Rational design and semisynthesis of betulinic acid analogues as potent topoisomerase inhibitors

Article abstract of DOI:10.1021/np900312u

Chemical transformation studies were conducted on betulinic acid (1), a common plant-derived lupane-type triterpene. Eleven new rationally designed derivatives of 1 (2-5 and 7-13) were synthesized based on docking studies and tested for their topoisomerase I and IIR inhibitory activity. Semisynthetic reactions targeted C-3, C-20, and C-28 in 1. Structures of the new compounds were confirmed by spectroscopic methods (1D and 2D NMR and MS). Compound 9, 3-O-[N-(phenylsulfonyl)carbamoyl-17β-N-(phenylsulfonyl)amide]betulinic acid, showed 1.5-fold the activity of CPT in a topoisomerase I DNA relaxation assay. Four out of 14 betulinic acid analogues (5, 9, 11, and 12) showed 1.5-fold the activity of etoposide in a topoisomerase II assay. The new analogues exhibited better cytotoxic activities against the human colon cancer cells SW948 and HCT-116 and the breast cancer cell line MDA-MB-231 compared to the parent (1). Betulinic acid (1) is a potential scaffold for the design of new topoisomerase I and IIα inhibitors.

Full text of DOI:10.1021/np900312u

J. Nat. Prod. 2009, 72, 1643–1650
1643
Rational Design and Semisynthesis of Betulinic Acid Analogues as Potent Topoisomerase
Inhibitors
Fatma M. Abdel Bar,†,‡ Mohammad A. Khanfar,† Ahmed Y. Elnagar,† Hui Liu,§ Ahmed M. Zaghloul,‡ Farid A. Badria,‡
Paul W. Sylvester,† Kadria F. Ahmad,‡ Kevin P. Raisch,§ and Khalid A. El Sayed*,†
Department of Basic Pharmaceutical Sciences, College of Pharmacy, UniVersity of Louisiana at Monroe, Monroe, Louisiana 71201,
Department of Pharmacognosy, Faculty of Pharmacy, Mansoura UniVersity, Mansoura 35516, Egypt, and Department of Radiation Oncology,
UniVersity of Alabama at Birmingham, Birmingham, Alabama 35294
ReceiVed May 20, 2009
Chemical transformation studies were conducted on betulinic acid (1), a common plant-derived lupane-type triterpene.
Eleven new rationally designed derivatives of 1 (2-5 and 7-13) were synthesized based on docking studies and tested
for their topoisomerase I and IIR inhibitory activity. Semisynthetic reactions targeted C-3, C-20, and C-28 in 1. Structures
of the new compounds were confirmed by spectroscopic methods (1D and 2D NMR and MS). Compound 9, 3-O-
[N-(phenylsulfonyl)carbamoyl-17ꢀ-N-(phenylsulfonyl)amide]betulinic acid, showed 1.5-fold the activity of CPT in a
topoisomerase I DNA relaxation assay. Four out of 14 betulinic acid analogues (5, 9, 11, and 12) showed 1.5-fold the
activity of etoposide in a topoisomerase II assay. The new analogues exhibited better cytotoxic activities against the
human colon cancer cells SW948 and HCT-116 and the breast cancer cell line MDA-MB-231 compared to the parent
(1). Betulinic acid (1) is a potential scaffold for the design of new topoisomerase I and IIR inhibitors.
Betulinic, boswellic, ursolic, and oleanolic acids are abundant
pentacyclic triterpenes that occur in many edible fruits and
vegetables.1 They have been reported to inhibit topoisomerases I
and IIR by competing with DNA for topoisomerase binding through
direct interaction via preventing topoisomerase-DNA complex
formation.1 Structure-activity relationship studies suggested that
the pentacyclic skeleton and carboxyl group were important
pharmacophores for topoisomerase inhibitory activity.1
Betulinic acid (1) is a common lupane-type pentacyclic triterpene
known to exhibit immunomodulation,2 inhibitory effects on
Epstein-Barr virus early antigen activation,3 and anti-HIV, anal-
gesic, anti-inflammatory, antibacterial, antimalarial, antihelmintic,
and antitumor activities.4 The mechanism of action of 1 has been
suggested to be a multitarget inducer of apoptosis.5,6 It suppressed
NF-κB activation through the inhibition of IκBR kinase and p65
phosphorylation induced by carcinogen and inflammatory stimuli.5
Thus, 1 inhibited the production of NF-κB-regulated gene products
such as cyclooxygenase-2 and matrix metalloproteinase-9.5-10
Topoisomerases are essential enzymes that relax DNA super-
coiling inside cells during processes such as replication, recombina-
tion, and transcription.11,12 They alter DNA topology by generating
transient breaks in the double helix.11,12 Eukaryotic type I topoi-
somerases are monomeric enzymes that alter topology by creating
transient single-stranded breaks in the DNA. Topoisomerase I
functions through three major steps: (a) nucleophilic attack of DNA
by the OH group of the active site of topoisomerase I (tyrosine
723, Ty 723), resulting in a covalent attachment of topoisomerase
I to the broken DNA strand and formation of a transient phospho-
tyrosine ester bond between the Tyr 723 and the DNA scissile strand
backbone phosphate; (b) DNA unwinding (strand passage or free
rotation); and (c) a second transesterification reaction resulting in
religation of the DNA strand and dissociation of the topoisomerase
I-DNA complex.14,15
II enzymes cut both strands of the double helix, each protomer
subunit contains tyrosine 805, Tyr 805 in the case of human
topoisomerase IIR.13,14 Topoisomerase II initiates DNA cleavage
by nucleophilic attack of the active site tyrosine on the phosphate
of the nucleic acid backbone.13 The active site’s tyrosine residue
of each topoisomerase II protomer subunit becomes covalently
linked to the newly generated 5′-terminal phosphate moiety on each
strand.13 This step is followed by strand passage and DNA
ligation.13
Under normal conditions, topoisomerase-DNA covalent com-
plexes are transient and found at very low levels.13-15 Stabilization
of these complexes generates DNA lesions, inducing cell arrest and
apoptosis.13-15 Compounds that inhibit the catalytic activity of
topoisomerase II can be either catalytic inhibitors, which decrease
the overall activity of the enzyme, or topoisomerase II poisons,
which increase the levels of topoisomerase II-DNA cleavage
complex.13
In this study, a series of new betulinic acid analogues (2-5 and
7-13) were designed, synthesized, and tested for their ability to
inhibit both types of topoisomerases.
Results and Discussion
Molecular Modeling and Docking. To design more active
analogues of 1 as topoisomerase inhibitors, a computer-aided
molecular modeling study was carried out within the active site of
the human topoisomerases I and IIR, starting with the crystal
structure of the enzyme bound to DNA (PDB 1t8i and 1bgw for
topoisomerases I and IIR, respectively).16,17 Several analogues of
1 with modified C-3, C-28, and C-20 were proposed and docked
to topoisomerase I and IIR active sites. The structures of the
topoisomerase I and II inhibitors camptothecin (CPT) and etoposide,
respectively, were used as reference molecules in docking and
modeling studies.
The active site of topoisomerase I is an arginine- and lysine-
rich pocket with condensed positive charge that allows electrostatic
interactions with the negatively charged DNA-phosphate backbone.
The C-28 COOH group of 1 represents the exclusive binding
pharmacophore, generating hydrogen-bonding interactions with the
structural skeleton that fit the catalytic residues Arg 488, Lys 532,
and Tyr 723, but it is too short to bridge to the other side of the
topoisomerase I-DNA binding pocket (Figure 1A). During topoi-
somerase I-catalyzed DNA supercoiling, Tyr 723 is involved in
Eukaryotic type II topoisomerases function as homodimers. The
enzyme unwinds, unknots, and untangles the genetic material by
generating transient double-stranded breaks in DNA.13,14 Since type
* To whom correspondence should be addressed. Tel.: 318-342-1725.
Fax: 318-342-1737. E-mail: elsayed@ulm.edu.
† University of Louisiana at Monroe.
‡ Mansoura University.
§ University of Alabama at Birmingham.
10.1021/np900312u CCC: $40.75  2009 American Chemical Society and American Society of Pharmacognosy
Published on Web 08/19/2009

1644 Journal of Natural Products, 2009, Vol. 72, No. 9
Bar et al.
