An improved stereocontrolled synthesis of pyochelin, siderophore of Pseudomonas aeruginosa and Burkholderia cepacia

Article abstract of DOI:10.1016/S0040-4020(99)00946-1

A considerably improved stereocontrolled synthesis of pyochelin, a hydroxyphenylthiazolinylthiazolidine type of siderophore common to most strains of Pseudomonas aeruginosa and Burkholderia cepacia is described. 2'- (2-Hydroxyphenyl)-2'-thiazoline-4'-carboxaldehyde, a key molecule involved in this synthesis has been prepared by reduction of 2'-(2-hydroxyphenyl)-2'- thiazoline-4'-(N-methoxy,N-methyl) carboxamide with lithium aluminium hydride. The aldehyde was further coupled with (R)-N-methylcysteine to yield pyochelin. Under the conditions reported, epimerization at the C-4' center was considerably diminished.

Full text of DOI:10.1016/S0040-4020(99)00946-1

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 249–256
An Improved Stereocontrolled Synthesis of Pyochelin,
Siderophore of Pseudomonas aeruginosa and Burkholderia cepacia
†
*
Adel Zamri and Mohamed A. Abdallah
´
´
´
Laboratoire de Chimie Microbienne, Departement des Recepteurs et Proteines Membranaires, UPR 9050 C.N.R.S,
´
´
Ecole Superieure de Biotechnologie de Strasbourg, Boulevard Sebastien Brant, F-67400, Illkirch, France
Received 10 August 1999; accepted 19 October 1999
Abstract—A considerably improved stereocontrolled synthesis of pyochelin, a hydroxyphenylthiazolinylthiazolidine type of siderophore
common to most strains of Pseudomonas aeruginosa and Burkholderia cepacia is described. 20-(2-Hydroxyphenyl)-20-thiazoline-40-carbox-
aldehyde, a key molecule involved in this synthesis has been prepared by reduction of 20-(2-hydroxyphenyl)-20-thiazoline-40-(N-methoxy,N-
methyl) carboxamide with lithium aluminium hydride. The aldehyde was further coupled with (R)-N-methylcysteine to yield pyochelin.
Under the conditions reported, epimerization at the C-40 center was considerably diminished. ᭧ 1999 Elsevier Science Ltd. All rights reserved.
Introduction
diastereoisomer 1b, they are (40R), (200S) and (400R). These
two diastereoisomeric forms were observed in synthetic
pyochelins as well.7
Pyochelin 1a is a hydroxyphenylthiazolinylthiazolidine
type of siderophore which was isolated for the first time
from iron-deficient cultures of Pseudomonas aeruginosa
ATCC 15692 by Liu and Shokrani.1 This siderophore was
later shown to be produced by a great number of strains of
Pseudomonas aeruginosa but also by many strains of
Burkholderia (ex Pseudomonas) cepacia,2 both species
being involved in severe lung infections occurring in cystic
fibrosis patients (Scheme 1).
Epimerization of the C-200 center has also been reported for
yersiniabactin 28 where it was assumed to occur as in
thiazolidine derivatives via a Schiff base.9–11
The structure of pyochelin was established later by Cox et
al.;3 it possesses a hydroxyphenylthiazoline group as well as
a carboxylic acid group, both involved in the chelation of
iron(III) with a 2:1 stoichiometry. Although its association
constant with iron(III) is very weak (estimated to be
5×105 MϪ1 l in methanol),4 the pyochelin-mediated iron
transport is very efficient.5
The first synthesis of pyochelin had been reported by
Ankenbauer et al.6 It is based on the condensation of (R)-
N-methylcysteine with aldehyde 4. This latter was prepared
after condensation of 2-hydroxybenzonitrile with (R)-
cysteine followed by reduction of acid 3a by thexylborane,
but the yield of this last step was unfortunately particularly
low (15%) due to the formation of a stable boron/thiazoline
complex (Scheme 2).
Natural pyochelin had initially been isolated as a mixture of
two steroisomers 1a and 1b resulting from epimerization of
the 200 center.6 The absolute configuration of the asymmetric
centers was deduced from the 3D structure of 4-methylpyo-
chelin determined by X-ray crystallography and by com-
parison of the 1H- and 13C NMR data of the same
4-methylpyochelin prepared by mutasynthesis. The absolute
configurations of the asymmetric centers of pyochelin 1a
are, respectively, (40R), (200R) and (400R) whereas for its
This was followed by cyclization of aldehyde 4 with N-
methylcysteine (prepared in two steps from (R)-cysteine)
which gave a mixture of diastereoisomers: pyochelin 1a
(40R,200R,400R), pyochelin 1b (40R,200S,400R) both corre-
sponding to natural pyochelin (after epimerization of the
C-200 center), together with pyochelin 1d (40S,200R,400R)
(no mention was made at that time of pyochelin 1c).7
Keywords: pyochelin; siderophores; thiazolines; thiazolidines.
* Corresponding author. Fax: ϩ33-3-88-65-52-34; e-mail: abdallah@
chimie.u-strasbg.fr
†
These results show how difficult it was to prepare pure
pyochelin in large amounts and preparing synthetic
Present address: FMIPA-UNRI, Kampus Bina Widya Km 12.5, Simpang
baru-Pekanbaru, Indonesia.
