Synthesis of pyochelin-norfloxacin conjugates

Article abstract of DOI:10.1016/j.bmcl.2006.11.005

Using synthetic functionalized analogues of pyochelin, a siderophore common to several pathogenic Pseudomonas and Burkholderia species, four fluoroquinolone-pyochelin conjugates were efficiently synthesized and evaluated for their biological activities.

Full text of DOI:10.1016/j.bmcl.2006.11.005

Bioorganic & Medicinal Chemistry Letters 17 (2007) 640–644
Synthesis of pyochelin–norfloxacin conjugates
´
´
Freddy Rivault, Clemence Liebert, Alain Burger, Franc¸oise Hoegy,
Mohamed A. Abdallah, Isabelle J. Schalk and Gae¨tan L. A. Mislin*
Me´taux et Microorganismes: Chimie, Biologie et Applications, UMR 7175-LC1 Institut Gilbert Laustriat CNRS-Universite´ Louis
Pasteur (Strasbourg I), ESBS, Boulevard Se´bastien Brant, F-67400 Illkirch, France
Received 6 October 2006; revised 31 October 2006; accepted 1 November 2006
Available online 6 November 2006
Abstract—Using synthetic functionalized analogues of pyochelin, a siderophore common to several pathogenic Pseudomonas and
Burkholderia species, four fluoroquinolone-pyochelin conjugates were efficiently synthesized and evaluated for their biological
activities.
Ó 2006 Elsevier Ltd. All rights reserved.
Pseudomonas aeruginosa and Burkholderia cepacia, two
opportunistic Gram-negative bacteria, are causes of se-
vere and often lethal lung infections especially for cystic
fibrosis affected or immunocompromised patients.1 The
low outer membrane permeability of these bacteria
and the antibiotic mediated selective pressure raised an
increasing number of multiresistant strains. Therefore,
it appears necessary to develop new therapeutic strate-
gies against these microorganisms.
excrete molecules called siderophores, able to chelate ir-
on(III) and compete with the host for this element.2 In
Gram-negative bacteria, the siderophore-iron(III) com-
plex is recognized by a specific outer membrane receptor
and then transported into the bacterial cell by a process
driven by the proton motive force of the inner
membrane.3,4 The iron uptake pathways can be used
in Trojan horse strategy in order to counter the low
outer membrane permeability of P. aeruginosa and
B. cepacia.5,6
In this context, iron uptake pathways developed by
P. aeruginosa and B. cepacia might become interesting
targets for new antibiotics against these two pathogenic
bacteria. Indeed, during infection, the bacterial prolifer-
ation requires high quantities of various nutrients.
Among them iron is one of the most crucial. Although
this metal is abundant in the earth crust, its bioavailabil-
ity is limited by the low solubility of iron(III) at physio-
logical pHs. Moreover, even if higher eukaryotes
contain substantial amounts of this crucial element,
the latter is so tightly associated with transport and stor-
age proteins that it is not freely available for pathogens.
Consequently, the level of free iron in biological fluids is
usually estimated to only 10À18 M. To overcome this vi-
tal problem, pathogenic microorganisms synthesize and
A few years ago we used pyoverdine, the chromopeptidic
siderophore of P. aeruginosa, as a vector to transport a
fluoroquinolone into the bacterial cells. The correspond-
ing adducts were shown to be transported inside the cells
and showed promising bactericidal activity.6 However
this strategy was of limited application since each P.
aeruginosa strain produces its own pyoverdine along
with a specific transporter. Moreover, very few cross-
feeding have been observed between these different
pyoverdines and their corresponding transporters. In
this respect and since all P. aeruginosa strains and almost
all B. cepacia genomovars excrete a common sidero-
phore, the pyochelin 1 (Fig. 1),7,8 this latter siderophore
OH
Keywords: Pseudomonas; Burkholderia; Iron uptake; Cystic fibrosis;
Siderophore; Pyochelin; Fluoroquinolone.
COOH
H
N
N
*
Corresponding author. Tel.: +33 3 90244727; fax: +33 3 90244829;
e-mail: mislin@esbs.u-strasbg.fr
´
Present address: Laboratoire de Chimie des Molecules Bioactives et
S
S
1
ˆ ´
des Aromes, UMR 6001 CNRS, Universite de Nice Sophia Antip-
olis, Parc Valrose, F-06108 Nice Cedex 2, France.