Figure 1. Detailed view of docked structures 1 (A), 9 (B), and 14 (C) with the corresponding interacting amino acids of topoisomerase I,
and 11 (D) with topoisomerase IIR binding site.
reversible formation of a 3′-O-phospho-tyrosine bond between the
active tyrosine site and the cleaved DNA backbone, and therefore
this particular amino acid is critical for the enzyme’s activity.18,19
The aromatic-sulfonyl substitutions at the C-3 OH and C-28
COOH moieties of 1 allow charge-transfer interaction of the
π-aromatic cloud with the guanidine and amine groups of arginine
and lysine residues, respectively (Table 3). The most active analogue
of 1 in the topoisomerase I-DNA relaxation assay (9) showed, in
addition to interaction of the C-3 side chain π-aromatic cloud
mentioned above, the insertion of the C-3 benzenesulfonyl car-
bamoyl moiety in the positively charged pocket via the H-bonding
interactions of the C-1′ carbamate carbonyl with Arg 488 and the
interaction of the sulfone group with Lys 587 and the critical
conserved catalytic residue Tyr 732 (Figure 1B). Furthermore, the
C-28 N-phenylsulfonylamide moiety in 9 allows bridging between
both sides of the topoisomerase I active site through a complex
H-bonding interaction of its sulfone group with Trp 412. These
electrostatic interactions were also generated in other aromatic-
substituted active analogues, 8 and 13. Therefore, the topoisomerase
I inhibitory activity of 8, 9, and 13 can be explained by the blocking
of this important catalytic residue, which drastically compromises
topoisomerase I activity.
Oxidation of the isopropenyl side chain to C-30 aldehyde in 14
enhanced the topoisomerase I binding affinity (Figure 1C). The C-30
aldehyde group creates H-bond interactions with Arg 488 and Tyr
723 (Figure 1C). The C-28 COOH group binds the Lys 587 residue
at one side, while the C-3 OH group binds Lys 436 at the other
side (Figure 1C).
Molecular modeling simulations were also applied to understand
the interactions of analogues of 1 with topoisomerase IIR. Starting
with the crystallographic structure of human topoisomerase IIR
available in the RCSB PDB database (code, 1zxm),20 various
possible pockets were identified. ATP, nucleotide, and DNA binding
sites were investigated and docked by analogues of 1. Unfortunately,
all docking attempts were unsuccessful, and severe crash or
inappropriate superficial bindings were observed. The binding site
of the eukaryotic yeast Saccharomyces cereVisiae topoisomerase
II was previously identified.21 Human and S. cereVisiae DNA
topoisomerase II sequences exhibit 47.3% identity and 68.2%
similarity.21 This good sequence similarity, particularly at the DNA-
binding domain, allowed molecular modeling simulations and
docking studies at the former domain.21 The conserved amino acids
at the DNA-binding domain, Asp 799, Arg 804, and the catalytic
residue Tyr 805, were used to direct the protomol generation, and
then the search grid was expanded to 15 Å to fit the binding of
bulky betulinic acid derivatives.
Similar to topoisomerase I, the S. cereVisiae topoisomerase II
DNA-binding domain is an arginine- and lysine-rich pocket with
condensed positive charge that permits a strong charge-transfer
interaction of the π-aromatic cloud of the active betulinic acid
analogues (5, 8, 9, 10, 11, and 13) with guanidine and amine groups
of arginine and lysine residues, respectively. Compound 11 showed
a strong H-bonding interaction with the catalytic residue Tyr 805,
which is a critical amino acid for the enzyme’s activity, through
the carbamoyl moiety (Figure 1D). The C-28 carboxyl group
bridged 11 across the DNA-binding domain and H-bonded to His
1012 and Arg 1016 (Figure 1D). The biphenyl ring in 11 was
sandwiched between two positively charged Lys residues, Lys 713
and Lys 812, creating strong electrostatic interactions. Docked poses
of compounds 5 and 12 are available in the Supporting Information
(Figures S26 and S27, respectively).
Thus, aromatic or olefinic substitutions at C-3 markedly increase
the binding affinity of 1 to both topoisomerse types, as indicated
by the relatively high docking scores of compounds 5 and 8-13
(Table 3). However, oxidation of the isopropenyl side chain to an
aldehyde, as in 3 and 14, enhanced the topoisomerase I docking
score without marked effects on topoisomerase IIR binding affinity
(Table 3). The C-30 aldehyde group in 14 created H-bond
interactions in the case of topoisomerase I with Arg 488 and Tyr
723 (Figure 1C). Thus, compounds 2-14 were synthesized and
tested for their in vitro topoisomerase I and IIR inhibitory activities.
Semisynthesis of Betulinic Analogues 2-14. Reaction of
m-chloroperbenzoic acid with 1 afforded products 2 and 3. The
HREIMS, 1H NMR, and 13C NMR data of 2 (Supporting Informa-
tion, Tables S1 and S2, respectively) indicated epoxidation of the
∆
20,29 exomethylene of 1. The 1H NMR spectrum of 2 showed two

Semisynthesis of Betulinic Acid Analogues
Journal of Natural Products, 2009, Vol. 72, No. 9 1645
Table 1. 1H NMR Data of Compounds 5, 9, 11, and 12a
δH
position
5
9
11
12
1
2
3
5
1.73, m
2.01, m
3.82, m
0.85, m
0.80, m, 1.54, m
1.39, m
4.29, dd (10.6, 5.5)
0.60, m
1.02, m, 1.71, m
1.61, m 1.78, m
4.50, brd (8.1)
0.83, m
0.95, m, 1.67, m
1.39, m, 1.63, m
4.33, dd (11.3, 4.0)
0.73, m
6
7
9
1.24, m, 1.57, m
1.34, m
1.24, m
1.23, m 1.42, m
1.22, m
1.04, m
1.40, m 1.53, m
1.38, m
1.25, m
1.27, m, 1.50, m
1.36, m 1.39, m
1.22, m
11
12
13
15
16
18
19
21
22
23
24
25
26
27
29
30
2′
1.18, m, 1.45, m
1.64, m
2.16, m
1.23, m 1.54, m
2.13, m, 1.57, m
1.41, m
1.11, m
1.28, m 1.43, m
1.15, m 1.70, m
2.17, ddd (12.4, 12.4, 3.3)
1.18, m 1.55, m
1.42, m 2.28, dt (12.8, 3.0)
1.61, m
3.00, ddd (10.3, 10.3, 4.4)
1.40, m, 1.98, m
1.44, m, 1.96, m
0.95, 3H, s
0.88,3H, s
0.87, 3H, s
0.95, 3H, s
0.99, 3H, s
4.63, brs, 4.76, brs
1.71, 3H, s
0.88, m, 1.25, m
0.98, m 1.64, m
2.13, ddd (12.4, 12.4, 3.3)
1.15, m 1.57, m
1.41, m 2.23, dt (12.8, 3.0)
1.55, m
2.96, ddd (11.0, 11.0, 5.1)
1.47, m, 1.96, m
1.45, m, 1.92, m
0.87, 3H, s
0.78,3H, s
0.82, 3H, s
0.91, 3H, s
0.96, 3H, s
4.59, brs, 4.72, brs
1.68, 3H, s
3.79, brs
5.79, ddd (16.8, 10.2, 5.1)
5.09, dd (10.3, 1.1) 5.15, dd (16.8, 1.5)
0.85, m 1.49, m
2.13, ddd (12.5, 12.5, 3.3)
1.01, m
1.47, m 1.92, dt (14.3, 3.0)
1.45, m
2.82, ddd (11.0, 11.0, 4.4)
1.26, m, 1.61, m
1.37, m, 1.75, m
0.70, 3H, s
0.73, 3H, s
0.72, 3H, s
0.47, 3H, s
0.82, 3H, s
2.82, m
1.86, m, 1.35, m
1.45, m, 2.02, m
0.93, s
0.75, s
0.79, s
0.91, s
0.93, s
4.58, brs, 4.70, brs
1.65, s
7.96, d (8.8)
6.87, d (8.8)
4.52, brs, 4.62, brs
1.57, 3H, s
3′
7.99, d (7.3)b
7.48, brs
7.54, d (8.8)
4′
7.51, dd (7.3, 1.5)c
7.61, dd (8.5, 1.5)d
7.51, dd (7.3, 1.5)c
7.99, d (7.3)b
5′
6.87, d (8.8)
7.99, d (8.8)
3.85, s
6′
7.54, d (8.8)
7.48, brs
7′
2′′
3′′
4′′
5′′
6′′
NH
NH
8.03, d (7.3)b
7.56, d (8.0)
7.41, dd (8.1, 7.3)
7.30, dd (7.3, 7.3)
7.41, dd (8.1, 7.3)
7.56, d (8.8)
6.81, brs
7.53, dd (7.3, 1.5)c
7.63, dd (8.4, 1.5)d
7.53, dd (7.3, 1.5)c
8.03, d (7.3)b
8.11, brs
8.87, brs
4.69, brs
a In CDCl3, at 400 MHz. b-d Interchangeable in the same column.