0040–4020/00/$ - see front matter ᭧ 1999 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(99)00946-1

250
A. Zamri, M. A. Abdallah / Tetrahedron 56 (2000) 249–256
Scheme 1.
pyochelin analogs has always been found to be a rather
difficult challenge, the only access to analogs being muta-
synthesis. Using this technique Ankenbauer et al. have
prepared three analogs of pyochelin: 5-fluoropyochelin 5,
4-methylpyochelin 6 and 6-azapyochelin 7. After incu-
bation of 13 analogs of salicylic acid with a mutant of
Pseudomonas aeruginosa SalϪ IA602, only three of them
(5-fluorosalicylic acid, 4-methylsalicylic acid and 3-hy-
droxypicolinic acid) gave the corresponding analog of pyo-
chelin. These analogs present biological activities quite
comparable to pyochelin in relation to the pyochelin-
mediated iron-transport in Pseudomonas aeruginosa.12
iron transport and the amount of these receptor proteins.16 In
addition, any study concerning these receptors implies
handling large amounts of pyochelin and/or some specially
designed analogs of pyochelin.
Therefore, after having carefully examined the previous
conditions reported for the synthesis of pyochelin, we intro-
duced several modifications leading to a considerably
improved synthesis of pyochelin which we describe in this
paper, and discuss the conditions leading to a complete
stereocontrolled synthesis.
Results and Discussion
Other analogs of pyochelin such as desferriferrithiocine
8
(from Streptomyces antibioticus),13 anguibactin
9
The key molecule for the synthesis of pyochelin 1 is alde-
hyde 4 which is then readily condensed with N-methyl-
cysteine. The aldehyde has already been obtained either
by reduction of the carboxylic acid group of thiazoline 3a
with thexylborane,17 or by reduction of ester 3b with
diisobutylaluminum hydride.7
(produced by Vibrio anguillarum)14 and yersiniabactin (or
yersiniophore) 2 excreted by Yersinia enterocolitica were
also reported.8,15
At present the pyochelin-dependent iron-transport system of
these bacteria is not very clearly understood. A few reports
have shown the existence of receptor proteins on the outer
membrane and also a direct relationship between the rate of
The first synthesis of thiazoline 3a has been described by
Mathur et al.18 condensing (R)-cysteine hydrochloride with
Scheme 2.

A. Zamri, M. A. Abdallah / Tetrahedron 56 (2000) 249–256
251
Scheme 3.
Scheme 4.
2-hydroxybenzonitrile in the presence of a base (piperidine
at pH ca. 9.0). Ankenbauer et al.6 then Rinehart et al.7 used
the same method but reported that the corresponding methyl
ester 3b had no optical activity.
The protection of the phenol group as its t-butyldi-
methylsilyl ether gave the protected aldehyde 11 with
very high yields (87%) (Scheme 4). Unfortunately the
deprotection of the TBDMS group at this stage with
tetrabutylammonium fluoride (TBAF) afforded a mixture
of colored compounds which after purification by
column chromatography on silica gel yielded the thiazole
aldehyde 12. Performing this deprotection after the final
step of the synthesis of pyochelin afforded a mixture of
colored compounds which were even more difficult to
purify.
With a view to understanding the reasons for this epimer-
ization we have prepared thiazoline 3a as described by
Mathur et al.18 and found a slight optical activity with an
[a]DˆϪ6.5Њ (cˆ0.9; MeOH/HCl 6N 2%). We performed
the same reaction according to Bergeron et al.19 in a 1:1
mixture of phosphate buffer 0.1M, pH 6.4 and methanol at
50ЊC for 4 days, and obtained thiazoline 3a with a yield as
high as 84% but with a very different optical rotation:
[a]DˆϪ190Њ (cˆ0.9; MeOH/HCl 6N 2%) (Scheme 3).
We assigned this large difference to the epimerization of
the C-40 asymmetric center caused by the very basic pH
conditions caused by piperidine (pH ca. 9.0).
The structure of thiazole aldehyde 12 was confirmed by
proton NMR and by mass spectrometry. The signals of
proton H-40 at 5.33 ppm and of the two protons H-50 at
3.39 and 3.69 ppm present in aldehyde 4 had completely
disappeared. Instead a signal at 8.1 ppm corresponding to
a single proton appears (H-50), whereas the aldehydic proton
shifts from 9.86 to 10.06 ppm. The FAB mass spectrum
shows a molecular peak at m=zˆ206 (MϩH)ϩ correspond-
ing to the aldehyde 12.
Esterification of thiazoline 3a with methanol/HCl at 50ЊC
under anhydrous conditions yielded ester 3b in 93% yield.
This was reduced at Ϫ78ЊC with DIBAH (2 equiv.) to yield
aldehyde 4 (61%). It was however accompanied by the
starting material in an 8:2 ratio (determined from TLC),
and during its purification by column chromatography on
silica gel epimerization of the carbon atom C-40 in the
position a to the aldehyde function occurred.20 Therefore
it was used without further purification.
We further attempted to prepare aldehyde 4 by oxidation of
alcohol 13 (itself obtained by reduction of ester 3b with
sodium borohydride in 60% yield). However this oxidation
step performed by different methods according to Swern,
Moffat or Collins always yielded aldehyde 12 (Scheme 5),
showing that aromatization of alcohol 13 is faster than its
oxidation into aldehyde 4.7,21
Attempts were made to improve the yield by modification of
the solvent, the amount of DIBAH and the reaction tempera-
ture: raising the temperature above 50ЊC or using more than
2 equiv. of DIBAH considerably increased the amount of
alcohol 10.
