Synthesis and biological evaluation of thiazolidine-2-one 1,1-dioxide as inhibitors of Escherichia coli β-ketoacyl-ACP-synthase III (FabH)

Article abstract of DOI:10.1016/j.bmcl.2006.11.067

A series of cyclic sulfones has been synthesized and their activity against β-ketoacyl-ACP-synthase III (FabH) has been investigated. The compounds are selectively active against Escherichia coli FabH (ecFabH), but not Mycobacterium tuberculosis FabH (mtFabH) or Plasmodium falciparum KASIII (PfKASIII). The activity against ecFabH ranges from 0.9 to >100 μM and follows a consistent general SAR trend. Many of the compounds were shown to have antimalarial activity against chloroquine (CQ)-sensitive (D6) P. falciparum (IC₅₀ = 5.3 μM for the most potent inhibitor) and some were active against E. coli (MIC = 6.6 μg/ml for the most potent inhibitor).

Full text of DOI:10.1016/j.bmcl.2006.11.067

Bioorganic & Medicinal Chemistry Letters 17 (2007) 879–883
Synthesis and biological evaluation of thiazolidine-2-one
1,1-dioxide as inhibitors of Escherichia coli
b-ketoacyl-ACP-synthase III (FabH)
Mamoun M. Alhamadsheh,a Norman C. Waters,b Donald P. Huddler,b
Mara Kreishman-Deitrick,b Galina Florovaa and Kevin A. Reynoldsa,*
aDepartment of Chemistry, Portland State University, Portland, OR 97207, USA
bDivision of Experimental Therapeutics, Walter Reed Army Institute of Research, Silver Spring, MD 20910, USA
Received 21 October 2006; revised 15 November 2006; accepted 20 November 2006
Available online 1 December 2006
Abstract—A series of cyclic sulfones has been synthesized and their activity against b-ketoacyl-ACP-synthase III (FabH) has been
investigated. The compounds are selectively active against Escherichia coli FabH (ecFabH), but not Mycobacterium tuberculosis
FabH (mtFabH) or Plasmodium falciparum KASIII (PfKASIII). The activity against ecFabH ranges from 0.9 to >100 lM and fol-
lows a consistent general SAR trend. Many of the compounds were shown to have antimalarial activity against chloroquine (CQ)-
sensitive (D6) P. falciparum (IC50 = 5.3 lM for the most potent inhibitor) and some were active against E. coli (MIC = 6.6 lg/ml for
the most potent inhibitor).
Ó 2006 Elsevier Ltd. All rights reserved.
b-Ketoacyl-ACP-synthase III (FabH) is a key condens-
ing enzyme of the dissociated type II fatty acid synthase
(FAS) system.1,2 It catalyzes a cysteine-mediated Claisen
condensation reaction between malonyl-ACP (MACP)
and an enzyme-bound acyl unit formed by initial trans-
acylation from a short-chain acyl-CoA primer to the ac-
tive site Cys112.2 The released b-ketoacyl-ACP product
is subsequently reduced and elongated by other (FAS)
condensing enzymes to generate a long-chain fatty acid.3
This type of FabH is ubiquitous in both Gram-positive
and -negative bacteria.4,5 It is also present in Plasmodi-
um falciparum and other parasitic organisms.6 An
unusual FabH is also present in Mycobacterium tubercu-
losis where it is required in the conversion of long-chain
fatty acids into mycolic acid.7,8 FabH bears no homolo-
gy to the condensing enzymes in the mammalian type I
FAS.9,10 FabH inhibitors, therefore, have potential as
general antibacterial, antiparasitic or even antimycobac-
terial agents.11
We have previously identified a series of potent FabH
inhibitors12 by searching the 3D National Cancer Insti-
tute (NCI) database for compounds bearing structural
similarities to thiolactomycin (TLM) (Fig. 1).13–15 One
set of these compounds were 1,2-dithiole-3-ones.12
Another set were cyclic sulfones such as compound 1a
(Fig. 1) which has been reported earlier as a herbicide
and fungicide.16 Here, we report our effort to further
study and develop more potent analogues of 1a.
