Synthesis and biological activity of enantiomeric pairs of 5-[(E)-cycloalk-2-enylidenemethyl]thiolactomycin congeners

Article abstract of DOI:10.1016/j.bmcl.2008.08.103

The title congeners were synthesized by employing our efficient synthetic route previously explored for preparing enantiomeric pairs of thiolactomycin and its 3-demethyl derivative. While all the synthesized congeners lacked in vitro antibacterial activit

Full text of DOI:10.1016/j.bmcl.2008.08.103

Bioorganic & Medicinal Chemistry Letters 18 (2008) 5598–5600
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and biological activity of enantiomeric pairs
of 5-[(E)-cycloalk-2-enylidenemethyl]thiolactomycin congeners
b
Kohei Ohata a, , Shiro Terashima
*
a Kyorin Pharmaceutical Co., Ltd, Discovery Research Laboratories, 2399-1 Nogi, Nogi-Machi, Shimotsuga, Tochigi 329-0114, Japan
b Sagami Chemical Research Center, 2743-1 Hayakawa, Ayase, Kanagawa 252-1193, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The title congeners were synthesized by employing our efficient synthetic route previously explored for
preparing enantiomeric pairs of thiolactomycin and its 3-demethyl derivative. While all the synthesized
congeners lacked in vitro antibacterial activity, some of the congeners bearing an (E)-cyclohept-2-eny-
lidenemethyl or an (E)-cyclooct-2-enylidenemethyl group were found to exhibit more potent type I
FAS inhibitory activity than (S)-3-demethylthiolactomycin having an unnatural configuration.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 29 July 2008
Revised 27 August 2008
Accepted 28 August 2008
Available online 31 August 2008
Keywords:
Antibiotic
Inhibitor of mammalian type I fatty acid
synthase
5-[(E)-Cycloalk-2-
enylidenemethyl]thiolactomycin
(R)-(+)-Thiolactomycin (TLM, 1), a thiolactone antibiotic iso-
lated from
soil bacterium, Nocardia sp.,1 shows moderate
tion of the side chain at the C5-position.8 In the course of our
continuing studies on the synthesis and biological activity of 1
and its congeners, we next paid attention to exploring the confor-
mational effects of the isoprenoid 1,3-diene moiety at the C5-posi-
tion of 1 on in vitro antibacterial and type I FAS inhibitory activity.
Accordingly, we designed enantiomeric pairs of 5-[(E)-cycloalk-2-
enylidenemethyl]-TLM congeners and their 3-demethyl deriva-
tives (4, ent-4, 5 and ent-5) bearing six- to eight-membered rings
as the next targets.9 In these compounds, the isoprenoid 1,3-diene
moiety for 1 is clearly locked into an s-trans configuration as a con-
sequence of their involvement in a ring system. It is also expected
that successful synthesis of 4, ent-4, 5 and ent-5 might further
demonstrate the flexibility and efficiency of our synthetic scheme
developed for 1 and its congeners.7,8 Here, we wish to report the
synthesis of 4, ent-4, 5 and ent-5 and their in vitro antibacterial
and type I FAS inhibitory activity. These studies disclosed novel as-
pects of the structural-activity relationships for TLM congeners
especially concerning the isoprenoid 1,3-diene moieties at the
C5-position.
a
in vitro activity against a number of pathogens, including Gram-
positive and Gram-negative bacteria,2 Mycobacterium tuberculosis3
and malaria parasites (Fig. 1).4 It has also been reported that 1
inhibits bacterial and plant type II fatty acid synthase (FAS).5 These
inhibitory activities are considered to explain the antibacterial and
antiparasitical properties observed for 1.5 Interestingly, Townsend
et al. recently disclosed that 1 and its derivatives also show inhib-
itory activity against mammalian type I FAS.6 Therefore, 1 and its
congeners attract much synthetic attention because they seem to
constitute promising drug candidates with a hitherto unexplored
mode of action for cancer and obesity treatments, as well as infec-
tive diseases.