For these reasons we used a method reported in 1983 by
Fehrentz and Castro22 for the preparation of aldehydes
derived from protected amino acids based on the reduction
Scheme 5.

252
A. Zamri, M. A. Abdallah / Tetrahedron 56 (2000) 249–256
Scheme 6.
Scheme 7.
Scheme 8.
of the corresponding N-methoxy-N-methyl hydroxamates
using excess lithium aluminum hydride (5 equiv.) at 0ЊC.
These aldehydes were obtained with excellent yields with-
out any epimerization.
have quantitatively derivatized the acids 1a–1d as their
methyl esters 15a–15d using trimethylsilyl diazomethane,27
these esters being easier to handle and purify than the
corresponding acids 1a–1d7 (Scheme 8).
We have adapted their reaction conditions to prepare alde-
hyde 4, and also modified the conditions of preparation of
hydroxamate 3c and the temperature of reduction of this
latter.
The chemical ionization mass spectrum of the mixture 15a–
15d showed fragmentations similar to those observed by
Cox et al.3
The proton NMR spectrum of the mixture of esters
presented great similarities with the spectrum reported by
Rinehart et al. (Table 1).7 In our hands a mixture of four
Diethyl cyanophosphonate (DECP) was used as a coupling
agent for acid 3a and N,O-dimethylhydroxylamine,23 and
afforded hydroxamate 3c in excellent yield (94%).
Table 1. Chemical shifts of the protons H-40, H-200 and H-400 of the diaster-
eoisomers of pyochelin methyl esters 15a–15d. The values in brackets
represent the coupling constants expressed in hertz
In order to prevent the secondary reaction yielding alcohol
13, the reduction of hydoxamate 3c was performed with
3 equiv. of lithium aluminum hydride (instead of 5 equiv.)
and at Ϫ20ЊC (instead of 0ЊC). Under these conditions the
reduction was complete after 20 min, yielding aldehyde 4
with no trace of starting material and no trace of alcohol 13
(Scheme 6).
Aldehyde 4 was condensed with (R)-N-methylcysteine
hydrochloride24 in a 4:1 mixture of ethanol and water in
the presence of potassium acetate. The yield of this step is
70%, and pyochelin was obtained as a mixture of four
diastereoisomers 1a–1d (Scheme 7).
H-40
H-200
H-400
15a
15b
15c
15d
5.08 t,d (8.7; 5.2)
5.08 t,d (8.7; 5.2)a
4.82 q (8.1)
4.51 d (5.2)
4.53 d (5.2)a
4.56 d (6.9)
4.56 d (6.9)a
5.02 d (4.7)
5.04 d (4.6)a
4.23 d (7.7)
4.24 d (7.7)a
3.65 d,d (8.7; 6.6)
3.65 d,d (9.1; 6.3)a
4.09 t (6.2)
4.92 q (8.2)a
4.09 t (6.0)a
5.13 t,d (9.3; 4.7)
5.12 t,d (9.1; 4.6)a
4.80 q (8.1)
4.14 d,d (4.8; 2.2)
4.13 d.d (6.4; 2.1)a
3.83 t (6.8)
The structure of pyochelins 1a–1d and the absolute config-
uration of the asymmetric centers were determined by
comparison of the proton NMR spectra and the mass spectra
of their methyl esters, with the corresponding reference
spectra described by Rinehart et al.7 In this respect we
4.79 q (8.0)a
3.83 t (6.8)a
a Data reported by Rinehart et al.7

A. Zamri, M. A. Abdallah / Tetrahedron 56 (2000) 249–256
253
diastereoisomers was obtained: 15a (40R,200R,400R), 15b
(40R,200S,400R), 15c (40S,200S,400R) and 15d (40S,200R,400R),
but in a ratio of 2:1:2:5, respectively (calculated from the
signals of proton H-40, H-200 and H-400). Rinehart et al.7 had
indicated a ratio of 4:1:1:4 for the corresponding acids
(determined by TLC). It is very likely that the starting
compound they used for the condensation reaction was
already epimerized at C-40 before the DIBAH reduction
affording aldehyde 4.
—a 0.4% solution of 2,4-dinitrophenylhydrazine in 2M
hydrochloric acid.
—a 10% solution of phosphomolybdic acid in ethanol.
The anhydrous solvents were obtained by distillation from
an appropriate drying agent: phosphoric anhydride for
methylene chloride and dimethylformamide; calcium
hydride for toluene, ether and acetonitrile; sodium for
methanol. The oxygen- or humidity-sensitive reactions
were performed under argon.
From these results, we have drawn the following
conclusions:
20-(2-Hydroxyphenyl)-20-thiazoline-40-carboxylic acid 3a
1. There is an epimerization at the C-40 (30%), occurring
probably during the condensation reaction of aldehyde 4
with N-methylcysteine in the presence of potassium acetate.
2. The diastereoisomers possessing the (40S) configuration
(15cϩ15d) represent 70% of the mixture, which is not
surprising since the starting product is (R)-cysteine and
that all care was taken to prevent epimerization of this
asymmetric center (Rinehart et al.7 had a total epimerization
of this center).
