Modification of the side chain of micromolide, an anti-tuberculosis natural product

Article abstract of DOI:10.1016/j.bmcl.2008.08.027

This paper describes a series of modifications of the side chain of micromolide, an anti-tuberculosis natural product. Most of the synthesized compounds showed significantly decreased activities, which suggests that the long aliphatic side chain of micromolide and its double bond are essential to its activity.

Full text of DOI:10.1016/j.bmcl.2008.08.027

Bioorganic & Medicinal Chemistry Letters 18 (2008) 5311–5315
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Modification of the side chain of micromolide, an anti-tuberculosis
natural product
Hai Yuan a, Rong He a, Baojie Wan b, Yuehong Wang b, Guido F. Pauli a,b, Scott G. Franzblau b,
Alan P. Kozikowski a,
*
a Drug Discovery Program, Department of Medicinal Chemistry and Pharmacognosy, University of Illinois at Chicago, 833 South Wood Street, Chicago, IL 60612, USA
b Institute for Tuberculosis Research, College of Pharmacy, University of Illinois at Chicago, 833 South Wood Street, Chicago, IL 60612, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
This paper describes a series of modifications of the side chain of micromolide, an anti-tuberculosis nat-
ural product. Most of the synthesized compounds showed significantly decreased activities, which sug-
gests that the long aliphatic side chain of micromolide and its double bond are essential to its activity.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 8 July 2008
Revised 6 August 2008
Accepted 8 August 2008
Available online 14 August 2008
Keywords:
Lactone
Micromolide
Natural product
SAR
Tuberculosis
Tuberculosis (TB), a chronic infectious disease caused by myco-
bacteria, primarily Mycobacterium tuberculosis, has threatened
mankind for thousands of years and remains one of the deadliest
diseases worldwide.1 TB is also known as one of the major AIDS-
associated infections. The morbidity caused by TB increases sub-
stantially when the immune system is impaired by HIV infection.2
The current treatment for TB requires a regimen of three to four
drugs for 6 to 9 months.3,4 The lengthy treatment period required
by the current therapies may cause significant toxic side effects
and/or lead to relapse as a consequence of poor patient compli-
ance. In addition, the emergence of drug-resistant TB, such as mul-
ti-drug-resistant TB (MDR-TB) and extensively drug-resistant TB
(XDR-TB) has contributed to the resurgence of tuberculosis.5 De-
spite the need for better TB therapies, no new TB drugs have been
introduced over the last 40 years. Therefore, new anti-TB agents
are urgently needed to shorten the current treatment protocol, to
combat drug-resistant TB, as well as to be compatible with antiret-
roviral drugs for HIV patients.
of a new drug. In addition, the high potency of micromolide against
TB, as well as its low molecular weight and simple structure, makes
it a promising lead compound. We first embarked on the synthesis
of 1 in order to reconcile its reported biological data and provide
larger quantities of 1 for additional animal tests. The total synthe-
sis of racemic 1 has been carried out in our laboratories by follow-
ing a literature procedure, with some modifications (Scheme 1).9
The desired c-lactone was obtained from the corresponding alde-
hyde and ethyl 3-bromoproionate by a SmI2-induced Barbier-type
reaction in the presence of hexamethylphosphoric triamide
(HMPA). Both isomers 1 and 1a in an approximate ratio of 95:5
were successfully separated by HPLC in the final step, which were
created by the Wittig synthesis. The further Pd/C-catalyzed reduc-
tion of racemic 1 provided 1c.
