Fluorine-substituted dihydrobicyclomycins: Synthesis and biochemical and biological properties

Article abstract of DOI:10.1016/j.bmc.2005.07.075

Many studies show that selective introduction of fluorine within pharmacological agents leads to improved activities. In this study, we determine the effects of aryl fluorine substitution in 5a-(benzylsulfanyl)- dihydrobicyclomycin (3) on the in vitro inh

Full text of DOI:10.1016/j.bmc.2005.07.075

Bioorganic & Medicinal Chemistry 13 (2006) 1–21
Fluorine-substituted dihydrobicyclomycins: Synthesis
and biochemical and biological properties
Boon-Saeng Park,a William Widgerb and Harold Kohna,*
aDivision of Medicinal Chemistry and Natural Products, School of Pharmacy, University of North Carolina,
Chapel Hill, NC 27599-7360, USA
bDepartment of Biology and Biochemistry, University of Houston, Houston, TX 77204-5001, USA
Received 1 July 2005; revised 26 July 2005; accepted 27 July 2005
Available online 10 October 2005
Abstract—Many studies show that selective introduction of fluorine within pharmacological agents leads to improved activities. In
this study, we determine the effects of aryl fluorine substitution in 5a-(benzylsulfanyl)-dihydrobicyclomycin (3) on the in vitro inhi-
bition of Escherichia coli rho-dependent ATPase activity. Compound 3 is an analog of bicyclomycin (1), which is the only known
selective inhibitor of rho, and 1 and 3 have comparable in vitro inhibitory activities. We demonstrate that aryl fluorine substitution
of 3 leads to increase in inhibitory activity but that the beneficial effects due to fluorine were dependent upon the site and number of
fluorine substituents. The bioactivities are rationalized in terms of the bond moment created by the aryl fluoride bond within the
5a-aryl dihydrobicyclomycin-rho-binding pocket. Use of this hypothesis led to the design of dihydrobicyclomycin derivatives with
superior in vitro rho inhibitory activities.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
function, and the signal that triggers the hydrolysis of
ATP that powers rho along the mRNA.10–13 Rho binds
mRNA at two sites. The primary, tight RNA-binding
site is located in the N-terminus and encompasses all
six subunits. The secondary RNA site binds mRNA less
tightly and spans the N and C subdomains along the
central hole within each subunit.
Bicyclomycin (1) is a structurally unique antibiotic1,2
with a novel mode of action.3–7 The molecular target
of 1 in Escherichia coli is the rho transcription termina-
tion factor.8 In E. coli, rho has been estimated to
actively terminate transcription in approximately half
of the open reading frames.9 Thus, the rho protein is
integral for the expression of many bacterial gene prod-
ucts and without rho the cell loses viability.10 The rho
termination pathway extends to most Gram-negative
organisms.10
Bicyclomycin inhibits rho function by preventing the
secondary site RNA-induced conformational changes
necessary for ATP hydrolysis.5,14 Moreover, the recent
X-ray crystallographic images of bicyclomycins bound
to rho show that the antibiotic occludes the nucleophilic
water molecule required for the conversion of ATP to
ADP and Pi.7
O
O
HN
1
N
H
5 HO
5a
HO
HO CH3
HO
H
O
Here, we examine the effect of fluorine substitution on
bicyclomycin bioactivity. Fluorine is found in many
pharmacologically active compounds15,16 including anti-
cancer,17–19 antiviral,20–22 antibiotic,23 cardiovascular,24
and anesthetic15,16 agents. No single factor accounts
for the beneficial pharmaceutical effects observed
with fluorine substitution, in part, because fluorine af-
fects many molecular properties that influence drug func-
tion and efficacy (e.g., lipophilicity, bioavailability,
chemical stability, metabolic reactivity, and receptor
binding).25
1
Rho is a homohexameric RNA/DNA helicase/translo-
case.11 mRNA is rhoÕs substrate serving as the recogni-
tion element for its binding, the template for its
Keywords: Bicyclomycins; Fluorine substitution; Inhibitory activity;
Bond moment.
*
Corresponding author. Tel.: +1 919 843 8112; fax: +1 919 843
7835; e-mail: harold_kohn@unc.edu
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.07.075

2
Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
We report the synthesis and evaluation of 12 fluorinat-
ed derivatives of 1 and document that fluorination can
improve the bicyclomycin rho inhibitory activity com-
pared with 1 in an in vitro assay. Our structure–activity
relationship (SAR) study revealed a distinctive fluorine
substitution pattern for rho inhibitory activity and we
provide a rational for this pattern in context of the
bicyclomycin-rho inhibition pathway. Support for this
hypothesis is gained by the design of bicyclomycin
derivatives with superior in vitro rho inhibitory
activities.
NH
R212
S
ADP
O
H2N NH2
P
HO
O
O
O
HN
Mg2+
OH
2
NH3
K336
H
O
HO
N
H
OH
H3C
NH2
E211
O
O
S
H3C
OH
H2N
OH
HO
O
O
NH3
K181
T323
O
D265
HN
R269
N
P180
S266
2. Results and discussion
2.1. Selection of fluorine-substituted compounds and
method of assessment
Figure 2. Bicyclomycin 2 contacts when bound to the rho lock washer
conformation.
Structurally, 1 can be divided into the C(1) triol group,
the [4.2.2]-bicyclic ring system, and the C(5)–C(5a)
exomethylene moiety (Fig. 1). We showed that modifica-
tions of the C(1) triol group26,27 and the [4.2.2]-bicyclic
ring system28,29 led to diminished inhibitory activities
in in vitro rho functional assays while select changes
of the C(5)–C(5a) exomethylene moiety in 130,31 provid-
ed compounds with similar or improved inhibitory
activities compared with 1. These findings indicated that
the C(1) triol and [4.2.2]-bicyclic ring were essential for
bicyclomycin-rho inhibitory activity while the C(5)–
C(5a) exomethylene moiety was not. The X-ray crystal-
lographic structures of 1 and 5a-((3-formyl)phenylsulfa-
nyl)-dihydrobicyclomycin7(2) bound to the opened, lock
washer conformation of rho showed why this is the case
(see Fig. 2 for 2). We found that 1 and 2 bind to a pocket
located adjacent to the ATP and the secondary RNA-
binding sites in the C-terminal half of rho. The bicyclo-
mycin-binding pocket (12 A (deep) · 10 A (wide)) is
formed by the P and R loops, the b7/a8 connector loop
of one monomer and helix a12 of the adjacent mono-
mer. The [4.2.2] piperazinedione ring and C(1) triol
group dock snuggly into this binding pocket. The C(1)
triol group extends into the bottom of the pocket where
the hydroxyl groups create a network of hydrogen
bonds with the nearby amino acid residues (Glu211,
Asp265, and Ser266). The 5a-exomethylene-substituent
in 1 projects into the hexameric hole. When the C(5)–
C(5a) exomethylene group in 1 is modified to give
5a-substituted dihydrobicyclomycins, such as 2, the
5a-moiety is positioned along side the center hole and
directed to the C terminus. Most important, the 5a-(3-
formyl)phenylsulfanyl substituent in 2 is contained with-
in a hydrophobic pocket situated between the Pro180
side chain in one subunit and the aliphatic side chain
of Lys336 from the adjacent subunit.
O
O
HN
H
S
N
HO
HO CH3
HO
H
HO
H
5a
O
H(O)C
2
We selected 5a-benzylsulfanyldihydrobicyclomycin (3)
as our parent compound since an earlier study showed
that 3 exhibited inhibitory activity comparable with 1
in the rho poly(C)-dependent ATPase assay.30 Thus,
beginning with 3 we asked whether aromatic fluorine-
substitution influenced the dihydrobicyclomycin inhibi-
tory activity. We felt confident that the modest steric
changes incurred with fluorine aromatic substitution25
of 3 would not adversely impact drug binding to rho be-
cause we had documented that increases in the size of
the appended aromatic ring in 3 gave dihydrobicycl-
omycins with excellent inhibitory activities.14 Signifi-
cantly, 3 is structurally similar to 2 used in the rho X-
ray crystallographic study.7 Distinguishing 3 from 2 is
the inclusion of a methylene group between the sulfur
atom and the phenyl ring.
˚
˚
O
O
HN
H
S
N HO
H
H
HO
5a
O
HO CH3
HO
3
Ten fluorine-substituted derivatives of 3 were selected
for synthesis and evaluation where the fluorine substitu-
ent was directly attached to the aromatic ring. They con-
sisted of the mono-fluoro 4–6, di-fluoro 7–9, tri-fluoro
10, 11, tetra-fluoro 12, and penta-fluoro 13 benzyl deriv-
atives. We also prepared the two trifluoromethyl analogs
of 3, 14, and 15, to evaluate the effect of alkyl fluorine
substitution.
Figure 1. The three structural sectors of bicyclomycin.

Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
3
In every case but one, we generated the benzylthio-
lates in situ from the corresponding benzylthiolace-
tates (i.e., 29, 30, and 32–40) under the conditions
of the Michael reaction. The benzylthiolacetates were
prepared from commercially available benzyl halides
and potassium thioacetate in acetone (Scheme 2).14
For 6, we used commercially available 4-fluorobenze-
nethiol in place of benzylthiolacetate 31 to prepare
19. TFA removal of the 20,30-acetonide group in
17–28 in the final step (Scheme 1) provided the de-
sired 4–15, respectively, in 33–52% overall yields.
The dihydrobicyclomycins were isolated as diastereo-
meric mixtures (ꢀ3:1–10:1). We were not able to sep-
arate the isomers by PTLC. For the structurally
related 5a-anilinodihydrobicyclomycin, X-ray crystal-
lographic analysis showed that the major isomer
corresponded to the C(5)–S isomer.30
O
O
HN
H
N
H
HO
HO CH3
HO
H
HO
O
ArCH2S
F
F
F
F
2'
10
Ar =
Ar =
Ar =
Ar =
Ar =
Ar =
4
5
6
7
Ar =
Ar =
Ar =
Ar =
Ar =
Ar =
F
F
3'
11
F
F
F
F
Satisfactory spectral data (IR, 1H NMR, 13C NMR, and
MS) were obtained for all new compounds. Distinctive
fluorine-coupled 13C NMR signals were observed for
all compounds consistent with the fluorine substitution
pattern in the aryl unit.
4' F
12
13
F
F
F
F
F
F
F
2.3. Biochemical evaluation of 4–15
F
F
Table 1 lists the I50 values of the fluorinated analogs
4–15 in the rho poly(C)-dependent ATPase assay.32
Each compound was evaluated against a concurrently
run sample of 1 and then the I50 values were normal-
ized to the reported I50 for 1 (60 lM).8 Inspection of
the data revealed important structure–activity patterns.
First, introduction of a fluorine substituent in 3 at
either the 30-(5) or the 40-(6) aryl position led to im-
F
14
CF3
8
F
CF3
CF3
F
F
15
9
proved inhibitory activity compared with
3 (I50
(lM): 3, 47; 5, 23; 6, 42). Second, successive introduc-
tion of fluorine substituents in the aromatic ring led to
progressive decreases in bioactivity (I50 (lM): 5, 23; 8,
31; 11, 50; 12, 73; 13, 108). Third, placement of fluo-
rine at the 20-aryl position provided dihydrobicycl-
omycins with lower rho inhibitory activities than the
corresponding 30-aryl derivatives (I50 (lM): 4, 60 vs.
5, 23; 7, 73 vs. 8, 31; 10, 59 vs. 11, 50). The two
trifluoromethyl derivatives of 3, 14 and 15, were less
active than 3. Once again, we observed a decrease in
inhibitory activity with increasing fluorine content
(I50 (lM): 14, 52; 15, 80). Together, these findings
showed that while fluorine substitution of 3 can
enhance dihydrobicyclomycin inhibition of rho (i.e.,
5, 6) the effects of this substitution on 3 bioactivity
were finely tuned to the site and the number of
fluorine residues.
We determined the effect of site-specific inclusion of
fluorine within the appended aryl unit in 3 by measuring
the inhibitory activity of 4–15 in the rho poly(C)-depen-
dent ATPase assay.32 This test has proven to be a reli-
able assay to measure bicyclomycin inhibition of
rho.26–31 We also determined the antimicrobial activity
of select fluorine-substituted dihydrobicyclomycins
against E. coli W3350 using a filter disk assay.33 Since
we did not measure the intracellular concentrations of
4–15 under the assay conditions, we did not use this
information in our SAR. Significantly, our use of a ra-
pid, biochemical test to measure the effects of fluorine
substitution on bioactivity rather than a cellular-based
assay removes several factors (i.e., bioavailability,
chemical stability, and metabolic reactivity) previously
shown to influence the efficacy of fluorine-substituted
compounds.25
We attempted to correlate the 4–15—rho inhibitory
activities with various physicochemical parameters
associated with fluorobenzenes. Table 1 lists the I50
values of 3–15 along with the available lipophilicity
constants35 and the dipole moments for the fluoro-
substituted benzene derivatives36–38 that corresponded
to the 5a-benzyl unit. We observed that increases in
the fluorine composition of the benzene analog led
to increased lipophilicity and, in most cases, an in-
crease in the I50 value (decreased rho inhibitory activ-
2.2. Synthesis and structural characterization
A three-step synthetic sequence was used to prepare
4–15 (Scheme 1). Key to our synthesis was the
Michael addition of the incipient thiolate anion to
bicyclomycin 20,30-acetonide (16).34 The Michael
acceptor was the enone generated upon hemiaminal
ring opening at C(6) of 16 under basic conditions.

