Total syntheses and evaluation of the siderophore functions of fimsbactin B and its analogs

Full text of DOI:10.1002/bkcs.10265

Note
DOI: 10.1002/bkcs.10265
BULLETIN OF THE
H. Ree et al.
KOREAN CHEMICAL SOCIETY
Total Syntheses and Evaluation of the Siderophore Functions of
Fimsbactin B and Its Analogs
*
Hwisoo Ree, Jimin Kim, Woon Young Song, Jae Eun Lee, and Hak Joong Kim
Department of Chemistry, College of Science, Korea University, Seoul 136-701, Republic of Korea.
*E-mail: hakkim@korea.ac.kr
Received December 11, 2014, Accepted January 10, 2015, Published online April 24, 2015
Keywords: Siderophore, Fimsbactin, Acinetobacter baumannii, Antibiotics, Drug resistance
Iron is an essential element for the survival of all forms of life,
in which it functions as a cofactor during key metabolic pro-
cesses. However, due to the low aqueous solubility of Fe(III),
most organisms possess dedicated systems for active assimi-
lation of iron from the aerobic environment. Siderophores
are small-molecule, high-affinity Fe(III) chelators produced
and utilized by most bacteria for this purpose. Currently, more
than 500 siderophores have been identified, and some have
been extensively studied to elucidate their physiological func-
tions andmechanisms.1,2 Becauseoftheessentialityofironfor
bacterial homeostasis and infection, siderophore-related
mechanisms have received considerable attention as promis-
ing, novel targets for the discovery of antibiotics.3 Among
the various approaches in this regard, development of sidero-
phore–drug conjugates (SDCs) has attracted particular
interest,3–6 in which the siderophores are utilized as carrier
vectors to deliver the conjugated drugs intracellularly.
Because the uptake of SDCs mediates designated transport
machineries, they have been considered tobeeffective intreat-
ing Gram-negative pathogens, particularly of which the drug
resistance is associated with the restricted outer membrane
(OM) permeability.7–9
AcinetobacterbaumanniiisaGram-negativebacterium and
one of the ESKAPE pathogens, referring to highly problem-
atic drug-resistant strains.10,11 Indeed, nosocomial and com-
munity infection of A. baumannii has been an increasing
cause of concern due to its rapid acquisition of resistance to
clinically employed antibiotics including carbapenems.12–14
Notably, it has been shown that A. baumannii features
intrinsically high OM permeability barrier compared to other
Gram-negative pathogens such as Escherichia coli and Pseu-
domonas aeruginosa.15 Thus, the employment of the SDC
strategy seems perfectly suited to overcome this recalcitrant
bacterium, and indeed a number of promising early results
in this direction have been reported.16,17
biosynthesis, was described. However, the biological and
physicochemical properties of fimsbactins have yet to be
established. In this regard, we have devised an effective syn-
thetic route to fimsbactins to study fimsbactin-based SDCs
effective for A. baumannii. Here we describe the successful
syntheses of fimsbactin B and its analogs, and also the in vivo
evaluation of their siderophore functions.
Our retrosynthetic analysis for fimsbactin B (2) is outlined
in Figure 1(b), in which the backbone construction is accom-
plished through an assembly of two fragments, oxazoline car-
boxylate7andtheputrescinederivative 8. Weoptedtoemploy
Mo(VI) oxide-catalyzed dehydrative cyclization, recently
developed by Ishihara et al.,21–23 for synthesis of an oxazoline
moiety, because this biomimetic approach would be advanta-
geous over other methods, since it would allow the establish-
ment of the correct stereochemistry from natural amino acid
derivatives with L-configuration.
The synthesis of amine fragment 8 was accomplished by
adopting Bergeron's reaction sequence24 with some modifica-
tions, initiated by the substitution of 4-chlorobutyronitrile (9)
with N-Boc-O-benzyloxyamine (Scheme 1(a)). The Boc pro-
tecting group of 10 was removed with trifluoroacetic acid, and
the following N-acetylation established the hydroxamate moi-
ety in an O-benzyl protected form. Then, chemoselective
hydrogenation of 11 using Raney-Ni as a catalyst allowed
reduction of a nitrile group to afford amine 12. The primary
amine 12 was then coupled with N,O-protected L-serine
1325 with EDC, and subsequent brief exposure to trifluoroace-
tic acid led to chemoselective Boc deprotection to afford the
desired fragment 8.
