified general procedure A.
The resultant material was purified by flash chromatography on
silica (ethyl acetate–petroleum ether, 2 : 1 then 4 : 1) to give 13c
(133 mg, 74%) as a white solid; mp decomp. >40 ЊC (Found: C,
57.63; H, 5.73; N, 11.21. C12H14N2O4 requires C, 57.59; H, 5.64;
N, 11.19%); νmax(CHCl3)/cmϪ1 1544.9, 1616.2, 1672.2, 1745.5,
3419.6; δH(500 MHz; CDCl3; Me4Si) 2.01–2.27 (4 H, m,
NCH2CH2CH2CH), 3.73 (3 H, s, CO2CH3), 3.86 (1 H, m,
NCH2aCH2CH2CH), 3.97 (1 H, m, NCH2bCH2CH2CH), 4.70
(1 H, m, NCH ), 6.69 (1 H, m, pyrrole 3-H), 6.92 (1 H, m,
pyrrole 4-H), 9.63 (1 H, s, CHO), 10.38 (1 H, br s, pyrrole NH );
δC(75 MHz; CDCl3; Me4Si) 25.3, 28.6 (NCH2CH2CH2CH),
48.5 (NCH2CH2CH2CH), 52.3 (CO2CH3), 60.2 (NCH), 113.2
(pyrrole 3-C), 119.3 (pyrrole 4-C), 130.7 (pyrrole 2-C), 133.3
(pyrrole 5-C), 159.5 (CONH), 172.3 (CO2), 180.1 (CHO); m/z
(EI) 250.0952 (Mϩ. C12H14N2O4 requires 250.0954), 218 (14%),
191 (47), 122 (26), 94 (12) and 70 (100).
N-(5-Hydroxymethyl-1H-pyrrole-2-carbonyl)-L-proline
methyl ester (14c). The pyrrole 13c (120 mg, 0.48 mmol, 1
equiv.) was reduced with zinc borohydride according to general
procedure B. The resultant material was purified by flash
chromatography on silica (ethyl acetate–acetone, 10 : 1) to give
14c (29 mg, 24%) as an orange oil; νmax(CHCl3)/cmϪ1 1596.9,
1743.5, 3010.7, 3436.9, 3691.5; δH(500 MHz; CDCl3; Me4Si)
2.03–2.24 (4 H, m, NCH2CH2CH2CH), 3.44 (1 H, br s,
CH2OH ), 3.73 (3 H, s, CO2CH3), 3.85 (1 H, m, NCH2aCH2-
CH2CH), 3.95 (1 H, m, NCH2bCH2CH2CH), 4.61 (2 H, s,
CH2OH), 4.73 (1H, m, NCH ), 6.13 (1 H, s, pyrrole 4-H), 6.60
(1 H, s, pyrrole 3-H), 10.75 (1 H, br s, pyrrole NH ); δC(75 MHz;
CDCl3; Me4Si) 25.2, 28.8 (NCH2CH2CH2CH), 48.8 (NCH2-
CH2CH2CH), 52.4 (CO2CH3), 57.4 (CH2OH), 60.2 (NCH),
108.2 (pyrrole 4-C), 113.8 (pyrrole 3-C), 124.6 (pyrrole 2-C),
136.9 (pyrrole 5-C), 161.3 (CON), 172.8 (CO2); m/z (EI)
252.1119 (Mϩ. C12H16N2O4 requires 252.1110), 252 (20%), 193
(48), 191 (16), 176 (14), 175 (76), 147 (11), 147 (11), 128 (77),
124 (62), 122 (12), 108 (21), 107 (83), 106 (100), 96 (11), 94 (16),
79 (61), 78 (64), 70 (90) and 68 (26).
N-(5-Hydroxymethyl-1H-pyrrole-2-carbonyl)-L-phenylalanine
methyl ester (14a). The pyrrole 13a (106 mg, 0.35 mmol, 1
equiv.) was reduced with zinc borohydride according to general
procedure B. The resultant material was purified by flash
chromatography on silica (ethyl acetate–petroleum ether, 4 : 1)
to give 14a (61 mg, 57%) as a tan solid; mp 105–107 ЊC (Found:
C, 63.42; H, 5.90; N, 9.38. C16H18N2O4 requires C, 63.57; H,
6.00; N, 9.27%); νmax(CHCl3)/cmϪ1 1525.6, 1637.5, 1739.7,
3035.7, 3435.0; δH(500 MHz; CDCl3; Me4Si) 2.97 (1 H,
br s, CH2OH ), 3.20 (2 H, m, CH2Ph), 3.74 (3 H, s, CO2CH3),
4.65 (2 H, s, CH2OH), 5.00 (1 H, m, NHCH ), 6.08 (1 H, t,
J = 3.2 Hz, pyrrole 4-H), 6.37 (1 H, d, J = 7.8 Hz, NHCH), 6.47
(1 H, t, J = 3.2 Hz, pyrrole 3-H), 7.12 (2 H, d, J = 7.3 Hz, ArH ),
7.27 (3 H, m, ArH ), 10.31 (1 H, br s, pyrrole NH ); δC(75 MHz;
CDCl3; Me4Si) 37.9 (CH2Ph), 52.4 (CO2CH3), 53.2 (NHCH),
57.2 (CH2OH), 108.1 (pyrrole 4-C), 110.9 (pyrrole 3-C), 124.5
Ethyl 5-(3-phenylpropionyl)-1H-pyrrole-2-carboxylate (15a).