Recently, we reported an efficient total synthesis of enantio-
meric pairs of 1 and its 3-demethyl derivative (3-demethyl-TLM
2) by employing novel deconjugative asymmetric
a-sulfenylation
of the chiral 3-( ,b, ,d-unsaturated-acyl)oxazolidin-2-one with a
a
c
methanethiosulfonate as a key step.7 The flexibility and efficiency
of the explored synthetic route were next demonstrated by the
successful synthesis of enantiomeric pairs of various structural
types of 5-vinyl-TLM congeners (3).8 From the biological activity
assay carried out using 3 and ent-3 along with 1, ent-1, 2 and
ent-2, it appeared evident that in vitro antibacterial and type I
FAS inhibitory activity of TLM congeners can be cleanly separated
by changing not only the structure but also the absolute configura-
According to the explored synthetic scheme,7 the synthesis of
(R)-5-[(E)-cycloalk-2-enylidenemethyl]-TLM and its 3-demethyl
analogs (4 and 5) commenced from
acids 7a–c, which were prepared from
a
a
,b-unsaturated carboxylic
,b-unsaturated aldehydes
6a–c by sequential Horner–Wadworth–Emmons reaction and alka-
line hydrolysis (Scheme 1).10 While 6a was commercially available,
6b and 6c were prepared from 1-nitromethylcycloheptene11 and
N0-cyclooctylidenetosylhydrazide,12 respectively, following the re-
ported procedures.11,12 Activation of 7a–c with pivaloyl chloride
followed by treating the formed mixed anhydrides with (R)-4-ben-
* Corresponding author. Fax: +81 0280 57 1293.
E-mail address: kouhei.oohata@mb.kyorin-pharm.co.jp (K. Ohata).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.08.103

K. Ohata, S. Terashima / Bioorg. Med. Chem. Lett. 18 (2008) 5598–5600
5599
R1
HO
R1
HO
HO
R1
3
3
3
R2
5
Me
O
5
Me
O
O
S
S
M5e
S
n
4: R1 = Me; 5: R1 = H
3: R1= Me or H;
R2= H, Me, Et, nPr, nBu or nHex
thiolactomycin (1): R1= Me
3-demethylthiolactomycin (2): R1 = H
a: n = 1; b: n = 2; c: n = 3.
Figure 1. Structures of thiolactomycin (1), 3-demethylthiolactomycin (2), 5-vinylthiolactomycin congeners (3), 5-[(E)-cycloalk-2-enylidenemethyl]thiolactomycin and 5-
[(E)-cycloalk-2-enylidenemethyl]-3-demethylthiolactomycin derivatives (4 and 5).
zyloxazolidin-2-one afforded (R)-N-acyloxazolidin-2-ones 8a–c.
With 8a–c in hand, the key asymmetric deconjugative -sulfenyla-
tion was attempted. Thus, treatments of 8a,b with 3,3-dimethoxy-
against type I FAS so far reported.6,21 Being different from 1 and
2, all the tested compounds were found to lack in vitro antibacte-
rial activity against all the tested strains of bacteria even though
4a–c and 5a–c bear the same (R)-configurations as 1 and 2. As
for type I FAS inhibitory activity, fairly comparable activity was ob-
served for ent-5a, 4b, ent-4b, 5b, ent-5b, ent-4c, 5c and ent-5c
whose isoprenoid 1,3-diene moieties are involved in a ring system.
In particular, ent-4b, 5b and ent-4c showed clearly more potent
type I FAS inhibitory activity than that of the previously synthe-
sized ent-2,7 and the activity of the most potent ent-4c was almost
twice as much as that of ent-2.
In conclusion, we have succeeded in the synthesis of six enan-
tiomeric pairs of (R)-5-[(E)-cycloalk-2-enylidenemethyl]-TLM
congeners (4, ent-4, 5 and ent-5) by employing our efficient syn-
thetic route previously established for the total synthesis of TLM
(1), 3-demethyl-TLM (2) and their enantiomers (ent-1 and ent-2).
From the results of in vitro antibacterial activity, it appeared evi-
dent that the free-rotational isoprenoid 1,3-diene moiety and the
(R)-configuration at the C5-position are essential for 1 and its
congeners to exhibit in vitro antibacterial activity. As for type I
FAS inhibitory activity, some 5-[(E)-cycloalk-2-enylidenemethyl]-
TLM congeners in which the isoprenoid 1,3-diene moieties are
locked into an s-trans configuration as a consequence of their
involvement in a ring system were found to exhibit inhibitory
activity equal to (for ent-5a, 4b, ent-5b, 5c and ent-5c) or a little
more potent (for ent-4b, 5b and ent-4c) than that of the previously
reported ent-2.7 The results on biological activity collected in these
studies should be useful for future attempts to design novel thio-
lactomycin congeners that might show more prominent and clini-
cally useful activity than TLM.
a
propyl methanethiosulfonate using NaHMDS as a base in the
presence of HMPA led to the asymmetric deconjugative
a-sulfeny-
lation, giving rise to the -sulfenylated products 9a,b highly dia-
a
stereoselectively (>90% de) and in good yields.13–16 However, the
reaction of 8c under the same conditions as for 8a,b was found
to give a lower yield of 9c. After some experimentation, it was dis-
closed that the use of KHMDS as a base in the presence of 18-
crown-6 was more effective, giving 9c in a highly diastereoselec-
tive manner (>90% de) in a good yield.13–16
The
a
-sulfenylated products 9a–c isolated in a pure state14
were transformed into benzyl esters 10a–c by imide-ester ex-
change using titanium benzyloxide.17 Acidic hydrolysis of the di-
methyl acetal moieties in 10a–c produced aldehydes 11a–c.