The preparation of oxazoline carboxylate 7 started from
the o-xylyl-protected 2,3-dihydroxybenzoic acid 15
(Scheme 1(b)). As proposed by Sakakura et al.,22 the o-xylyl
protection was crucial for the success of fimsbactin synthesis,
because the employment of other protective groups such as
dibenzyl had detrimental effects on the downstream modifica-
tionsduetothebulkiness.26–28 Afteramidecoupling ofacid15
with L-Thr-OCH3, a dehydrative oxazoline formation to pro-
vide 17 was successfully accomplished by treatment of 16
with a catalytic amount of MoO2(TMHD)2 in toluene under
azeotropic refluxconditions. Interestingly, whenthesamepre-
cursor (16) was treated with Burgess’ reagent, production of
For a while, acinetobactin has been known as the sole sid-
erophore produced and utilized by A. baumannii,18,19 but
recently an alternative set of siderophores called fimsbactins
were isolated from A. baumannii ATCC 17978 as well as non-
pathogenic Acinetobacter baylyi ADP1 strains by Proschak
et al.20 In their report, structural characterization of six fims-
bactins (A–F) (Figure 1(a)), as well as a proposal for their
Bull. Korean Chem. Soc. 2015, Vol. 36, 1520–1523
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Wiley Online Library
1520

BULLETIN OF THE
Note
KOREAN CHEMICAL SOCIETY
an epimer of 17 with 9(S),10(S)-syn-configuration was
observed, clearly demonstrating the mechanistic difference
between these two oxazoline formation conditions.21 The
ester moiety in 17 was then hydrolyzed with KOTMS,29
and the subsequent amide formation with 8 proceeded seam-
lessly with no hint of epimerization at C-10. The introduction
of the second catechol group involved the removal of the silyl
protection of 18 using HF•pyridine complex followed by car-
bodiimide-promoted esterification with 15, giving rise to the
penultimate intermediate 19. Finally, completion of total syn-
thesis of fimsbactin B (2) was realized by global deprotection
under palladium-catalyzed hydrogenolysis conditions.
Although the exact mode of the Fe(III) binding by fimsbac-
tinsisunknown, formationoftheFe(III)complexwithanocta-
hedral geometry for fimsbactin A (1), B (2), and C (3) is
conceivable involving two catechol and one hydroxamate
moieties, each as a bidentate ligand. However, other fimsbac-
tins, D(4), E(5), andF(6), aredeficient inoneofthoseligands,
raising a question whether they would indeed function as side-
rophores. To address this issue, we pursued the syntheses of
9-methyl variants of fimsbactin D (27) and F (21) by slight
modifications of the synthetic scheme for fimsbactin B (2).
As delineated in Scheme 2(a), 9-CH3-fimsbactin F (21) could
be readily prepared by hydrogenolytic deprotection of the
intermediate 20. The synthesis of 9-CH3-fimsbactin D (27)
began with the amide formation between N-monoacetyl-1,3-
propanediamine (23)30 and N,O-protected L-serine derivative
2225 followed by chemoselective removal of the Boc group
(Scheme 2(b)). The employment of the TBDPS moiety in
22 was advantageous over TBS in 13, because the isolation
of the resulting polar free amine intermediate was more facile
due to the higher lipophilicity of TBDPS. In addition, gratify-
ingly, the bulkiness of TBDPS had no detrimental effects on
the EDC-mediated coupling with 7 to afford 25, and the
TBDPS ether could be readily cleaved by treatment of TBAF.
Finally, the attachment of the catechol ester at C-15 followed
by global deprotection under hydrogenolysis conditions suc-
cessfully led to the formation of 9-CH3-fimsbactin D (27)
albeit in a low yield of 23%, likely due to the low solubility
of 27 in methanol.
(a)
(b)
Figure 1. (a) Structures of fimsbactin A–F. (b) Retrosynthesis for the
total synthesis of fimsbactin B (2).