A solution of pyrrole ester 1112 (2.91 g, 21 mmol, 1 equiv.)
in dry 1,2-dichloroethane (35 cm3) was added in a dropwise
fashion over 5 min to a stirred suspension of hydrocinnamoyl
chloride (7.04 g, 42 mmol, 2 equiv.) and zinc chloride (5.70 g,
42 mmol, 2 equiv.) in dry 1,2-dichloroethane (35 cm3), cooled to
0 ЊC under an inert atmosphere. The suspension was heated to
50 ЊC for 1 h, after which the resultant solution was poured
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 0 3 – 2 1 1 0
2107

View Article Online
onto ice–water and the organic layer was separated. The aque-
ous layer was extracted with ethyl acetate (40 cm3), and the
combined organic fractions were washed with aqueous satur-
ated sodium hydrogen carbonate (30 cm3), aqueous saturated
brine (30 cm3), dried (MgSO4), and the solvent was removed by
evaporation under reduced pressure. The crude material was
dissolved in ethyl acetate (50 cm3), washed with 1 M aqueous
sodium hydroxide (4 × 20 cm3) to remove excess hydrocin-
namoyl chloride, dried (MgSO4), and the solvent was removed
by evaporation under reduced pressure. The resultant material
was purified by flash chromatography on silica (ethyl acetate–
petroleum ether, 1 : 4) to give 15a (2.0 g, 35%) as a cream
residue; νmax(KBr)/cmϪ1 1548.7, 1604.7, 1662.5, 1710.7, 2987.5,
3006.8, 3429.2; δH(500 MHz; CDCl3; Me4Si) 1.36 (3 H, t, J = 7.1
Hz, CH2CH3), 3.04 (2 H, t, J = 7.3 Hz, PhCH2CH2), 3.13 (2 H,
m, PhCH2CH2), 4.35 (2 H, q, J = 7.2 Hz, CH2CH3), 6.81 (1 H,
dd, J = 2.9, 3.9 Hz, pyrrole H), 6.86 (1 H, dd, J = 2.4, 3.9 Hz,
pyrrole H), 7.18–7.30 (5 H, m, ArH ), 9.89 (1 H, br s, pyrrole
NH ); δC(75 MHz; CDCl3; Me4Si) 14.2 (CH2CH3), 30.2
(PhCH2CH2), 40.0 (PhCH2CH2), 61.1 (CH2CH3), 115.4, 115.5
(pyrrole 3-C and 4-C), 126.2, 128.3, 128.5, 141.8 (ArC), 127.2
(pyrrole 2-C), 133.7 (pyrrole 5-C), 160.3 (CO2), 190.3 (CH2CO);
m/z (EI) 271.1220 (Mϩ. C16H17NO3 requires 271.1208), 271
(85%), 226 (17), 198 (27), 166 (63), 139 (66), 120 (100), 104 (40)
and 91 (48).
3-C and 4-C), 127.9 (pyrrole 2-C), 135.1 (pyrrole 5-C), 161.2
(CO2H), 188.2 (CH3CO); m/z (EI) 153.0426 (Mϩ. C7H7NO3
requires 153.0426), 153 (45%), 138 (21) and 120 (100).