Compounds 11a–c were sequentially subjected to retro-Michael
reaction using Cs2CO3 as a base and acylation with propionyl or
acetyl chloride, furnishing
a-acylthio esters 12a–c or 13a–c,
respectively. Dieckmann condensation of 12a–c and 13a–c with
LiHMDS gave rise to (R)-5-[(E)-cycloalk-2-enylidenemethyl]-TLM
4a–c and its 3-demethyl derivatives 5a–c.18 By the same synthetic
scheme, the enantiomers of 4a–c and 5a–c (ent-4a–c and ent-5a–c)
were prepared from ent-8a–c obtained from 7a–c and (S)-4-ben-
zyloxazolidin-2-one.18
With 4a–c, ent-4a–c, 5a–c and ent-5a–c in hand, their in vitro
antibacterial activity against various strains of bacteria19 and
inhibitory activity against mammalian type I FAS20 were evaluated.
The results are summarized in Table 1 along with those for 1, ent-1,
2, ent-2, ciprofloxacin (CPFX) and C75, the only potent inhibitor
O
O
O
O
CHO
CO2H
α
N
a
b
c
O
S
Me
N
O
n
n
n
n
Ph
Ph
7
6
MeO
OMe
9
8
CO2Bn
CO2Bn
Me
CO2Bn
g
d
S
Me
f
e
S
S
Me
n
4 and 5
n
n
O
CHO
11
R
MeO
OMe
10
12: R = Me
13: R= H
a: n = 1; b: n = 2; c: n = 3.
Scheme 1. Reagents and conditions: (a) (EtO)2P(O)CHMeCO2Et, tBuOLi, hexane, rt, then 10% NaOH aq, EtOH, 50 °C, 84% for 7a, 69% for 7b, 81% for 7c; (b) tBuCOCl, Et3N, THF,
ꢀ15 °C, then LiCl, (R)-4-benzyl-2-oxazolidinone, rt, 81% for 8a, 90% for 8b, 67% for 8c; (c) NaHMDS, HMPA, ꢀ78 °C or KHMDS, 18-crown-6, ꢀ78 °C, then 3,3-dimethoxypropyl
methanethiosulfonate, ꢀ78 to 0 °C, 56% for 9a, 31% for 9b, 50% for 9c; (d) Ti(OiPr)4, BnOH, 70 °C, 76% for 10a, 69% for 10b, 87% for 10c; (e) 6% HCl aq, THF, rt, 87% for 11a, 92%
for 11b, 96% for 11c; (f) Cs2CO3, EtOH, 0 °C, then CH3CH2COCl or CH3COCl, Et3N, CH2Cl2, 0 °C, 66% for 12a, 68% for 12b, 73% for 12c, 68% for 13a, 63% for 13b, 67% for 13c; (g)
LiHMDS, THF, ꢀ78 to 0 °C, 84% for 4a, 89% for 4b, 66% for 4c, 99% for 5a, 64% for 5b, 46% for 5c.

5600
K. Ohata, S. Terashima / Bioorg. Med. Chem. Lett. 18 (2008) 5598–5600
Table 1
In vitro antibacterial and mammalian type I FAS inhibitory activity of enantiomeric pairs of TLM and its congeners (1, ent-1, 2, ent-2, 4a-c, ent-4a-c, 5a-c and ent-5a-c)
Compound
In vitro antibacterial activity, MIC (
lg/mL)
Mammlian type I FAS inhibitory activity,
IC50 (ng/mL) HepG2 14C
S. aureus Smith
M. catarrhalis ATCC 25238
H. influenzae IID983
TLM (1)
ent-1
2
ent-1
4a
ent-4a
5a
ent-Sa
4b
ent-4b
5b
ent-Sb
4c
ent-4c
5c
ent-5c
CPFXa
C75c
128
>128
>128
>128
>64
>64
>64
>64
>64
>64
>64
>64
>128
>128
>128
>128
0.063
N.T.b
0.25
>128
16
2
>80
43.7
>80
19.0
>80
>80
57.0
18.9
25.4
14.9
13.3
22.6
41.8
11.6
19.5
24.6
N.T.b
7.4
N.T.b
32
>128
>64
>64
>64
>64
>64
>64
>64
>64
64
N.T.b
>64
>64
>64
>64
>64
>64
>64
>64
>128
>128
>128
>128
0.008
N.T.b
64
128
64
0.031
N.T.b
a
Ciprofloxacin.