Possible modifications of 1 were then investigated. Despite its
high in vitro anti-TB activity, the high lipophilicity (ClogP = 6.28)10
of 1 is out of the range of Lipinski’s Rule of Five and, therefore, may
limit its therapeutic potential.11 Micromolide is composed of a
c-
Micromolide, ((À)-Z-9-octadecene-4-olide, (À)-1, Fig. 1), a nat-
ural product isolated from the stem bark of Micromelum hirsutum
(Rutaceae), has been reported to show good in vitro anti-TB activ-
lactone ring and a monounsaturated aliphatic side chain. The long
aliphatic side chain of micromolide contributes significantly to its
high lipophilicity. Therefore, a strategy to decrease the lipophilicity
of micromolide and, consequently, to improve its bioavailability is
to introduce polar groups into the aliphatic side chain. In this com-
munication, a series of modifications of the side chain of micromo-
lide are described.
ity (MIC: 1.5 l
g/mL).6 While the anti-TB carbazole alkaloids of
Micromelum hirsutum have already been followed-up in analog
studies,7,8 no efforts have been made thus far to utilize 1 as scaffold
Various functional groups were thus introduced into the side
chain as displayed in Figure 2. The synthesized analogues include
* Corresponding author. Tel.: +1 312 996 7577.
E-mail address: kozikowa@uic.edu (A.P. Kozikowski).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.08.027

5312
H. Yuan et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5311–5315
as lower ClogP values. Acetylenes 13a–b were also prepared to
O
study the importance of the double bond on activity. Another strat-
egy we used to decrease the lipophilicity was to shorten the ali-
phatic chain as in compound 14.
O
(-)-1
Figure 1. Structure of (À)-micromolide.
The synthesis of isoxazolines 8a–d was carried out using a
1,3-dipolar cycloaddition (Scheme 2).17 Condensation of alde-
hydes 15a–b with hydroxylamine hydrochloride provided oximes
16a–b, which were converted to oxime chlorides 17a–b by treat-
ment with N-chlorosuccinimide. Lactones 19a–b were prepared
from 18a–b by treatment with 1-morpholino-2-trimethylsilyl
acetylene followed by aqueous KHF2.18 Cycloaddition of 17a–b
with 19a–b gave 8a–d. A similar cycloaddition of 19a with ethyl
chlorooximidoacetate provided ethyl ester 9b, which was subse-
quently hydrolyzed to give acid 9a (Scheme 3). Compounds 9c–h
were obtained by coupling 9a with corresponding amines and
alcohols.
isoxazoline, isoxazole, amide, acetylene, and a shorter chain ana-
logue. Compounds containing isoxazoline12,13 or isoxazole14–16
moieties have been previously reported as potential anti-TB
agents. Therefore, we prepared 8a–d and 10a to study whether
the isoxazoline or isoxazole rings can be combined with the lac-
tone structure of micromolide. In addition to alkyl chains, esters
and amides were attached to the isoxazoline and isoxazole rings
to introduce more diversity (9a–h, 10b–c). Amide analogues 11
and 12a–b were selected for their simplicity in structure as well
c
OTBDPS
a
b
OTBDPS
d
OH
O
HO
O
HO
e
4
2
3
OH
( )6
OTBDPS
6
5
O
O
f
O
O
( )7
( )6
O
+
7
racemic 1
1a
O
g
1b
Scheme 1. Reagents and conditions: (a) Et3N, t-BuPh2SiCl, DMAP, CH2Cl2, rt; (b) PDC, CH2Cl2, rt; (c) CH3(CH2)8P(Ph3)3+BrÀ, BuLi, THF, 78 °C; (d) Bu4NF, THF, rt; (e) PDC, CH2Cl2,
rt; (f) BrCH2CH2COOEt, SmI2 in THF, HMPA; (g) Pd/C, CH3OH, rt.
N
O
O
( )m
( )n
O
8a: m = 5, n = 4
8b: m = 6, n = 4
8c: m = 5, n = 3
8d: m = 6, n = 3
O
N
O
O
R
O
N
R
10a: R = n-C7H15
O
O
10b: R = COOEt
10c: R = CONH-
O
n
-C6H13
9e: R = NHBz
9f: R = OBz
9g: R = NEt2
9a: R = OH
9b: R = OEt
9c: R = NH-
Isoxazoles
n
-C6H13
9d: R = O-
n-C6H13
9h: R = O-t-Bu
Isoxazolines
O
O
(-)-1
R
O
n
-C8H17
O
O
O
N
H
O
11
13a: R =
13b: R =
n
n
-C8H17
-C7H15
H
N
R
O
O
O
O
n-C7H15
O
12a: R =
12b: R =
n
n
-C8H17
-C7H15
14
Acetylene and Short Chain Analogues
Amides
Figure 2. Analogues of micromolide selected for synthesis.