N-[5-(3-Phenylpropionyl)-1H-pyrrole-2-carbonyl]-L-phenyl-
alanine methyl ester (17a). The pyrrole acid 16a (150 mg,
0.62 mmol, 1 equiv.) was coupled with -phenylalanine methyl
ester hydrochloride (146 mg, 0.68 mmol, 1.1 equiv.) according
to modified general procedure A. The resultant material was
purified by flash chromatography on silica (ethyl acetate–
petroleum ether, 2 : 3) to give 17a (175 mg, 70%) as a brown
residue (Found: C, 71.37; H, 6.15; N, 7.10. C24H24N2O4 requires
C, 71.27; H, 5.98; N, 6.93%); νmax(KBr)/cmϪ1 1529.4, 1548.7,
1668.3, 1741.6, 3008.7, 3425.3; δH(500 MHz; CDCl3; Me4Si)
3.03 (2 H, m, PhCH2CH2), 3.14 (2 H, m, PhCH2CH2), 3.21
(2 H, m, CH2Ph), 3.75 (3 H, s, CO2CH3), 5.05 (1 H, m, NHCH ),
6.46 (2 H, m, pyrrole 3-H and NHCH), 6.78 (1 H, dd, J = 2.4,
3.9 Hz, pyrrole 4-H), 7.10 (2 H, m, ArH ), 7.17–7.30 (8 H, m,
ArH ), 10.00 (1 H, br s, pyrrole NH ); δC(75 MHz; CDCl3;
Me4Si) 30.3 (PhCH2CH2), 37.9 (CHCH2Ph), 40.0 (PhCH2-
CH2), 52.5 (CO2CH3), 53.1 (NHCH), 110.4 (pyrrole 3-C),
115.5 (pyrrole 4-C), 126.2, 127.2, 128.3, 128.5, 128.6, 129.3,
135.6, 140.9 (ArC), 129.7 (pyrrole 2-C), 133.3 (pyrrole 5-C),
159.4 (CONH), 171.9 (CO2Me), 190.0 (CH2CO); m/z (EI)
404.1745 (Mϩ. C24H24N2O3 requires 404.1736), 370 (33%), 265
(15), 242 (29), 238 (16), 226 (100), 214 (23), 208 (23), 120 (22)
and 91 (81).
Ethyl 5-acetyl-1H-pyrrole-2-carboxylate (15b). The pyrrole
1112 (4.0 g, 29 mmol, 1 equiv.) was treated with acetyl chloride
(4.51 g, 57 mmol, 2 equiv.) as described for 15a. Washing with
1 M aqueous sodium hydroxide was not required in this case.
The resultant material was purified by flash chromatography on
silica (ethyl acetate–petroleum ether, 1 : 2) to give 15b (1.93 g,
37%) as a yellow solid; mp 54–57 ЊC (lit.,16 60–60.5 ЊC); δH(500
MHz; CDCl3; Me4Si) 1.36 (3 H, t, J = 7.3 Hz, CH2CH3), 2.46
(3 H, s, COCH3), 4.34 (2 H, q, J = 7.2 Hz, CH2CH3), 6.83 (1 H,
t, J = 3.2 Hz, pyrrole H), 6.87 (1 H, t, J = 3.2 Hz, pyrrole H),
9.82 (1 H, br s, pyrrole NH ).
N-[5-Acetyl-1H-pyrrole-2-carbonyl]-L-phenylalanine methyl
ester (17b). The pyrrole acid 16b (200 mg, 1.31 mmol, 1 equiv.)
was coupled with -phenylalanine methyl ester hydrochloride
(310 mg, 1.44 mmol, 1.1 equiv.) according to modified general
procedure A. The resultant material was purified by flash
chromatography on silica (ethyl acetate–petroleum ether, 2 : 1)
to give 17b (375 mg, 91%) as a cream solid. An analytical
sample was obtained by diffusion of petroleum ether into a
solution of 17b dissolved in ethyl acetate; mp 103–105 ЊC
(Found: C, 65.13; H, 5.84; N, 8.77. C17H18N2O4 requires C,
64.96; H, 5.77; N, 8.91%); νmax(KBr)/cmϪ1 1546.8, 1654.8,
1741.6, 2956.7, 3031.9, 3112.9, 3352.1; δH(500 MHz; CDCl3;
Me4Si) 2.42 (3 H, s, CH3CO), 3.20 (2 H, m, CH2Ph), 3.75 (3 H,
s, CO2CH3), 5.06 (1 H, m, NHCH ), 6.52 (1 H, dd, J = 2.9,
3.9 Hz, pyrrole 3-H), 6.67 (1 H, d, J = 7.8 Hz, NHCH), 6.88
(1 H, dd, J = 2.7, 4.2 Hz, pyrrole 4-H), 7.10 (2 H, m, ArH ), 7.25
(3 H, m, ArH ), 10.11 (1 H, br s, pyrrole NH ); δC(75 MHz;
CDCl3; Me4Si) 25.8 (CH3CO), 37.9 (CH2Ph), 52.5 (CO2CH3),
53.2 (NHCH), 110.9 (pyrrole 3-C), 116.3 (pyrrole 4-C), 127.2,
128.6, 129.2, 135.7 (ArC), 130.0 (pyrrole 2-C), 133.6 (pyrrole
5-C), 159.4 (CONH), 172.1 (CO2), 188.6 (CH3CO); m/z (EI)
314.1271 (Mϩ. C17H18N2O4 requires 314.1267), 280 (33%), 162
(24), 152 (73), 136 (100), 120 (32), 118 (15) and 91 (19).