The inhibitors were prepared following a general ap-
proach described by Aitken et al. and Le Core et al. as
shown in Scheme 1.17,18 The required N-substituted
O
O
OH
S
O
O
S
N
Keywords: FabH; Fatty acid biosynthesis; Inhibition; SAR;
Antimalarial.
TLM
1a
*
Corresponding author. Tel.: +1 503 725 3886; fax: +1 503 725
9525; e-mail: reynoldsk@pdx.edu
Figure 1.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.067

880
M. M. Alhamadsheh et al. / Bioorg. Med. Chem. Lett. 17 (2007) 879–883
R
H
N
OH
F
CHO
+
N
R
9
10
a
S
S
HO
O
CHO
2
3
b
c
8
R
HN
R
O
N
S
N
S
HO
2g
O
O
Scheme 3. Reagents and conditions: (a) K2CO3, DMF, 100 °C; (b) i—
benzene, NH2CH2CH2OH; ii—NaBH4, MeOH.
O
O
4
1
The compounds synthesized (N-substituted analogues of
1, 3, and 4) were tested against E. coli FabH (ecFabH).
Preliminary SAR for these analogues against ecFabH
are shown in Table 1. Investigations focused primarily
on the N-substitution and the oxidation state of the sul-
fur moiety. Decreasing the oxidation state of parent
compound 1a to the thiazolidin-2-one 4a resulted in a
13-fold decrease in the activity against ecFabH, while
no activity was observed for the thiazolidine-2-thiones
3a at 100 lM. Replacing the N-substituted benzyl of
1a with a phenethyl group (1b) resulted in a ꢀ2-fold de-
crease in activity. The lower oxidation state of 3b gave
similar activity pattern to the one observed for 3a.
Replacing the benzyl group with simple alkyl moieties
(3c,1c, 3d, and 1d) resulted in a major loss of biological
activity (IC50 > 100 lM). Introducing a 3-pyridyl ring
Scheme 1. Reagents and conditions: (a) CS2, aq NaOH; (b) KMnO4
(3 equiv), PhCO2H, benzyltriethylammonium ClÀ, CH2Cl2–H2O; (c)
same as (b) except KMnO4 (5 equiv).
aminoethanols 2b and 2e–g were readily obtained by
condensation of aldehydes 5–8 with ethanolamine to
give the corresponding imines which were subsequently
reduced by sodium borohydride (NaBH4) in methanol
(Schemes 2 and 3). Aldehyde 8 was prepared by nucleo-
philic aromatic substitution of 4-fluorobenzaldehyde 10
by phenol 9 (Scheme 3). Aminoethanols 2 were subject-
ed to reaction with carbon disulfide (CS2) in aqueous
sodium hydroxide to give N-substituted thiazolidine-2-
thiones 3.17,18 Under phase-transfer conditions, the oxi-
dation of 3 to the corresponding thiazolidin-2-ones 4
and thiazolidine-2-one 1,1-dioxides 1 was controlled
using different amounts of KMnO4 (3 and 5 equiv,
respectively).18 The desired products were purified and
Table 1. IC50 against ecFabH and P. falciparum (D6)
1
fully characterized by H NMR, 13C NMR, and high-
R
R
R
resolution mass spectroscopy (HRMS).