(a)
BocNHOBn
NaH, NaI
CH3
1.TFA, CH2Cl2, rt
NC
NBoc
OBn
NC
N
O
NC
Cl
2.Ac2O, pyr, rt
2-step 66%
DMF, 85 ºC
70%
11 OBn
9
10
OTBS
13
OTBS
CH3
BocHN
CO2H
H2, Raney-Ni, NH3
CH3
O
H
H2N
N
N
O
EDC, HOBt
i-Pr2NEt, DMF, rt
2 steps 64%
BocHN
N
CH3OH, H2O, rt
98%
12
OBn
O
OBn
14
OTBS
CH3
TFA, CH2Cl2
H
N
Xyl
OH
(a)
O
CH3
O
H2N
N
O
O
O
H
rt, 71%
H2, Pd/C
N
O
OBn
HN
N
O
N
H
N
8
9-CH3-Fimsbactin F (21)
O
OBn
CH3OH, rt
85%
(b)
CH3
20
Xyl
Xyl
L-Thr-OCH3
EDC, HOBt
(b)
OTBDPS
OTBDPS
OH
O
O
(S)
O
O
CO2CH3
MoO2(TMHD)2
toluene
CO2CH3
CH3
H
N
EDC, HOBt
H
H
10
N
O
H2N
CH3
N
N
CH3
+
O
O
BocHN
BocHN
i-Pr2NEt, DMF, rt
CH3
DMF, rt
82%
9
95%
O
OH
azeotropic reflux
82%
O
O
O
23
O
(R)
22
24
CO2H
17
16
15
Xyl
OTBDPS
Xyl
1. TFA
CH2Cl2, rt, 74%
O
OTBS
H
N
Xyl
O
O
H
H
1. TBAF, THF, rt
O
CH3
O
O
O
O
O
8, EDC
N
N
CH3
KOTMS
N
N
H
N 10
O
CO2K
CH3
N
O
N
N
2. 15, EDC
DMF, rt
CH2Cl2, rt
60%
2. 7, EDC
CH2Cl2, rt, 51%
H
O
O
Et2O, rt
66%
O
OBn
O
CH3
CH3
25
2 steps 49%
18
7
O
O
O
Xyl
O
Xyl
Xyl
O
O
O
Xyl
O
O
O
15
O
H2, Pd/C
O
CH3
1. HF·pyr, THF, rt, 67%
O
O
H
N
H
N
O
O
H2, Pd/C
H
9-CH3-Fimsbactin D (27)
CH3
N
2
N
O
N
O
N
N
N
O
CH3OH, rt
23%
2. 15, EDC, DMAP,
CH2Cl2, rt, 87%
CH3OH, rt
77%
H
H
O
OBn
26
CH3
CH3
19
Scheme 1. (a) Synthesis of amine fragment 8. (b) Completion of total
synthesis of fimsbactin B (2).
Scheme 2. (a) Synthesis of 9-CH3-fimsbactin F (21). (b) Synthesis of
9-CH3-fimsbactin D (27).
Bull. Korean Chem. Soc. 2015, Vol. 36, 1520–1523
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de 1521

BULLETIN OF THE
Note
KOREAN CHEMICAL SOCIETY
this role of 1 has received indirect support from the observa-
tion that its biosynthesis is regulated in response to the envi-
ronmental iron availability.20 To address this issue, we
attempted to synthesize fimsbactin A (1) by slight modifica-
tions of Scheme 1 utilized for the preparation of fimsbactin
B (2). Unfortunately, however, the amide formation between
8 and 9-desmethyl-7 led to epimerization at C-10 in contrast to
the case in fimsbactin B synthesis. Thus, we are currently pur-
suing the optimization of this key step to resolve the stereose-
lectivity issue, of which establishment would be crucial not
only for confirmation of the biological function of fimsbactin
A (1) but also eventuallyfor the design of effective fimsbactin-
based SDCs overcoming the OM barrier of drug-resistant
A. baumannii.
Figure 2. Evaluation of the siderophore functions of fimsbactin B (2)
and its analogs (21 and 27) based on the agar diffusion assay. Paper
disks were placed on iron-deficient agar plates (250 μM 2,20-dipyri-
dyl) overlaid with A. baylyi ADP1, and each was treated with 5 μL of
the sample in DMSO of the indicated concentration. After 6 h of incu-
bation at 37 ꢀC, formation of a zone surrounding each disk was
monitored.