5-(3-Phenylpropionyl)-1H-pyrrole-2-carboxylic acid (16a). A
stirred solution of the pyrrole ester 15a (801 mg, 2.95 mmol,
1 equiv.) and potassium hydroxide (663 mg, 11.8 mmol, 4
equiv.) in 2 : 1 water–tetrahydrofuran (20 cm3) was heated to
40–50 ЊC for 2 h. After cooling to rt, the solution was washed
with ether (20 cm3), acidified to pH ∼1 with concentrated aque-
ous hydrochloric acid, and extracted with ether (5 × 20 cm3).
The combined ethereal fractions were dried (MgSO4) and the
solvent was removed by evaporation under reduced pressure to
give 16a (616 mg, 86%) as a brown solid; mp 181–183 ЊC;
νmax(KBr)/cmϪ1 1500.5, 1550.7, 1670.2, 1701.1, 2366.5, 2597.9,
3197.8; δH(500 MHz, acetone-d6,) 3.11 (2 H, t, J = 7.6 Hz,
PhCH2CH2), 3.32 (2 H, t, J = 7.6 Hz, PhCH2CH2), 6.97 (1 H,
dd, J = 2.4, 3.9 Hz, pyrrole H), 7.15 (1 H, dd, J = 2.4, 3.9 Hz,
pyrrole H), 7.29 (1 H, m, ArH ), 7.39 (4 H, m, ArH ), 11.23 (1 H,
br s, pyrrole NH ); δC(75 MHz, acetone-d6) 30.1 (PhCH2CH2),
39.9 (PhCH2CH2), 115.5, 115.8, 126.1, 128.5, 128.6, 141.6,
127.9 (pyrrole 2-C), 134.8 (pyrrole 5-C), 161.2 (CO2H), 190.0
(CH2CO); m/z (EI) 243.0900 (Mϩ. C14H13NO3 requires
243.0895), 243 (57%), 143 (60), 120 (100), 113 (63), 104 (72) and
94 (75).
(R,S )-N-[5-(1-Hydroxy-3-phenylpropyl)-1H-pyrrole-2-
carbonyl]-L-phenylalanine methyl ester (18a). The pyrrole 17a
(112 mg, 0.28 mmol, 1 equiv.) was reduced with zinc boro-
hydride according to general procedure B. The resultant
material was purified by flash chromatography on silica (ethyl
acetate–petroleum ether, 1 : 1) to give a mixture of 17a and 18c
(41 mg); δH(18c from mixture; 500 MHz; CDCl3; Me4Si) 1.95
(2 H, m, CH2), 2.63 (4 H, m, 2 × CH2), 3.18 (2 H, m, CH2Ph),
3.71 (3 H, s, CO2CH3), 5.03 (1 H, m, NHCH ), 5.93 (1 H, m,
pyrrole 4-H), 6.23 (1 H, d, J = 7.8 Hz, NHCH), 6.45 (1 H, m,
pyrrole 3-H), 7.11–7.29 (10 H, m, ArH ), 9.35 (1 H, br s, pyrrole
NH ); m/z (EI) 390.1953 (Mϩ. C24H26N2O3 requires 390.1943),
390 (10%), 356 (18), 228 (38), 212 (100), 208 (18), 124 (59), 120
(13), 107 (15), 91 (55) and 79 (31).