oxygenated protons at δ 2.55 (H2-29), which correlated in HETCOR
with the methylene carbon at δ 57.0 (C-29). Protons H2-29 showed
2J-HMBC correlations with the quaternary oxygenated carbon at
δC 60.8 (C-20) and 3J-HMBC correlations with the C-30 methyl at
δC 18.0 and the C-19 methine carbon at δC 45.5. The R configuration
at C-20 was indicated by comparing its optical rotation sign ([R]D25
-10.0) with those of the related 3ꢀ,28-acetoxy-29-lupanal epimers
(20R, [R]2D5 -103; 20S, [R]2D5 +70).22 Thus, 2 was determined to
be 20(R)-20(29)-epoxybetulinic acid.
The HREIMS and data of 3 suggested a C-29-isopropanal
functionality. The NMR data of 3 (Tables S1 and S2) indicated
the replacement of the exomethylene H2-29 protons of 1 with an
aldehyde proton doublet (δΗ 9.81 and δC 207.0). The COSY
spectrum of 3 showed coupling of H-29 with H-20 (δH 2.54), which
in turn was coupled to the methyl doublet H3-30 (δH 1.10) and
H-19 (δH 2.40), confirming a 19-isopropanal side chain. The C-20R
configuration of C-20 was indicated on the basis of the negative
sign of optical rotation.22 Comparison of the 13C NMR data of 3
with messagenic acid C (20S-3ꢀ-hydroxy-29-oxolupan-28-oic acid)
supported the fact that the compounds are C-20 epimers.23
Therefore, 3 was determined to be 20R-3ꢀ-hydroxy-29-oxolupan-
28-oic acid.
carbothioic group. This was confirmed via observing the 3J-HMBC
correlations of the proton multiplets at δ 1.59 and 2.16 (H-16a and
H-16b, respectively) and the proton signals at δ 1.43 and 1.99
(H-22a and H-22b, respectively) with the quaternary thiocarbonyl
C-28 (δC 203.2). Protons H2-16 and H2-22 also showed 2J-HMBC
correlations with the quaternary C-17 (δC 64.9). Thus, the structure
of 4 was determined to be 28-thiobetulinic acid.
1
The H and 13C NMR data of 5 (Tables 1 and 2) suggested a
coupling product with Lawesson’s reagent. The HREIMS data of
5 suggested the molecular formula C37H55O3PS3, indicating a
thiobetulinic acid-4-methoxyphenyl phosphonate ester. The 13C
NMR data (Table 2) showed a downfield shift of C-3 (δC 86.5)
versus that of 1 (+9.2 ppm), indicating esterification of this position.
3
This was further supported by the J-HMBC correlation of the
methyl singlets H3-23 and H3-24 (δH 0.93 and 0.75, respectively)
with C-3. The 1H NMR spectrum (Table 1) showed an OCH3 signal
3
at δ 3.85, which showed J-HMBC correlation with the aromatic
quaternary carbon C-4′ (δC 162.7), confirming the presence of a
4-methoxyphenyl moiety. Aromatic doublets, due to H-2′ and H-6′
(δH 7.96 and δ 7.99), showed 3J-HMBC correlations with C-4′ and
COSY couplings with the signal at δ 6.87 (H-3′/5′). H-3′/H-5′
3
showed a J-HMBC correlation with C-1′ (δ 136.5). The thiocar-
Reaction of 1 with Lawesson’s reagent afforded compounds 4
and 5, while the same reaction with betulonic acid (6) afforded 7.
The HREIMS of 4 showed a molecular ion consistent with the
molecular formula C30H48O2S, and a pattern of ions with intensity
ratios of 95:0.75:4.2, corresponding to the three abundant isotopes
of a single sulfur atom. The 1H and 13C NMR data of 4 (Tables S1
and S2) were similar to those of 1, except that downfield shifts of
C-28 and C-17 (+20.0 and +7.9 ppm, respectively) compared to
those of 1 indicated replacement of the C-28 carboxylic with a
bonyl signals appeared at δ 203.2 (C-28) and 64.9 (C-17) in a
similar pattern to 4. Thus, the structure of 5 was as indicated.
Oxidation of 1 using with CrO3 afforded the known betulonic
acid (6). Reaction of 6 with Lawesson’s reagent afforded compound
1
7, C30H46O2S, by HREIMS. The H and 13C NMR spectra of 7
(Tables S1 and S2) showed a close similarity to those of 6, except
for the downfield shifted quaternary carbons, C-28 (+20.4) and
C-17 (+8.4), which indicated the replacement of the carboxylic
group with a carbothioic-O-acid group. The chemical shift of C-3

1646 Journal of Natural Products, 2009, Vol. 72, No. 9
Bar et al.
Table 2. 13C NMR Data of Compounds 5, 9, 11, and 12a
The 1H NMR spectrum of 9 (Table 1) showed 10 aromatic
protons, indicating two phenyl substitutions. The downfield shifts
of H-3 and C-3 signals in 9 versus those of 1 suggested carbam-
oylation at C-3 (Tables 1 and 2). The 13C NMR data (Table 2)
showed an upfield shift of C-28 (-9.7 ppm) compared to that of 1,
suggesting a change at this position. The reaction of COOH groups
with sulfonyl isocyanates to produce N-acylsulfonylamides has been
documented.24 A broad singlet at δH 8.87 was assigned to the NH
proton of an N-phenylsulfonylamide moiety at C-28 (δC 174.0).
δC
position
5
9
11
12
1
2
3
4
38.6, CH2
26.0, CH2
86.5, CH
39.0, qC
38.3, CH2
23.5, CH2
85.1, CH
38.0, qC
38.5, CH2
24.2, CH2
82.5, CH
38.1, qC
38.1, CH2
24.2, CH2
81.4, CH
38.4, qC
5
6
7
8
55.4, CH
18.5, CH2
34.3, CH2
40.8, qC
55.4, CH
18.1, CH2
34.2, CH2
40.6, qC
55.6, CH
18.3, CH2
34.3, CH2
40.8, qC
55.5, CH
18.2, CH2
34.3, CH2
40.7, qC
2
This was based on its J-HMBC correlation with the C-28 amide
carbonyl. The HREIMS data (C43H58N2O7S2) were consistent with
phenylsulfonylcarbamoyl and N-phenylsulfonylamide moieties.
NMR assignments were based on HETCOR and HMBC data.
9
50.5, CH
37.1, qC
50.5, CH
37.0, qC
50.5, CH
37.2, qC
50.4, CH
37.2, qC
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
1′
20.9, CH2
25.5, CH2
37.5, CH
42.5, qC
29.5, CH2
33.5, CH2
64.9, qC
20.9, CH2
25.4, CH2
37.2, CH
42.4, qC
29.3, CH2
32.6, CH2
57.1, qC
21.0, CH2
25.5, CH2
38.5, CH
42.5, qC
29.8, CH2
32.3, CH2
56.5, qC
20.9, CH2
25.5, CH2
38.5, CH
42.5, qC
29.8, CH2
32.2, CH2
56.5, qC
49.3, CH
47.0, CH
150.5, qC
30.6, CH2
37.2, CH2
28.0, CH3
16.1, CH3
16.3, CH3
16.6, CH3
14.8, CH3
182.0, qC
109.8, CH2
19.4, CH3
156.8, qC
43.5, CH2
134.8, CH
116.0, CH2
3
Protons H-4′/6′ (δH 7.51) showed J-HMBC correlation with the
quaternary carbon at δ 138.9 (C-2′), while H-3′/7′ (δH 7.99) showed
3J-HMBC correlation with the aromatic methine carbon at δ 133.9
3
(C-5′). Protons H-3′′/5′′ (δH 7.53) showed J-HMBC correlation
with the quaternary carbon at δ 138.8 (C-1”), while H-2′′/6′′ (δH
8.03) showed a similar correlation with the methine carbon at δ
133.8 (C-4′′). Thus, the structure of 9 was determined to be as
shown.