4
Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
O
O
O
O
HN
HN
ArCH2SC(O)CH3
H
R
N
H
N HO
H
H
6
HO
HO
H
CH3
HO
aq. MeOH
"pH" 10.5
O
R'O CH3
R"O
O
S
O
O
R'
R"
CH3O
OCH3
R',R" = C(CH3)2
17-28
R',R" = H
1
TFA
aq. MeOH
p-TsOH
DMF
R',R" = H
4-15
R',R" = C(CH3)2
16
F
F
F
F
2'
10, 23 Ar =
4, 17 Ar =
F
F
F
3'
11, 24 Ar =
5, 18 Ar =
6, 19 Ar =
7, 20 Ar =
8, 21 Ar =
9, 22 Ar =
F
F
F
4' F
12, 25 Ar =
13, 26 Ar =
14, 27 Ar =
15, 28 Ar =
F
F
F
F
F
F
F
F
F
F
F
CF3
CF3
F
CF3
F
Scheme 1. Preparation of 4–15.
ity). A similar finding was observed when the inhibi-
tory data were compared with the available lipophi-
licities for the corresponding fluorine-substituted
toluene derivatives (data not shown).36–38 The only
noticeable exceptions to this pattern were 5 and 6.
These findings were surprising in view of the project-
ed hydrophobic pocket7 for the 5a-benzyl substituent
in 3–15 (see Fig. 2 for 2). These trends suggested that
factors other than lipophilicity influenced the bioac-
tivity of dihydrobicyclomycins 3–15.
1.40 D. Similarly, the I50 values for 4–7, 10, and
13 ranged from 23 to 108 lM, yet the dipole mo-
ments for the corresponding benzene analogs were
comparable to one another (1.37–1.51 D). A similar
observation was found when we used the molecular
dipole moments for the available meta-substituted
toluene derivatives36,40,41 in place of the benzene ana-
logs (data not shown). The differential I50 values for
the mono-fluoro derivatives 4–6 suggested that the
bioactivities for this class of compounds were gov-
erned by site-specific biomolecular (protein, RNA)
interactions.
When the poly(C)-dependent I50 values for 3–13 were
compared with the molecular dipole moments for the
corresponding substituted benzene derivatives,36–38 we
observed no apparent correlation (Table 1). The
molecular dipole moment is a resultant of the mo-
ments associated with all the bonds in the entire
molecule.39 We found a near 3-fold range in inhibi-
tory activities for the isomeric mono-fluoro dihydro-
bicyclomycins 4–6 (I50 (lM): 4, 60; 5, 23; 6, 42)
where the dipole moment for flurobenzene is
2.4. Synthesis and biochemical evaluation of the non-
fluorinated dihydrobicyclomycins
We considered if a defined bond moment within the
aryl substituent of dihydrobicyclomycins 4–15 affected
the inhibition process. Since 5 was the most active fluo-
rine-substituted dihydrobicyclomycin, we focused on
interactions at the 30-benzyl site. Fluorine substitution

Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
5
K+-SC(O)CH3
O
O
HN
H
ArCH2-SC(O)CH3
ArCH2-Hal
N
H
HO
R'O CH3
R"O
H
HO
29-40
O
(3-X)ArCH2S
F
F
F
F
F
2'
41 R', R" = H, X = Cl
29
35
36
Ar =
Ar =
Ar =
Ar =
Ar =
Ar =
Ar =
Ar =
Ar =
Ar =
Ar =
Ar =
42 R', R" = H, X = CN
43 R', R" = H, X = OCH3
44 R', R" = H, X = NO2
F
3'
30
31
32
F
45 R', R" = H, X = OH
F
F
F
46 R', R" = C(CH3)2, X = Cl
47 R', R" = C(CH3)2, X = CN
48 R', R" = C(CH3)2, X = OCH3
49 R', R" = C(CH3)2, X = NO2
4' F
37
38
F
F
F
F
F
F
F
(3-X)ArCH2SC(O)CH3
50 X = Cl
F
F
F
33
39
F
CF3
51 X = CN
52 X = OCH3
53 X = NO2
F
F
CF3
CF3
34
40
derivatives were the most potent inhibitors (I50 ꢀ16 lM)
while the 30-Cl (41) and the 30-CN (42) analogs were the
weakest inhibitors (I50 = 61–86 lM).
Scheme 2. Preparation of benzylthiolacetates.
at this site may enhance binding of the dihydrobicyclo-
mycin to rho or perturb the rho–RNA interaction nec-
essary for protein function, or both, leading to a loss of
activity. To test the importance of this interaction, we
prepared 5a-(benzylsulfanyl)-dihydrobicyclomycins that
contained either a chloro (41), cyano (42), methoxy
(43), or nitro (44) substituent at the 30-benzyl position.
Attempts to prepare the corresponding 30-hydroxy
derivative 45 were unsuccessful (data not shown). The
rho inhibitory activities of 41–44 were compared with
those of 3 and 5. We expected that our choice of aryl
substituents would lead to significant differences in the
magnitude of the bond moment at the 30-site in 5, and
41–44.
In agreement with the I50 values for the fluoro-substi-
tuted dihydrobicyclomycins 4–13, the bioactivities for
3, 5, and 41–44 did not correlate with the lipophilicity
constants for the corresponding benzene and toluene
(data not shown) derivatives (Table 2).35 For example,
we found that the 30-chloro derivative 41 was nearly 3-
fold less active than the 30-fluoro analog 5 (I50 (lM): 5,
23; 41, 61), even though chlorobenzene is more lipo-
philic than fluorobenzene. Similarly, we were not able
to correlate the dihydrobicyclomycin inhibitory activi-
ties of 3, 5, and 41–44 with the molecular dipole
moment for the corresponding benzene36 and meta-
substituted toluene42–45 (data not shown) derivatives.
The I50 for 30-cyano derivative 42 was more than five
times higher than the 30-nitro analog 44 (I50 (lM):
42, 86; 44, 16), despite the fact that both benzonitrile
and nitrobenzene exhibited comparable molecular di-
pole moments (i.e., l (D): benzonitrile, 3.98; nitroben-
zene, 4.21). Similarly, 5 and 43 displayed enhanced rho
inhibitory activity (i.e., I50 (lM): 5, 23; 43, 17)
compared with 41 (I50 (lM): 41, 61), yet the dipole mo-
ments of the corresponding aromatic analogs, chloro-
benzene, fluorobenzene, and anisole were nearly the
same (1.32–1.57 D).
Compounds 41–44 were synthesized by the same meth-
od used for 3–15 (Schemes 1 and 2). Thus, we prepared
46–49 by treating 16 with benzylthiolacetates 50–53,
respectively, under basic conditions, and then the 20,30-
acetonide group was removed (TFA) to provide dihyd-
robicyclomycins 41–44. In Table 2, we list the inhibitory
activities for 3, 5, and 41–44 in the poly(C)-dependent
ATPase assay, the lipophilicity constants,35 and the di-
pole moments36,42–45 for the corresponding substituted
benzene analogs, the Hammett rmeta value46 for the 30-
aryl substituent, and various atomic or group electro-
negativity values47–50 reported for the selected 30-aryl
substituents. The I50 values for 41–44 ranged from 16
to 86 lM where the 30-OCH3 (43) and the 30-NO2 (44)
We attempted to correlate the inhibitory activities for
3, 5, and 41–44 with the Hammett rmeta values derived
from the linear free energy relationship that correlates

6
Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
Table 1. Biochemical and biological activities of fluorine-substituted dihydrobicyclomycins 4–15 and physicochemical values of fluorobenzenes
O
O
HN
H
N
H
HO
HO CH3
HO
H
HO
O
ArCH2S
Compound
Ar
Biochemical activity
Lipophilicity
(benzene analog)
logP
Dipole moment
(benzene analog)
(D)
I50 (lM)a
Actual (BCM)b
Normalized valuec
3
4
H
2-F
67 (85)
73 (73)
29 (77)
50 (71)
88 (72)
44 (85)
46 (85)
70 (70)
58 (70)
88 (72)
127 (70)
60 (70)
94 (71)
47
60
23
42
73
31
32
59
50
73
108
52
80
2.13d
2.27d
5
3-F
4-F
1.40d,e
6
7
2,4-Di-F
3,4-Di-F
3,5-Di-F
2,3,6-Tri-F
3,4,5-Tri-F
2,3,5,6-Tetra-F
2,3,4,5,6-Penta-F
4-CF3
—
1.51f
2.59e,f
—
1.40f
3.00f
2.47f
1.42f
2.56h
1.37h
8
2.37d
—
—
9
10
11
12
13
14
15
2.41g
—
2.53d
3.01d
—
3,5-Di-CF3
a Inhibitory activity measured using the rho poly(C)-dependent ATPase assay.32 The I50 value is the average 50% inhibition concentration determined
from duplicate tests. The corresponding value obtained from bicyclomycin in a concurrently run experiment is provided in parentheses.
b BCM = bicyclomycin.
c Values for dihydrobicyclomycin after bicyclomycin value in a concurrently tested sample was normalized to 60 lM.
d Ref. 35.
e Ref. 36.
f Ref. 37.
g Value is for the 2,3,5-trifluorobenzene found in Ref. 35.
h Ref. 38.
Table 2. Biochemical and biological properties of 30-substituted 5a-benzylsulfanyldihydrobicyclomycins and physicochemical values of substituted
benzenes
O
O
HN
H
N
H
HO
HO CH3
HO
H
HO
X
O
CH2S
d
Compound
X
Biochemical activity
I50 (lM)a
Actual (BCM)b Normalized valuec
Lipophilicity Dipole moment
(benzene analog) (benzene analog)
rm
Electronegativity scales
Marriotte Inamotof Mullayg
logP
(D)
3
5
H
67 (85)
29 (77)
90 (88)
125 (88)
25 (85)
19 (69)
6 (82)
47
23
61
86
17
16
5
2.13h
2.27h
2.89h
1.56h
2.11h
1.85h
0.24h
0
0
2.00
F
Cl
1.40i
1.57i
3.98i
1.32i
4.00i
0.56i
0.34 0.52
0.37 0.28
0.56 0.31
0.12 0.44
0.71 0.40
0.37 0.18
3.10
2.37
3.20
3.54
3.42
2.82
4.73
41
42
43
44
54
CN
OCH3
NO2
COOH
3.46
4.03
4.08
3.15
a Inhibitory activity measured using the rho poly(C)-dependent ATPase assay.32 The I50 value is the average 50% inhibition concentration determined
from duplicate tests. The corresponding value obtained from bicyclomycin in a concurrently run experiment is provided in parentheses.
b BCM = bicyclomycin.
c Value for dihydrobicyclomycin after bicyclomycin value run in a concurrently tested sample was normalized to 60 lM.
d Ref. 46.
e Ref. 47.
f Ref. 48.
g Ref. 49.
h Ref. 35.
i Ref. 36.

Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
7
the inductive effect of meta-aryl substituents on the ion-
ization of benzoic acid derivatives46 (Table 2). Using
the rmeta scale, we found that substituents (F, CN,
and NO2) that promoted ionization (rmeta = 0.56–
0.71) both led to increased (I50 (lM): 5, 23; 44, 16)
and decreased (I50 (lM): 42, 86) dihydrobicyclomycin
inhibitory activities compared with that of 3 (I50
(lM): 2, 47). Furthermore, we observed that while
the 30-fluoro (5), 30-methoxy (43), and 30-nitro (44)
aryl-substituted dihydrobicyclomycins were among the
most potent analogs evaluated (I50: ꢀ16–23 lM), the
rmeta values for the 30-fluoro and 30-methoxy substitu-
ents (i.e., 5, 0.34; 43, 0.12) were considerably lower
than the 30-nitro moiety 44 (0.71). A similar result
was observed when the dihydrobicyclomycin bioactivi-
ties were compared with the Taft r* values51 (data not
shown).
of a weak hydrogen bond interaction at the binding
site (rho-X-H—(30-F)Ar-dihydrobicyclomycin), then a
similar but stronger interaction would have occurred
with either the 30-methoxy moiety in 43 or the 30-nitro
unit in 44 leading to an increase in inhibitory activity.
When the inhibitory activities for the 30-fluoro (5), 30-
methoxy (43), and the 30-nitro (44) compounds were
compared, we found that they were all similar (I50
(lM): 4, 23; 43, 17; 44, 16). Moreover, there are few
reports documenting hydrogen bonding to aryl and
alkyl fluorides.57–63
We concluded that the 30-X bond moment in 5a-aryl
dihydrobicyclomycins was an important structural
determinant that affected rho inhibitory activities.
Inspection of the X-ray crystallographic image for 2
bound to the lock washer conformation of rho7 suggest-
ed two plausible explanations for the proposed bond
moment effect on bioactivity. First, close to the 30-for-
myl group in 2 is Lys181. We have reported that incuba-
tion of 2 with rho followed by NaBH4 reduction led to
reductive amination of Lys181.64 Thus, a favorable
bond moment interaction between the carbonyl unit in
2 and the protonated Lys181 may promote 2 binding
to rho prior to imine formation. Dihydrobicyclomycin
2 exhibited an apparent I50 of 4 lM in the rho
poly(C)-dependent ATPase assay.64 Similarly, the bond
moments created by the 30-fluorine (5), 30-methoxy (43),
and 30-nitro (44) groups in these dihydrobicyclomycins
may be stabilized by the positively charged Lys181 lead-
ing to enhanced rho binding. This hypothesis suggests
that the observed increase in bioactivity for select 5a-ar-
yl dihydrobicyclomycins is due to increased dihydrobi-
cyclomycin binding.12 The second explanation
considers the impact of dihydrobicyclomycin 5 and
41–44 binding on RNA-mediated ATP catalysis. We
have shown that the I50 value for 1 decreases in rho
functional ATPase assays with RNA substrates that
bind less tightly to the secondary site,12 suggesting that
1 is more able to disrupt RNA-mediated ATP catalysis
for weaker-binding RNAs. The aldehyde group in 2 is
wedged between Pro180 of one subunit and Lys336 of
the adjacent subunit. We have previously shown that
Lys336 is part of the extended secondary RNA tracking
site that consists of a strip of positively charged amino
acid residues that line the central hole on each subunit.12
A significant bond moment at the 30-aryl substituent in 3
may electrostatically perturb (weaken) the binding of the
RNA phosphate backbone to the secondary site and dis-
rupt ATP catalysis. Since we have not measured the dis-
sociation constants of the dihydrobicyclomycin–rho
complexes, we cannot at this time distinguish these
two inhibitory pathways.
Next, we considered whether the bond moment39,52 cre-
ated by the asymmetry of the charge distribution with-
in the 30-X benzyl bond in 5 and 41–44 accounted for
the dihydrobicyclomycinÕs bioactivities. Studies have
shown that a qualitative correlation exists between
the bond moment values and the differences in electro-
negativities of the bonded atoms.39,52 Numerous empir-
ical and theoretical electronegativity scales have been
developed to quantify the ability of an atom or func-
tional group to attract electrons.47–50,53–56 Unfortunate-
ly, there is no agreement of which scale provides the
best description of electronegativity53 and efforts have
been made to account for the atomÕs (groupÕs) molecu-
lar environment and the effect of multiple bonds on
electronegativity.49 In Table 2, we list the I50 values
for 3, 5, and 41–44 versus the atom or group electro-
negativity values in the Marriott,47 Inamoto,48 and
Mullay49,50 scales. We observed a qualitative correla-
tion where atoms or groups with increased electroneg-
ativities (higher values) exhibited enhanced (lower) I50
values. For example, the Mullay electronegativities
for the nitro-, methoxy-, and fluoro-moieties ranged
from 0.40 to 0.52 and dihydrobicyclomycins with these
30-aryl substituents exhibited inhibitory activities that
ranged from 16 to 23 lM (I50 (lM); 5, 23; 43, 17; 44,
16). Correspondingly, the chloro- and cyano-substitu-
ents have lower electronegativities (0.28–0.31) and
dihydrobicyclomycins with these 30-aryl substituents
displayed higher I50 values (I50 (lM); 41, 61; 42, 86).
Similar results were obtained using the Marriott and
Inamoto values, except that the Mullay scale
appeared to provide a better correlation for the 30-
CN benzyl analog 42. This improved fit of the data
may reflect Mullay and co-workers inclusion of reso-
nance effects in their assigned value for the cyano
group.49 Noteworthy, none of the scales provided a
linear correlation of inhibitory activity with
electronegativity.
Our identification of the importance of the 30-aryl moi-
ety in substituted 3 led us to further test the importance
of this site. We rationalized that introduction of a 30-car-
boxyl group in 3 to give 54 would lead to the corre-
sponding 30-carboxylate species under the rho poly(C)-
dependent ATPase conditions (pH 7.9).32 Generation
of a discrete negative charge should either enhance
dihydrobicyclomycin binding or perturb RNA-binding
to the secondary site, or both, leading to a low I50 value.
We also considered if dihydrobicyclomycin 3, 5, and
41–44 binding to rho was promoted by a hydrogen
bond donor interaction from rho to the 30-aryl substi-
tuent. The biochemical I50 values for 5 and 41–44 do
not support this interaction. If the enhanced inhibitory
activity of 5 compared with that of 3 is a manifestation

8
Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
Correspondingly, we predicted that the 40-carboxyl iso-
mer 55 would have diminished inhibitory activities com-
pared with 54. Synthesis of 54 and 55 followed the same
protocols (Schemes 1 and 2) previously used and re-
quired our initial preparation of 56 and 57 from 16 using
58 and 59, respectively.
why the inhibitory activities were dependent on the site
of fluorine substitution within the mono-fluorine dihyd-
robicyclomycins 4–6. We can also rationalize why de-
creased inhibitory activities were observed for the
multi-fluorine-substituted dihydrobicyclomycins 8–13
compared with 5. The introduction of additional elec-
tron-withdrawing fluorine substituents on the aromatic
ring in 5 to give 8–13 reduces the bond moment at the
30-benzyl site, and diminishes either the site-specific bond
moment stabilizing interaction of the 30-fluoro substitu-
ent with Lys181 or the adverse interaction with the
RNA phosphate backbone. Our data set led to the suc-
cessful prediction that dihydrobicyclomycin 54 would
function as a potent rho inhibitor. These findings, which
support a site-specific, bond moment interaction between
the 3-fluoro moiety in 5 with activated rho, may help ex-
plain, in part, the beneficial properties observed for other
fluorine-substituted ligands15–24 and their cognate
receptors.
O
O
HN
H
N
H
HO
R'O CH3
R"O
H
HO
O
(X)ArCH2S
54 R', R" = H, X = 3'-CO2H
55 R', R" = H, X = 4'-CO2H
56 R', R" = C(CH3)2, X = 3'-CO2H
57 R', R" = C(CH3)2, X = 4'-CO2H
(X)
ArCH2SC(O)CH3
58 X = 3'-CO2H
59 X = 4'-CO2H
4. Experimental
4.1. General methods
The I50 for 54 was 5 lM (Table 2) making this com-
pound, with 2,64 the most potent rho inhibitors discov-
ered to date. In agreement with the proposed site-
specific interaction, we found that the 40-substituted car-
boxyl derivative 55 was less effective in inhibiting rho
poly(C)-ATPase activity (I50 (lM): 54, 5; 55, 31). This
difference in inhibitory activities mirrored those ob-
served for 5 and 6 (I50 (lM): 5, 23; 6, 42). We were un-
able to correlate the bioactivity for 54 with published
electronegativity tables since most scales provide values
for the unionized carboxylic acid residue. For example,
in the Mullay electronegativity scale49 the carboxyl moi-
ety is 3.15, thus indicating that it is less electronegative
than a cyano (3.46) substituent.
1H (300 MHz) and 13C (75 MHz) NMR were recorded
on a Varian VXR300 spectrometer. Low-resolution
and high-resolution (CI) mass spectral studies were
run at the University of Texas at Austin Department
of Chemistry by Dr. M. Moini and by Dr. Dana Reed
at the University of Minnesota Department of Chemis-
try. Thin-layer chromatographies were run on precoated
silica gel slides (20 · 20 cm; Sigma Z12272-6). Bicyclo-
mycin was purified by silica gel chromatography (3·)
using 20% MeOH–CHCl3 as the eluant solvent prior
to use in biological experiments. Rho protein was isolat-
ed from E. coli AR 120 containing the overexpressing
plasmid p39-ASE, which has been corrected for the
K155E substitution seen in the original p39-AS plas-
mid.65 His-tagged wild-type rho was expressed and puri-
fied using the pET-14b vector previously described.66
Rho concentration was measured according to the
bicinchoninic acid (BCA) method. Poly(C) concentra-
tions were estimated by determining the average size
of the RNA using denaturing PAGE electrophoresis.
[c-32P]ATP was purchased from Perkin Elmer (Boston,
MA), and nucleotides were obtained from Sigma. Poly-
ethyleneimine (PEI) thin-layer chromatography (TLC)
plates used for ATPase assays were purchased from
J.T. Baker, Inc. (Phillipsburg, NJ).
The antibacterial activities of fluorine-substituted 5a-
(benzylsulfanyl)-dihydrobicyclomycins (5, 8, 11, 12,
and 14) against W3350 E. coli were determined using
the filter disk assay.33 The compounds tested showed
no detectable antibiotic activities (MIC > 32 mg/mL).
We have observed similar findings for other dihydrobi-
cyclomycins that exhibited pronounced inhibitory activ-
ities in rho functional assays.14,30
3. Conclusion
4.2. General procedure 1. Preparation of 5a-substituted
dihydrobicyclomycin 20,30-acetonides (methods A and B)
Selective fluorine substitution of 5a-benzylsulfanyldihyd-
robicyclomycin (3) has permitted us to probe the micro-
environment surrounding the 5a-(benzylsulfanyl)-
dihydrobicyclomycin-binding pocket in rho. We docu-
mented that incorporation of either an electronegative
atom or group at the 30-aryl site leads to improved rho
inhibitory activity and project that this increase in bioac-
tivity stems from either a beneficial bond moment stabi-
lizing interaction with the nearby protonated Lys181 or
the disruption of RNA binding to the secondary RNA
tracking site in rho, or both. These hypotheses explain
To a methanolic solution (3–10 mL) of 16 under Ar was
added the substituted benzylthioacetate (3–4 equiv,
method A) or the substituted benzylthiol (3–4 equiv,
method B) and then the ‘‘pH’’ was adjusted to 10.5 with
aqueous 1.0 N NaOH. The mixture was stirred at room
temperature (2–72 h) until no starting material remained
(TLC analysis). The ‘‘pH’’ of the mixture was adjusted
to 7.0 with aqueous 0.1 N HCl. The solvent was re-
moved in vacuo and the residue was redissolved in

Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
9
MeOH and purified by preparative TLC (10% MeOH–
CHCl3) to afford the desired product.
4.6. 5a-(3-Fluorobenzylsulfanyl)-dihydrobicyclomycin 20,30-
acetonide (18)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(3-fluorobenzyl) ester (30)
(62 mg, 0.44 mmol) gave 18 as a mixture of diastereo-
mers (ꢀ3:1): yield, 47 mg (70%); Rf 0.73 (10% MeOH–
CHCl3); FT-IR (KBr) 3304 (br), 2986, 1681, 1404,
1046 cmꢁ1; 1H NMR (CD3OD) for the major diastereo-
mer, d 1.35 (s, 3H, C(20)CH3), 1.41 (s, 3H, C(CH3)2),
1.44 (s, 3H, C(CH3)2), 1.83–2.14 (m, 2H, C(4)HH0,
C(4)HH0), 2.06–2.21 (m, 1H, C(5)H), 2.12 (d,
J = 11.8 Hz, 1H, C(5a)HH0), 3.14 (d, J = 11.8 Hz, 1H,
C(5a)HH0), 3.69 (s, 2H, C(5b)H2), 3.70 (d, J = 8.2 Hz,
1H, C(30)HH0), 3.76 (d, J = 8.2 Hz, 1H, C(3)HH0),
3.87 (d, J = 8.2 Hz, 1H, C(3)HH0), 4.43 (d, J = 8.2 Hz,
1H, C(30)HH0), 4.03 (s, 1H, C(10)H), 7.02–7.13 (m,
2H, ArH), 7.22–7.29 (m, 1H, ArH), 7.36–7.43 (m, 1H,
4.3. General procedure 2. Preparation of substituted
5a-benzylsulfanyldihydrobicyclomycins
To a 50% aqueous methanolic solution (2–10 mL) con-
taining the substituted 5a-benzylsulfanyldihydrobicyclo-
mycin 20,30-acetonide was added TFA (2–10 drops), and
then the solution was stirred at room temperature (2–
48 h) until no starting material remained (TLC analysis).
The solvent was removed in vacuo and the residue was
purified by preparative TLC (10–20% MeOH–CHCl3)
to afford the desired product.
4.4. General procedure 3. Preparation of benzylthio-
acetates 29–30, 32–40, 50–53, 58, and 60
1
ArH); H NMR (CD3OD) for the minor diastereomer,
d 4.08 (s, 1H, C(10)H), the remaining peaks overlapped
with nearby signals and were not detected; the structural
To an acetone solution (10–20 mL) of the desired aro-
matic benzyl halide under Ar was added potassium thi-
oacetate (1.2 equiv). The mixture was heated to reflux
(0.5–4 h) until no starting material remained (TLC anal-
ysis). The solvent was removed in vacuo and H2O
(20 mL) was added. The reaction mixture was extracted
with CH2Cl2 (2 · 10 mL), washed with brine, dried
(MgSO4), and concentrated in vacuo. The product was
purified by silica gel column chromatography using 5–
20% EtOAc–hexanes.
1
assignments were in agreement with the H–1H COSY
experiment; 13C NMR (CD3OD) for the major
diastereomer, 23.1, 26.0, 27.4, 29.6, 30.7, 35.8, 51.4,
63.0, 72.3, 72.4, 82.9, 85.6, 88.0, 110.8, 113.8
(d, J = 21.8 Hz), 115.9 (d, J = 21.7 Hz), 125.2
(d, J = 3.4 Hz), 130.3 (d, J = 8.0 Hz), 142.2 (d,
J = 6.9 Hz), 163.4 (d, J = 243.9 Hz, C(300)), 167.4,
171.0 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 27.6, 72.5, 82.8, 88.2, 111.9, 166.3, 173.0 ppm, the
remaining peaks overlapped with nearby signals and were
not detected; MS (ꢁCI) 483 [Mꢁ1]ꢁ; Mr (ꢁCI) 483.160 00
[Mꢁ1]ꢁ (calcd for C22H28FN2O7S 483.160 13 [Mꢁ1]ꢁ).
4.5. 5a-(2-Fluorobenzylsulfanyl)-dihydrobicyclomycin 20,30-
acetonide (17)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(2-fluorobenzyl) ester (29)
(62 mg, 0.44 mmol) gave 17 as a mixture of diastereo-
mers (ꢀ3:1): yield, 49 mg (72%); Rf 0.71 (10% MeOH–
CHCl3); FT-IR (KBr) 3303, 2897, 1687, 1403,
1043 cmꢁ1; 1H NMR (CD3OD) for the major diastereo-
mer, d 1.36 (s, 3H, C(20)CH3), 1.43 (s, 3H, C(CH3)2),
1.45 (s, 3H, C(CH3)2), 1.85–2.37 (m, 3H, C(4)HH0,
C(4)HH0, C(5)H), 2.17 (d, J = 7.7 Hz, 1H, C(5a)HH0),
3.16 (d, J = 7.7 Hz, 1H, C(5a)HH0), 3.71 (d,
4.7. 5a-(4-Fluorobenzylsulfanyl)-dihydrobicyclomycin
20,30-acetonide (19)
Using General procedure 1(B), 16 (50 mg, 0.15 mmol)
and 4-fluorobenzenethiol (62 mg, 0.44 mmol) gave 19
as a mixture of diastereomers (ꢀ3:1): yield, 45 mg
(75%); Rf 0.70 (10% MeOH–CHCl3); FT-IR (KBr)
3368 (br), 2953, 1691, 1406, 1042 cmꢁ1 1H NMR
;
(CD3OD) for the major diastereomer, d 1.27 (s, 3H,
C(20)CH3), 1.33 (s, 3H, C(CH3)2), 1.36 (s, 3H,
C(CH3)2), 1.75–2.20 (m, 2H, C(4)HH0, C(4)HH0),
1.97–2.12 (m, 1H, C(5)H), 2.03 (d, J = 11.7 Hz, 1H,
C(5a)HH0), 3.05 (d, J = 11.7 Hz, 1H, C(5a)HH0), 3.59
(s, 2H, C(5b)H2), 3.59–3.71 (m, 1H, C(3)HH0), 3.62 (d,
J = 8.2 Hz, 1H, C(30)HH0), 3.70–3.84 (m, 1H,
C(3)HH0), 4.00 (s, 1H, C(10)H), 4.36 (d, J = 8.2 Hz,
1H, C(30)HH0), 6.89–6.95 (m, 2H, ArH), 7.23–7.28 (m,
J = 8.2 Hz,
1
H, C(30)HH0), 3.69–3.76 (m, 1H,
C(3)HH0), 3.73 (s, 2H, C(5b)H2), 3.81–3.95 (m, 1H,
C(3)HH0), 4.44 (d, J = 8.2 Hz, 1H, C(30)HH0), 4.09 (s,
1H, C(10)H), 7.03–7.14 (m, 2H, ArH), 7.23–7.30 (m,
1H, ArH), 7.36–7.44 (m, 1H, ArH); 1H NMR (CD3OD)
for the minor diastereomer, d 4.08 (s, 1H, C(10)H), the
remaining peaks overlapped with nearby signals and
were not detected; the structural assignments were in
agreement with the 1H–1H COSY experiment; 13C
NMR (CD3OD) for the major diastereomer, 24.1,
26.0, 27.4, 29.3, 29.5, 31.1, 51.5, 62.7, 72.3, 72.4, 82.9,
85.6, 88.0, 110.8, 115.2 (d, J = 21.8 Hz), 124.4 (d,
J = 4.5 Hz), 126.4 (d, J = 14.9 Hz), 129.2 (d,
J = 8.1 Hz), 131.5 (s), 161.0 (d, J = 243.9 Hz), 167.3,
171.8 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 27.5, 62.6, 82.8, 88.2, 110.9 ppm, the remaining
peaks overlapped with nearby signals and were not
detected; MS (+ESI) 507 [M+Na]+; Mr (+ESI)
507.1600 [M+Na]+ (calcd for C22H29FN2NaO7S
507.1572 [M+Na]+).
1
2H, ArH); H NMR (CD3OD) for the minor diastereo-
mer, d 4.00 (s, 1H, C(10)H), the remaining peaks over-
lapped with nearby signals and were not detected; the
structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for the
major diastereomer, 23.8, 26.0, 27.4, 29.5, 30.6, 35.5,
51.3, 62.7, 72.2, 72.3, 82.9, 85.6, 88.0, 110.8, 115.2 (d,
J = 20.0 Hz, C(300), C(500)), 131.1 (d, J = 8.0 Hz, C(200),
C(600)), 135.2 (d, J = 3.4 Hz, C(100)), 162.4 (d,
J = 246.2 Hz, C(400)), 167.4, 170.8 ppm; 13C NMR
(CD3OD) for the minor diastereomer, 27.5, 35.9, 62.9,
72.4, 82.8, 88.2, 110.9, 166.2, 172.8 ppm, the remaining

10
Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
peaks overlapped with nearby signals and were not
detected; MS (ꢁCI) 483 [Mꢁ1]ꢁ; Mr (ꢁCI) 483.160 83
[Mꢁ1]ꢁ (calcd for C22H28FN2O7S 483.160 13 [Mꢁ1]ꢁ).
diastereomer, 27.3, 34.8, 82.5, 110.5 ppm, the remaining
peaks overlapped with nearby signals and were not
detected; MS (+CI) 503 [M+1]+; Mr (+CI) 503.167 89
[M+1]+ (calcd for C22H29F2N2O7S 503.166 36 [M+1]+).
4.8. 5a-(2,4-Difluorobenzylsulfanyl)-dihydrobicyclomycin
20,30-acetonide (20)
4.10. 5a-(3,5-Difluorobenzylsulfanyl)-dihydrobicyclomycin
20,30-acetonide (22)
Using General procedure 1(A), 16 (40 mg, 0.12 mmol)
and thioacetic acid S-(2,4-difluorobenzyl) ester (32)
(88 mg, 0.44 mmol) gave 20 as a mixture of diastereomers
(ꢀ3:1): yield, 32 mg (53%); Rf 0.69 (10% MeOH–CHCl3);
FT-IR (KBr) 3308 (br), 2950, 1689, 1403, 1041 cmꢁ1; 1H
NMR (CD3OD) for the major diastereomer, d 1.28 (s, 3
H, C(20)CH3), 1.34 (s, 3H, C(CH3)2), 1.37 (s, 3H,
C(CH3)2), 1.75–2.28 (m, 3H, C(4)HH0, C(4)HH0,
C(5)H), 2.07 (d, J = 10.2 Hz, 1H, C(5a)HH0), 3.09 (d,
J = 10.2 Hz, 1H, C(5a)HH0), 3.61 (d, J = 8.1 Hz, 1H,
C(30)HH0), 3.62 (s, 2H, C(5b)H2), 3.67–3.89 (m, 2H,
C(3)HH0, C(3)HH0), 4.01 (s, 1H, C(10)H), 4.36 (d,
J = 8.1 Hz, 1H, C(30)HH0), 6.79–6.87 (m, 1H, ArH),
7.29–7.37 (m, 2H, ArH); 1H NMR (CD3OD) for the min-
or diastereomer, d 3.95 (s, 1H, C(10)H), the remaining
peaks overlapped with nearby signals and were not
detected; the structural assignments were in agreement
with the 1H–1H COSY experiment; 13C NMR (CD3OD)
for the major diastereomer, 23.9, 25.2, 27.8, 28.5, 30.3,
30.8, 51.2, 62.4, 72.0, 72.1, 82.6, 85.4, 87.8, 103.6 (t,
J = 26.3 Hz), 110.6, 111.1 (dd, J = 21.0, 4.0 Hz), 122.4
(dd, J = 13.4, 4.3 Hz), 125.3 (dd, J = 6.3, 3.4 Hz), 161.2
(dd, J = 248.2, 12.0 Hz), 162.5 (dd, J = 248.7, 12.0 Hz),
167.1, 170.5 ppm; 13C NMR (CD3OD) for the minor dia-
stereomer peaks overlapped with nearby signals and were
not detected; MS (+CI) 503 [M+1]+; Mr (+CI) 503.165 47
[M+1]+ (calcd for C22H29F2N2O7S 503.166 36 [M+1]+).
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(3,5-difluorobenzyl) ester (34)
(88 mg, 0.44 mmol) gave 22 as a mixture of diastereo-
mers (ꢀ3:1): yield, 44 mg (60%); Rf 0.65 (10% MeOH–
CHCl3); FT-IR (KBr) 3303 (br), 2987, 1690, 1402,
1047 cmꢁ1; 1H NMR (CD3OD) for the major diastereo-
mer, d 1.29 (s, 3H, C(20)CH3), 1.36 (s, 3H, C(CH3)2),
1.38 (s, 3H, C(CH3)2), 1.78–1.90 (m, 1H, C(4)HH0),
1.92–2.33 (m, 2H, C(4)HH0, C(5)H), 2.07 (d,
J = 10.8 Hz, 1H, C(5a)HH0), 3.06 (d, J = 10.8 Hz, 1H,
C(5a)HH0), 3.63 (s, 2H, C(5b)H2), 3.64 (d, J = 8.3 Hz,
1H, C(30)HH0), 3.68–3.91 (m, 2H, C(3)HH0, C(3)HH0),
4.03 (s, 1H, C(10)H), 4.38 (d, J = 8.3 Hz, 1H, C(30)HH0),
1
6.71–6.77 (m, 2H, ArH), 6.85–6.95 (m, 1H, ArH); H
NMR (CD3OD) for the minor diastereomer, d 3.91 (s,
1H, C(10)H), the remaining peaks overlapped with near-
by signals and were not detected; the structural assign-
ments were in agreement with the 1H–1H COSY
experiment; 13C NMR (CD3OD) for the major diaste-
reomer, 23.6, 25.2, 27.1, 29.4, 30.4, 35.2, 51.1, 62.4,
72.1, 72.3, 82.6, 85.3, 88.0, 102.0 (t, J = 26.4 Hz),
110.7, 111.9 (dd, J = 17.7, 7.7 Hz, 2C), 143.8 (t,
J = 9.2 Hz), 163.3 (dd, J = 246.2, 13.4 Hz, C(300),
C(500)), 167.1, 170.6 ppm; 13C NMR (CD3OD) for the
minor diastereomer peaks overlapped with nearby
signals and were not detected; MS (+ESI) 525
[M+Na]+; Mr (+ESI) 525.1464 [M+Na]+ (calcd for
C22H28F2N2NaO7S 525.1478 [M+Na]+).
4.9. 5a-(3,4-Difluorobenzylsulfanyl)-dihydrobicyclomycin
20,30-acetonide (21)
4.11. 5a-(2,3,6-Trifluorobenzylsulfanyl)-dihydrobicyclomy-
cin 20,30-acetonide (23)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(3,4-difluorobenzyl) ester (33)
(88 mg, 0.44 mmol) gave 21 as a mixture of diastereo-
mers (ꢀ3:1): yield, 45 mg (63%); Rf 0.65 (10% MeOH–
CHCl3); FT-IR (KBr) 3300 (br), 2932, 1689, 1402,
1049 cmꢁ1; 1H NMR (CD3OD) for the major diastereo-
mer, d 1.35 (s, 3H, C(20)CH3), 1.44 (s, 3H, C(CH3)2),
1.45 (s, 3H, C(CH3)2), 1.83–2.40 (m, 3H, C(4)HH0,
C(4)HH0, C(5)H), 2.17 (d, J = 11.3 Hz, 1H, C(5a)HH0),
3.22 (d, J = 11.3 Hz, 1H, C(5a)HH0), 3.65 (d, J = 8.3 Hz,
1H, C(30)HH0), 3.70 (s, 2H, C(5b)H2), 3.76–3.98 (m, 2H,
C(3)HH0, C(3)HH0), 4.09 (s, 1H, C(10)H), 4.44 (d,
J = 8.3 Hz, 1H, C(30)HH0), 6.23–6.31 (m, 2H, ArH),
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(2,3,6–trifluorobenzyl) ester (35)
(97 mg, 0.44 mmol) gave 23 as a mixture of diastereo-
mers (ꢀ3:1): yield, 44 mg (59%); Rf 0.60 (10%
MeOH–CHCl3); FT-IR (KBr) 3306 (br), 2988, 1689,
1405, 1046 cmꢁ1 1H NMR (CD3OD) for the major
;
diastereomer, d 1.30 (s, 3H, C(20)CH3), 1.38 (s, 3H,
C(CH3)2), 1.39 (s, 3H, C(CH3)2), 1.78–1.90 (m, 1H,
C(4)HH0), 1.92–2.30 (m, 1H, C(4)HH0), 2.05–2.33 (m,
1H, C(5)H), 2.20 (d, J = 12.6 Hz, 1H, C(5a)HH0),
3.22 (d, J = 12.6 Hz, 1H, C(5a)HH0), 3.62 (d,
J = 8.4 Hz, 1H, C(30)HH0), 3.67–3.94 (m, 2H,
C(3)HH0, C(3)HH0), 3.72 (s, 2H, C(5b)H2), 4.01 (s,
1H, C(10)H), 4.38 (d, J = 8.4 Hz, 1H, C(30)HH0),
1
7.39 (s, 1H, ArH); H NMR (CD3OD) for the minor
diastereomer,
d
4.08 (s, 1H, C(10)H), 4.45 (d,
J = 7.3 Hz, 1H, C(30)HH0), the remaining peaks over-
lapped with nearby signals and were not detected; the
structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for the
major diastereomer, 23.8, 25.7, 27.2, 29.4, 30.3, 34.9,
51.1, 62.4, 72.1, 72.2, 82.6, 85.3, 87.7, 110.6, 117.0 (d,
J = 9.2 Hz), 117.8 (d, J = 9.2 Hz), 125.5 (dd, J = 9.2,
3.4 Hz), 136.7 (dd, J = 9.2, 3.4 Hz), 149.5 (dd,
J = 246.2, 13.1 Hz), 150.2 (dd, J = 246.2, 12.5 Hz),
167.1, 170.6 ppm; 13C NMR (CD3OD) for the minor
1
6.86–6.93 (m, 1H, ArH), 7.09–7.11 (m, 1H, ArH); H
NMR (CD3OD) for the minor diastereomer, d 4.02
(s, 1H, C(10)H), 4.40 (d, J = 8.4 Hz, C(30)HH0), the
remaining peaks overlapped with nearby signals and
were not detected; the structural assignments were in
agreement with the 1H–1H COSY experiment; 13C
NMR (CD3OD) for the major diastereomer, 23.7,
26.3, 27.7, 29.9, 30.7, 32.2, 51.9, 63.0, 72.6, 72.6, 83.1,
85.9, 88.3, 109.9 (ddd, J = 24.0, 6.9, 4.6 Hz), 111.1,

Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
11
114.9 (dd, J = 20.1, 10.3 Hz), 117.9 (dd, J = 21.1,
13.4 Hz), 147.9 (ddd, J = 246.1, 13.8, 3.6 Hz), 149.3
(ddd, J = 249.1, 11.8, 7.8 Hz), 157.1 (ddd, J = 246.2,
5.7, 2.3 Hz), 167.5, 171.0 ppm; 13C NMR (CD3OD)
for the minor diastereomer, 27.8, 62.9, 72.7, 82.9,
88.5, 111.2, 166.7, 173.1 ppm, the remaining peaks
overlapped with nearby signals and were not detected;
MS (ꢁCI) 519 [Mꢁ1]ꢁ; Mr (ꢁCI) 519.141 80 [Mꢁ1]ꢁ
(calcd for C22H26F3N2O7S 519.141 28 [Mꢁ1]ꢁ).
(s, 1H, C(10)H), the remaining peaks overlapped with
nearby signals and were not detected; the structural
1
assignments were in agreement with the H–1H COSY
experiment; 13C NMR (CD3OD) for the major diaste-
reomer, 24.0, 26.0, 27.4, 29.7, 28.3, 30.9, 51.3, 62.7,
72.2, 72.3, 82.8, 85.6, 88.0, 110.8, 112.6 (ddd, J = 19.5,
4.3, 3.4 Hz), 123.7–123.9 (m), 138.2–139.4 (m), 141.5–
142.7 (m), 144.5–145.5 (m), 147.8–148.7 (m), 168.0,
171.3 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 27.5, 82.7, 88.3, 110.9, 166.8, 173.4 ppm, the
remaining peaks overlapped with nearby signals and
were not detected; MS (ꢁCI) 537 [Mꢁ1]ꢁ; Mr (ꢁCI)
537.131 81 [Mꢁ1]ꢁ (calcd for C22H25F4N2O7S 537.131
86 [Mꢁ1]ꢁ).
4.12. 5a-(3,4,5-Trifluorobenzylsulfanyl)-dihydrobicyclomy-
cin 20,30-acetonide (24)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(3,4,5-trifluorobenzyl) ester (36)
(97 mg, 0.44 mmol) gave 24 as a mixture of diastereo-
mers (ꢀ3:1): yield, 46 mg (61%); Rf 0.65 (10% MeOH–
CHCl3); FT-IR (KBr) 3296 (br), 2988, 1688, 1472,
1043 cmꢁ1; 1H NMR (CD3OD) for the major diastereo-
mer, d 1.28 (s, 3H, C(20)CH3), 1.33 (s, 3H, C(CH3)2),
1.36 (s, 3H, C(CH3)2), 1.79–1.87 (m, 1H, C(4)HH0),
1.97–2.14 (m, 1H, C(4)HH0), 2.04 (d, J = 11.5 Hz, 1H,
C(5a)HH0), 2.16–2.27 (m, 1H, C(5)H), 2.99 (d,
J = 11.5 Hz, 1H, C(5a)HH0), 3.59 (s, 2H, C(5b)H2),
3.62 (d, J = 8.2 Hz, 1H, C(30)HH0), 3.67–3.74 (m, 1H,
C(3)HH0), 3.78–3.90 (m, 1H, C(3)HH0), 4.02 (s, 1H,
C(10)H), 4.35 (d, J = 8.2 Hz, 1H, C(30)HH0), 7.04 (d,
J = 8.2 Hz, 1H, ArH), 7.09 (d, J = 8.2 Hz, 1H, ArH);
1H NMR (CD3OD) for the minor diastereomer, d 4.01
(s, 1H, C(10)H),the remaining peaks overlapped with
nearby signals and were not detected; the structural
4.14. 5a-(2,3,4,5,6-Pentafluorobenzylsulfanyl)-dihydrobi-
cyclomycin 20,30-acetonide (26)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(2,3,4,5,6-pentafluorobenzyl) ester
(38) (111 mg, 0.44 mmol) gave 26 as a mixture of diaste-
reomers (ꢀ3:1): yield, 52 mg (65%); Rf 0.61 (10%
MeOH–CHCl3); FT-IR (KBr) 3303 (br), 2958, 1688,
1392, 1045 cmꢁ1; 1H NMR (CD3OD) for the major dia-
stereomer, d 1.26 (s, 3H, C(20)CH3), 1.30 (s, 3H,
C(CH3)2), 1.34 (s, 3H, C(CH3)2), 1.78–1.85 (m, 1H,
C(4)HH0), 2.03–2.11 (m, 1H, C(4)HH0), 2.07 (d,
J = 10.2 Hz, 1H, C(5a)HH0), 2.11–2.33 (m, 1H,
C(5)H), 3.15 (d, J = 10.2 Hz, 1H, C(5a)HH0), 3.56 (d,
J = 8.1 Hz, 1H, C(30)HH0), 3.66 (s, 2H, C(5b)H2),
3.78–3.86 (m, 2H, C(3)HH0, C(3)HH0), 4.03 (s, 1H,
1
1
assignments were in agreement with the H–1H COSY
C(10)H), 4.34 (d, J = 8.1 Hz, 1H, C(30)HH0); H NMR
experiment; 13C NMR (CD3OD) for the major diaste-
reomer, 23.9, 25.9, 27.4, 29.7, 30.5, 35.0, 51.2, 62.7,
72.3, 72.4, 82.8, 85.5, 88.0, 110.9, 113.5 (dd, J = 16.0,
6.8 Hz, C(200), C(600)), 135.7–136.7 (m, C(100)), 138.9
(ddd, J = 248.5, 10.3, 3.9 Hz, C(400)), 150.5 (ddd,
J = 249.6, 10.3, 3.9 Hz, C(300), C(500)), 167.3,
170.8 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 26.0, 27.6, 34.8, 82.7, 85.6, 88.3, 110.9, 166.3,
171.8 ppm, the remaining peaks overlapped with nearby
signals and were not detected; MS (+ESI) 543 [M+Na]+;
(CD3OD) for the minor diastereomer, d 4.02 (s, 1H,
C(10)H), the remaining peaks overlapped with nearby
signals and were not detected; the structural assignments
1
were in agreement with the H–1H COSY experiment;
13C NMR (CD3OD) for the major diastereomer, 23.0,
24.0, 25.8, 27.2, 29.7, 32.0, 51.6, 62.8, 72.1, 72.3, 82.8,
85.5, 88.0, 110.8, 113.9–114.1 (m), 135.9–136.8 (m),
138.9–140.0 (m, 2C), 143.5–147.3 (m, 2C), 167.1,
170.7 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 26.1, 27.8, 29.3, 30.8, 62.6, 82.6, 85.6, 88.1, 110.9,
166.5, 172.8 ppm, the remaining peaks overlapped with
nearby signals and were not detected; MS (ꢁCI) 555
[Mꢁ1]ꢁ; Mr (ꢁCI) 555.121 86 [Mꢁ1]ꢁ (calcd for
C22H24F5N2O7S 555.122 44 [Mꢁ1]ꢁ).
Mr
C22H27F3N2NaO7S 543.1384 [M+Na]+).
(+ESI)
543.1396
[M+Na]+
(calcd
for
4.13. 5a-(2,3,5,6-Tetrafluorobenzylsulfanyl)-dihydrobicy-
clomycin 20,30-acetonide (25)
4.15. 5a-(4-Trifluoromethylbenzylsulfanyl)-dihydrobicyclo-
mycin 20,30-acetonide (27)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(2,3,5,6–tetrafluorobenzyl) ester
(37) (104 mg, 0.44 mmol) gave 25 as a mixture of diaste-
reomers (ꢀ3:1): yield, 46 mg (59%); Rf 0.63 (10%
MeOH–CHCl3); FT-IR (KBr) 3301 (br), 2988, 1691,
1380, 1046 cmꢁ1; 1H NMR (CD3OD) for the major dia-
stereomer, d 1.36 (s, 3H, C(20)CH3), 1.43 (s, 3H,
C(CH3)2), 1.45 (s, 3H, C(CH3)2), 1.81–2.00 (m, 1H,
C(4)HH0), 2.02–2.28 (m, 2H, C(4)HH0, C(5)H), 2.16
(d, J = 8.2 Hz, 1H, C(5a)HH0), 3.08 (d, J = 8.2 Hz,
1H, C(5a)HH0), 3.67–4.01 (m, 2H, C(3)HH0,
C(3)HH0), 3.71 (d, J = 8.2 Hz, 1H, C(30)HH0), 3.73 (s,
2H, C(5b)H2), 4.10 (s, 1H, C(10)H), 4.45 (d,
J = 8.2 Hz, 1H, C(30)HH0), 7.22–7.32 (m, 1H, ArH);
1H NMR (CD3OD) for the minor diastereomer, d 3.99
Using General procedure 1(A), 16 (40 mg, 0.12 mmol)
and thioacetic acid S-(4-trifluoromethyl) ester (39)
(84 mg, 0.36 mmol) gave 27 as a mixture of diastereo-
mers (ꢀ3:1): yield, 44 mg (68%); Rf 0.70 (10% MeOH–
CHCl3); FT-IR (KBr) 3312 (br), 2948, 1667, 1401,
1043 cmꢁ1; 1H NMR (CD3OD) for the major diastereo-
mer, d 1.35 (s, 3H, C(20)CH3), 1.42 (s, 3H, C(CH3)2),
1.44 (s, 3H, C(CH3)2), 1.82–1.99 (m, 1H, C(4)HH0),
2.13–2.17 (m, 1H, C(4)HH0), 2.18 (d, J = 8.3 Hz, 1H,
C(5a)HH0), 2.17–2.27 (m, 1H, C(5)H), 3.15 (d,
J = 8.3 Hz, 1H, C(5a)HH0), 3.72 (d, J = 8.2 Hz, 1H,
C(30)HH0), 3.74–3.82 (m, 2H, C(3)HH0, C(3)HH0),
3.88 (s, 2H, C(5b)H2), 4.08 (s, 1H, C(10)H), 4.26 (d,