N.T., not tested.
See the text.
b
c
Acknowledgments
10. The (E)-configuration of 7a was determined by observing NOESY between two
olefinic protons in its 1H NMR spectrum. The structures of 7b,c were assigned
by comparing their 1H NMR spectra with that of 7a.
11. Tamura, R.; Sato, M.; Oda, D. J. Org. Chem. 1986, 51, 4368.
12. Naya, A.; Ishikawa, M.; Matsuda, K.; Ohwaki, K.; Saeki, T.; Noguchi, K.; Ohtake,
N. Bioorg. Med. Chem. 2004, 11, 875.
We are grateful to Dr. T. Ishizaki of Kyorin Pharmaceutical Co.,
Ltd for his extensive support and Dr. Y. Fukuda of Kyorin Pharma-
ceutical Co., Ltd for his valuable suggestions and discussions. The
in vitro antibacterial and mammalian type I FAS inhibitory activity
assay were carried out by Drs. M. Takei and M. Tsunoda of Kyorin
Pharmaceutical Co., Ltd, to whom the authors’ thanks are due.
13. In the asymmetric deconjugative
a-sulfenylation of 8a–c, formation of
small amounts of the -sulfenylated (S)-diastereomer and the
a
a-
sulfenylated (Z)-isomer [(R)- or (S)-diastereomer] was always observed.
Their formation ratios estimated by the 1H NMR spectra and/or HPLC
analysis are as follows: 9:(S)-isomer:(Z)-isomer; 12:1:1 for 9a; 14:1:4 for
9b; 12:1:1 for 9c.
References and notes
1. Oishi, H.; Noto, T.; Sasaki, H.; Suzuki, K.; Hayashi, T.; Okazaki, H.; Ando, K.;
Sawada, M. J. Antibiot. 1982, 35, 391.
2. (a) Noto, T.; Miyakawa, S.; Oishi, H.; Endo, H.; Okazaki, H. J. Antibiot. 1982, 35,
401; (b) Miyakawa, S.; Suzuki, K.; Noto, T.; Harada, Y.; Okazaki, H. J. Antibiot.
1982, 35, 411.
3. (a) Kremer, L.; Douglas, J. D.; Baulard, A. R.; Morehouse, C.; Guy, M. R.; Alland,
D.; Dover, L. G.; Lakey, J. H.; Jacobs, W. R.; Brennan, P. J.; Minnikin, D. E.; Besra,
G. S. J. Biol. Chem. 2000, 275, 16857; (b) Slayden, R. A.; Lee, R. E.; Armour, J. W.;
Cooper, A. M.; Orme, I. M.; Brennan, P. J.; Besra, G. S. Antimicrob. Agents
Chemother. 1996, 40, 2813; (c) Kim, P.; Zhang, Y.-M.; Shenoy, G.; Nguyen, Q.-A.;
Boshoff, H. I.; Manjunatha, U. H.; Goodwin, M. B.; Lonsdale, J.; Price, A. C.;
Miller, D. J.; Duncan, K.; White, S. W.; Rock, C. O.; Barry, C. E., III; Dowd, C. S. J.
Med. Chem. 2006, 49, 159.
4. (a) Waller, R. F.; Ralph, S. A.; Reed, M. B.; Su, V.; Douglas, J. D.; Minnikin, D. E.;
Cowman, A. F.; Besra, G. S.; McFadden, G. I. Antimicrob. Agents Chemother. 2003,
47, 297; (b) Jones, S. M.; Urch, J. E.; Brun, R.; Harwood, J. L.; Berry, C.; Gilbert, I.
H. Bioorg. Med. Chem. 2004, 12, 683.
14. Pure samples of 9a–c were obtained by sequential separation with column
chromatography (SiO2:solvent; hexane/AcOEt = 4/1 for 9a; hexane/
AcOEt = 5/1 for 9b; hexane/AcOEt = 5/1 for 9c) and HPLC. The HPLC
conditions were as follows: 9a [Daicel Chiralpak IA,
hexane/EtOH = 90/10, followed by Daicel Chiralpak IA,
u
u
2.0 cm ꢁ 25 cm,
2.0 cm ꢁ 25 cm,
hexane/iPrOH = 97/3]; 9b [Daicel Chiralpak IA,
EtOH = 91/9]; 9c [Daicel Chiralpak IA,
iPrOH = 80:10:10, followed by Daicel Chiralpak IC,
hexane/iPrOH = 75/25].