According to Mathur et al.18 2-Hydroxy-2-benzonitrile
(6.0 g, 50 mmol), (R)-cysteine hydrochloride (8.0 g,
50 mmol) and sodium bicarbonate (4.2 g, 50 mmol) were
dissolved in absolute ethanol (50 ml) and heated under
reflux for 30 min. The pH was adjusted to 9.0 by addition
of piperidine and the suspension stirred ovenight. Hot water
(10 ml) was added and the pH was carefully adjusted to 4.0
by addition of acetic acid. After filtration, the precipitate
was washed with cold water, then with ethanol, dried
under reduced pressure to yield a clear yellow powder
(9.48 g, 42.5 mmol, 85%).
3. The diastereoisomers possessing the (200R) configuration
(15aϩ15d) also represent 70% of the mixture and not
50% as would be expected since the C-200 asymmetric
center is formed after addition of N-methylcysteine on a
planar aldehydic carbonyl group. This result indicates that
this last step is performed with asymmetric induction
created by the C-40 center, which is not finally so surprising
since all the molecules participating in the reaction are
chiral.
Mp: 268–270ЊC (lit. 269ЊC18).
[a]DˆϪ6.5Њ (c 0.9 methanol/HCl 6N 2%).
According to Bergeron et al.19 2-Hydroxy-2-benzonitrile
(5.33 g, 44.8 mmol) and (R)-cysteine (10.83 g, 89.5 mmol)
were dissolved in a 1:1 (v/v) mixture of methanol and phos-
phate buffer pH 6.4 (200 ml). The mixture was stirred for
4 days at 50ЊC, filtered and concentrated under reduced
pressure. Water (50 ml) was added and the solution
extracted with methylene chloride (3×25 ml). The organic
phase was washed with water (2×10 ml), then with brine
(10 ml) and dried over sodium sulfate. After evaporation of
the solvent under reduced pressure a yellow solid was
obtained (8.4 g, 37.6 mmol, 84%).
4. The esterification reaction having no effect on the chiral
centers, pyochelins 1a–1d are obtained in the same propor-
tions as their corresponding methyl esters.
Experimental
General indications
The melting points (mps) were measured in a capillary tube
using a Bu¨chi SMP-20 instrument and are not corrected. The
proton NMR spectra were measured on Bruker AC-200
(200 MHz), AM-400 (400 MHz), ARX-500 (500 MHz),
AMX 500 (500 MHz) or with gradient instruments. The
mass spectra (MS) were determined using either an LKB
9000S or a Thomson THN 208 spectrometer. The FAB (fast
atom bombardment) spectra were performed using a ZAB-
HF instrument (VG Analytical, Manchester, UK). The rela-
tive intensities of the signals are stated in parentheses after
the mass m/z of the fragment considered. Optical rotations
were determined with a Perkin–Elmer 241 MC instrument.
The elemental analyses (microanalyses) were performed in
Strasbourg by the Service de Microanalyse of the Institut
Charles Sadron. The column chromatographies were
performed using Merck 9385 silica gel (Darmstadt,
Germany) 40–63 mm mesh. Thin layer chromatographies
(TLCs) were performed using silica gel analytical plates
Merck 5715 (F254) of 0.25 mm thickness. The detection on
TLC plates was performed by UV light at 254 or 365 nm or
using a spray and heating the plates. The sprays used were:
Mp: 116–117ЊC.
[a]DˆϪ190Њ (c 0.9 methanol/HCl 6N 2%).
MS (EI): m=zˆ223 (55); 178 (100); 119 (17.2); 59 (38.04).
1H NMR (200 MHz; CDCl3): 3.69 (m; 2-H, H-50); 5.42 (t;
Jˆ8.12 Hz, H-40); 6.89 (t; Jˆ7.5 Hz, 1-H, H-3); 7.03 (d;
Jˆ8.5 Hz, 1-H, H-4); 7.43 (m; 5-H, H-5 and H-6).
13C NMR (100 MHz; CDCl3): 33.6 (C-50); 76.3 (C-40);
115.84 (C-1); 117.38 (C-3); 119.09 (C-5); 130.8 (C-6);
133.8 (C-4); 159.12 (C-20); 174.7 (C-2); 175.14 (COOH).
Calculated for C10H9NO3S C% 53.81 H% 4.04 N% 6.28 O% 21.52
Found
C% 53.84 H% 3.98 N% 6.23 O% 21.34
20-(2-Hydroxyphenyl)-20-thiazoline-40-carboxaldehyde 4
—a 0.3% solution of ninhydrin in a mixture of butanol/
acetic acid (97:3, v/v).
Methyl-20-(2-hydroxyphenyl)-2-thiazoline-40-carboxy-
late 3b. To a solution of anhydrous hydrochloric acid in

254
A. Zamri, M. A. Abdallah / Tetrahedron 56 (2000) 249–256
methanol (prepared from 20 ml methanol and 5 ml acetyl
chloride at 0ЊC), thiazoline 3a (2.95 g, 13.2 mmol) was
added and the mixture heated at 50ЊC for 24 h. After
evaporation of the solvent, 50 ml ethyl acetate were
added, and the organic phase washed with water
(2×10 ml), then with brine, dried over sodium sulfate,
filtered and evaporated under reduced pressure to yield a
yellow oil (2.91 g, 12.28 mmol, 93%).