5-Acetyl-1H-pyrrole-2-carboxylic acid (16b). The pyrrole 15b
(1.90 g, 10 mmol, 1 equiv.) was hydrolysed with potassium
hydroxide (2.35 g, 42 mmol, 4 equiv.) as described for 15a to
give 16b (1.50 g, 93%) as a yellow solid; mp 185–187 ЊC (Found:
C, 54.93; H, 4.55; N, 9.07. C7H7NO3 requires C, 54.90; H, 4.61;
N, 9.15%); νmax(KBr)/cmϪ1 1546.8, 1670.2, 2360.7, 2511.1,
2584.4, 3124.5, 3436.9; δH(500 MHz, acetone-d6,) 2.57 (3 H, s,
CH3CO), 6.98 (1 H, dd, J = 2.4, 3.9 Hz, pyrrole H), 7.11 (1 H,
dd, J = 2.4, 3.9 Hz, pyrrole H), 11.20 (1 H, br s, pyrrole NH );
δC(75 MHz, acetone-d6) 25.5 (CH3CO), 115.5, 116.2 (pyrrole
Further elution gave a 1 : 1 mixture of epimers 18a (24 mg,
21%) as a light yellow oil; νmax(KBr)/cmϪ1 1521.7, 1602.7,
1641.3, 1741.6, 3012.6, 3030.0, 3436.9; δH(500 MHz; CDCl3;
Me4Si) 2.13 (2 H, m, PhCH2CH2), 2.75 (2 H, m, PhCH2CH2),
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 0 3 – 2 1 1 0
2108

View Article Online
3.20 (2 H, m, CHCH2Ph), 3.71 and 3.72 (each 3 H, s, CO2CH3,
isomers A and B), 4.75 (1 H, m, CHOH), 5.02 (1 H, m,
NHCH ), 6.02 (1 H, m, pyrrole 4-H), 6.42 (1 H, d, J = 7.3 Hz,
NHCH), 6.48 (1 H, m, pyrrole 3-H), 7.11–7.29 (10 H, m, ArH ),
10.67 and 10.70 (each 1 H, br s, pyrrole NH, isomers A and B);
δC(75 MHz; CDCl3; Me4Si) 31.8 (PhCH2CH2), 38.2 (CHCH2-
Ph), 38.6 and 38.7 (PhCH2CH2, isomers A and B), 52.4
(CO2CH3), 53.1 (NHCH), 67.1 and 67.1 (CHOH, isomers A
and B), 106.4 and 106.5 (pyrrole 4-C, isomers A and B), 110.7
(pyrrole 3-C), 124.0 (pyrrole 2-C), 125.9, 127.2, 128.4, 128.5,
128.6, 129.3, 135.6, 141.6 (ArC), 140.1 (pyrrole 5-C), 161.1
(CONH), 172.0 (CO2Me); m/z (EI) 388.1800 (Mϩ Ϫ H2O.
C24H24N2O3 requires 388.1787), 388 (17%), 226 (25), 210 (100),
208 (54), 180 (18) and 91 (23).
(pyrrole 2-C), 141.5 (pyrrole 5-C), 161.9 (CO2); m/z (EI)
183.0897 (Mϩ. C9H13NO3 requires 183.0895), 183 (43%), 168
(56), 165 (25), 140 (24), 136 (14), 122 (100), 120 (54), 94 (12), 91
(17) and 65 (28).
(R,S )-5-(1-Hydroxyethyl)-1H-pyrrole-2-carboxylic
acid
(19b). A stirred solution of the pyrrole ester 19a (575 mg, 3.13
mmol, 1 equiv.), in 1 : 1 tetrahydrofuran–water (10 cm3) under
an inert atmosphere, was warmed to 40–50 ЊC. Sodium hydrox-
ide (377 mg, 9.42 mmol, 3 equiv.) was added, and the progress
of the reaction was monitored by thin layer chromatography
(ethyl acetate–petroleum ether, 1 : 1). After 24 h, the solution
was cooled to rt, washed with ether (10 cm3), cooled to 0 ЊC and
acidified to pH ∼4 by the dropwise addition of 1 M aqueous
hydrochloric acid. The resulting solution was extracted with
ether (3 × 15 cm3), the combined ethereal extracts were washed
with aqueous saturated brine (15 cm3), dried (MgSO4), and the
solvent was removed by evaporation under reduced pressure to
give 19b (85 mg, 17%) as a purple solid. This product was used
without further purification; mp 110–115 ЊC; νmax(KBr)/cmϪ1
1651.0, 3230.5, 3419.6; δH(300 MHz, acetone-d6) 1.59 (3 H, d,
J = 6.3 Hz, CH3CHOH), 4.43 (1 H, br s, CHOH ), 5.02 (1 H, m,
CH3CHOH), 6.20 (1 H, t, J = 3.2 Hz, pyrrole 4-H), 6.87 (1
H, t, J = 3.2 Hz, pyrrole 3-H), 10.47 (1 H, br s, pyrrole NH );
δC(75 MHz, acetone-d6) 23.2 (CH3CHOH), 63.3 (CH3CHOH),
105.9 (pyrrole 4-C), 115.6 (pyrrole 3-C), 121.8 (pyrrole 5-C),
143.2 (pyrrole 2-C), 161.8 (CO2H); m/z (EI) 155.0579 (Mϩ.
C7H9NO3 requires 155.0582), 155 (40%), 140 (56), 137 (46), 122
(100), 119 (64), 112 (14), 91 (69), 69 (15) and 65 (50).