Figure 1. Structure of pyochelin 1.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.005

F. Rivault et al. / Bioorg. Med. Chem. Lett. 17 (2007) 640–644
641
appeared to be more suitable to vectorize antibiotics the
more so as synthetic pyochelin as well as its functional-
ized analogues are now readily available.9
labile one which could release the norfloxacin under
physiological conditions.
The synthesis of the fluoroquinolone fitted with a stable
spacer arm starts with the condensation of norfloxacin
11 with succinic anhydride.6 The free succinic carboxyl-
ate of the resulting diacid 12 was then regioselectively
converted to its pentafluorophenyl ester 13 using EDCI
(Scheme 2).15,16
The present article describes the synthesis of pyochelin–
norfloxacin conjugates and the first evaluation of their
bactericidal activities.
The phenol and the carboxylate function of pyochelin 1
are known to be required for iron chelation and recog-
nition by FptA receptor.10,11 Therefore, these two chem-
ical functions appeared to be not suitable for the
conjugation with an antibiotic. In order to minimize
interactions with these crucial groups, we synthesized
and described three pyochelin analogues, functionalized
with a Teoc protected amine function in position 5 of
the aromatic ring (Scheme 1: compounds 2, 3, and
4).9a This new function was further used to connect
the synthetic pyochelin to an antibiotic fitted with a
spacer arm. The next step was thus the deprotection of
the three pyochelin analogues 2, 3, and 4. Unfortunate-
ly, the Teoc group removal using the most common
fluoride ion sources was unsuccessful,12,13 leading gener-
ally to complex mixtures predominantly composed of
starting material. To overcome this situation, three
new functionalized pyochelins 5,9b 6, and 7, whose
amine function is protected with a tert-butyloxycarbonyl
(Boc) group, were synthesized. The protecting group
removal on molecules 5, 6, and 7 appeared to be much
easier: in the presence of 6% TFA in dichloromethane,
the corresponding ammonium salts 8, 9, and 10 were ob-
tained very efficiently. A higher proportion of TFA or a
prolonged reaction time led often to complex mixtures
mainly composed of degradation products (Scheme 1).
In the case of the synthesis of the labile spacer arm-
antibiotic block, norfloxacin 11 was first reacted with
chloromethylchloroformate and gave compound 14 with
O
O
N
F
COOH
F
COOH
i
N
N
N
HOOC
N
HN
11
12 (85 %)
O
ii
O
F
COOH
O
N
N
N
C6F5O
13 (51 %)
O
Scheme 2. Synthesis of the norfloxacin–succinic spacer arm block 13.
Reagents and conditions: (i) succinic anhydride, pyridine, DMSO,
95 °C; (ii) pentafluorophenol, EDCI, dioxane, 60 °C.
O
O
F
COOH
F
COOH
The compounds 8, 9, and 10 were not very stable and
were therefore used as quickly as possible in the cou-
pling reaction with the pentafluorophenyl esters of the
antibiotic-linker units. The antibiotic chosen in a first
approach was norfloxacin 11, a fluoroquinolone active
against P. aeruginosa that acts by inhibiting the bacterial
topoisomerase.14 The spacer arms used to link the drug
to the functionalized pyochelins were of two types: a
succinic linker, stable in physiological condition, and a
i
N
N
N
N
Cl
O
N
HN
11
14 (80 %)
O
O
N
ii
F
COOH
COOH
O
N
O
O
O
N
O
15 (55 %)
O
O
O
O
O
N
OH
OH
iii
F
COOH
H
COOH
H
N
N
N
N
RHN
O
O
N
S
S
O
N
S
S
HO
NHR
16 (98 %)
2 R = Teoc
5 R = Boc
3 R = Teoc
6 R = Boc
9 R = H2+CF3COO-
i
i
8 R = H2+CF3COO-
O
N
F
COOH
iv
OH
COOH
H
N
N
N
O
O
N
S
C6F5O
S
17 (44 %)
O
O
4 R = Teoc
7 R = Boc
10 R = H2+CF3COO-
NHR
i
Scheme 3. Synthesis of the norfloxacin-hydrolyzable spacer arm block
17. Reagents and conditions: (i) chloromethylchloroformate, 1,8-
bis(N,N-dimethylamino)naphthalene, CHCl3, 20 °C; (ii) succinic acid
mono-tert-butyl ester, Ag2CO3, DMF, 95 °C; (iii) TFA, CH2Cl2,
20 °C; (iv) EDCI, pentafluorophenol, dioxane, 20 °C.