N
N
N
O
S
O
S
S
S
H
N
O
3
1
O
4
CHO
a
R
Compound EcFabH
P. falciparum
IC50 (lM) D6 IC50 (lM)
HO
5
6
2b
Ph–CH2
a
3a
4a
1a
3b
1b
3c
1c
3d
1d
3e
4e
3f
>100
13.7
1.1
61.6
90.9
67.6
55.5
21.6
>100
>100
>100
>100
73.4
78.4
55.6
67.1
49.4
5.3
CHO
H
N
Ph–CH2–CH2
>100
1.9
a
b
CH3
>100
>100
>100
>100
>100
20.7
>100
10.9
0.9
N
c
CH3CH2CH2
N
2e
HO
d
3-pyridyl–CH2
H
N
CHO
PhO
e
Ph–O–Ph–CH2
a
f
4f
HO
1f
C5H9–Ph–O–Ph–CH2 3g
PhO
2f
7
17.7
14.4
2.4
g
4g
1g
11.1
10.8
Scheme 2. Reagents: (a) i—benzene, NH2CH2CH2OH; ii—NaBH4,
MeOH.

M. M. Alhamadsheh et al. / Bioorg. Med. Chem. Lett. 17 (2007) 879–883
881
Figure 2. Alignment of mtFabH, ecFabH, and PfKASIII sequences (performed using ClustalW)19 encompassing the enzyme active site and the walls
and mouth of the CoA binding channel. Catalytic residues are indicated in red. ( ), (:), and (Æ) indicate identity, conserved substitution, and semi-
*
conserved substitution, respectively.
(4e) in place of the phenyl (4a) resulted in decrease of
activity. Based on this initial SAR data, we decided to
retain the benzyl moiety and extend it at the para-posi-
tion with another phenoxy group (3f, 4f, and 1f). This
modification resulted in a slight improvement in bioac-
tivity compared to our parent compound 1a. Interest-
ingly, the correlation between the oxidation state of
these analogues and bioactivity was in full agreement
with the one observed for the rest of the inhibitors
(i.e., a higher oxidation gave higher bioactivity). Extend-
ing the size of the N-substituted aryl moiety of 1f by an
extra cyclopentyl group (3g, 4g, and 1g) resulted in ꢀ2-
fold decrease in activity (1g compared to 1a). From
these SAR studies we conclude that the most potent ec-
FabH inhibitors require both the sulfur in the oxidized
sulfone form and an aromatic substitution at the 3-N
position. While there is some flexibility in the N-substi-
tuent size, there appears to be an upper size limit.
and ecFabH were incubated under conditions that
resulted in greater than 85% enzyme inhibition. Dialysis
of the inhibited enzyme for 12 h in 50 mM phosphate
buffer (pH 7.4 at 4 °C) resulted in a ꢀ90% restoration
of its activity (no restoration of activity was observed
when these experiments were carried out with irrevers-
ible inhibitors). These data demonstrated the mode of
binding of 1a as a reversible inhibitor.
The activity of the 17 compounds against M. tubercu-
losis FabH (mtFabH), and P. falciparum KASIII
(PfKASIII), was also evaluated. Surprisingly, no inhi-
bition activity was observed for any of the compounds
(at 100 lM). Thus for the most potent 1 lM inhibitors
of the ecFabH, there is a decrease of more than two
orders of magnitude for inhibition of the other two
FabH enzymes. Attempts to determine the selective
mode of binding of these inhibitors with the ecFabH
through generation of a cocrystal structure were unsuc-
cessful. Likewise docking studies failed to produce a
compelling case for a specific mode of binding to the
A reversibility study was carried out to determine the
mode of binding of 1a with ecFabH. Compound 1a
Figure 3. Surface representations of the conserved CoA binding channels of mtFabH (PDB ID 1U6S), ecFabH (PDB ID 1MZS), and PfKASIII
(from homology model) created using Pymol.20 All representations are shown in approximately the same orientation, looking at the opening of the
channel.

882
M. M. Alhamadsheh et al. / Bioorg. Med. Chem. Lett. 17 (2007) 879–883
ecFabH. Nonetheless, a comparison of the crystal
structures of the ecFabH, mtFabH and a homology
model of the PfKASIII revealed some structural differ-
ences which might account for the differing sensitivity
to these and potentially other inhibitors. The three
enzymes are 45% identical and 66% conserved in the
region comprising the entire CoA binding channel
(see alignment in Fig. 2). The channel itself is structur-
ally quite similar; however, some differences do exist at
the surfaces surrounding the mouth of the channel.