(R,S )-N-[5-(1-Hydroxyethyl)-1H-pyrrole-2-carbonyl]-L-
phenylalanine methyl ester (18b). Method 1. A stirred solution
of the pyrrole 17b (150 mg, 0.48 mmol, 1 equiv.), dissolved in
dry tetrahydrofuran (20 cm3) under an inert atmosphere, was
cooled to Ϫ78 ЊC (dry ice–acetone). Lithium borohydride (21
mg, 0.95 mmol, 2 equiv.) was added, and the resultant solution
was stirred at Ϫ78 ЊC for 1 h, then warmed to rt and stirred for
an additional 1 h. Water (10 cm3) was carefully added to quench
the reaction, and the solution was extracted with ethyl acetate
(3 × 10 cm3). The combined organic phases were washed with
saturated aqueous brine (10 cm3), dried (MgSO4) and the sol-
vent was removed by evaporation under reduced pressure. The
resultant material was purified by flash chromatography on
silica (ethyl acetate–petroleum ether, 4 : 1) to give impure 18b
(70 mg, 46%) which could not be purified further; δH(from mix-
ture; 500 MHz; CDCl3; Me4Si) 1.53 (3 H, m, CH3CHOH), 3.20
(2 H, m, CH2Ph), 3.74 (3 H, s, CO2CH3), 4.90 (1 H, m, CHOH),
5.03 (1 H, m, NHCH ), 6.00 (1 H, s, pyrrole 4-H), 6.48 (1 H, t,
J = 2.7 Hz, pyrrole 3-H), 6.68 (1 H, m, NHCH), 7.15 (2 H, d,
J = 6.8 Hz, ArH ), 7.26 (m, 3 H, m, ArH ), 10.71 and 10.73 (each
1 H, br s, pyrrole NH, isomers A and B); m/z (EI) 316.1423
(Mϩ. C17H20N2O4 requires 316.1423), 298 (11%), 138 (12), 136
(61), 120 (100), 91 (18) and 65 (25).
N-(5-Formyl-1H-pyrrole-2-carbonyl)-L-phenylalanyl-L-leucine
methyl ester (21). To a stirred suspension of -phenylalanyl--
leucine hydrate 20a (200 mg, 0.67 mmol) in dry tetrahydrofuran
(10 cm3) at rt was added a 12-fold excess of an ethereal solution
of diazomethane in a dropwise fashion over 10 min. After 15 h,
the solvent was removed by evaporation under reduced pressure
to give the methyl ester 20b (202 mg, quantitative yield) as
a light brown oil. This was immediately dissolved in dry di-
chloromethane (6 cm3) at rt under an inert atmosphere, and to
this was added the pyrrole acid 12b14 (87 mg, 0.63 mmol, 1
equiv.), benzotriazol-1-yloxytris(dimethylamino)phosphonium
hexafluorophosphate (306 mg, 0.69 mmol, 1.1 equiv.), and N,N-
diisopropylethylamine (89 mg, 0.69 mmol, 1.1 equiv.). The reac-
tion mixture was stirred at rt for 16 h, after which the solution
was diluted with ethyl acetate (15 cm3), washed with 1 M aque-
ous hydrochloric acid (10 cm3), aqueous saturated sodium
hydrogen carbonate (10 cm3), aqueous saturated brine (10 cm3),
dried (MgSO4) and the solvent was removed by evaporation
under reduced pressure. The resultant material was purified by
flash chromatography on silica (ethyl acetate–petroleum ether,
3 : 1), followed by diffusion of petroleum ether into a solution
of the title product dissolved in ethyl acetate, to give 21 (135
mg, 52%) as a tan solid; mp 100–103 ЊC; νmax(KBr)/cmϪ1 1558.4,
1635.5, 1674.1, 1743.5, 2958.6, 3286.5; δH(500 MHz; CDCl3;
Me4Si) 0.85 (6 H, m, CH(CH3)2), 1.43–1.61 (3 H, m, CH2CH ),
3.13 (2 H, m, CH2Ph), 3.73 (3 H, s, CO2CH3), 4.56 (1 H, m, Leu
αH ), 5.06 (1 H, q, J = 7.3 Hz, Phe αH ), 6.64 (1 H, dd, J = 2.4,
3.9 Hz, pyrrole 3-H), 6.90 (2 H, m, pyrrole 4-H and NHCH),
7.19 (6 H, m, ArH and NHCH), 9.61 (1 H, s, CHO), 10.68 (1 H,
br s, pyrrole NH ); δC(75 MHz; CDCl3; Me4Si) 22.1, 22.6
(CH(CH3)2), 24.8, 41.6 (CH2CH), 38.8 (CH2Ph), 51.0 (Leu
αCH), 52.5 (CO2CH3), 54.7 (Phe αCH), 111.3 (pyrrole 3-C),
119.6 (pyrrole 4-C), 126.9, 128.4, 129.4, 136.4 (ArC), 131.3
(pyrrole 2-C), 134.1 (pyrrole 5-C), 159.7 (PyrCONH), 171.3
(CHCONH), 173.1 (CO2), 180.3 (CHO); m/z (ES) 414.2031
(Mϩ ϩ H. C22H28N3O5 requires 414.2029), 414 (100%), 293 (62),
288 (31), 268 (20), 247 (20), 225 (20) and 146 (29).