Scheme 1. The nine functionalized synthetic pyochelins 2, 3, 4 (Teoc
protected), 5, 6, 7 (Boc-protected), and 8, 9, 10 (ammonium salts).
Reagents and conditions: (i) TFA, CH2Cl2, 20 °C.

642
F. Rivault et al. / Bioorg. Med. Chem. Lett. 17 (2007) 640–644
O
F
COOH
O
N
N
N
HN
18
(96%)
O
S
S
N
N
i, ii
HOOC
HO
6
i, iii
O
O
HN
S
S
O
O
F
COOH
N
O
N
N
N
HOOC
HO
N
19
(90%)
O
Scheme 4. Synthesis of prodrugs 18 and 19. Reagents and conditions: (i) TFA, CH2Cl2, 20 °C; (ii) 13, DIPEA, dry DMF, 20 °C; (iii) 17, DIPEA, dry
DMF, 20 °C.
O
F
COOH
N
N
O
N
HN
20
(78%)
O
S
S
i, ii
N
N
HOOC
HO
7
O
i, iii
O
HN
O
F
COOH
S
S
O
N
N
N
N
HOOC
O
N
21
HO
(89%)
O
Scheme 5. Synthesis of prodrugs 20 and 21. Reagents and conditions: (i) TFA, CH2Cl2, 20 °C; (ii) 13, DIPEA, dry DMF, 20 °C; (iii) 17, DIPEA, dry
DMF, 20 °C.
an excellent yield. The reaction of chloride 14 with tert-
butyl hemisuccinate in the presence of silver carbonate
gave the corresponding tert-butyl ester derivative 15.17
This acylation method produced the clean desired com-
pound 15 easily and efficiently, and was preferred to the
previously published method.6 The tert-butyl protecting
group was removed with TFA and the resulting free car-
boxylic acid 16 was then converted to the activated ester
17 (Scheme 3).18
were obtained from 6 in 96% and 90% yield, respective-
ly, over two steps (Scheme 4).
Similarly, the conjugates 20 and 21 containing the ali-
phatic extension were isolated from 7 in 78% and 89%
yield, respectively, over two steps (Scheme 5).
Unfortunately we were not able to isolate the conjugates
derived from the pyochelin analogue 5. The amine func-
tion of the intermediate 8 seemed to be not nucleophilic
enough to react efficiently with activated esters 13 or 17.
The coupling reactions between functionalized pyoche-
lins 9 and 10, and activated esters 13 and 17 were carried
out in dry DMF in the presence of DIPEA.19 The con-
jugates 18 and 19 containing the acetylenic extension
As a preliminary biological evaluation, the four
pyochelin–norfloxacin adducts were then tested against

F. Rivault et al. / Bioorg. Med. Chem. Lett. 17 (2007) 640–644
643
sis) for financial support and for providing a postdoctor-
al fellowship to Dr. F. Rivault.
0.60
0.40
0.20
0.00
References and notes
1. (a) Shepp, D. H.; Tang, I. T.; Ramundo, M. B.; Kaplan,
M. K. J. Acquir. Immune Defic. Syndr. 1994, 7, 823; (b)
Jacques, I.; Derelle, J.; Weber, M.; Vidailhet, M. Eur. J.
Pediatr. 1998, 157, 427.
2. For a review on siderophores, see: Pattus, F.; Abdallah,
M. A. J. Chin. Chem. Soc. 2000, 47, 1.
3. For a review on iron uptake in pathogenic bacteria, see:
Braun, V. Int. J. Med. Microbiol. 2001, 291, 67.
4. For a review on iron uptake in Pseudomonas aeruginosa,
see: Poole, K.; McKay, G. A. Front. Biosci. 2003, 8, D661.
5. (a) Lin, Y. M.; Helquist, P.; Miller, M. J. Synthesis 1999,
1510; (b) Budzikiewicz, H. Curr. Topics Med. Chem. 2001,
1, 73.