49.7, CH
46.4, CH
150.2, qC
30.3, CH2
37.8, CH2
28.8, CH3
16.3, CH3
16.8, CH3
16.1, CH3
14.6, CH3
203.2, qC
109.9, CH2
19.4, CH3
136.5, qC
131.9, CH
113.8, CH
162.7, qC
114.0, CH
132.0, CH
49.8, CH
45.9, CH
150.1, qC
30.3, CH2
36.7, CH2
27.8, CH3
16.4, CH3
16.2, CH3
15.7, CH3
14.5, CH3
174.0, qC
109.8, CH2
19.4, CH3
150.6, qC
138.9,b qC
128.1,c CH
129.0, CH
133.9,d CH
129.0, CH
128.1,c CH
138.8,b qC
128.3,c CH
129.0, CH
133.8,d CH
129.0, CH
128.3,c CH
49.3, CH
47.1, CH
150.5, qC
30.7, CH2
37.2, CH2
28.1, CH3
16.2, CH3
16.3, CH3
16.8, CH3
14.8, CH3
182.9, qC
109.9, CH2
19.5, CH3
155.3, qC
137.5, qC
119.0, CH
127.7, CH
136.2, qC
127.7, CH
119.0, CH
140.6, qC
126.9, CH
128.9, CH
127.1, CH
128.9, CH
126.9, CH
Compound 10 had the molecular formula C41H55NO4, indicating
1
the presence of a naphthyl carbamoyl moiety. The H NMR data
of 10 showed a downfield shift of H-3 (+1.3 ppm), suggesting
carbamoylation at C-3. The 13C NMR spectrum revealed line
broadening of five carbon signals at 21 °C. In the HMQC spectrum,
C-3 showed a cross-peak with H-3 (δH 4.50) and a broad carbon
signal at δ 83.0. The methyl singlets H3-23 and H3-24 supported
3
this assignment via their J-HMBC correlations to C-3. Similarly,
the proton signals at δ 7.64 and 7.90 showed cross-peaks in HMQC
with broad carbon signals at δ 123.3 and 119.0 (C-5′ and C-6′).
H-4′ and H-5′ confirmed this assignment through their HMBC
correlations with C-5′ and C-6′, respectively. The HMBC spectrum
revealed a fourth very broad quaternary carbon signal at δ 136.0
(C-2′), which was correlated with both H-4′ and H-9′. The line
broadening of these carbons may be attributed to the steric effect
of the bulky naphthyl carbamoyl moiety.25 Thus, the structure of
10 was determined to be as shown.
2′
3′
4′
5′
6′
7′
1”
2”
3”
4”
5”
6”
Analysis of HREIMS of 11 (C43H57NO4) indicated the presence
1
of a biphenyl carbamoyl moiety. The H NMR data of 11 (Table
1) showed nine aromatic protons and a downfield shifted H-3,
confirming biphenyl carbamoylation at C-3. Due to the steric effect
of the bulky biphenyl carbamoyl group,25 the 13C NMR spectrum
of 11 revealed line broadening of carbon signals for C-3, C-1′, and
C-3′/7′. Assignments of these carbons were based on extensive
analysis of HETCOR and HMBC data. The biphenyl system was
confirmed via the 3J-HMBC correlations of the proton doublet H-3′/
7′ (δH 7.43) with the quaternary carbon signal at δ 140.6 (C-1′′)
and the proton doublet H-2′′/6′′ (δH 7.56) with the quaternary carbon
C-2′ (δC 137.5). Thus, the structure of 11 was determined to be
the 3-O-[N-(biphenyl)-p-carbamoyl] derivative of betulinic acid.
Compound 12 had the molecular formula C34H53NO4, and 13C NMR
data (Tables 1 and 2) indicated that it was a C-3-O-[(N-allyl)-
carbamoyl] derivative of 1. H-3′ (δH 5.79) showed COSY couplings
with H-4′a (δH 5.09) and H-4′b (δH 5.15) and the broad singlet
H2-2′ (δH 3.79), indicating an allyl side chain. The nitrogenated
methylene carbon C-2′ (δC 43.5) was assigned on the basis of its
3J-HMBC correlation with H2-4′. The olefinic methine carbon C-3′
a In CDCl3, at 100 MHz. Carbon multiplicities were determined by
APT experiment. qC ) quaternary, CH ) methine, CH2 ) methylene,
CH3 ) methyl carbons. b-d Interchangeable in the same column.
(δC 218.4) was similar to that of 6. The methyl singlets due to
3
H3-23 and H3-24 showed J-HMBC correlations with C-3. This
confirmed the ketone at C-3 and the structure of 7 to be
28-thiobetulonic acid.
Carbamoylation of 1 by reaction with different isocyanates
afforded compounds 8-12. The HREIMS data of 8 showed a
molecular ion peak at m/z 589.4139 (C38H55NO4 and 12 degrees of
unsaturation). Downfield shifts of the H-3 and C-3 signals in 8
(+1.22 and +4.2 ppm, respectively) compared to those of 1
1
suggested possible carbamoylation at C-3. The H and 13C NMR
data supported this and were comparable to those of 1 with
additional C-3-O-[N-(benzyl)carbamoyl] moiety signals. The car-
bonyl carbon at δC 156.9 was assigned C-1′ on the basis of its
2J-HMBC correlation with the H2-2′ multiplet (δΗ 4.32). The H2-
2′ protons also showed 2J-HMBC correlations with C-3′ (δC 138.7)
and 3J-HMBC correlations with the methine carbons C-4′ and C-8′
2
(δC 134.8) showed a J-HMBC correlation with H2-4′. Thus, the
structure of 12 was determined to be as shown.
The HREIMS and NMR data of 13 suggested benzenesulfonation
3
1
(δC 127.6). The proton doublet H2-4′/8′ (δΗ 7.27) showed a J-
of 1 at C-3. The H NMR spectrum of 13 showed five aromatic
protons. The 13C NMR data showed three methine carbons at δ
127.7 (C-2′/6′), 129.1 (C-3′/5′), and 133.4 (C-4′) in addition to a
quaternary at δ 137.9 (C-1′), confirming the benzenesulfonyl
HMBC correlation with the methine carbon C-6′ (δH 127.5), while
3
the H2-5′/7′ (δΗ 7.31) signal showed a J-HMBC correlation with
the quaternary carbon C-3′, confirming the 3-O-[N-(benzyl)car-
bamoyl] side chain. Thus, the structure of 8 was proved to be as
indicated.
3
moiety. The HMBC spectrum showed J-HMBC correlations of
the methyl singlets H3-23 and H3-24 (δΗ 0.77 and 0.76, respectively)

Semisynthesis of Betulinic Acid Analogues
Journal of Natural Products, 2009, Vol. 72, No. 9 1647
Table 3. Inhibition of Cell Proliferation, Topoisomerases I and IIR Inhibitory Activity, and Virtual Binding Affinity of Betulinic Acid
(1) and Analogues 2-14
IC50 [µM]a
topoisomerase I
inhibition
topoisomerase I
total docking
scorec
topoisomerase
IIR inhibition
[100 µM]b
topoisomerase
IIR total docking
scorec
compound
SW948
HCT-116
MDA-MB-231
[100 µM]b
1
2
3
4
5
6
7
8
10.50 ( 3.07
>25
13.46 ( 0.22
>25
15.24
4.08
0.17
>25
9.83
9.77
>25
0.45
0.92
>25
>25
0.65
0.80
1.37
ND
1+
1+
2+
1+
2+
0
1+
2+
3+
1+
1+
1+
1+
2+
2+
ND
4.03
3.90
4.27
3.30
4.40
3.23
3.30
4.46
5.09
4.05
4.65
4.50
5.18
5.07
4.89
ND
0
0
0
2+
3+
0
1+
2+
3+
2+
3+
3+
2+
0
2.66
2.91
2.99
3.05
5.31
2.94
3.03
5.63
3.80
5.65
4.90
5.28
4.75
2.78
ND
>25
>25
>25
>25
>25
>25
2.61 ( 0.41
>25
19.17 ( 0.37
>25
>25
22.24 ( 2.40
13.34 ( 3.00
>25
9
10
11
12
13
14
CPT
Etoposide
17.37 ( 0.24
>25
>25
>25
17.62 ( 1.30
6.84 ( 1.04
14.88 ( 0.84
0.16 ( 0.06
1.3 ( 0.2
17.73 ( 2.06
21.26 ( 1.65
4.38 ( 0.04
0.13 ( 0.03
1.7 ( 0.2
ND
2+
ND
4.62
a Concentration of betulinic acid analogues that inhibit 50% cell proliferation (IC50). b Topoisomerase inhibitory activity was scored as 3+ (strong),
2+ (moderate), 1+ (weak), and 0 (none), with the positive control drug CPT (topoisomerase I assay) or etoposide (topoisomerase II assay) scoring 2+.
c Virtual binding affinity of compounds 1-14 for human topoisomerase I (PDB code: 1t8i) or topoisomerase IIR (PDB code: 1bgw) using SYBYL 8.1
Surflex-Docking, the total score was expressed in -log(Kd) units to represent binding affinities.
with the downfield methine carbon signals at δ 91.5 (C-3),
confirming benzenesulfonation at C-3.