Experimental
FimsbactinB(2). Acatalyticamount of palladium onactivated
carbon (10 wt%) was suspended in a solution of compound 19
(43.0 mg, 0.0487 mmol) in methanol (2 mL) at room tempera-
ture,andthisreactionmixturewaschargedwithhydrogengasin
a balloon (1 atm). After 2 h, the reaction mixture was filtered
through a pad of celite to remove the palladium catalyst, and
the filtrate was concentrated under reduced pressure to afford
the desired product fimsbactin B (2) (22.0 mg, 0.0374 mmol)
in 77% yield without further purification.
With three fimsbactins, 2, 21, and 27, in hand, their sider-
ophore functions were evaluated based on the agar diffusion
assay, as shown in Figure 2.31 At the outset, we had predicted
that only fimsbactin B (2) would promote the cellular growth
around the paper disc by functioning as a siderophore because
of its possession of all three bidentate ligands, in contrast to the
analogs of fimsbactin D (27) and F (21) lacking the hydroxa-
mateorthecatecholmoiety, respectively. However, tooursur-
prise, not only all of these three compounds failed to elicit any
growth promotion under the iron-limiting conditions, but fims-
bactin B (2) and 9-CH3-fimsbactin D (27) wereinstead foundto
displayinhibitoryactivityinaconcentration-dependentmanner
indicated by the clear zone formation around the disc (also see
Figure S1 in Supporting Information). Among them, the inhib-
itory activity of 2 appeared to be more pronounced than that of
27, judged by comparing the sizes of the clear zones (diameters
of ~19 mm vs. 12 mm, respectively, at each 20 mM). These
unexpected antagonistic effects of 2 and 27 on the growth of
A. baylyi is likely due to their metal chelating ability, further
depletingthe iron inthegrowthmedia, theobservationofwhich
is not unprecedented.32
1H-NMR (500 MHz, DMSO-d6) δ 11.83 (s, 1H), 8.63 (d,
J = 8.1 Hz, 1H), 8.30 (s, 1H), 7.14 (d, J = 7.9 Hz, 1H), 7.06
(d, J = 7.7 Hz, 1H), 6.99 (br s, 2H), 6.73 (t, J = 7.7 Hz, 1H),
6.59 (t, J = 7.8 Hz, 1H), 4.87 (m, 1H), 4.74 (q, J = 6.6 Hz,
1H), 4.64–4.55 (m, 2H), 4.39 (dd, J = 10.9, 6.9 Hz, 1H),
3.46 (t, J = 6.7 Hz, 2H), 3.16–3.06 (m, 2H), 1.96 (s, 3H),
1.53–1.46 (m, 2H), 1.45 (d, J = 6.3 Hz, 3H), 1.41–1.36 (m,
2H); 13C-NMR (125 MHz, DMSO-d6) δ 170.19, 169.82,
168.91, 167.85, 165.95, 149.49, 148.34, 146.03, 145.77,
120.84, 119.77, 119.46, 118.77, 118.65, 117.90, 112.77,
110.25, 78.65, 73.35, 64.46, 51.51, 46.47, 38.49, 26.11,
23.75, 20.50, 20.39. HR-MS (ESI-TOF) m/z for
[C27H32N4NaO11]+ ([M + Na]+): calcd 611.1960, found
611.1960.
Acknowledgments. This work was supported by Basic
Here we reported the first total synthesis of fimsbactin B (2)
as well as its analogs 9-methyl variants of fimsbactin D (27)
and F (21). Surprisingly, in vivo biological assays unveiled
that none of them is capable of assimilating iron to promote
the cellular growth under the iron-deficient conditions, pre-
sumably due to their failure to interact with the cellular fims-
bactin uptake machineries (e.g., FbsN, FbsO, and FbsP).20
This unexpected finding, contrasted with their proposed phys-
iological roles, suggests that fimsbactin A (1), the major
metabolite (>90%) isolated from A. baylyi, would be the gen-
uine siderophore for its producer. This implication is rather
striking considering that the structural difference between 1
and 2 is merely the methyl substituent at C-9. Certainly, this
claim still awaits experimental corroboration of the physiolog-
ical function of fimsbactin A (1) as the siderophore, although
Science Research Program through the National Research
Foundation
of
Korea
(NRF-2012R1A1A1042665,
NRF20100020209) and the TJ Science Fellowship of POSCO
TJ Park Foundation.
Supporting Information. The detailed information includ-
ing synthetic procedures, spectral data for all new compounds,
and additional results of the in vivo cell-based assays is avail-
able free of charge via the Internet at http://kcsnet.or.kr
References
1. M. Miethke, M. A. Marahiel, Microbiol. Mol. Biol. Rev. 2007,
71, 413.