6. Hennard, C.; Truong, Q. C.; Desnottes, J. F.; Paris, J. M.;
Moreau, N. J.; Abdallah, M. A. J. Med. Chem. 2001, 44,
2139.
OD600
0
5
10
time (h)
15
20
25
Figure 2. Growth of wild-type P. aeruginosa in the presence of the
prodrugs.20 P. aeruginosa PAO1 cells at an OD600 of 0.1 were
incubated at 30 °C in succinic medium (black line),21 or in the presence
of 10 lM norfloxacin (j, black line) conjugate 18 (s, dotted line),
conjugate 19 (d, dotted line), conjugate 20 (n, grey line) or conjugate
21 (m, grey line) for 25 h. The cell growth was followed at 600 nm. All
growth experiments were performed at least in triplicate.
7. (a) Liu, P. V.; Shokrani, F. Infect. Immun. 1978, 22, 878;
(b) Cox, C. D.; Graham, R. J. Bacteriol. 1979, 137, 357; (c)
Cox, C. D. J. Bacteriol. 1980, 142, 581; (d) Cox, C. D.;
Rinehart, K. L.; Moore, M. L.; Cook, J. C. Proc. Natl.
Acad. Sci. U.S.A. 1981, 78, 4256.
8. Darling, P.; Chan, M.; Cox, A. D.; Sokol, P. A. Infect.
Immun. 1998, 66, 874.
´
9. (a) Rivault, F.; Schons, V.; Liebert, C.; Burger, A.; Sakr,
P. aeruginosa PAO1 strain (Fig. 2). Only the labile-arm
conjugates 19 and 21 showed a lethal activity against P.
aeruginosa as pronounced as for norfloxacin. For the
stable arm-containing conjugates 18 and 20, the growth
curves showed no significant inhibition. Apparently the
dissociation of the siderophore-antibiotic conjugate is
necessary for observing a bactericidal activity of these
conjugates. Therefore the norfloxacin, when bound to
pyochelin by a non-hydrolyzable spacer arm, may be
either not able to reach the DNA gyrase in the cyto-
plasm or is too sterically hindered by the siderophore
moiety to interact and inhibit this crucial enzyme. These
results are quite similar to those obtained previously
with conjugates derived from the siderophore
pyoverdine.6
E.; Abdallah, M. A.; Schalk, I. J.; Mislin, G. L. A.
Tetrahedron 2006, 62, 2247; (b) Zamri, A.; Schalk, I. J.;
Pattus, F.; Abdallah, M. A. Bioorg. Med. Chem. Lett.
2003, 13, 1147; (c) Zamri, A.; Abdallah, A. M. Tetrahe-
dron 2000, 56, 249.
10. Tseng, C. F.; Burger, A.; Mislin, G. L. A.; Schalk, I. J.;
Yu, S. S. F.; Chan, S. I.; Abdallah, M. A. J. Biol. Inorg.
Chem. 2006, 11, 419.
11. (a) Cobessi, D.; Celia, H.; Pattus, F. J. Mol. Biol. 2005,
352, 893; (b) Mislin, G. L. A.; Hoegy, F.; Cobessi, D.;
Poole, K.; Rognan, D.; Schalk, I. J. J. Mol. Biol. 2006,
357, 1437.
12. Carpino, L. A.; Tsao, J. H.; Ringsdorf, H.; Fell, E.;
Hettrich, G. J. Chem. Soc. Chem. Commun. 1978, 8, 358.
13. Roush, W. R.; Coffey, D. S.; Madar, D. J. Am. Chem. Soc.
1997, 119, 11331.
14. For a recent review on fluoroquinolones, see: Mitscher, L.
A. Chem. Rev. 2005, 105, 559.
15. Schueller, C. M.; Manning, D. D.; Kiessling, L. L.
Tetrahedron Lett. 1996, 37, 8853.
In summary, we have synthesized four unprecedented
pyochelin–norfloxacin conjugates. The preliminary bio-
logical tests against a wild-type strain of Pseudomonas
aeruginosa show a bactericidal activity for two of these
four conjugates. Further studies are under way in our
laboratory in order to investigate if these compounds
are Trojan horse conjugates transported by the pyoch-
elin uptake pathway or if these conjugates behave as
classical antibiotic prodrugs. In addition, complementa-
ry biological evaluations with different pathogenic clini-
cal CF strains of Pseudomonas aeruginosa and
Burkholderia cepacia will be investigated.