Oxidation of 1 with SeO2 afforded compound 14, which proved
to be the known messagenic acid F.15 It is worth noting that the
presence of the C-30 aldehyde group restricted rotation of the C-20
side chain, which causes severe line broadening of several nearby
carbons (C-18, C-19, C-20, C-21, and C-29), resulting in their
disappearance into noise.25 This led to the incorrect assignment of
some of these carbon signals in the published NMR data.23 The
HMBC spectrum of 14 showed 3J-HMBC correlation of the proton
signals at δ 5.88 (H-29b) and 6.26 (H-29b) with the methine carbons
C-30 (δC 195.4) and C-19 (δC 38.2), leading to the unambiguous
assignment of these carbons. The 2J-HMBC correlation of the
aldehyde proton singlet H-30 (δH 9.46) with the broad carbon signal
at δ 157.2 confirmed the assignment as C-20. H-22b (δH 1.89),
3
H-21a (δH 1.92), and H-16b (δH 2.23) showed J-HMBC correla-
Figure 2. (A) Catalytic inhibition of topoisomerase I by betulinic
acid derivatives. Lane 1, supercoiled plasmid DNA alone (250 ng);
lane 2, same as lane 1 + 0.5% DMSO; lane 3, same as lane 1 +
topoisomerase I (2U); lane 4, topoisomerase I (2U) + 100 µM
camptothecin and DNA. Lanes 5-18, topoisomerase I (2U) + DNA
and 100 µM betulinic acid analogues: lane 5 (3); lane 6 (14); lane
7 (8); lane 8 (2); lane 9 (9); lane 10 (6); lane 11 (10); lane 12 (1);
lane 13 (11); lane 14 (7); lane 15 (12); lane 16 (4); lane 17 (13);
lane 18 (5). (B) Topoisomerase II inhibitory assay by betulinic acid
derivatives. Lane 1, kinetoplast DNA (kDNA) alone; lane 2, same
as lane 1 + topoisomerase II (2U); lane 3, the same as lane 2 +
100 µM etoposide. Lanes 4-17, topoisomerase II (2U) + kDNA
and 100 µM betulinic acid analogues: lane 4 (3); lane 5 (14); lane
6 (8); lane 7 (2); lane 8 (9); lane 9 (6); lane 10 (10); lane 11 (1);
lane 12 (11); lane 13 (7); lane 14 (12); lane 15 (4); lane 16 (13);
lane 17 (5).
tions with the broad carbon signal at δ 53.1, leading to the
assignment of this carbon as C-18. Finally, the 2J-HMBC correlation
of H-22b with the broad signal at δ 25.0 suggested the assignment
of this as C-21.
Topoisomerase I-DNA Relaxation Assay.26 Inhibition of
topoisomerase function constitutes a useful strategy for the iden-
tification of potential antitumor agents. Betulinic acid derivatives
were tested using topoisomerase I-DNA relaxation assay and
compared to the known natural topoisomerase I inhibitor CPT.
Reaction mixtures were analyzed by agarose gel electrophoresis.
Figure 2A and Table 3 show the catalytic inhibition of topoi-
somerase I by 1 and derivatives 2-14 as shown in lanes 5-18 at
100 µM in comparison to 100 µM camptothecin (lane 4).
Compounds 3, 5, 8, and 14 (Figure 2aA, lanes 5, 18, 7, and 6,
respectively) showed topoisomerase I inhibitory activities compa-
rable to CPT. Compound 9 (lane 9) showed the highest inhibitory
activity toward relaxation of supercoiled DNA with activity greater
than that of CPT.
selected analogues (Table 3) and indicated the potential of the
readily available natural triterpene betulinic acid (1) as a scaffold
for the design of additional topoisomerase inhibitors.
Inhibition of Decatenation of kDNA by Topoisomerase
II Enzyme.26 Compounds 1-14 were also screened for their ability
to inhibit topoisomerase II enzymatic activity, with etoposide as a
positive control.26 Nine of the compounds exhibited inhibition of
topoisomerase II catalytic activity (Figure 2bB and Table 3). Of
these, four compounds (5, 9, 11, and 12) showed 1.5-fold better
inhibition of the catalytic activity of topoisomerase II compared to
etoposide (Figure 2B, lanes 17, 8, 12, and 14, respectively).
The above results show a rational correlation of topoisomerase
I and II inhibitory activities with the virtual binding affinity of the
Inhibition of Cell Proliferation Activity. Although colon
cancer cells were reported to be almost completely refractory to
treatment with 1 and its derivatives,27 most of new analogues
showed IC50 values < 20 µM (Table 3). Analogue 13, with a
benzenesulfonyl moiety at C-3, showed good activity against
SW948 colon cancer cells (IC50 6.84 µM). Analogues 6 and 14
showed good activity against HCT-116 colon cancer cells (IC50
2.61 and 4.38 µM, respectively).

1648 Journal of Natural Products, 2009, Vol. 72, No. 9
Scheme 1. Preparation of Compounds 2 and 3a
Bar et al.
a Reagents and conditions: (i) m-CPBA, CH2Cl2, 1 h, rt.
Scheme 2. Preparation of Compounds 4, 5, and 7a
MDA-MB-231 breast cancer cells showed less resistance to the
test compounds than the colon cancer cell lines. Compounds 3, 8,
9, 13, and 14 showed remarkable activities against MDA-MB-231
cells, with IC50’s of 0.17, 0.45, 0.92, 0.80, and 1.37 µM,
respectively, compared to 1 (IC50 15.24 µM). The most potent
compound, 3, is a dihydrobetulinic acid analogue, which has been
reported to be more cytotoxic and a better topoisomerase I inhibitor
than 1.8,9 The marked variations in cytotoxicity results compared
to topoisomerase I inhibitory activities of the derivatives may
indicate that 1 is a multitarget inducer of apoptosis.
Compound 4, with a C-28 thiocarboxylic acid group, was tested
by the Developmental Therapeutics Program (DTP) against the 60
human cell line panel at a 10 µM dose. These results (Figure S25,
NSC749378) showed better cytotoxicity than 1, indicating that a
C-28 thiocarboxy group is better than a carboxy group for cytotoxic
activity.
a Reagents and conditions: (i) Lawesson’s reagent, toluene, 110 °C, 3 h.
Plant Material. Betulinic acid (1) was isolated and purified from a
methanolic extract of leaves of Melaleuca ericifolia grown in Egypt.4
Preparation of Compounds 2 and 3.28 To a solution of 1 (90 mg,
0.2 mmol) in CH2Cl2 (2.5 mL) was added 70% m-chloroperoxybenzoic
acid (m-CPBA) (59 mg, 0.24 mmol, Scheme 1). The solution was stirred
for 1 h at room temperature, after which 10% aqueous Na2SO3 (10
mL) was added. The aqueous solution was extracted twice with CHCl3.
The combined organic layers were washed with saturated NaHCO3 and
brine and then dried under vacuum. The crude product was chromato-
graphed on silica gel 60 using n-hexane-EtOAc (8:2) to give 2 (60
mg, 66%) and 3 (10 mg, 11%).