2. R. C. Hider, X. Kong, Nat. Prod. Rep. 2010, 27, 637.
Bull. Korean Chem. Soc. 2015, Vol. 36, 1520–1523
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de 1522

BULLETIN OF THE
Note
KOREAN CHEMICAL SOCIETY
3. T. L. Foley, A. Simeonov, Expert Opin. Drug Discov. 2012,
7, 831.
17. T. A. Wencewicz, M. J. Miller, J. Med. Chem. 2013,
56, 4044.
4. C. Ji, R. E. Juárez-Hernández, M. J. Miller, Future Med. Chem.
2012, 4, 297.
18. S. Yamamoto, N. Okujo, Y. Sakakibara, Arch. Microbiol. 1994,
162, 249.
5. M. G. P. Page, Ann. N. Y. Acad. Sci. 2013, 1277, 115.
6. A. Górska, A. Sloderbach, M. P. Marszałł, Trends Pharmacol.
Sci. 2014, 35, 442.
19. S. Yamamoto, N. Okujo, H. Kataoka, S. Narimatsu, J. Health
Sci. 1999, 45, 297.
20. A. Proschak, P. Lubuta, P. Grün, F. Löhr, G. Wilharm, V. De
Berardinis, H. B. Bode, ChemBioChem 2013, 14, 633.
21. A. Sakakura, S. Umemura, R. Kondo, K. Ishihara, Adv. Synth.
Catal. 2007, 349, 551.
22. A. Sakakura, S. Umemura, K. Ishihara, Chem. Commun.
2008, 3561.
23. A. Sakakura, R. Kondo, S. Umemura, K. Ishihara, Tetrahedron
2009, 65, 2102.
7. H. Nikaido, Science 1994, 264, 382.
8. M. N. Alekshun, S. B. Levy, Cell 2007, 128, 1037.
9. A. H. Delcour, Biochim. Biophys. Acta 2009, 1794, 808.
10. H. W. Boucher, G. H. Talbot, J. S. Bradley, J. E. Edwards,
D. Gilbert, L. B. Rice, M. Scheld, B. Spellberg, J. Bartlett, Clin.
Infect. Dis. 2009, 48, 1.
11. J. N. Pendleton, S. P. Gorman, B. F. Gilmore, Expert Rev. Anti
Infect. Ther. 2013, 11, 297.
12. L. Dijkshoorn, A. Nemec, H. Seifert, Nat. Rev. Microbiol. 2007,
5, 939.
24. R. Bergeron, J. McManis, Tetrahedron 1990, 46, 5881.
25. H.-J. Lim, M. H. Jung, I.-Y. C. Lee, W. K. Park, Bull. Korean
Chem. Soc. 2006, 27, 1371.
13. N. C. Gordon, D. W. Wareham, Int. J. Antimicrob. Agents 2010,
35, 219.
14. I. Abbott, G. M. Cerqueira, S. Bhuiyan, A. Y. Peleg, Expert Rev.
Anti Infect. Ther. 2013, 11, 395.
15. J. Vila, S. Marti, J. Sánchez-Céspedes, J. Antimicrob. Che-
mother. 2007, 59, 1210.
16. P. G. Higgins, D. Stefanik, M. G. P. Page, M. Hackel, H. Seifert,
J. Antimicrob. Chemother. 2012, 67, 1167.
26. P. J. Maurer, M. J. Miller, J. Am. Chem. Soc. 1983, 105, 240.
27. J. Hu, M. J. Miller, J. Am. Chem. Soc. 1997, 119, 3462.
28. Y. Xu, M. J. Miller, J. Org. Chem. 1998, 63, 4314.
29. E. D. Laganis, B. L. Chenard, Tetrahedron Lett. 1984, 25, 5831.
30. W. Tang, S. Fang, Tetrahedron Lett. 2008, 49, 6003.
31. A. J. Bäumler, K. Hantke, J. Bacteriol. 1992, 174, 1029.
32. T. Zheng, J. L. Bullock, E. M. Nolan, J. Am. Chem. Soc. 2012,
134, 18388.
Bull. Korean Chem. Soc. 2015, Vol. 36, 1520–1523
© 2015 Korean Chemical Society, Seoul & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.bkcs.wiley-vch.de 1523