Method 2. The pyrrole acid 19b (20 mg, 0.13 mmol, 1 equiv.)
was coupled with -phenylalanine methyl ester hydrochloride
(31 mg, 0.14 mmol, 1.1 equiv.) according to general procedure
A, except in the workup whereby 0.1 M aqueous hydrochloric
acid was used instead of the 3 M aqueous hydrochloric acid.
The resultant material was purified by flash chromatography on
silica (ethyl acetate–petroleum ether, 4 : 1) to give 18b (22 mg,
54%) as a colourless oil.
(R,S )-Ethyl
5-(1-hydroxyethyl)-1H-pyrrole-2-carboxylate
(19a). A stirred solution of the pyrrole ester 12a13 (400 mg, 2.39
mmol, 1 equiv.) in dry tetrahydrofuran (40 cm3) under an argon
atmosphere was cooled to Ϫ23 ЊC (dry ice–carbon tetrachlor-
ide). Methyllithium (1.50 cm3 of a 1.6 M solution in ether, 2.39
mmol, 1 equiv.) was added over 30 min, and the solution was
stirred at Ϫ23 ЊC for 2.5 h. After this, a further 1.50 cm3 of the
methyllithium solution was added over 30 min, and the solution
was again stirred at Ϫ23 ЊC for 2.5 h. The resultant solution was
poured into ether–ice, and once the ice had melted the layers
were separated. The organic layer was washed with aqueous
saturated brine (25 cm3), water (2 × 25 cm3), then dried
(MgSO4), and the solvent was removed by evaporation under
reduced pressure to give 19a (418 mg, 95%) as a brown oil
(Found: C, 58.59; H, 7.22; N, 7.60. C9H13NO3 requires C, 59.00;
H, 7.15; N, 7.65%); νmax(KBr)/cmϪ1 1693.4, 2983.7, 3020.3,
3446.6; δH(500 MHz; CDCl3; Me4Si) 1.34 (3 H, t, J = 7.1 Hz,
CH2CH3), 1.55 (3 H, d, J = 6.8 Hz, CH3CHOH), 3.08 (1 H,
br s, CHOH ), 4.29 (2 H, q, J = 7.2 Hz, CH2CH3), 4.96 (1 H, q,
J = 6.5 Hz, CHOH), 6.05 (1 H, m, pyrrole 4-H), 6.83 (1 H, m,
pyrrole 3-H), 9.77 (1 H, br s, pyrrole NH ); δC(75 MHz; CDCl3;
Me4Si) 14.4 (CH2CH3), 23.2 (CH3CHOH), 60.4 (CH2CH3),
64.0 (CHOH), 106.3 (pyrrole 4-C), 115.9 (pyrrole 3-C), 121.8
N-(5-Hydroxymethyl-1H-pyrrole-2-carbonyl)-L-phenylalanyl-
L-leucine methyl ester (22). The pyrrole 21 (82 mg, 0.20 mmol, 1
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 0 3 – 2 1 1 0
2109

View Article Online
3 A. D. Abell and J. C. Litten, J. Chem. Soc., Chem. Commun., 1998,
919.
4 A. D. Abell, B. K. Nabbs and A. R. Battersby, J. Org. Chem., 1998,
63, 8163.
5 A. D. Abell, B. K. Nabbs and A. R. Battersby, J. Am. Chem. Soc.,
1998, 120, 1741.
6 A. Kashima, Y. Inoue, S. Sugio, I. Maeda, T. Nose and
Y. Shimohigashi, Eur. J. Biochem., 1998, 255, 12.
7 W. Hinz, R. A. Jones and T. Anderson, Synthesis, 1986, 620.
8 R. M. Silverstein, E. E. Ryskiewicz, C. Willard and R. Koehler,
J. Org. Chem., 1955, 20, 668.
9 W. C. Groutas, M. A. Stanga and M. J. Brubaker, J. Am. Chem. Soc.,
1989, 111, 1931.