20-(2-t-Butyldiphenylsilyloxyphenyl)-20-thiazoline-40-
carboxaldehyde 11
To a solution of ester 3b (459 mg, 1.3 mmol) in anhydrous
toluene (5 ml) at Ϫ78ЊC under argon, a 1.5M solution of
DIBAH in toluene (1.3 ml, 1.95 mmol) was added dropwise
during 8 min with a syringe. After 1 h, 2 ml methanol then a
saturated solution of aqueous ammonium chloride (10 ml)
was added. The temperature was raised to 0ЊC and water
(20 ml) added. The organic phase was conserved and the
aqueous phase was extracted with ether (3×15 ml). All the
organic phases were pooled and washed with brine and dried
over sodium sulfate. After evaporation of the solvents a
colorless oil was obtained (362 mg, 1.13 mmol, 87%).
Rf 0.51 (methylene chloride–acetone 75:25, v/v).
1H NMR (200 MHz; CDCl3): 3.58 (m; 2-H, H-50); 3.81 (s;
1-H, OMe); 5.12 (t; Jˆ9.28 Hz, H-40); 6.87 (t; Jˆ8.3 Hz,
1-H, H-3); 6.97 (d; Jˆ7.4 Hz, 1-H, H-5); 7.28 (dd; Jˆ1.3 and
7.6 Hz, 1-H, H-4); 7.78 (d,t; Jˆ1.3 and 7.7 Hz, 1-H, H-6).
13C NMR (100 MHz; CDCl3): 33.6 (C-50); 52.8 (C–OMe);
76.6 (C-40); 116.2 (C-1); 118.9 (C-3); 119.09 (C-5); 130.6
(C-6); 133.5 (C-4); 159.2 (C-20); 174.3 (C-2); 170.6
(COOH).
Rf 0.45 (hexane–ethyl acetate 7:3, v/v; detection either with
the phosphomolybdic acid or with the 2,4-dinitro-
phenylhydrazine sprays).
1H NMR (200 MHz; CDCl3): 0.13 (s; 6-H, Si(Me)2); 0.89
(s; 9-H, t-butyl); 3.57–3.80 (m; 2-H, H-50); 5.25 (dd;
Jˆ6.15 and 9.8 Hz, 1-H, H-40); 6.85–7.03 (m; 2-H, H-3
and H-5); 7.13–7.4 (m; 2-H, H-4, H-6).
Reduction of ester 3b with DIBAH into aldehyde 4
To a solution of ester 3b (864 mg, 3.64 mmol) in anhydrous
ether (20 ml) at Ϫ78ЊC under argon, a 1.5M solution of
DIBAH 1.5M in toluene (4.5 ml, 6.75 mmol) was added
dropwise during 8 min with a syringe. After 2 h, methanol
(3 ml) then a saturated solution of ammonium chloride
(17 ml) were added. The temperature was raised to 0ЊC
and water (30 ml) was added. The organic phase was
conserved and the aqueous phase was extracted with ether
(3×15 ml). All the organic phases were pooled and washed
with brine and dried over sodium sulfate. After evaporation
of the solvents, a yellow resin was obtained (459 mg,
2.21 mmol, 61% as a 8:2 mixture of aldehyde 4 and ester
3b according to NMR).
Deprotection of the TBDMS function of aldehyde 11
To a solution of aldehyde 11 (233 mg, 0.73 mmol) in tetra-
hydrofuran (3 ml) at 0ЊC, 1.4 ml of a 1M solution of tetra-
butylammonium fluoride in tetrahydrofuran (1.4 mmol),
was added dropwise. After 30 min, the solvents were evapo-
rated and the dark brown residue was purified by column
chromatography on silica gel (3 g) and eluted with methyl-
ene chloride to yield 20-(2-hydroxyphenyl)-20-thiazole-40-
carboxaldehyde 12 (59 mg, 0.29 mmol, 40%).
Mp: 134–136ЊC (lit. 135–135.5ЊC)21.
Rf 0.75 (methanol–methylene chloride 9:1, v/v).
FAB-MS: 206.1 (MϩH)ϩ.
For the characteristics of aldehyde 4 see the following.
Attempts to synthesize aldehyde 4 after protection of the
phenol group as a t-butyldiphenylsilyl ether
1H NMR (200 MHz; CDCl3): 6.88–7.08 (m; 2-H, 3-H,
H-4); 7.3–7.6 (m; 2-H, H-5 and H-6); 8.1 (s; 1-H, H-50);
10.02 (s; 1-H, aldehydic proton); 11.5 (s; 1-H, –OH).
Methyl-20-(2-t-butyldiphenylsilyloxyphenyl)-20-thiazo-
line-40-carboxylate 10. Imidazole (782 mg, 11.5 mmol)
and t-butyldiphenylsilyl chloride (832 mg, 5.52 mmol)
were added to a solution of ester 3b (1.098 g, 4.6 mmol) in
dimethylformamide (5 ml). The mixture was stirred at room
temperature for 24 h. Water (25 ml) was added and the
mixture extracted with ethyl acetate (3×10 ml). The organic
phases were pooled, washed with brine (3×10 ml), dried over
sodium sulfate and evaporated under reduced pressure. The
crude compound (1.8 g) was purified by column chromato-
graphy on silica gel (50 g), and eluted with a 9/1 mixture of
hexane–acetone 9/1 (v/v). After evaporation, a colorless oil
was obtained (1.4 g, 4 mmol, 87%).