These differences are highlighted in Figure 3. First,
all three enzymes contain a patch of four variable res-
idues distal to the opening of the channel (magenta in
Fig. 3). The character of all three patches is principally
hydrophobic, but the key difference among the patches
is the nature of the polar/charged amino acid in each
(K233, E219, and N273 for mtFabH, ecFabH, and
PfKASIII, respectively). Closer to the mouth of the
channel, three variable amino acids (indicated in yellow
in Fig. 3) were also identified. The most striking of
these (P224 in mtFabH, N210 in ecFabH, and K264
in PfKASIII) is located directly adjacent to the channel
opening. Conformational restrictions imposed by pro-
line in mtFabH may contribute to the poor inhibitor
binding observed in this study. The long lysine side
chain in PfKASIII provides a rather different surface
in this region and may restrict access to the CoA bind-
ing channel. The steric bulk surrounding the opening
of the PfKASIII channel is further increased by the
presence of N197 and N262. In the mtFabH these res-
idues are G162 and E222, with the latter also contrib-
uting to steric bulk. In ecFabH the equivalent positions
are occupied by small aliphatic residues (G152 and
A208). In summary, there are clear structural differenc-
es between the enzymes at the mouth of the CoA bind-
ing channel. These variable residues may limit access to
the CoA binding channel for PfKASIII and mtFabH
as compared to the ecFabH enzyme.
In conclusion, we have identified and synthesized a new
series of ecFabH inhibitors. Preliminary SAR studies
indicated that the most potent inhibitors require both
the sulfur in the oxidized sulfone form and a limited-size
aromatic substitution at the 3-N position. The selectivity
of these inhibitors (against ecFabH and not mtFabH or
PfKASIII) could provide structural insights and under-
standing of the different specificities of these three
enzymes.
Acknowledgments
We are grateful to Dr Martin Smilkstein (Portland Vet-
erans Affairs Medical Center) for performing the anti-
malarial assays. We also thank Mr. Sarbjot Sachdeva
(Portland State University) for performing the mtFabH
assay. We thank Dr. Sean Prigge (Johns Hopkins Uni-
versity) for providing the PfKASIII homology model
and Ms. Patricia Lee (Walter Reed Army Institute of
Research) for performing the PfKASIII assay. Funding
for this research was generously provided by the Nation-
al Institutes of Health (AI52230).
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2006.11.067.
References and notes
1. Clough, R. C.; Matthis, M.; Barnum, S. R.; Jaworski, J.
G. J. Biol. Chem. 1992, 267, 20992.
2. Tsay, J. T.; Oh, W.; Larson, T. J.; Jackowski, S.; Rock, C.
O. J. Biol. Chem. 1992, 267, 6807.
3. Edwards, P.; Nelsen, J. S.; Metz, J. G.; Dehesh, K. FEBS
Lett. 1997, 402, 62.
4. Qiu, X.; Janson, C. A.; Konstantinidis, A. K.; Nwagwu,
S.; Silverman, C.; Smith, W. W.; Khandekar, S.;
Lonsdale, J.; Abdel-Meguid, S. S. J. Biol. Chem. 1999,
274, 36465.
The activity of the 17 compounds against both E. coli
and P. falciparum was determined. In the case of E. co-
li, only 1a and 1b showed ability to inhibit the growth
of E. coli with
a concentration below 50 lg/ml
(MIC = 6.6 lg/ml for 1a and 16.5 lg/ml for 1b). An
E. coli strain, carrying an ecFabH expression plasmid,
grown under appropriate induction conditions also
displayed an MIC value for 1a of 6.6 lg/ml or less.
The absence of increase in resistance to 1a when the
ecFabH was overexpressed in E. coli suggests that ec-
FabH is not the sole target for this class of inhibitors.