H. Yuan et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5311–5315
5313
Isoxazoles 10a–c were synthesized in a similar manner to that
described for the preparation of the isoxazoline analogues (Scheme
4). The key intermediate 22 was prepared from 19a via aldehyde
20. Cycloaddition reactions of acetylene 22 with 17b and ethyl
chlorooximidoacetate provided 10a and 10b, respectively. Hexyl
ester 10c was made from 10b via the acid 23.
The synthesis of 12a–b is shown in Scheme 6. Compound 30 was
prepared from bromide 28 via azide 29. Epoxidation of 30 with
MCPBA gave 31, which was converted to lactone 32. Deprotection
of 32 with hydrochloric acid provided amine 33. Amides 12a–b
were prepared by coupling 33 with nonanoic acid and octanoic
acid, respectively.
Amide 11 was synthesized from 24 (Scheme 5). Epoxidation of
24 with MCPBA provided 25, which was then converted to lactone
26. Hydrolysis of 26 with potassium hydroxide provided acid 27.
Compound 11 was obtained by coupling 27 with 1-octanamine.
The synthesis of the acetylene analogues 13a–b is shown in
Scheme 7. Acetylenic coupling of 34a–b with 28 provided 35a–b.
Epoxidation of 35a–b followed by treatment with 1-morpholino-
2-trimethylsilyl acetylene followed by aqueous KHF2 gave
O
( )n
c
O
( )n
O
18a: n = 4
18b: n = 3
19a-b
Cl
N
O
a
b
d
OH
( )m
N
( )m
O
( )m
O
N
OH
( )m
( )n
8a-d
O
15a: m = 5
15b: m = 6
16a-b
17a-b
Scheme 2. Synthesis of 8a–d. Reagents and conditions: (a) NH2OH–HCl, Na2CO3, rt; (b) N-chlorosuccinimide, 40–50 °C; (c) i—1-morpholino-2-trimethylsilyl acetylene,
BF3ÁEt2O, 0 °C; ii—KHF2, rt; (d) Et3N, 0 °C to rt.
R1
N O
N
O
R2
( )4
O
O
c
d
9c, 9e, 9g
N O
N O
a
b
EtO
HO
O
O
O
( )4
O
( )4
O
( )4
19a
O
O
O
N O
9b
9a
O
O
( )4
O
R
O
9d, 9f, 9h
Scheme 3. Synthesis of 9a–h. Reagents and conditions: (a) ethyl chlorooximidoacetate, Et3N, 0 °C to rt; (b) NaOH, 0 °C to rt; (c) R1R2NH, BOP, DMAP, Et3N, rt; (d) ROH, DCC,
DMAP, rt.
O N2O
MeO P C C Me
OMe
21
N O
a
b
c
O
O
O
( )6
O
O
( )4
19a
( )4
( )4
20
O
O
O
O
22
10a
d
N O
N O
N O
f
e
EtO
HO
O
O
O
O
( )4
( )4
( )4
O
O
n-C6H13
O
O
O
O
10b
23
10c
Scheme 4. Synthesis of 10a–c. Reagents and conditions: (a) OsO4, 2,6-lutidine, KIO4, rt; (b) 21, K2CO3, rt; (c) 17b, Et3N, 0 °C to rt; (d) ethyl chlorooximidoacetate, Et3N, 0 °C to
rt; (e) NaOH, 0 °C to rt; (f) n-C6H12OH, DCC, DMAP, rt.