12
Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
J = 8.2 Hz, 1H, C(30)HH0), 7.45–7.53 (m, 2H, ArH),
7.58–7.68 (m, 2H, ArH); H NMR (CD3OD) for the
3.94 (s, 1H, C(10)H), 4.44 (d, J = 8.3 Hz, 1H, C(30)HH0),
7.21–7.32 (m, 3H, ArH), 7.37 (s, 1H, ArH); H NMR
1
1
minor diastereomer, d 4.07 (s, 1H, C(10)H), the remain-
ing peaks overlapped with nearby signals and were not
detected; the structural assignments were in agreement
with the 1H–1H COSY experiment; 13C NMR (CD3OD)
for the major diastereomer, 23.7, 25.7, 27.2, 29.4, 30.5,
35.4, 51.1, 62.4, 72.1, 72.2, 82.6, 85.3, 87.7, 110.5,
123.6-123.7 (m, CF3), 125.6–125.7 (m), 129.2 (2C),
132.8 (2C), 140.6, 167.0, 170.6 ppm; 13C NMR
(CD3OD) for the minor diastereomer 27.3, 28.6, 28.8,
35.2, 110.7, 166.0, 172.7 ppm, the remaining peaks over-
lapped with nearby signals and were not detected; MS
(ꢁCI) 533 [Mꢁ1]ꢁ; Mr (ꢁCI) 533.157 67 [Mꢁ1]ꢁ (calcd
for C23H28F3N2O7S 533.157 93 [Mꢁ1]ꢁ).
(CD3OD) for the minor diastereomer, d 3.10 (d,
J = 9.8 Hz, 1H, C(5a)HH0), 4.09 (s, 1H, C(10)H), 4.47
(d, J = 8.3 Hz, 1H, C(30)HH0), the remaining peaks
overlapped with nearby signals and were not detected;
the structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for the
major diastereomer, 24.1, 26.0, 27.4, 29.5, 30.6, 35.6,
51.4, 62.6, 72.2, 72.3, 82.8, 85.5, 88.0, 110.8, 127.2,
127.7, 129.2, 130.6, 134.3, 141.7, 167.3, 170.7 ppm; 13C
NMR (CD3OD) for the minor diastereomer, 25.9,
27.4, 72.2, 82.7, 85.6, 88.2, 110.9, 127.3, 127.8, 129.5,
130.3, 134.4, 166.2, 172.8 ppm, the remaining peaks
overlapped with nearby signals and were not detected;
MS (+ESI) 523 [M+Na]+; Mr (+ESI) 523.1305
[M+Na]+ (calcd for C22H29ClN2NaO7S 523.1277
[M+Na]+).
4.16. 5a-(2,5-Bis-trifluoromethylbenzylsulfanyl)-dihydrobi-
cyclomycin 20,30-acetonide (28)
Using General procedure 1 (A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(2,5-bis-trifluoromethylbenzyl) es-
ter (40) (62 mg, 0.44 mmol) gave 28 as a mixture of dia-
stereomers (ꢀ3:1): yield, 61 mg (70%); Rf 0.50 (10%
MeOH–CHCl3); FT-IR (KBr) 3296 (br), 2990, 1697,
1380, 1047 cmꢁ1; 1H NMR (CD3OD) for the major dia-
stereomer, d 1.28 (s, 3H, C(20)CH3), 1.36 (s, 3H,
C(CH3)2), 1.37 (s, 3H, C(CH3)2), 1.75–1.96 (m, 1H,
C(4)HH0), 2.02–2.24 (m, 2H, C(4)HH0, C(5)H), 2.05
(d, J = 13.2 Hz, 1H, C(5a)HH0), 3.02 (d, J = 13.2 Hz,
1H, C(5a)HH0), 3.63 (d, J = 8.2 Hz, 1H, C(30)HH0),
3.64 (s, 2H, C(5b)H2), 3.67–3.92 (m, 2H, C(3)HH0,
C(3)HH0), 4.02 (s, 1H, C(10)H), 4.36 (d, J = 8.2 Hz,
1H, C(30)HH0), 7.75 (s, 1H, ArH), 7.89 (s, 2H, ArH);
1H NMR (CD3OD) for the minor diastereomer, d 4.01
(s, 1H, C(10)H), the remaining peaks overlapped with
nearby signals and were not detected; the structural
4.18. 5a-(3-Cyanobenzylsulfanyl)-dihydrobicyclomycin
20,30-acetonide (47)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(3-cyanobenzyl) ester (51)
(84 mg, 0.44 mmol) gave 47 as a mixture of diastereo-
mers (ꢀ3:1): yield, 42 mg (56%); Rf 0.63 (10% MeOH–
CHCl3); FT-IR (KBr) 3302 (br), 2986, 1689, 1402,
1047 cmꢁ1; 1H NMR (CD3OD) for the major diastereo-
mer, d 1.36 (s, 3H, C(20)CH3), 1.42 (s, 3H, C(CH3)2),
1.44 (s, 3H, C(CH3)2), 1.83–2.31 (m, 2H, C(4)HH0,
C(4)HH0), 2.05–2.31 (m, 1H, C(5)H), 2.12 (d,
J = 12.6 Hz, 1H, C(5a)HH0), 3.14 (d, J = 12.6 Hz, 1H,
C(5a)HH0), 3.74 (s, 2H, C(5b)H2), 3.75 (d, J = 8.2 Hz,
1H, C(30)HH0), 3.78–3.97 (m, 2H, C(3)HH0, C(3)HH0),
4.10 (s, 1H, C(10)H), 4.44 (d, J = 8.2 Hz, 1H, C(30)HH0),
1
7.46–7.74 (m, 3H, ArH), 7.76 (s, 1H, ArH); H NMR
1
assignments were in agreement with the H–1H COSY
(CD3OD) for the minor diastereomer, d 3.13 (d,
J = 9.4 Hz, 1H, C(5a)HH0), 4.06 (s, 1H, C(10)H), 4.44
(d, J = 7.7 Hz, 1H, C(30)HH0), the remaining peaks
overlapped with nearby signals and were not detected;
the structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for the
major diastereomer, 24.5, 26.0, 27.4, 29.6, 30.5, 35.1,
51.2, 62.7, 72.2, 72.3, 82.9, 85.5, 88.0, 110.8, 112.5,
118.8, 129.8, 130.9, 132.8, 134.1, 141.3, 167.4,
171.0 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 27.5, 35.0, 72.2, 82.8, 85.6, 88.2, 111.9, 129.6, 131.0,
132.9, 134.2, 166.2, 172.7 ppm, the remaining peaks
overlapped with nearby signals and were not detected;
no signals were observed in the ESI and CI MS.
experiment; 13C NMR (CD3OD) for the major diaste-
reomer, 24.0, 26.0, 27.4, 29.8, 30.5, 34.8, 51.2, 62.7,
72.3, 72.4, 82.7, 85.6, 88.0, 110.8, 120.8–120.9 (m, 2C,
CF3), 125.6, 129.8–129.9 (m), 131.2–132.6 (m, 2C),
143.0, 167.3, 170.7 ppm; 13C NMR (CD3OD) for the
minor diastereomer, 27.5, 28.5, 29.1, 34.6, 82.7, 88.2,
166.2, 172.8 ppm, the remaining peaks overlapped with
nearby signals and were not detected; MS (ꢁCI) 601
[Mꢁ1]ꢁ; Mr (ꢁCI) 601.145 06 [Mꢁ1]ꢁ (calcd for
C24H27F6N2O7S 601.144 32 [Mꢁ1]ꢁ).
4.17. 5a-(3-Chlorobenzylsulfanyl)-dihydrobicyclomycin
20,30-acetonide (46)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(3-chlorobenzyl) ester (50)
(88 mg, 0.44 mmol) gave 46 as a mixture of diastereo-
mers (ꢀ3:1): yield, 42 mg (56%); Rf 0.65 (10% MeOH–
CHCl3); FT-IR (KBr) 3297 (br), 2985, 1683, 1404,
1048 cmꢁ1; 1H NMR (CD3OD) for the major diastereo-
mer, d 1.36 (s, 3H, C(20)CH3), 1.42 (s, 3H, C(CH3)2),
1.46 (s, 3H, C(CH3)2), 1.84–2.28 (m, 3H, C(4)HH0,
C(4)HH0, C(5)H), 2.13 (d, J = 11.1 Hz, 1H, C(5a)HH0),
3.15 (d, J = 11.1 Hz, 1H, C(5a)HH0), 3.69 (s, 2H,
C(5b)H2), 3.70 (d, J = 8.3 Hz, 1H, C(30)HH0), 3.69–
3.78 (m, 1H, C(3)HH0), 3.84–3.93 (m, 1H, C(3)HH0),
4.19. 5a-(3-Methoxybenzylsulfanyl)-dihydrobicyclomycin
20,30-acetonide (48)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(3-methoxybenzyl) ester (52)
(62 mg, 0.44 mmol) gave 48 as a mixture of diastereo-
mers (ꢀ3:1) along with small amount of unreacted 16.
PTLC purification of the product mixture gave 48: yield,
36 mg (48%); Rf 0.50 (10% MeOH–CHCl3); FT-IR
(KBr) 3303 (br), 2986, 1689, 1405, 1047 cmꢁ1 1H
;
NMR (CD3OD) for the major diastereomer, d 1.35 (s,
3H, C(20)CH3), 1.42 (s, 3H, C(CH3)2), 1.44 (s, 3H,

Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
13
C(CH3)2), 1.81–2.14 (m, 3H, C(4)HH0, C(4)HH0,
C(5)H), 2.12 (d, J = 10.9 Hz, 1H, C(5a)HH0), 3.16 (d,
J = 10.9 Hz, 1H, C(5a)HH0), 3.63–4.02 (m, 2H,
C(3)HH0, C(3)HH0), 3.69 (s, 2H, C(5b)H2), 3.70 (d,
J = 8.2 Hz, 1H, C(30)HH0), 3.78 (s, 3H, ArOCH3), 4.08
(s, 1H, C(10)H), 4.42 (d, J = 8.2 Hz, 1H, C(30)HH0),
6.76 (d, J = 8.2 Hz, 1H, ArH), 6.90 (s, 1H, ArH), 6.94
(d, J = 8.2 Hz, 1H, ArH), 7.18 (t, J = 8.2 Hz, 1H,
1041 cmꢁ1; 1H NMR (CD3OD) for the major diastereo-
mer, d 1.37 (s, 3H, C(20)CH3), 1.44 (s, 3H, C(CH3)2),
1.47 (s, 3H, C(CH3)2), 1.85-2.18 (m, 4H, C(4)HH0,
C(4)HH0, C(5)H, C(5a)HH0), 3.16 (d, J = 10.4 Hz, 1H,
C(5a)HH0), 3.63 (d, J = 8.2 Hz, 1H, C(30)HH0), 3.70–
3.77 (m, 1H, C(3)HH0), 3.75 (s, 2H, C(5b)H2), 3.89–
3.96 (m, 1H, C(3)HH0), 4.10 (s, 1H, C(10)H), 4.45 (d,
J = 8.2 Hz, 1H, C(30)HH0), 7.30–7.35 (m, 1H, ArH),
7.41 (d, J = 7.7 Hz, 1H, ArH), 7.85 (d, J = 7.7 Hz, 1H,
1
ArH); H NMR (CD3OD) for the minor diastereomer,
1
d 3.13 (d, J = 9.4 Hz, 1H, C(5a)HH0), 4.06 (s, 1H,
C(10)H), 4.44 (d, J = 7.7 Hz, 1H, C(30)HH0), the remain-
ing peaks overlapped with nearby signals and were not
detected; the structural assignments were in agreement
with the 1H–1H COSY experiment; 13C NMR (CD3OD)
for the major diastereomer, 23.8, 25.8, 27.2, 29.2, 30.4,
36.2, 51.3, 54.6, 62.3, 72.1, 72.2, 82.7, 85.3, 87.7, 110.6,
112.7, 114.4, 121.3, 129.4, 140.5, 160.2, 167.3,
170.5 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 23.7, 25.7, 27.3, 36.1, 62.5, 72.3, 111.7, 114.5, 121.4,
129.5, 160.3, 166.0, 172.6 ppm, the remaining peaks
overlapped with nearby signals and were not detected
ArH), 8.00 (s, 1H, ArH); H NMR (CD3OD) for the
minor diastereomer,
d 3.12 (d, J = 9.4 Hz, 1H,
C(5a)HH0), 4.09 (s, 1H, C(10)H), 4.47 (d, J = 6.6 Hz,
1H, C(30)HH0), the remaining peaks overlapped with
nearby signals and were not detected; the structural
assignments were in agreement with the H–1H COSY
1
experiment; 13C NMR (CD3OD) for the major diaste-
reomer, 23.7, 25.8, 27.2, 29.4, 30.8, 36.2, 51.3, 62.7,
72.2, 72.3, 82.7, 85.3, 87.7, 110.6, 127.7, 127.9, 130.0,
130.9, 138.1, 138.5, 166.7, 170.5, 207.0 ppm; 13C NMR
(CD3OD) for the minor diastereomer, 62.5 ppm, the
remaining peaks overlapped with nearby signals and
were not detected; MS (ESI) 533 [M+Na]+; Mr (+ESI)
533.1546 [M+Na]+ (calcd for C23H30N2NaO9S
533.1564).
4.20. 5a-(3-Nitrobenzylsulfanyl)-dihydrobicyclomycin 20,30-
acetonide (49)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(3-nitrobenzyl) ester (53) (74 mg,
0.44 mmol) gave 49 as a mixture of diastereomers
(ꢀ3:1): yield, 31 mg (40%); Rf 0.49 (10% MeOH–
CHCl3); FT-IR (KBr) 3302 (br), 2996, 1687, 1402,
1045 cmꢁ1; 1H NMR (CD3OD) for the major diastereo-
mer, d 1.32 (s, 3H, C(20)CH3), 1.37 (s, 3H, C(CH3)2),
1.45 (s, 3H, C(CH3)2), 1.86–2.43 (m, 4H, C(4)HH0,
C(4)HH0, C(5)H, C(5a)HH0), 3.08 (d, J = 13.2 Hz, 1H,
C(5a)HH0), 3.66 (d, J = 9.3 Hz, 1H, C(30)HH0), 3.67–
4.02 (m, 2H, C(3)HH0, C(3)HH0), 3.89 (s, 2H,
C(5b)H2), 4.07 (s, 1H, C(10)H), 4.12 (d, J = 9.3 Hz,
1H, C(30)HH0), 7.57 (t, J = 8.2 Hz, 1H, ArH), 7.78 (d,
J = 8.2 Hz, 1H, ArH), 8.12 (d, J = 8.2 Hz, 1H, ArH),
4.22. 5a-(2-Fluorobenzylsulfanyl)-dihydrobicyclomycin (4)
Using General procedure 2, 17 (55 mg, 0.07 mmol) gave
4 as a mixture of diastereomers (ꢀ3:1): yield, 49 mg
(72%); Rf 0.45 (20% MeOH–CHCl3); FT-IR (KBr)
3302 (br), 2951, 1679, 1401, 1044 cmꢁ1 1H NMR
;
(CD3OD) for the major diastereomer, d 1.32 (s, 3H,
C(20)CH3), 2.02–2.26 (m, 3H, C(4)HH0, C(4)HH0,
C(5)H), 2.21 (d, J = 9.6 Hz, 1H, C(5a)HH0), 3.17 (d,
J = 9.6 Hz, 1H, C(5a)HH0), 3.45 (d, J = 11.5 Hz, 1H,
C(30)HH0), 3.65 (d, J = 11.5 Hz, 1H, C(30)HH0), 3.68–
3.78 (m, 1H, C(3)HH0), 3.72 (s, 2H, C(5b)H2), 3.87–
3.94 (m, 1H, C(3)HH0), 4.01 (s, 1H, C(10)H), 7.02–7.13
(m, 2H, ArH), 7.22–7.29 (m, 1H, ArH), 7.36–7.44 (m,
1
1
8.26 (s, 1H, ArH); H NMR (CD3OD) for the minor
1H, ArH); H NMR (CD3OD) for the minor diastereo-
diastereomer, d 3.12 (d, J = 9.4 Hz, 1H, C(5a)HH0),
4.17 (d, J = 8.8 Hz, 1H, C(30)HH0), the remaining peaks
overlapped with nearby signals and were not detected;
the structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for the
major diastereomer, 24.9, 26.8, 28.2, 30.5, 31.9, 36.3,
52.3, 63.5, 73.2, 73.3, 83.7, 86.4, 88.8, 111.7, 122.9,
124.8, 130.6, 136.1, 142.7, 149.8, 168.1, 171.5 ppm; 13C
NMR (CD3OD) for the minor diastereomer, 26.5,
27.1, 27.3, 31.9, 62.5, 73.4, 86.4, 89.9, 111.8, 122.8,
142.8, 149.7, 173.6 ppm, the remaining peaks over-
lapped with nearby signals and were not detected; MS
(ESI) 534 [M+Na]+; Mr (+ESI) 534.1538 [M+Na]+
(calcd for C22H29N3NaO9S 534.1517 [M+Na]+).
mer, d 3.96 (s, 1 H, C(10)H), the remaining peaks over-
lapped with nearby signals and were not detected; the
structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for the
major diastereomer, 23.4, 29.2, 29.4, 31.0, 51.4, 62.4,
67.7, 71.3, 77.3, 82.9, 88.5, 115.5 (d, J = 21.8 Hz),
124.4 (d, J = 3.4 Hz), 126.4 (d, J = 14.9 Hz), 129.2 (d,
J = 8.0 Hz), 131.5 (d, J = 4.5 Hz), 161.4 (d,
J = 246.2 Hz), 168.0, 171.3 ppm; 13C NMR (CD3OD)
for the minor diastereomer, 61.1, 82.7, 88.8 ppm, the
remaining peaks overlapped with nearby signals and
were not detected; MS (+ESI) 467 [M+Na]+; Mr
(+ESI) 467.1281 [M+Na]+ (calcd for C19H25FN2NaO7S
467.1259 [M+Na]+).
4.21. 5a-(3-Carboxybenzylsulfanyl)-dihydrobicyclomycin
20,30-acetonide (56)
4.23. 5a-(3-Fluorobenzylsulfanyl)-dihydrobicyclomycin
(5)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(3-carboxybenzyl) ester (58)
(62 mg, 0.44 mmol) gave 56 as a mixture of diastereo-
mers (3:1): yield, 23 mg (35%); Rf 0.33 (30% MeOH–
CHCl3); FT-IR (KBr) 3259 (br), 2986, 1697, 1404,
Using General procedure 2, 18 (30 mg, 0.06 mmol) gave
5 as a mixture of diastereomers (ꢀ3:1): yield, 16 mg
(59%); Rf 0.43 (20% MeOH–CHCl3); FT-IR (KBr)
3416 (br), 2934, 1697, 1404, 1046 cmꢁ1 1H NMR
;
(CD3OD) for the major diastereomer, d 1.26 (s, 3H,