The in vitro antimalarial activity of these analogues
against chloroquine (CQ)-sensitive (D6) P. falciparum
was also evaluated (Table 1).21 In this case, 13 of
the 17 inhibitors had shown activity with an IC50 val-
ue below 100 lM. Increasing the size of the aromatic
N-substituent resulted in ꢀ10-fold improvement of
the antimalarial activity compared to the parent com-
pound 1a (IC50 for 3g = 5.3 lM compared to IC50 for
1a = 67.7 lM). There was no pattern regarding the
oxidation state of the sulfur. As 1g, 3g, and 4g do
not inhibit the PfKASIII, the antimalarial activity of
these compounds must arise through some other
mechanism.
5. Revill, W. P.; Bibb, M. J.; Scheu, A. K.; Kieser, H. J.;
Hopwood, D. A. J. Bacteriol. 2001, 183, 3526.
6. Prigge, S. T.; He, X.; Gerena, L.; Waters, N. C.; Reynolds,
K. A. Biochemistry 2003, 42, 1160.
7. Choi, K. H.; Kremer, L.; Besra, G. S.; Rock, C. O. J. Biol.
Chem. 2000, 275, 28201.
8. Scarsdale, J. N.; Kazanina, G.; He, X.; Reynolds, K. A.;
Wright, H. T. J. Biol. Chem. 2001, 276, 20516.
9. Chirala, S. S.; Kuziora, M. A.; Spector, D. M.; Wakil, S.
J. J. Biol. Chem. 1987, 262, 4231.
10. Wakil, S. J. Biochemistry 1989, 28, 4523.
11. Wang, J.; Soisson, S. M.; Young, K.; Shoop, W.; Kodali,
S.; Galgoci, A.; Painter, R.; Parthasarathy, G.; Tang, Y.
S.; Cummings, R.; Ha, S.; Dorso, K.; Motyl, M.;
Jayasuriya, H.; Ondeyka, J.; Herath, K.; Zhang, C.;
Hernandez, L.; Allocco, J.; Basilio, A.; Tormo, J. R.;
Genilloud, O.; Vicente, F.; Pelaez, F.; Colwell, L.; Lee, S.
H.; Michael, B.; Felcetto, T.; Gill, C.; Silver, L. L.;
Hermes, J. D.; Bartizal, K.; Barrett, J.; Schmatz, D.;
Becker, J. W.; Cully, D.; Singh, S. B. Nature 2006, 441,
358.

M. M. Alhamadsheh et al. / Bioorg. Med. Chem. Lett. 17 (2007) 879–883
883
12. He, X.; Reeve, A. M.; Desai, U. R.; Kellogg, G. E.;
Reynolds, K. A. Antimicrob. Agents Chemother. 2004, 48,
3093.
13. Dolak, L. A.; Castle, T. M.; Truesdell, S. E.; Sebek, O. K.
J. Antibiot. 1986, 39, 26.
14. Nishida, I.; Kawaguchi, A.; Yamada, M. J. Biochem.
1986, 99, 1447.
15. Heath, R. J.; White, S. W.; Rock, C. O. Prog. Lipid Res.
2001, 40, 467.
16. Gaul, R. J.; Fremuth, W. J. US Patent 19591102, 1961.
17. Delaunay, D.; Toupet, L.; Le Corre, M. J. Org. Chem.
1995, 60, 6604.
18. Aitken, R. A.; Armstrong, D. P.; Galt, R. H.; Mesher, S.
T. J. Chem. Soc., Perkin Trans. 1 1997, 14, 2139.
19. Thompson, J. D.; Higgins, D. G.; Gibson, T. J. Nucleic
Acids Res. 1994, 22, 4673.
20. DeLano, W. L. T. P. M. G. S. D. S., San Carlos, CA,
USA. http://www.pymol.org.
21. Smilkstein, M.; Sriwilaijaroen, N.; Kelly, J. X.; Wilairat, P.;
Riscoe, M. Antimicrob. Agents Chemother. 2004, 48, 1803.