O
O
a
b
c
d
COOEt
( )4
O
COOEt
( )4
O
O
n
-C8H17
COOEt
( )4
COOH
( )4
( )4
O
N
O
O
H
24
25
26
27
11
Scheme 5. Synthesis of 11. Reagents and conditions: (a) MCPBA, 0 °C to rt; (b) i—1-morpholino-2-trimethylsilyl acetylene, BF3ÁEt2O, 0 °C; ii—KHF2; (c) KOH, rt; (d)
nC8H17NH2, BOP, DMAP, Et3N, rt.

5314
H. Yuan et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5311–5315
Table 1
O
a
b
c
PhthN
In vitro activities for 1 analogues against M. tuberculosis H37Rv
Br
N3
PhthN
( )4
( )4
( )4
( )4
Compound
MW
MIC (
MABA
l
M)
MIC (lg/ml)
ClogP
tPSA (Å2)
28
29
30
31
LORA
H
d
e
f
O
(À)-1
( )-1
1a
280.45
280.45
280.45
282.46
1.5
1.93
5.54
6.1
6.1
6.1
6.6
26.3
26.3
26.3
26.3
N
R
O
O
H2N
O
PhthN
( )4
6.90
19.7
>128
124
O
( )4
O
( )4
O
1b
>128
12a: R =
12b: R =
n
n
-C8H17
-C7H15
32
33
Scheme 6. Synthesis of 12a–b. Reagents and conditions: (a) NaN3, rt; (b) phthalic
anhydride, NBu4+CNÀ, rt; (c) MCPBA, 0 °C to rt; (d) i—1-morpholino-2-trimethylsilyl
acetylene, BF3ÁEt2O, 0 °C; ii—KHF2, rt; (e) HCl, reflux; (f) RCOOH, BOP, DMAP,
Et3N, rt.
Table 2
In vitro activities for isoxazoline analogues against M. tuberculosis H37Rv
N O
O
( )n
R
O
acetylene 13a–b. Selective reduction of 13a–b using Lindlar’s cata-
lyst under carefully controlled conditions provided ( )-1 and the
short chain analogue 14. This is also an alternative pathway to syn-
thesize ( )-1.
All of the synthesized compounds were evaluated by the MABA
(microplate Alamar Blue assay)19 method for their activities
against M. tuberculosis strain H37Rv. Some were also tested against
non-replicating cultures using the LORA (low oxygen recovery as-
say) method, but showed no activity.20 The MIC values using the
MABA assay for these compounds are presented in Tables 1–4.
ClogP values for these compounds were calculated using the pro-
gram KOWWIN,10 and the tPSA values were calculated using the
program Molinspiration property calculator.21
Compound
n
R
MIC (
MABA
l
M)
ClogP
tPSA (Å2)
8a
8b
8c
8d
9a
9b
9c
9d
9e
9f
4
4
3
3
4
4
4
4
4
4
4
4
–n-C6H13
–n-C7H15
–n-C6H13
–n-C7H15
–COOH
>128
>128
>128
>128
>128
>128
>128
63.1
>128
>128
>128
>128
4.59
5.08
4.10
4.59
2.51
3.28
4.18
5.25
3.43
4.50
3.41
4.65
47.9
47.9
47.9
47.9
85.2
74.2
77.0
74.2
77.0
74.2
68.2
74.2
–COOEt
–CONH-n-C6H13
COO-n-C6H13
CONHBz
COOBz
CONEt2
9g
9h
COO-t-Bu
MIC values of synthesized ( )-1, 1a, and 1b are shown in Table 1
and were compared with that of (À)-1 isolated from M. hirsutum.
The E isomer 1a is less active than the Z isomer 1. Compound 1c
without a double bond in the side chain shows a dramatic loss of
activity. These results suggest that the conformation and existence
of the double bond in side chain is crucial to achieving good anti-
TB activity. Compared with (À)-1, the racemic form of 1 still retains
a reasonable level of activity, and has the advantage of being easier
to synthesize. Based on these considerations, we prepared the lac-
tone analogues 8–14 in racemic form instead of in optically pure
form for SAR studies. Racemic micromolide ( )-1 was used as a ref-
erence for activity.