Attempts to synthesize aldehyde 4 by oxidation of
alcohol 13
20-(2-Hydroxyphenyl)-200-thiazoline-40-methanol 13.
A
mixture of ester 3b (894 mg, 3.77 mmol), sodium boro-
hydride (371 mg, 9.77 mol) and tetrahydrofuran (12 ml)
was heated under reflux, and methanol (3 ml) was added
within 15 min. The mixture was cooled down to 15ЊC,
water (1 ml) was added dropwise, the stirring was continued
for 10 min and after another addition of water (15 ml), it was
extracted with ether (3×15 ml). The organic phase was
washed with brine and dried over sodium sulfate. After filtra-
tion and evaporation of the solvents under reduced pressure
an oil was obtained (690 mg). It was purified by column
chromatography on silica gel (20 g), and afforded after
elution with methylene chloride–methanol 9:1 (v/v) and
evaporation, a yellowish oil (474 mg, 2.27 mmol, 60%).
Rf 0.66 (hexane–ethyl acetate 7:3, v/v).
1H NMR (200 MHz; CDCl3): 0.26 (s; 6-H, Si(Me)2); 0.97
(s; 9-H, Si-t-butyl); 3.57 (m; 2-H, H-50); 3.81 (s; 1-H, OMe);
5.12 (t; Jˆ9.36 Hz, H-40); 6.85 (d,d; Jˆ1 and 8.23 Hz, 1-H,
H-3); 6.94 (t; Jˆ9.2, 1-H, H-5); 7.28 (m; 1-H, H-4); 7.75
(dd; Jˆ1.8 and 7.8 Hz, 1-H, H-6).

A. Zamri, M. A. Abdallah / Tetrahedron 56 (2000) 249–256
255
Rf 0.43 (methylene chloride–methanol 9:1, v/v).
Reduction of hydroxamate 3c with lithium aluminum
hydride. Hydroxamate 3c (231 mg, 0.87 mmol) was
dissolved in ether (10 ml) and the solution cooled down to
Ϫ20ЊC. Lithium aluminum hydride (94 mg, 2.46 mmol)
was added. After 20 min at Ϫ20ЊC (until the starting
material had disappeared) methanol (1 ml) then a saturated
aqueous solution of ammonium chloride (10 ml) and finally
a 5% (v/v) solution of sulfuric acid (5 ml) were successively
added. The organic phase was conserved and the aqueous
phase extracted with ethyl acetate (3×10 ml). The
organic phases were pooled, washed with brine, dried over
sodium sulfate and evaporated under reduced pressure. 20-
(2-Hydroxyphenyl)-20-thiazoline-40-carboxaldehyde 4 (163 mg,
0.79 mmol, 90%) was obtained as a yellow foam.
1H NMR (200 MHz; CD3OD): 3.32–3.52 (m; 2-H, H-50);
4.82 (m; 1-H, H-40); 6.85–6.94 (m; 2-H, H-3 and H-5);
7.13–7.4 (m; 2-H, H-4, H-6).
Oxidation of alcohol 13. To a solution of oxalyl chloride
(0.24 ml, 2.6 mmol) and DMSO (0.34 ml, 4.4 mmol) in
methylene chloride (30 ml) at Ϫ78ЊC, a solution of alcohol
13 (244 mg, 1.17 mmol) in methylene chloride (30 ml) was
added dropwise in 20 min. Triethylamine (2 ml, 14 mmol)
was then added, the temperature raised to 25ЊC before
addition of water (20 ml). The organic phase was washed
with brine, dried over sodium sulfate, evaporated to yield
240 mg of a yellow solid product. It was purified by column
chromatography on silica gel (5 g) and afforded after elution
with a mixture of methylene chloride and ether 99:1
(v/v), 20-(2-hydroxyphenyl)-20-thiazole-40-carboxaldehyde
12 (96 mg, 0.46 mmol, 39%).
Rf 0.45 (methylene chloride–methanol 9:1, v/v; detection
either with the phosphomolybdic acid or with the 2,4-di-
nitrophenylhydrazine sprays).
1H NMR (200 MHz; CDCl3): 3.39 (t; Jˆ10 Hz, 2-H, H-50);
3.69 (dd; Jˆ7.6 and 11.4 Hz, H-50); 5.33 (dd; Jˆ7.8 and
9.1 Hz, 1-H, H-40); 7.50–7.62 (m; 3-H, H-4, H-5 and H-6);
9.84 (s; 1-H, aldehydic proton); 12.2 (s; 1-H, –OH).
For the characteristics of aldehyde 12 see above.
Synthesis of aldehyde 4 by reduction of hydroxamate 3c
(R)-N-Methylcysteine 14 was prepared according to the
method of Blondeau et al.25 by reduction of (R)-4-thiazo-
lidinyl carboxylic acid using sodium in liquid ammonia at
Ϫ78ЊC (68%). This latter acid was synthesized by the
method of Ratner et al.26 by condensation of cysteine hydro-
chloride with formaldehyde in 73% yield.