10 J. Laskowski, Jr., M. Tashiro, M. W. Empie, S. J. Park, I. Kato,
W. Ardelt and M. Wieczorek, in Proteinase Inhibitors: Medical and
Biological Aspects, ed. N. Katunuma, H. Umezawa and H. Holzer,
Japan Scientific Societies Press, Tokyo, 1983, p 55.
11 B. Imperiali and R. H. Abeles, Biochemistry, 1987, 26, 4474.
12 D. M. Bailey, R. E. Johnson and N. F. Albertson, Org. Synth., 1971,
51, 100.
equiv.) was reduced with lithium borohydride (9 mg, 0.40
mmol, 2 equiv.) as described for 18b (method 1). The resultant
material was purified by flash chromatography on silica (ethyl
acetate), followed by diffusion of petroleum ether into a solu-
tion of the title product in ethyl acetate, to give 22 (56 mg, 68%)
as a cream solid; mp decomp. >85 ЊC; νmax(KBr)/cmϪ1 1541.0,
1624.0, 1664.5, 1735.8, 2958.6, 3267.2; δH(500 MHz; CDCl3;
Me4Si) 0.84 (6 H, m, CH(CH3)2), 1.44–1.59 (3 H, m, CH2CH ),
3.11 (2 H, m, CH2Ph), 3.69 (3 H, s, CO2CH3), 4.59 (3 H, m, Leu
αH and CH2OH), 5.05 (1 H, q, J = 7.7 Hz, Phe αH), 6.04 (1 H, t,
J = 2.9 Hz, pyrrole 4-H), 6.53 (1 H, m, pyrrole 3-H), 7.03 (1 H,
br s, NHCH), 7.16 (5 H, m, ArH ), 7.25 (1 H, br s, NHCH),
10.41 (1 H, br s, pyrrole NH ); δC(75 MHz; CDCl3; Me4Si) 22.0,
22.6, 24.7, 41.5 (CH2CH), 38.7 (CH2Ph), 50.8 (Leu αCH), 52.3
(CO2CH3), 54.5 (Phe αCH), 57.5 (CH2OH), 108.1 (pyrrole
4-C), 111.0 (pyrrole 3-C), 124.9 (pyrrole 2-C), 126.7, 128.4,
129.4, 136.3 (ArC), 136.7 (pyrrole 5-C), 161.2 (PyrCONH),
172.0 (CHCONH), 173.4 (CO2); m/z (ES) 438.2023 (Mϩ ϩ Na.
C22H29N3NaO5 requires 438.2005), 438 (100%), 416 (66) and
271 (63).
13 D. M. Wallace, S. H. Leung, M. O. Senge and K. M. Smith, J. Org.
Chem., 1993, 58, 7245.
14 M. K. A. Khan, K. J. Morgan and D. P. Morrey, Tetrahedron, 1966,
22, 2095.
15 Z.-Q. Tian, B. B. Brown, D. P. Mack, C. A. Hutton and
P. A. Bartlett, J. Org. Chem., 1997, 62, 514.
Acknowledgements
16 M. Tani, T. Ariyasu, C. Nishiyama, H. Hagiwara, T. Watanabe,
Y. Yokoyama and Y. Murakami, Chem. Pharm. Bull., 1996, 44, 48.
17 B. Imperiali and R. H. Abeles, Biochemistry, 1986, 25, 3760.
18 E. J. Breaux and M. L. Bender, FEBS Lett., 1975, 56, 81.
19 D. Le Nguyen and B. Castro, in Peptide Chemistry 1987, ed. T. Shiba
and S. Sakakibara, Protein Research Foundation, Osaka, 1988,
p 231.
This work was supported by a Royal Society of New Zealand
Marsden grant. D. C. M. was personally funded by a Bright
Futures Top Achiever Doctoral Scholarship. We also thank
Professor John Blunt for help with the difference 13C NMR
spectroscopy.
20 W. L. F. Armarego and D. D. Perrin, in Purification of Laboratory
Chemicals, Butterworth-Heinemann, Oxford, 1996.
21 R. J. P. Cannell, S. J. Kellam, A. M. Owsianka and J. M. Walker,
Planta Med., 1988, 54, 10.
22 I. Schechter and A. Berger, Biochem. Biophys. Res. Commun., 1967,
27, 157.
References
1 A. D. Abell and B. K. Nabbs, Bioorg. Med. Chem., 2001, 9, 621.
2 R. M. Stroud, Sci. Amer., 1974, 231, 74.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 0 3 – 2 1 1 0
2110