Table 2 shows the MIC values for the isoxazoline analogues,
including the alkyl side chain analogues, 8a–d, as well as the car-
boxylic acid, esters, and amides, 9a–h. Most of these compounds
showed little or no activity against M. tuberculosis strain H37Rv ex-
cept the hexyl ester 9d, which exhibited a very modest MIC value
Table 3
In vitro activities for isoxazole analogues against M. tuberculosis H37Rv
N O
O
( )4
R
O
Compound
R
MIC (
MABA
l
M)
ClogP
tPSA (Å2)
10a
10b
10c
–n-C7H15
63.4
>128
>128
5.01
1.84
3.81
52.3
78.6
78.6
–COOEt
–COO-n-C6H13
tive than ( )-1 (MIC = 6.90 lM), was shown to be the most active
one among the synthesized analogues.
According to Tables 2–4, most of the synthesized analogues
showed considerably reduced activity against M. tuberculosis strain
of 63.1
shown in Table 3. Compound 10a showed a modest MIC value of
63.4 M, while no significant activities were observed for esters
lM in MABA. MIC values for isoxazole analogues 10a–c are
l
10b–c. Amide analogies 11 and 12a–b were also found to be inac-
tive (Table 4). Both of the acetylene analogues 13a and 13b also
showed weak activities (MIC values of 44.0 lM and 70.2 lM,
respectively). The acetylene 13b with a shorter side chain was less
active than 13a. This matches observations of a recent SAR study of
anti-TB plant polyacetylenes, which showed that factors other than
the triple bonds contribute significantly to the pharmacophore.22
H37Rv compared with the activity of ( )-1. These results suggest
that the double bond present in the side chain of 1 is essential
for its activity. Replacement of this double bound with alternative
functional groups results in reduced activity. In addition, in view of
the comparison between 13a and 13b, as well as between ( )-1 and
14, it is suggested that decrease of the side chain length also leads
to reduced activity. However, the chain length may not be as cru-
cial for activity as the presence of a double bond.
The short chain analogue 14 (MIC = 14.7
l
M), although still less ac-
R
R
R
a
b
c
d
O
O
O
R
H
O
O
( )4
( )4
( )4
( )4
R
34a: R = n-C H
8
17
35a-b
36a-b
13a-b
(+)-1, 14
34b: R = n-C H
-
7
15
Scheme 7. Synthesis of 13a–b, ( )-1 and 14. Reagents and conditions: (a) i—n-BuLi, HMPA, 78 °C; ii—28, 78 °C to rt; (b) MCPBA, 0 °C to rt; (c) i—1-morpholino-2-trimethylsilyl
acetylene, BF3ÁEt2O, 0 °C; ii—KHF2, rt; (d) H2, Lindlar’s cat, 10 °C.

H. Yuan et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5311–5315
5315
Table 4
Supplementary data
In vitro activities for amide, acetylene, and shorter chain analogues against M.
tuberculosis H37Rv
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.08.027.
R
O
( )4
O
References and notes
Compound
R
MIC (
MABA
l
M)
ClogP
tPSA (Å2)
1. World Health Organization. In Global Tuberculosis Control. WHO Report 2006:
Geneva, Switzerland, 2006.
2. Iademarco, M. F.; Castro, K. G. Semin. Respir. Infect. 2003, 18, 225.
3. British Thoracic Society. Br. J. Dis. Chest. 1984, 78, 330.
4. Ormerod, L. P.; Horsfield, N. Br. J. Dis. Chest. 1987, 81, 268.
5. World Health Organisation. In Report of the meeting of the WHO Global Task
Force on XDR-TB.: Geneva, Switzerland, 2007.
6. Ma, C.; Case, R. J.; Wang, Y.; Zhang, H. J.; Tan, G. T.; Van Hung, N.; Cuong, N. M.;
Franzblau, S. G.; Soejarto, D. D.; Fong, H. H.; Pauli, G. F. Planta Med. 2005, 71,
261.