u
2.0 cm ꢁ 25 cm, hexane/
u
2.0 cm ꢁ 25 cm, hexane/MTBE/
u
2.0 cm ꢁ 25 cm,
15. The absolute stereochemistries of newly created asymmetric centers for 9a–c
were assigned to have an (R)-configuration by comparing their 1H NMR spectra
with that of the corresponding a-sulfenylated intermediate for the synthesis of
1.7 Some representative data are as follows: 1H NMR(CDCl3): 5.44 ppm
[(Me)(SR)C–CH@C] for 9a; 5.50 ppm [(Me)(SR)C–CH@C] for the (S)-
diastereomer of 9a; 5.72 ppm [(Me)(SR)C–CH@C] for the synthetic
intermediate of 1; 5.77 ppm [(Me)(SR)C–CH@C] for the (S)-diastereomer of
the synthetic intermediate of 1.
5. (a) Hayashi, T.; Yamamoto, O.; Sasaki, H.; Kawaguchi, A.; Okazaki, H. Biochem.
Biophys. Res. Commun. 1983, 115, 1108; (b) Nishida, I.; Kawaguchi, A.; Yamada,
M. J. Biochem. (Tokyo) 1986, 99, 1447.
6. McFadden, J. M.; Medghalchi, S. M.; Thupari, J. N.; Pinn, M. L.; Vadlamudi, A.;
Miller, K. I.; Kuhajda, F. P.; Townsend, C. A. J. Med. Chem. 2005, 48, 946.
7. Ohata, K.; Terashima, S. Tetrahedron Lett. 2006, 47, 2787.
8. Ohata, K.; Terashima, S. Bioorg. Med. Chem. Lett. 2007, 17, 4070.
9. In addition to 4a–c, 5a–c and their enantiomers (ent-4a–c and ent-5a–c), 5-
[(E)-cyclopent-2-enylidenemethyl]-TLM congeners (4d, ent-4d, 5d and ent-5d)
were designed and their synthesis was attempted following the same synthetic
scheme shown in Scheme 1. Although the synthetic scheme smoothly
proceeded to the stage of 10 (n = 0), aldehyde 11 (n = 0) derived from 10
(n = 0) was found to be very unstable, probably due to facile isomerization to
the cyclopentadiene derivative and subsequent intramolecular Diels–Alder
reaction (Ohata, K.; Terashima, S., to be published).
16. The
a-sulfenylated products 9a,b bearing (R)-configuration clearly showed
NOESY between two olefinic protons in their 1H NMR spectra. The
structures of (Z)-isomers of 9a,b were determined by observing the
absence of NOESY between two olefinic protons in their 1H NMR spectra.
Assignment of the structures for (E)- and (Z)-isomer of 9c was performed
by comparing their 1H NMR spectra with those of the corresponding
isomers of 9a,b.
17. Evans, D. A.; Britton, T. C.; Ellman, J. A.; Dorow, R. L. J. Am. Chem. Soc. 1990, 112,
4011.
18. The physical data of 4a–c and 5a–c are as follows: 4a: colorless crystals, mp
111–113 °C, ½a 2D7
ꢂ
+ 128° (c 0.30, MeOH); 4b: light brown oil, ½a D28
+ 86.3° (c
ꢂ
0.24, MeOH); 4c: colorless crystals, mp 105–111 °C, ½ ꢂ + 77.3° (c 0.30,
a 2D7
MeOH); 5a: colorless solid, mp 105.5–110 °C, ½a D25
+ 138° (c 0.20, MeOH); 5b:
ꢂ
colorless crystals, mp 85–89 °C, ½a D27
+ 137° (c 0.21, MeOH); 5c: colorless
ꢂ
crystals, mp 131–137 °C, ½a D27
+ 88.9° (c 0.30, MeOH).
ꢂ
19. Determination of MICs by agar dilution methods was performed according to
the guidelines M7–A6 of the Clinical and Laboratory Standards Institute, 2003.
20. Fatty acid synthesis was evaluated measuring incorporation of [1-14C] acetate
into cellular fatty acid as previously described with some modifications. See,
Murakami, K.; Tobe, K.; Ide, T.; Mochizuki, T.; Ohashi, M.; Akanuma, Y.; Yazaki,
Y.; Kadowaki, T. Diabetes 1998, 47, 1841.
R1
HO
O
S
Me
4d: R1 = Me; 5d: R1 = H
21. Kuhajda, F. P.; Pizer, E. S.; Li, J. N.; Mani, S.; Frehywot, G. L.; Townsend, C. A.
Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 3450.