16. Synthesis of activated ester 13: Diacid 126 (96 mg,
0.23 mmol) and pentafluorophenol (84 mg, 0.27 mmol)
were dissolved in dioxane (2 mL). The solution was cooled
at 0 °C and EDCI (52 mg, 0.27 mmol) was added. The
reaction mixture was heated to 60 °C for 4 h and then the
solution was allowed to cooled down to room temperature
and evaporated to dryness. The crude product was
purified by chromatography on a silica gel column
(CH2Cl2 to CH2Cl2/MeOH:95/5), to give product 13
(68 mg, 0.12 mmol, 51%) as a white powder. Rf 0.29
(Acetone/AcOEt/AcOH:50/48/2); 1H NMR (CDCl3,
300 MHz): d (ppm) = 8.68 (s, 1H); 8.10 (d, J = 12.8 Hz,
1H); 6.71 (d, J = 7.0 Hz, 1H); 4.32 (q, J = 7.2 Hz, 2H);
3.91–3.88 (m, 2H); 3.77–3.74 (m, 2H); 3.55–3.38 (m, 2H);
3.30–3.27 (m, 2H); 3.09 (t, J = 6.5 Hz, 2H); 2.83 (t,
J = 6.5 Hz, 2H); 1.59 (t, J = 7.2 Hz, 3H); 13C NMR
(CDCl3, 75 MHz): d (ppm) = 177.1; 169.3; 169.0; 167.2;
153.6 (d, J = 250 Hz); 147.4, 145.7 (d, J = 10 Hz); 137.1,
Acknowledgments
The authors thank the C.N.R.S (Centre National de la
Recherche Scientifique) and the association Vaincre la
Mucoviscidose (French association against cystic fibro-

644
F. Rivault et al. / Bioorg. Med. Chem. Lett. 17 (2007) 640–644
121.2 (d, J = 7 Hz); 113.1 (d, J = 23 Hz); 108.5; 104.4;
50.3; 50.2; 49.9; 49.6; 49.5; 28.7; 27.9; 14.6; ESI-TOF m/z
(ppm) = 8.69 (s, 1H); 8.10 (d, J = 12.8 Hz, 1H); 6.83 (d,
J = 6.8 Hz, 1H); 5.85 (s, 2H); 4.31 (q, J = 7.3 Hz, 2H); 3.73
(br s, 4H); 3.28 (br s, 4H); 3.03 (dd, J = 5.3; 6.0 Hz, 2H);
586 (M+H+).
17. Synthesis of compound 15: Halide 14 (300 mg, 0.73 mmol)
was dissolved in DMF (7.8 mL) and then succinic acid
mono-tert-butyl ester (254 mg, 1.46 mmol) and silver
carbonate (603 mg, 2.19 mmol) were added successively.
The reaction mixture was heated to 95 °C for 4 h and then
the solution was cooled down to room temperature and
was filtered through Celite. Celite was thoroughly washed
with a mixture CH2Cl2/MeOH:90/10 and the resulting
filtrate evaporated to dryness. After adsorption on silica,
the crude residue was purified by chromatography on a
silica gel column (CH2Cl2 to CH2Cl2/MeOH:96/4). After
crystallization from CH2Cl2/cyclohexane, compound 15
(220 mg, 0.40 mmol, 55%) was obtained as a white
powder. Rf 0.74 (CH2Cl2/MeOH:95/5); 1H NMR
(DMSO-d6, 300 MHz): d (ppm) = 8.96 (s, 1H); 7.94 (d,
J = 13.2 Hz, 1H); 7.21 (d, J = 7.1 Hz, 1H); 5.73 (s, 2H);
4.59 (q, J = 7.1 Hz, 2H); 3.61 (br s, 4H); 3.35 (br s, 4H);
2.58–2.43 (m, 4H); 1.41 (t, J = 7.1 Hz, 3H); 1.37 (s, 9H);
13C NMR (CDCl3, 75 MHz): d (ppm) = 176.8; 171.5;
171.2; 167.0; 153.4 (d, J = 252 Hz); 153.4; 147.2; 145.7 (d,
J = 10 Hz); 137.1; 120.7 (d, J = 7 Hz); 112.6 (d,
J = 23 Hz); 108.2; 104.4; 80.9; 80.5; 49.8; 49.6; 43.9; 43.5;
30.0; 29.2; 28.1; 14.5.