Conclusions. Eleven new rationally designed analogues of 1
were synthesized on the basis of molecular modeling studies and
tested for their topoisomerase I and IIR inhibitory activity.
Semisynthetic reactions targeted C-3, C-28, and C-20 in 1.
Compounds 3, 5, 8, 9, and 14 showed topoisomerase I inhibitory
activities comparable to or better than CPT. Nine out of 14
compounds inhibited topoisomerase II catalytic activity, of which
5, 9, 11, and 12 showed 1.5-fold better activity than etoposide.
Aromatic or olefinic substitutions at C-3 markedly increased the
binding affinity toward both topoisomerse types, while oxidation
of the isopropenyl side chain to an aldehyde enhanced the
topoisomerase I activity without marked effects on topoisomerase
IIR binding affinity. The readily available natural product betulinic
acid is a potential scaffold for the design of potent topoisomerase
inhibitors.
Preparation of Compounds 4, 5, and 7.29 Equimolar quantities of
1 (100 mg) or 6 (50 mg) and Lawesson’s reagent (88 or 44 mg,
respectively) were dissolved in 3 or 1.5 mL of toluene (Scheme 2).
Each mixture was stirred and heated at 110 °C for 3 h. The reaction
was then quenched by addition of water (10 mL). The mixture was
extracted with ethyl acetate (3 × 10 mL). The ethyl acetate extract
was evaporated under vacuum, and the crude product was purified over
Si gel 60 using an n-hexane-EtOAc (9:1), isocratic, to give 4 (48 mg,
48%) and 7 (34 mg, 68%), or n-hexane-EtOAc (4:6) to give 5 (17
mg, 17%).
Experimental Section
Oxidation of 1 to 6. A solution of betulinic acid (100 mg) in a
mixture of toluene (1 mL) and acetonitrile (0.5 mL) was stirred with
CrO3 (100 mg) at room temperature for 1 h (Scheme 1).30 Water was
then added, and the mixture was extracted with CHCl3 (3 × 10 mL).
The chloroform extract was dried over anhydrous NaSO4 and evaporated
under vacuum. The solid product was purified over Si gel 60 using
n-hexane-EtOAc (9:1) to give betulonic acid 6 (80 mg, 80%).30
Preparation of Carbamates 8-12.31 To a solution of 1 (100 mg)
in toluene (3 mL) was added an equivalent amount of benzyl isocyanate
(29.0 mg), allyl isocyanate (18.3 mg), 4-biphenyl isocyanate (42.9 mg),
1-naphthyl isocyanate (37.2 mg), or benzene sulfonyl isocyanate (40.7
mg), and the solutions were separately mixed with 15 µL of Et3N and
refluxed for 1 h (Scheme 3). Water (10 mL) was then added, and each
General Experimental Procedures. Optical rotations were deter-
mined on a Rudolph Research Analytical Autopol III polarimeter. IR
1
spectra were recorded on a Varian 800 FT-IR spectrophotometer. H
and 13C NMR spectra were recorded in CDCl3, using TMS as an internal
standard, on a JEOL Eclipse NMR spectrometer operating at 400 MHz
for 1H and 100 MHz for 13C. HREIMS and HRESMS experiments were
conducted at the University of Michigan on a Micromass LCT
spectrometer. TLC analyses were carried out on precoated silica gel
60 F254 500 µm TLC plates, using the developing systems
n-hexane-EtOAc (9:1, 8:2, or 6:4). Silica gel 60 (EMD Chemicals
Inc.), 70-230 mesh, or silica gel (Natland International Corporation),
230-400 mesh, was used for column chromatography.

Semisynthesis of Betulinic Acid Analogues
Journal of Natural Products, 2009, Vol. 72, No. 9 1649
Scheme 3. Preparation of Compounds 8-14a
ability of each drug to inhibit topoisomerase I activity was scored as
3+ (strong), 2+ (moderate), 1+ (weak), and 0 (none), with the positive
control drug CPT scoring 2+.
Inhibition of Decatenation of kDNA by Topoisomerase II
Enzyme. Compounds 1-14 were screened for the ability to inhibit
the decatenation of kDNA using a Topoisomerase II assay kit
(TopoGen, Inc., Port Orange, FL).26 Briefly, 100 µM of each compound
or etoposide was incubated with kinetoplast DNA (kDNA) and 2U
topoisomerase II enzyme for 15 min at 37 °C.26 The reactions were
stopped and separated by 1% agarose gel electrophoresis containing
ethidium bromide (0.5 µg/mL). After destaining in water, the kDNA,
nicked circular DNA, and relaxed circular DNA were detected and
quantified using the Kodak Gel Logic 200 imaging system and
Molecular Imaging software. The data were then analyzed using Excel
(Microsoft, Inc., Redmond, WA). The ability of each tested compound
to inhibit topoisomerase II activity was scored as 3+ (strong), 2+
(moderate), 1+ (weak), and 0 (none), with the positive control drug
etoposide scoring 2+.
Cell Proliferation Assays. Antiproliferative activities of 1 and
analogues on colon cancer cells were assessed using the XTT microtiter
plate tetrazolium-based cell proliferation assay (R & D Systems Inc.,
Minneapolis, MN). For the XTT assay, 4000 cells were plated per well
into a 96-well plate. After 1-day growth at 37 °C, various concentrations
of the drug were added per well. The plates were incubated for an
additional 3 days and then analyzed using a plate reader (Molecular
Devices Corp., Sunnyvale, CA) after the addition of XTT reagent as
described by the manufacture’s protocol. Each drug concentration was
done in triplicate, and the results were analyzed using a linear-quadratic
equation to determine the concentration of drug that inhibited 50% of
the cell proliferation (IC50). The inhibition of cell proliferation assays
were repeated two or three times, and the mean ( SE of the IC50 for
each drug was determined. All compounds were compared to the IC50
determinations of camptothecin.
The cytotoxic properties of betulinic acid analogues on MDA-MB-
231 were assessed using the MTT cytotoxicity assay described before.32
Briefly, about 15 000 cells/well were plated into a 96-well plate.
Inoculates were allowed a preincubation period of 1 day at 37 °C. All
test compound doses were evaluated in triplicate. Incubation of the
cells with the test compounds lasted for 3 days at 37 °C. At the end of
the incubation period, the cells were assayed using MTT, Trevogen.
Aliquots (10 µL) of the MTT stock solution were added to each well
and incubated for 4 h. Then, formazan produced was dissolved by the
addition of 100 µL/well of detergent reagent, Trevogen, followed by a
further incubation at room temperature overnight in the dark. The optical
density (OD) was measured at 580 nm using a BioTeck microplate
reader against a blank prepared from cell-free cultures. The number of
cells/well was calculated using a calibration curve. The IC50 values
were calculated from appropriate dose-response curves.
Molecular Modeling and Docking. Three-dimensional structure
building and all modeling were performed using the SYBYL program
package,33 version 8.1, installed on Dell desktop workstations equipped
with a dual 2.0 GHz Intel Xeon processor running the Red Hat
Enterprise Linux (version 5) operating system. Conformations of each
compound were generated using Confort conformational analysis.
Energy minimizations were performed using the Tripos force field with
a distance-dependent dielectric and the Powell conjugate gradient
algorithm with a convergence criterion of 0.01 kcal/(mol A).34 Partial
atomic charges were calculated using the semiempirical program
MOPAC 6.0 and applying the AM1.35
Surflex-Dock program version 2.0 interfaced with SYBYL 8.1 was
used to dock the betulinic acid analogues to the topoisomerase I and II
active sites.36,37 Surflex-Dock employs an idealized active site ligand
(protomol) as a target to generate putative poses of molecules or
molecular fragments.38,39 These poses were scored using the Ham-
merhead scoring function.38,39 Surflex-Dock uses an empirically derived
scoring function that is based on the binding affinities of protein-ligand
complexes based on their X-ray structures. Surflex-Dock provides
accurate scoring function that is derived from the known binding affinity
data and negative training data to reduce false positive binding
scores.38,39 The scoring function includes the following terms: hydro-
phobic, polar, repulsive, entropic, and solvation energies. Surflex-Dock
scores are expressed in -log 10(Kd) units to represent the binding
affinities.38,39
a Reagents and conditions: (i) R1-NdCdO/toluene, Et3N, reflux, 1 h; (ii)
benzenesulfonyl chloride-pyridine, 48 h, rt; (iii) SeO2/dioxane, 24 h, rt.
reaction product was extracted with EtOAc (3 × 10 mL). The ethyl
acetate extract was dried over anhydrous Na2SO4 and evaporated under
vacuum. The crude products were then purified over Si gel 60 using
n-hexane-EtOAc (9:1 or 8:2) to give 8 (90 mg, 90%), 9 (98 mg, 98%),
10 (71 mg, 71%), 11 (93.5 mg, 93.5%), and 12 (53 mg, 53%).