11
–CONH-n-C8H17
–NHCO-n-C8H17
–NHCO-n-C7H15
–C„C-n-C8H17
–C„C-n-C7H15
>128
>128
>128
44.0
70.2
14.7
3.49
3.49
3.00
5.78
5.29
5.79
55.4
55.4
55.4
26.3
26.3
26.3
12a
12b
13a
13b
14
–cis-CH@CH-n-C7H15
7. Forke, R.; Krahl, M. P.; Krause, T.; Schlechtingen, G.; Knoelker, H.-J. Synlett 2007,
2, 268.
8. Choi, T. A.; Czerwonka, R.; Froehner, W.; Krahl, M. P.; Reddy, K. R.; Franzblau, S.
G. K.; Hans-Joachim ChemMedChem 2006, 1, 812.
9. Cosse, A. A.; Bartelt, R. J.; James, D. G.; Petroski, R. J. J. Chem. Ecol. 2001, 27, 1841.
10. KOWWIN demo version. Syracuse Research Corporation http://
www.syrres.com/esc/est_kowdemo.htm.
11. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Deliv. Rev.
2001, 46, 3.
12. Nicholas, G. M.; Eckman, L. L.; Newton, G. L.; Fahey, R. C.; Raya, S.; Bewley, C. A.
Bioorg. Med. Chem. 2003, 11, 601.
13. Tangallapally, R. P.; Sun, D.; Rakesh, N. B.; Lee, R. E. B.; Lenaerts, A. J. M.;
Meibohm, B.; Lee, R. R. E. Bioorg. Med. Chem. 2007, 17, 6638.
14. Doshi, R.; Kagthara, P.; Parekh, H. Indian J. Chem. B 1999, 38B, 348.
15. Kachhadia, V. V.; Patel, M. R.; Joshi, H. S. J. Sci. I. R. Iran 2004, 15, 47.
16. Mao, J.; Wan, B.; Wang, Y.; Franzblau, S. G.; Kozikowski, A. P. ChemMedChem
2007, 2, 811.
17. Huisgen, R. Angew. Chem., Int. Ed. Engl. 1963, 2, 565.
18. Movassaghi, M.; Jacobsen, E. N. J. Am. Chem. Soc. 2002, 124, 2456.
19. Collins, L.; Franzblau, S. G. Antimicrob. Agents Chemother. 1997, 41, 1004.
20. Cho, S. H.; Warit, S.; Wan, B.; Hwang, C. H.; Pauli, G. F.; Franzblau, S. G.
Antimicrob. Agents Chemother. 2006, 51, 1380.
Among the synthesized compounds, 9d, 10a, 13a–b, and 14
showed moderate activities, although they are less active than
( )-1. However, these compounds are less lipophilic. These five
compounds were calculated to be 3- to 19-fold less lipophilic than
( )-1 and comply better with Lipinski’s Rule of Five. The decreased
lipophilicities may result in better bioavailability and compensate
for the lost in in vitro activities.
In summary, a series of modifications to the side chain of
micromolide have been made. Most of the synthesized compounds
showed significantly decreased activities, which suggests that the
long aliphatic side chain of micromolide and its double bond con-
tribute to its activity. Despite the reduced in vitro activities, some
moderately active analogues showing lower lipophilicities and lar-
ger PSA values were identified, and these compounds may possess
improved ADME parameters when studied in vivo.
21. Molinspiration property engine v2007.04. Molinspiration Cheminformatics.
(http://www.molinspiration.com/cgi-bin/properties).
22. Deng, S.; Wang, Y.; Inui, T.; Chen, S.-N.; Farnsworth, N. R.; Cho, S.; Franzblau, S.
G.; Pauli, G. F. Phytother. Res. 2008, 22, 878.
Acknowledgment
This work was supported by 2006 UIC-ITR award.