20-(2-Hydroxyphenyl)-20-thiazoline-40-(N-methoxy,N-
methyl) carboxamide 3c. To a solution of thiazoline 3a
(2.23 g, 10 mmol), N,O-dimethylhydroxylamine hydro-
chloride (1.07 g, 11 mmol) and diethylcyanophosphonate
(2.0 g, 11 mmol) in DMF (15 ml) at 0ЊC, diisopropyl-
ethylamine (3.65 ml, 21 mmol) was added. The mixture
was stirred for 30 min at 0ЊC, then another 30 min at
room temperature. A 3% solution of citric acid (90 ml)
was added and the mixture filtered. The precipitate was
conserved and the filtrate extracted twice with a mixture
of toluene and ethyl acetate (1:1, v/v). The organic phase
was washed with water, then with brine, dried over
sodium sulfate, filtered and evaporated. The residue
and the former precipitate were pooled and dried
under reduced pressure to yield a yellow powder (2.33 g,
8.9 mmol, 89%) which was crystallized from methylene
chloride.
Synthesis of pyochelin 1
20-(2-Hydroxyphenyl)-20-thiazoline-40-carboxaldehyde
4
(154 mg, 0.74 mmol) was dissolved in a mixture of ethanol
(15 ml) and water (4 ml) and potassium acetate (0.4 g,
5.1 mmol) and (R)-N-methylcysteine (0.2 g, 1.12 mmol)24
were added. After 24 h, water (50 ml) was added and the
mixture extracted with hexane (3×10 ml). The aqueous
phase was acidified to pH 5.0, extracted with ethyl acetate
(3×10 ml), washed with brine and dried over sodium
sulfate. After filtration and removal of the solvents a yellow
foam was obtained (168 mg, 0.52 mmol, 70%).
Rf 0.68 (methylene chloride–methanol 9:1, v/v).
Mp: 46–48ЊC.
I
II
III
IV
MS (CI, NH3ϩ): (MϩH)ϩˆ267.2 (28); 237.2 (5); 236.2 (12);
Rf 0.42 0.35 0.26 0.16 (Butanol:water:methanol:hexane 2:2:1:1)
178 (100).
Detection by UV light (254 and 366 nm): Compounds I and
IV gave a bright blue fluorescence whereas II and III were
less intensely fluorescent. The four compounds gave a
brown color after spraying with a 0.1M solution of ferric
chloride in methanol.
1H NMR (200 MHz; CDCl3): 3.3 (s; 3-H, CH3); 3.48 (dd;
Jˆ9.2 and 11 Hz, 2-H, H-50); 3.81 (s; 1-H, OMe); 5.70 (t;
Jˆ8.8 Hz, 1-H, H-40); 6.88 (t; Jˆ7.7 Hz, 1-H, H-3); 6.98 (d;
Jˆ8.3 Hz, 1-H, H-5); 7.34 (dd; Jˆ1.4 and 7.3 Hz, 1-H,
H-4); 7.42 (d,t; Jˆ1.6 and 7.8 Hz, 1-H, H-6).
13C NMR (100 MHz; CDCl3): 32.64 (C-50); 32.92
(N-CH3); 61.87 (O-CH3); 74.7 (C-40); 116.2 (C-1); 117.2
(C-3); 119.02 (C-5); 130.6 (C-6); 133.5 (C-4); 159.2
(C-20); 174.3 (C-2).
Preparation of the methyl esters of pyochelin 15
Pyochelin (123 mg, 0.38 mmol) was dissolved in 5 ml of a
4:1 mixture of benzene and methanol (v/v), and trimethyl-
silyldiazomethane was added (0.25 ml, 0.5 mmol). After
15 min, the solvents were evaporated under reduced
pressure and the residue was purified by column chromato-
graphy on silica gel (4 g). The product was eluted with
Calculated for C12H14N2O3S C% 54.12 H% 5.30 N% 10.52 O% 18.02
Found
C% 54.10 H% 5.33 N% 10.52 O% 18.13

256
A. Zamri, M. A. Abdallah / Tetrahedron 56 (2000) 249–256
1H-NMR (200 MHz; CDCl3).
Protons
I
II
III
IV
3
4
7.00 (d; 10 Hz)
7.32 (m)
7.00 (d; 10 Hz)
7.32 (m)
7.00 (d; 10 Hz)
7.32 (m)
7.00 (d; 10 Hz)
7.32 (m)
5
6.83 (t; 7.56 Hz)
7.40 (m)
5.08 (d,t, 5.2, and 8.7)
3.51 (m)
4.51 (d; 5.2)
3.65 (dd; 6.6 and 8.7)
3.01–3.1 (m)
2.58 (s)
6.83 (t; 7.56 Hz)
7.40 (m)
6.83 (t; 7.56 Hz)
7.40 (m)
5.13 (d,t; 4.7, and 9.3)
3.41–3.47 (m)
5.02 (d;4.7)
4.14 (dd; 2.2 and 4.8)
3.28 (m)
2.58(s)
6.83 (t; 7.56 Hz)
7.40 (m)
4.80 q (8.1)
3.51 (m)
4.23 (d;7.7)
3.83 (t; 6.8)
3.24–3.36 (m)
2.58(s)
6
40
4.82 q (8.1)
3.42–3.46 (m)
4.56 (d; 6.9)
4.09 (t; 6.2)
3.18–3.24 (m)
2.58(s)
50
200
400
500
N-CH3
O-CH3
3.78 (s)
3.78 (s)
3.78 (s)
3,78 (s)
9. Poncticelli, F.; Marinello, E.; Missale, M. C. Org. Magn. Res.
1982, 20, 138–140.
a mixture of hexane–ethyl acetate, and after evaporation of
the solvents, a yellow foam (127 mg, 0.375 mmol, 98%) as a
mixture of four diastereoisomers: I, II, III, IV, was obtained.