18. Synthesis of activated ester 17: To a solution of tert-butylic
ester 15 (99 mg, 0.18 mmol) in dichloromethane (7 mL),
TFA (3 mL) was added. After 3 h at 23 °C, the mixture
was evaporated to dryness and the residue was triturated
in diethyl ether. After filtration, compound 16 (87 mg,
0.18 mmol, 98%) was obtained as a white powder and was
used without any further purification. Diacid 16 (50 mg,
0.10 mmol) and pentafluorophenol (37 mg, 0.20 mmol)
were dissolved in dioxane (1 mL). The solution was cooled
to 0 °C and EDCI (23 mg, 0.12 mmol) was added. The
reaction mixture was stirred for 4 h at 24 °C and then the
solution was evaporated to dryness. The crude product
was purified by chromatography on a silica gel column
(CH2Cl2 to CH2Cl2/MeOH:95/5) and gave product 17
(29 mg, 0.05 mmol, 44%) as a white powder. Rf 0.51
(CH2Cl2/MeOH:95/5); 1H NMR (CDCl3, 300 MHz): d
2.85 (dd, J = 5.3; 6.0 Hz, 2H); 1.59 (t, J = 7.3 Hz, 3H); 13
C
NMR (CDCl3, 75 MHz): d (ppm) = 177.1; 170.7; 168.4;
167.3; 159.9; 153.6 (d, J = 250 Hz); 153.4; 147.4; 145.8 (d,
J = 10 Hz); 137.1; 121.2 (d, J = 7 Hz); 113.1 (d,
J = 23 Hz); 108.5; 104.4; 80.8; 49.9; 49.7; 43.9; 43.6; 28.8;
28.2; 14.6; ESI-TOF m/z 660 (M+H+) 682 (M+Na+) 698
(M+K+).
19. General procedure for norfloxacin–pyochelin conjugates 18,
19, 20, 21: To a solution of Boc-protected amine 6 (or 7)
(0.10 mmol) in dichloromethane (5 mL), TFA (0.30 mL)
was added successively. After 3 h at room temperature, the
mixture was evaporated to dryness and the residue was
triturated in diethyl ether. After filtration, compound 9 (or
10) was obtained as yellow powder which was used
without any further purification. Compound 9 (or 10) was
taken up in dry DMF (5 mL) and activated ester 13 (or 17)
(0.10 mmol) and DIPEA (0.15 mmol) was added. After
18 h at room temperature (20–24 °C), the reaction mixture
was evaporated to dryness and then triturated in diethyl
ether. After filtration, the desired conjugate 18 (96%), 19
(90%), 20 (78%) or 21 (89%) was obtained. Compound
(18) ESI-TOF m/z 779 (M+H+); HRMS Calcd for
C37H40FN6O8S2: 779.2328. Found 779.2336. Compound
(19) ESI-TOF m/z 853 (M+H+); HRMS Calcd for
C39H42FN6O11S2: 853.2332. Found 853.2331. Compound
(20) ESI-TOF m/z 783 (M+H+); HRMS Calcd for
C37H44FN6O8S2: 783.2641. Found 783.2640. Compound
(21) ESI-TOF m/z 857 (M+H+); HRMS Calcd for
C39H46FN6O11S2: 857.2645. Found 857.2648.
20. Bacterial strains and culture conditions: P. aeruginosa
PAO1 cells were grown overnight in a succinate medium at
30 °C.21 For measurements of bacterial growth, an over-
night culture was diluted to an OD600 of 0.1 in succinate
medium and was incubated at 30 °C in the presence of
either 10 lM norfloxacin, or of prodrugs 18, 19, 20 or 21
for 25 h. The cell growth was followed at 600 nm.
21. Demange, P.; Wendenbaum, S.; Linget, C.; Mertz, C.;
Cung, M. T.; Dell, A.; Abdallah, M. A. Biol. Metals 1990,
3, 155.