Preparation of Compound 13. To a solution of 1 (100 mg) in
pyridine (3 mL) was added benzenesulfonyl chloride (40 mg), and the
mixture was stirred for 48 h at room temperature (Scheme 3). The
reaction was stopped by addition of water and extracted with EtOAc
(3 × 10 mL). The combined EtOAc extract was dried over anhydrous
Na2SO4 and evaporated under vacuum. The crude product was purified
over Si gel 60 using n-hexane-EtOAc (9.5:0.5) to give 13 (28 mg,
28%).
Preparation of Compound 14. SeO2 (50 mg) was added to a
solution of 1 (100 mg) in anhydrous dioxane (5 mL) (Scheme 3). The
reaction mixture was stirred for 24 h at room temperature. Water was
then added, and the mixture was extracted with CHCl3 (3 × 10 mL).
The combined CHCl3 extract was dried over anhydrous Na2SO4 and
evaporated under vacuum. The crude product was chromatographed
on Si gel 60 using n-hexane-EtOAc (8:2) to give 14 (40 mg, 40%).
Cell Lines. Cells were purchased from the American Type Culture
Collection (ATCC, Manassas, VA) and propagated as recommended
by ATCC. HCT-116 cells were maintained on modified McCoy’s 5a
medium, and SW948 cells were maintained on Leibovitz’s L-15
medium. Both culture media were supplemented with 10% heat-
inactivated FBS and 2 mM L-glutamine. The human breast cancer cell
line MDA-MB-231 was maintained in RPMI 1640 medium supple-
mented with 10% heat-inactivated fetal bovine serum and 25 mM
HEPES. Cell cultures were maintained at 37 °C in a humidified
atmosphere of 5% CO2 in an incubator.
Inhibition of DNA Relaxation by Topoisomerase I Enzyme.
Topoisomerase I drug screening kit (TopoGen, Inc., Port Orange, FL)
was used to determine the inhibitory activity of 1-14 to block or reduce
topoisomerase I DNA relaxation activity. For this assay, 250 ng of
supercoiled plasmid DNA was added to the assay buffer (10 mM Tris-
HCl, pH 7.9, 1 mM EDTA, 150 mM NaCl, 0.1% BSA, 0.1 mM
spermidine, and 5% glycerol), followed by tested compound or
camptothecin at a final concentration of 100 µM. Equal concentrations
of DMSO (0.5%) were maintained in each reaction mixture so as not
to produce solvent-mediated inhibition of topoisomerase I activity.
Topoisomerase I (2U) was added to the assay mixture, and reactions
were carried out at 37 °C for 30 min, then terminated by addition of
1% SDS (sodium dodecyl sulfate). The reaction mixtures were digested
with proteinase K (50 µg/mL) for 30 min at 37 °C followed by
chloroform-isoamyl alcohol extraction. The aqueous phase was
collected, and DNA was separated by electrophoresis in a 1% agarose
gel in TAE buffer (40 mM Tris-acetate, 1 mM EDTA) at 1 V/cm for
16 h at room temperature. The agarose gels were stained with ethidium
bromide (0.5 µg/mL) and extensively destained in water, and the DNA
bands were visualized by transillumination with UV light (320 nm)
and quantified using the Kodak Gel Logic 200 imaging system and
molecular imaging software (Eastman Kodak Co., Rochester, NY). The

1650 Journal of Natural Products, 2009, Vol. 72, No. 9
Bar et al.
The 3D structures of topoisomerases I and II were taken from
the Brookhaven Protein Databank (PDB codes: 1t8i and 1bgw,
respectively).16,17 Topoisomerase I was prepared by removal of DNA
double helix and removal of the phosphate group at Tyr 723, followed
by minimization.
in S. cereVisiae topoisomerase II DNA-binding domain. This material
is available free of charge via the Internet at http://pubs.acs.org.
References and Notes
(1) Syrovets, T.; Bc¨hele, B.; Gedig, E.; Slupsky, R. J.; Simmet, T. Mol.
Pharm 2000, 58, 71–81.
(2) Yun, Y.; Han, S.; Park, E.; Yim, D.; Lee, S.; Lee, C. K.; Cho, K.;
Kim, K. Arch. Pharm Res 2003, 26, 1087–1095.
20(R)-20(29)-Epoxybetulinic Acid (2): amorphous solid, [R]D25
-10.0 (c 0.43, CHCl3); IR νmax (CHCl3) 3511, 3188, 2946, 2869, 1710,
1461, 1377, 1136, 1107, 983, 901 cm-1; 1H and 13C NMR, see Tables
S1 and S2, Supporting Information; HREIMS m/z 472.3570 [M]+ (calcd
for C30H48O4, 472.3553).
(3) Takasaki, M.; Konoshima, T.; Tokuda, H.; Masuda, K.; Arai, Y.;
Shiojima, K.; Ageta, H. Biol. Pharm. Bull. 1999, 22, 606–610.
(4) (a) Yogeeswari, P.; Siram, D. Curr. Med. Chem. 2005, 12, 657–666.
(b) Abdel Bar, F. M.; Zaghloul, A. M.; Bachawal, S. V.; Sylvester,
P. W.; Ahmad, K. F.; El Sayed, K. A. J. Nat. Prod. 2008, 71, 1787–
1790.
(5) Takada, Y.; Agrawal, B. B. J. Immunol. 2003, 171, 3278–3286.
(6) Fluda, S.; Scaffidi, C.; Susin, S. A.; Krammer, P. H.; Kroemer, G.;
Peter, M. E.; Debatin, K. M. J. Biol. Chem. 1998, 273, 33942–33948.
(7) Tan, Y. M.; Yu, R.; Pezzuto, J. M. Clin. Cancer Res. 2003, 9, 2866–
2875.
(8) Chowdhury, A. R.; Mandal, S.; Mittra, B.; Sharma, S.; Mukhopadhyay,
S.; Majumder, H. K. Med. Sci. Monit. 2002, 8, BR254–265.
(9) Ganguly, A.; Das, B.; Roy, A.; Sen, N.; Dasgupta, S. B.; Mukho-
padhayay, S.; Majumder, H. K. Cancer Res. 2007, 67, 11848–11858.
(10) Chowdhury, A. R.; Mandal, S.; Goswami, A. Mol. Med. 2003, 9, 26–
36.
(11) Wang, J. C. Annu. ReV. Biochem. 1996, 65, 635–692.
(12) Champoux, J. J. Annu. ReV. Biochem. 2001, 70, 369–413.
(13) Deweese, J. E.; Osheroff, N. Nucleic Acids Res. 2009, 37, 738–748.
(14) Hsiang, Y. H.; Liu, L. F. Cancer Res. 1988, 48, 1722–1726.
(15) Jaxel, C.; Kohn, K. W.; Wani, M. C.; Wall, M. E.; Pommier, Y. Cancer
Res. 1989, 49, 5077–5082.
(16) Staker, B. L.; Feese, M. D.; Cushman, M.; Pommier, Y.; Zembower,
D.; Stewart, L.; Burgin, A. B. J. Med. Chem. 2005, 48, 2336–2345.
(17) Berger, J. M.; Gamblin, S. J.; Harrison, S. C.; Wang, J. C. Nature
1996, 379, 225–232.
(18) Stewart, L.; Redinbo, M. R.; Qiu, X.; Hol, W. G. J.; Champoux, J. J.
Science 1998, 279, 1534–1540.
(19) Champoux, J. J. J. Bio. Chem. 1981, 256, 4805–4809.
(20) Wei, H.; Ruthenburg, A. J.; Bechis, S. K.; Verdine, G. L. J. Biol.
Chem. 2005, 280, 37041–37047.
(21) Serge, C.-F.; Hugues-Olivier, B.; Anne, G.-L.; Arsene, D. G. P.;
Olivier, M.; Remy, H.; Serge, F. J. Med. Chem. 2004, 47, 6840–6853.