10. Somogy, L. Liebigs Ann. Chem. 1985, 657–876.
11. Chiba, T.; Sakaki, J.; Kobayashi, S.; Furuya, T.; Inukai, N.;
Kaneko, C. Chem. Pharm. Bull. 1989, 37, 877–882.
12. Ankenbauer, R. G.; Staley, A. L.; Rinehart, K. L.; Cox, C. D.
Proc. Natl Acad. Sci. USA 1991, 88, 1878–1882.
MS (CI, NH3ϩ): m/z 339.1 [MϩH]ϩ (100); 279.1
[MϪCOOMe] (4); 178.1 [C9H8NOS]ϩ (16); 160.1
[(C6H10NO2S]ϩ (34); 101.1 [C4H6NS]ϩ (4).
¨
13. Naegeli, H. U.; Zahner, H. Helv. Chim. Acta 1980, 63, 1400–
1406.
14. Jalal, M. A. F.; Hossain, M. B.; van der Helm, D.; Sanders-
Loehr, J.; Actis, L. A.; Crosa, J. H. J. Am. Chem. Soc. 1989, 111,
292–296.
15. Chambers, C. E.; McIntyre, D. D.; Mouck, M.; Sokol, P. A.
BioMetals 1996, 9, 157–167.
16. Heinrichs, D. A.; Young, L.; Poole, K. Infect. Immun. 1991,
59, 3680–3684.
17. Brown, C. B.; Heim, P.; Yoon, N. M. J. Org. Chem. 1972, 37,
2942–2950.
18. Mathur, K. B.; Iyer, R. N.; Dhar, M. L. J. Sci. Ind. Res. 1962,
21B, 34–37.
19. Bergeron, R. J.; Wiegand, J.; Dionis, J. B.; Egli-Karmakka,
M.; Frei, J.; Huxley-Tencer, A.; Peter, H. J. Med. Chem. 1991, 34,
2072–2078.
Acknowledgements
`
`
We thank the Ministere des Affaires Etrangeres for a grant
for one of us (A.Z.), and the Centre National de la
Recherche Scientifique for financial support (ARI Chimie-
Biologie). We also express our thanks to Roland Graff for
the determination of NMR, and to Dr J. Philip Poyser
(AstraZeneca Pharmaceuticals, Alderley Park, Cheshire)
for assistance in checking the manuscript.
20. Jurczak, J.; Golebiowski, A. Chem. Rev. 1989, 89, 149–164.
21. Cuppels, D. A.; Stipanovic, R. D.; Stoessl, A.; Stothers, J. B.
Can. J. Chem. 1987, 65, 2126–2130.
References
22. Fehrentz, J. A.; Castro, B. Synthesis 1983, 676–678.
23. Takuma, S.; Hamada, Y.; Shioiri, T. Chem. Pharm. Bull. 1982,
30, 3147–3153.
24. (R)-N-Methylcysteine 14 was prepared according to the
method of Blondeau et al.25 by reduction of (R)-4-thiazolidinyl
carboxylic acid using sodium in liquid ammonia at Ϫ78ЊC
(68%). This latter acid was synthesized by the method of Ratner
et al.26 by condensation of cysteine hydrochloride with formalde-
hyde in 73% yield.
25. Blondeau, P.; Berse, C.; Gravel, D. Can. J. Chem. 1967, 45,
49–52.
26. Ratner, S.; Clarke, H. T. J. Am. Chem. Soc. 1937, 59, 200–206.
27. Hashimoto, N.; Aoyama, T.; Shiori, T. Chem. Pharm. Bull.
1981, 29, 1475–1478.
1. Liu, P. V.; Shokrani, F. Infect. Immun. 1978, 22, 878–890.
2. Sokol, P. A. J. Clin. Microbiol. 1986, 23, 560–562.
3. Cox, C. D.; Rinehart, K. L.; Moore, M. L.; Cook, J. C. Proc.
Natl Acad. Sci. USA 1981, 78, 4256–4260.
4. Cox, C. D.; Graham, R. J. Bacteriol. 1979, 137, 357–364.
5. Cox, C. D. J. Bacteriol. 1980, 142, 581–587.
6. Ankenbauer, R. G.; Toyokuni, T.; Staley, A. L.; Rinehart, K. L.
J. Bacteriol. 1988, 170, 5344–5351.
7. Rinehart, K. L.; Staley, A. L.; Wilson, S. R.; Ankenbauer, R. G.;
Cox, C. D. J. Org. Chem. 1995, 60, 2786–2791.
¨
8. Drechsel, H.; Stephan, H.; Lotz, R.; Haag, H.; Zahner, H.;
Hantke, K.; Jung, G. Liebigs Ann. Chem. 1995, 1727–1733.