(22) Vystrcil, A.; Pouzar, V.; Kerˇcˇek, V. Collect. Czech. Chem. Commun.
1973, 38, 3902–3911.
20R-3ꢀ-Hydroxy-29-oxolupan-28-oic Acid (3): amorphous solid,
[R]2D5 -25.0 (c 0.40, CHCl3); IR νmax (CHCl3) 3510, 3155, 2947, 2869,
1716, 1694, 1456, 1364, 1134, 1029, 865 cm-1; 1H and 13C NMR, see
Tables S1 and S2, Supporting Information; HREIMS m/z 472.3541
[M]+ (calcd for C30H48O4, 472.3553).
28-Thiobetulinic Acid (4): amorphous solid, [R]2D5 +1.7 (c 0.12,
CHCl3); IR νmax 3605, 2991, 2947, 2868, 1694, 1642, 1598, 1462, 1378,
1
1113, 981, 893, 834 cm-1; H and 13C NMR, see Tables S1 and S2,
Supporting Information; HREIMS m/z 471.3298 [M - H]- (calcd for
C30H47O2S, 471.3297).
3-O-ꢀ-[Mercapto(4-methoxyphenyl)phosphorothioyloxy]betulinic Acid
(5): amorphous solid, [R]2D5 -90 (c 0.14, CHCl3); IR νmax 3405, 3020,
1
2975, 2947, 2866, 1698, 1597, 1460, 1114, 975, 892, 838 cm-1; H
and 13C NMR, see Tables 1 and 2; HREIMS m/z 673.2980 [M - H]-
(calcd for C37H54O3PS3, 673.2973).
28-Thiobetulonic Acid (7): amorphous solid, [R]2D5 +29.8 (c 0.57,
CHCl3); IR νmax 3515, 2989, 2946, 2867, 1696, 1647, 1460, 1378, 1115,
969, 949, 894 cm-1; 1H and 13C NMR, see Tables S1 and S2, Supporting
Information; HREIMS m/z 469.3162 [M - H]- (calcd for C30H45O2S,
469.3140).
3-O-[N-(Benzyl)carbamoyl]betulinic Acid (8): amorphous solid,
[R]2D5 +17.4 (c 3.2, CHCl3); IR νmax (CHCl3) 3450, 3316, 3161, 2989,
2946, 2870, 1712, 1698, 1642, 1496, 1456, 1376, 1135, 1106, 976,
1
891 cm-1; H and 13C NMR, see Table S3, Supporting Information;
HREIMS m/z 589.4139 [M]+ (calcd for C38H55NO4, 589.4131).
3-O-[N-(Phenylsulfonyl)carbamoyl-17ꢀ-N-(phenylsulfonyl)amide]-
betulinic Acid (9): amorphous solid, [R]2D5 +8.4 (c 1.76, CHCl3); IR
νmax 3390, 3264, 3076, 2949, 2872, 1744, 1723, 1451, 1402, 1351,
1
1294, 1162, 1090, 963, 887, 841, 604, 574 cm-1; H and 13C NMR,
see Tables 1 and 2; HREIMS m/z 777.3604 [M - H]- (calcd for
C43H57N2O7S2, 777.3607).
(23) Mac´ıas, F. A.; Simonet, A. M.; Galindo, J. C. G.; Pacheco, P. C.;
Sa´nchez, J. A. Phytochemistry 1998, 49, 709–717.
(24) Manabe, S.; Sugioka, T.; Ito, Y. Tetrahedron Lett. 2007, 48, 787–
789.
(25) Burns, D.; Reynolds, F. W.; Buchanan, G.; Reese, B. P.; Enriquez,
G. R. Magn. Reson. Chem. 2000, 38, 488–493.
(26) Shinkre, B. A.; Raisch, K. P.; Fan, L.; Velu, S. E. Bioorg. Med. Chem.
Lett. 2007, 17, 2890–2893.
(27) Fluda, S.; Friesen, C.; Los, M.; Scaffidi, C.; Mier, W.; Benedict, M.;
Nunez, G.; Krammer, P. H.; Peter, M. E.; Debatin, K. M. Cancer
Res. 1997, 57, 4956–4964.
(28) Ballini, R.; Marcantoni, E.; Torregiani, E. J. Nat. Prod. 1997, 60, 505–
509.
3-O-[N-(1-Naphthyl)carbamoyl]betulinic Acid (10): amorphous
solid, [R]2D5 +14.4 (c 0.24, CHCl3); IR νmax, 3432, 3178, 2989, 2949,
1
2873, 1724, 1699, 1493, 1363, 1105, 1005, 974, 891, 600 cm-1; H
and 13C NMR, see Table S3, Supporting Information; HREIMS m/z
648.4039 [M + Na]+ (calcd for C41H55NO4Na, 648.4029).
3-O-[N-(Biphenyl)-p-carbamoyl]betulinic Acid (11): amorphous
solid, [R]2D5+24.0 (c 0.30, CHCl3); IR νmax 3436, 3139, 2949, 2873,
1726, 1696, 1591, 1505, 1377, 1316, 1057, 973, 891, 839, 552 cm-1
;
1H and 13C NMR, see Tables 1 and 2; HREIMS m/z 674.4197 [M +
Na]+ (calcd for C43H57NO4Na, 674.4185).
(29) Jesberger, M.; Davis, T. P.; Barner, L. Synthesis 2003, 13, 1929–
1958.
(30) Kim, D. S. H. L.; Pezzuto, J. M.; Pisha, E. Bioorg. Med. Chem. Lett.
1998, 8, 1707–1712.
(31) El Sayed, K. A.; Laphookhieo, S.; Yousaf, M.; Prestridge, J. A.;
Shirode, A. B.; Wali, V. B.; Sylvester, P. W. J. Nat. Prod. 2008, 71,
117–122.
(32) Sarek, J.; Klinot, J.; Dzubak, P.; Klinotova, E.; Noskova, V.; Krecek,
V.; Korinkova, G.; Thomson, J. O.; Janostakova, A.; Wang, S.;
Parsons, S.; Fischer, P. M. J. Med. Chem. 2003, 46, 5402–5415.
(33) SYBYL Molecular Modeling Software, version 8.0; Tripos Associates:
St. Louis, MO, 2007; http://www.tripos.com.
3-O-[N-(Allyl)carbamoyl]betulinic Acid (12): amorphous solid,
[R]2D5 +19.4 (c 0.54, CHCl3); IR νmax 3454, 3151, 2949, 2873, 1699,
1505, 1455, 1376, 1275, 1136, 1106, 977, 891 cm-1; 1H and 13C NMR,
see Tables 1 and 2; HRESIMS m/z 562.3891 [M + Na]+ (calcd for
C34H53NO4Na, 562.3872).
3-ꢀ-O-(Benzenesulfonyloxy)betulinic Acid (13): amorphous solid,
[R]2D5 +15.4 (c 0.228, CHCl3); IR νmax 2979, 2949, 2870, 1698, 1798,
1731, 1642, 1450, 1359, 1175, 1098, 989, 939, 905, 591 cm-1; 1H and
13C NMR, see Table S3, Supporting Information; HREIMS m/z
595.3453 [M - H]- (calcd for C36H51O5S, 595.3457).
(34) Clark, M.; Cramer, R. D., III; van Opdenbosch, N. Comput. Chem.
1989, 10, 982–1012.
(35) Stewart, J. J. J. Comput. Aided Mol. Des. 1990, 4, 1–105.
(36) Kellenberger, E.; Rodrigo, J.; Muller, P.; Rognan, D. Proteins 2004,
57, 225–242.
(37) Jain, A. N. J. Med. Chem. 2003, 46, 499–511.
(38) Welch, W.; Ruppert, J.; Jain, A. N. Chem. Biol. 1996, 3, 449–462.
(39) Ruppert, J.; Welch, W.; Jain, A. N. Protein Sci. 1997, 6, 524–533.
Acknowledgment. The Egyptian Government is acknowledged for
the fellowship support to F.M.A.B.
1
Supporting Information Available: H and 13C NMR spectra of
compounds 2-5 and 7-14, full 1H and 13C NMR data assignments of
2-4, 7, 8, 10, and 13, one dose (10.0 µM) mean graphs for 1 and 4
tested by DTP (Developmental Therapeutics Program) against the 60
human cell line panel, and the docked poses of compounds 5 and 12
NP900312U