14
Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
C(20)CH3), 1.84–2.28 (m, 3H, C(4)HH0, C(4)HH0,
C(5)H), 2.10 (d, J = 10.4 Hz, 1H, C(5a)HH0), 3.04 (d,
J = 10.4 Hz, 1H, C(5a)HH0), 3.44 (d, J = 11.5 Hz, 1H,
C(30)HH0), 3.55 (d, J = 11.5 Hz, 1H, C(30)HH0), 3.61–
3.67 (m, 1H, C(3)HH0), 3.63 (s, 2H, C(5b)H2), 3.80–
3.89 (m, 1H, C(3)HH0), 3.95 (s, 1H, C(10)H), 6.84-6.95
(m, 1H, ArH), 7.01–7.09 (m, 2H, ArH), 7.19–7.27 (m,
1H, C(30)HH0), 3.64 (s, 2H, C(5b)H2), 3.65–3.92 (m,
2H, C(3)HH0, C(3)HH0), 4.00 (s, 1H, C(10)H), 6.82–
6.90 (m, 2H, ArH), 7.32–7.40 (m, 1H, ArH); H NMR
1
(CD3OD) for the minor diastereomer, d 3.95 (s, 1H,
C(10)H), the remaining peaks overlapped with nearby
signals and were not detected; the structural assignments
1
were in agreement with the H–1H COSY experiment;
1
1H, ArH); H NMR (CD3OD) for the minor diastereo-
13C NMR (CD3OD) for the major diastereomer, 23.1,
28.5, 29.4, 30.6, 51.1, 60.9, 67.4, 71.0, 77.1, 82.6, 88.3,
103.6 (t, J = 25.7 Hz), 111.0 (dd, J = 21.2, 3.4 Hz),
122.4 (dd, J = 14.9, 4.0 Hz), 132.2 (dd, J = 5.7,
4.0 Hz), 161.3 (dd, J = 249.1, 12.0 Hz), 162.4 (dd, J =
246.7, 12.0 Hz), 167.8, 171.0 ppm; 13C NMR (CD3OD)
for the minor diastereomer, 28.9, 82.4, 88.6, 167.7,
173.0 ppm, the remaining peaks overlapped with
nearby signals and were not detected; MS (+ESI) 485
[M+Na]+; Mr (+ESI) 485.1172 [M+Na]+ (calcd for
C19H24F2N2NaO7S 485.1164).
mer, d 3.96 (s, 1H, C(10)H), the remaining peaks over-
lapped with nearby signals and were not detected; the
structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for the
major diastereomer, 23.4, 29.2, 30.6, 35.8, 51.3, 62.4,
67.7, 71.3, 77.3, 82.9, 88.6, 113.8 (d, J = 21.8 Hz),
115.9 (d, J = 21.8 Hz), 125.1 (d, J = 3.4 Hz), 130.3 (d,
J = 8.0 Hz), 142.3 (d, J = 6.9 Hz), 163.4 (d,
J = 245.1 Hz), 168.0, 171.3 ppm; 13C NMR (CD3OD)
for the minor diastereomer, 82.7, 88.9 ppm, the remain-
ing peaks overlapped with nearby signals and were not
detected; MS (ꢁCI) 443 [Mꢁ1]ꢁ; Mr (ꢁCI) 443.128 12
[Mꢁ1]ꢁ (calcd for C19H24FN2O7S 443.128 83 [Mꢁ1]ꢁ).
4.26. 5a-(3,4-Difluorobenzylsulfanyl)-dihydrobicyclomycin (8)
Using General procedure 2, 21 (40 mg, 0.08 mmol) gave
8 as a mixture of diastereomers (ꢀ3:1): yield, 20 mg
(53%); Rf 0.35 (20% MeOH–CHCl3); FT-IR (KBr)
4.24. 5a-(4-Fluorobenzylsulfanyl)-dihydrobicyclomycin (6)
Using General procedure 2, 19 (45 mg, 0.09 mmol) gave 6
as a mixture of diastereomers (ꢀ3:1): yield, 25 mg (65%);
Rf 0.40 (20% MeOH–CHCl3); FT-IR (KBr) 3304 (br),
2986, 1691, 1404, 1046 cmꢁ1;1H NMR (CD3OD) for
the major diastereomer, d 1.25 (s, 3H, C(20)CH3), 1.98–
2.00 (m, 2H, C(4)HH0, C(4)HH0), 2.06 (d, J = 11.5 Hz,
1H, C(5a)HH0), 2.15–2.23 (m, 1H, C(5)H), 3.02 (d,
J = 11.5 Hz, 1H, C(5a)HH0), 3.03 (d, J = 11.7 Hz, 1H,
C(30)HH0), 3.56–3.65 (m, 1H, C(3)HH0), 3.58 (d,
J = 11.7 Hz, 1H, C(30)HH0), 3.60 (s, 2H, C(5b)H2),
3.72–3.87 (m, 1H, C(3)HH0), 3.95 (s, 1H, C(10)H), 6.91–
3401 (br), 2976, 1701, 1401, 1043 cmꢁ1 1H NMR
;
(CD3OD) for the major diastereomer, d 1.25 (s, 3H,
C(20)CH3), 1.83–2.22 (m, 3H, C(4)HH0, C(4)HH0,
C(5)H), 2.09 (d, J = 8.8 Hz, 1H, C(5a)HH0), 2.99 (d,
J = 8.8 Hz, 1H, C(5a)HH0), 3.43 (d, J = 10.7 Hz, 1H,
C(30)HH0), 3.62–3.90 (m, 2H, C(3)HH0, C(3)HH0),
3.63 (d, J = 10.7 Hz, 1H, C(30)HH0), 3.70 (s, 2H,
C(5b)H2), 3.95 (s, 1H, C(10)H), 7.07–7.16 (m, 2H,
ArH), 7.16–7.23 (m, 1H, ArH); 1H NMR (CD3OD)
for the minor diastereomer, d 3.94 (s, 1H, C(10)H), the
remaining peaks overlapped with nearby signals and
were not detected; the structural assignments were in
agreement with the 1H–1H COSY experiment; 13C
NMR (CD3OD) for the major diastereomer, 23.1,
29.0, 30.2, 35.1, 51.0, 61.0, 67.4, 71.0, 77.1, 82.4, 88.3,
117.1 (d, J = 9.2 Hz), 117.9 (d, J = 9.2 Hz), 125.5 (dd,
J = 9.2, 3.4 Hz), 136.7 (dd, J = 9.2, 3.4 Hz), 149.6 (dd,
J = 248.8, 13.1 Hz), 150.3 (dd, J = 248.2, 12.5 Hz),
167.7, 171.0 ppm; 13C NMR (CD3OD) for the minor
diastereomer peaks overlapped with nearby signals and
were not detected; MS (ꢁFAB) 461 [Mꢁ1]ꢁ; Mr
(ꢁFAB) 461.118 14 [Mꢁ1]ꢁ (calcd for C19H23F2N2O7S
461.119 41 [Mꢁ1]ꢁ).
1
6.97 (m, 2H, ArH), 7.24–7.30 (m, 2H, ArH); H NMR
(CD3OD) for the minor diastereomer, d 3.94 (s, 1H,
C(10)H), the remaining peaks overlapped with nearby sig-
nals and were not detected; the structural assignments
1
were in agreement with the H–1H COSY experiment;
13C NMR (CD3OD) for the major diastereomer, 23.4,
29.2, 30.5, 35.5, 51.3, 62.7, 67.7, 71.3, 77.3, 82.9, 88.5,
115.2 (d, J = 20.0 Hz, C(300), C(500)), 130.5 (d,
J = 8.0 Hz, C(200), C(600)), 135.3 (d, J = 2.3 Hz, C(100)),
162.4 (d, J = 243.9 Hz, C(400)), 168.0, 171.3 ppm; 13C
NMR (CD3OD) for the minor diastereomer, 62.6, 72.4,
82.8, 88.8, 166.7, 173.3 ppm, the remaining peaks over-
lapped with nearby signals and were not detected; MS
(ꢁCI) 443 [Mꢁ1]ꢁ; Mr (ꢁCI) 443.127 96 [Mꢁ1]ꢁ (calcd
for C19H24FN2O7S 443.128 83 [Mꢁ1]ꢁ).
4.27. 5a–(3,5–Difluorobenzylsulfanyl)–dihydrobicyclomy-
cin (9)
4.25. 5a-(2,4-Difluorobenzylsulfanyl)-dihydrobicyclo-
mycin (7)
Using General procedure 2, 22 (22 mg, 0.04 mmol) gave
9 as a mixture of diastereomers (ꢀ3:1): yield, 12 mg
(61%); Rf 0.41 (20% MeOH–CHCl3); FT-IR (KBr)
Using General procedure 2, 20 (30 mg, 0.06 mmol) gave
7 as a mixture of diastereomers (ꢀ3:1): yield, 17 mg
(63%); Rf 0.39 (20% MeOH–CHCl3); FT-IR (KBr)
3301 (br), 2938, 1687, 1392, 1045 cmꢁ1 1H NMR
;
(CD3OD) for the major diastereomer, d 1.25 (s, 3H,
C(20)CH3), 1.82–1.94 (m, 1H, C(4)HH0), 1.95–2.01 (m,
1H, C(4)HH0), 2.09 (d, J = 10.9 Hz, 1H, C(5a)HH0),
2.11–2.19 (m, 1H, C(5)H), 3.02 (d, J = 10.9 Hz, 1H,
C(5a)HH0), 3.43 (d, J = 11.0 Hz, 1H, C(30)HH0), 3.61
(s, 2H, C(5b)H2), 3.63 (d, J = 11.0 Hz, 1H, C(30)HH0),
3.65–3.79 (m, 2H, C(3)HH0, C(3)HH0), 3.95 (s, 1H,
C(10)H), 6.69–6.77 (m, 1H, ArH), 6.87–6.93 (m, 2H,
3381 (br), 2941, 1688, 1401, 1044 cmꢁ1 1H NMR
;
(CD3OD) for the major diastereomer, d 1.27 (s, 3H,
C(20)CH3), 1.96–2.03 (m, 2H, C(4)HH0, C(4)HH0),
2.01–2.08 (m, 1H, C(5)H), 2.15 (d, J = 9.3 Hz, 1H,
C(5a)HH0), 3.08 (d, J = 9.3 Hz, 1H, C(5a)HH0), 3.45
(d, J = 11.3 Hz, 1H, C(30)HH0), 3.60 (d, J = 11.3 Hz,

Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
15
ArH); 1H NMR (CD3OD) for the minor diastereomer, d
3.62 (s, 2H, C(5b)H), 3.93 (s, 1H, C(10)H), the remaining
peaks overlapped with nearby signals and were not
detected; the structural assignments were in agreement
with the 1H–1H COSY experiment; 13C NMR (CD3OD)
for the major diastereomer, 23.2, 29.1, 30.1, 35.1, 51.0,
60.9, 67.4, 71.1, 77.1, 82.6, 88.3, 102.0 (t, J = 26.3 Hz,
C(400)), 111.8 (dd, J = 17.6, 7.9 Hz, C(200), C(600)), 143.8
(t, J = 9.2 Hz, C(100)), 163.3 (dd, J = 247.0, 13.1 Hz,
C(300), C(500)), 167.7, 171.1 ppm; 13C NMR (CD3OD)
for the minor diastereomer peaks overlapped with near-
by signals and were not detected; MS (ꢁFAB) 461
[Mꢁ1]ꢁ; Mr (ꢁFAB) 461.120 73 [Mꢁ1]ꢁ (calcd for
C19H23F2N2O7S 461.119 41 [Mꢁ1]ꢁ).
remaining peaks overlapped with nearby signals and
were not detected; the structural assignments were in
agreement with the 1H–1H COSY experiment; 13C
NMR (CD3OD) for the major diastereomer, 23.4,
29.3, 30.3, 34.9, 51.2, 62.7, 67.6, 71.3, 77.3, 82.7, 88.6,
113.4 (dd, J = 15.0, 6.3 Hz, C(200), C(600)), 133.7 (dd,
J = 11.5, 7.9 Hz, C(100)), 138.7 (dt, J = 245.5, 17.0 Hz,
C(400)), 151.4 (ddd, J = 243.6, 11.3, 3.4 Hz, C(300),
C(500)), 167.3, 170.8 ppm; 13C NMR (CD3OD) for the
minor diastereomer, 82.8, 88.7, 166.9 ppm, the remain-
ing peaks overlapped with nearby signals and were not
detected; MS (ꢁCI) 479 [Mꢁ1]ꢁ; Mr (ꢁCI) 479.110 83
[Mꢁ1]ꢁ (calcd for C19H22F3N2O7S 479.109 98 [Mꢁ1]ꢁ).
4.30. 5a-(2,3,5,6-Tetrafluorobenzylsulfanyl)-dihydrobicy-
clomycin (12)
4.28. 5a-(2,3,6-Trifluorobenzylsulfanyl)-dihydrobicyclo-
mycin (10)
Using General procedure 2, 25 (40 mg, 0.07 mmol) gave
12 as a mixture of diastereomers (ꢀ3:1): yield, 21 mg
(58%); Rf 0.39 (20% MeOH–CHCl3); FT-IR (KBr)
Using General procedure 2, 23 (40 mg, 0.08 mmol) gave
10 as a mixture of diastereomers (ꢀ3:1): yield, 23 mg
(62%); Rf 0.41 (20% MeOH–CHCl3); FT-IR (KBr) 3384
3303 (br), 2958, 1688, 1392, 1045 cmꢁ1 1H NMR
;
1
(br), 2939, 1681, 1406, 1038 cmꢁ1; H NMR (CD3OD)
(CD3OD) for the major diastereomer, d 1.25 (s, 3H,
C(20)CH3), 1.85–2.31 (m, 3H, C(4)HH0, C(4)HH0,
C(5)H), 2.13 (d, J = 11.5 Hz, 1H, C(5a)HH0), 3.01 (d,
J = 11.5 Hz, 1H, C(5a)HH0), 3.44 (d, J = 11.5 Hz, 1H,
C(30)HH0), 3.59 (d, J = 11.5 Hz, 1H, C(30)HH0), 3.66
(s, 2H, C(5b)H2), 3.67–3.91 (m, 2H, C(3)HH0,
C(3)HH0), 3.95 (s, 1H, C(10)H), 7.14–7.23 (m, 1H,
for the major diastereomer, d 1.27 (s, 3H, C(20)CH3),
1.90–2.00 (m, 2H, C(4)HH0, C(4)HH0), 2.03–2.44 (m,
1H, C(5)H), 2.24 (d, J = 12.6 Hz, 1H, C(5a)HH0), 3.18
(d, J = 12.6 Hz, 1H, C(5a)HH0), 3.44 (d, J = 11.4 Hz,
1H, C(30)HH0), 3.60 (s, 2H, C(5b)H2), 3.61–3.93 (m,
2H, C(3)HH0, C(3)HH0), 3.62 (d, J = 11.4 Hz, 1H,
C(30)HH0), 3.96 (s, 1H, C(10)H), 6.86–6.94 (m, 1H,
1
ArH); H NMR (CD3OD) for the minor diastereomer,
1
ArH), 7.09–7.20 (m, 1H, ArH); H NMR (CD3OD) for
d 3.01 (d, J = 11.5 Hz, 1H, C(5a)HH0), 3.96 (s, 1H,
C(10)H), the remaining peaks overlapped with nearby
signals and were not detected; the structural assignments
the minor diastereomer,
d 3.48 (d, J = 11.4 Hz,
C(30)HH0), 3.66 (d, J = 11.4 Hz, C(30)HH0), 4.00 (s, 1H,
C(10)H), the remaining peaks overlapped with nearby sig-
nals and were not detected; the structural assignments
1
were in agreement with the H–1H COSY experiment;
13C NMR (CD3OD) for the major diastereomer, 23.4,
28.2, 29.3, 30.7, 51.2, 62.4, 67.7, 71.3, 77.3, 82.8, 88.6,
112.5 (dt, J = 19.5, 4.3 Hz), 123.7–123.9 (m), 138.2–
142.7 (m, 2C), 144.5–148.7 (m, 2C), 168.0, 171.3 ppm;
13C NMR (CD3OD) for the minor diastereomer, 82.6,
88.8, 166.9, 173.4 ppm, the remaining peaks overlapped
with nearby signals and were not detected; MS (ꢁCI)
497 [Mꢁ1]ꢁ; Mr (ꢁCI) 497.101 67 [Mꢁ1]ꢁ (calcd for
C19H21F4N2O7S 497.100 56 [Mꢁ1]ꢁ).
were in agreement with the H–1H COSY experiment;
1
13C NMR (CD3OD) for the major diastereomer, 23.3,
29.2, 30.8, 31.8, 51.5, 63.0, 67.7, 71.3, 77.3, 82.8, 88.6,
111.2 (ddd, J = 25.2, 4.5, 3.4 Hz), 116.1 (dd, J = 19.5,
9.1 Hz), 117.9 (dd, J = 21.1, 16.8 Hz), 147.7 (ddd,
J = 246.7, 12.8, 3.8 Hz), 148.9 (ddd, J = 242.8, 12.7,
7.8 Hz), 156.8 (ddd, J = 243.9, 5.7, 3.4 Hz), 167.9,
171.3 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 62.9, 82.6, 88.7, 167.0, 173.4 ppm, the remaining
peaks overlapped with nearby signals and were not
detected; MS (ꢁCI) 479 [Mꢁ1]ꢁ; Mr (ꢁCI) 479.114 24
[Mꢁ1]ꢁ (calcd for C19H22F3N2O7S 479.109 98 [Mꢁ1]ꢁ).
4.31. 5a-(2,3,4,5,6-Pentafluorobenzylsulfanyl)-dihydrobi-
cyclomycin (13)
Using General procedure 2, 26 (40 mg, 0.07 mmol) gave
13 as a mixture of diastereomers (ꢀ3:1): yield, 28 mg
(71%); Rf 0.39 (20% MeOH–CHCl3); FT-IR (KBr) 3301
4.29. 5a-(3,4,5-Trifluorobenzylsulfanyl)-dihydrobicyclo-
mycin (11)
1
(br), 2946, 1686, 1391, 1041 cmꢁ1; H NMR (CD3OD)
Using General procedure 2, 24 (30 mg, 0.06 mmol) gave
11 as a mixture of diastereomers (ꢀ3:1): yield, 18 mg
(68%); Rf 0.46 (20% MeOH–CHCl3); FT-IR (KBr)
for the major diastereomer, d 1.25 (s, 3H, C(20)CH3),
1.83–2.22 (m, 3H, C(4)HH0, C(4)HH0, C(5)H), 2.07 (d,
J = 12.3 Hz, 1H, C(5a)HH0), 3.10 (d, J = 12.3 Hz, 1H,
C(5a)HH0), 3.44 (d, J = 11.3 Hz, 1H, C(30)HH0), 3.60
(d, J = 11.3 Hz, 1H, C(30)HH0), 3.65–3.92 (m, 2H,
C(3)HH0, C(3)HH0), 3.75 (s, 2H, C(5b)H2), 3.95 (s, 1H,
C(10)H); 1H NMR (CD3OD) for the minor diastereomer
peaks overlapped with nearby signals and were not detect-
ed; the structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for the
major diastereomer, 23.5, 29.4, 30.9, 31.8, 51.7, 61.3,
67.8, 71.5, 77.5, 82.8, 88.8, 113.7–113.9 (m), 137.8–138.8
(m), 138.9–140.0 (m, 2C), 141.5–147.3 (m, 2C), 168.2,
3302 (br), 2947, 1678, 1391, 1042 cmꢁ1 1H NMR
;
(CD3OD) for the major diastereomer, d 1.26 (s, 3H,
C(20)CH3), 1.85–2.15 (m, 2H, C(4)HH0, C(4)HH0),
2.10 (d, J = 10.4 Hz, 1H, C(5a)HH0), 2.15–2.25 (m,
1H, C(5)H), 2.98 (d, J = 10.4 Hz, 1H, C(5a)HH0), 3.44
(d, J = 11.0 Hz, 1H, C(30)HH0), 3.60 (s, 2H, C(5b)H2),
3.61–3.93 (m, 3H, C(3)HH0, C(3)HH0, C(30)HH0), 4.00
(s, 1H, C(10)H), 7.11–7.21 (m, 2H, ArH); 1H NMR
(CD3OD) for the minor diastereomer, d 1.22 (s, 1H,
C(20)H), 3.63 (d, J = 11.0 Hz, 1H, C(30)HH0), the

16
Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
171.4 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 23.0, 62.4, 78.1, 82.7, 88.9, 167.1, 173.4 ppm, the
remaining peaks overlapped with nearby signals and were
not detected; MS (ꢁCI) 515 [Mꢁ1]ꢁ; Mr (ꢁCI) 515.092 43
[Mꢁ1]ꢁ (calcd for C19H20F5N2O7S 515.091 14 [Mꢁ1]ꢁ).
4.34. 5a-(3-Chlorobenzylsulfanyl)-dihydrobicyclomycin (41)
Using General procedure 2, 46 (30 mg, 0.06 mmol) gave
41 as a mixture of diastereomers (ꢀ3:1): yield, 14 mg
(52%); Rf 0.45 (20% MeOH–CHCl3); FT-IR (KBr)
1
3259, 1697, 1404, 1041 cmꢁ1; H NMR (CD3OD) for
4.32. 5a-(4-Trifluoromethylbenzylsulfanyl)-dihydrobicy-
clomycin (14)
the major diastereomer, d 1.40 (s, 3H, C(20)CH3), 2.01–
2.23 (m, 3H, C(4)HH0, C(4)HH0, C(5)H), 2.14 (d,
J = 10.4 Hz, 1H, C(5a)HH0), 3.19 (d, J = 10.4 Hz, 1H,
C(5a)HH0), 3.58 (d, J = 11.5 Hz, 1H, C(30)HH0), 3.73
(d, J = 11.5 Hz, 1H, C(30)HH0), 3.75–3.97 (m, 2H,
C(3)HH0, C(3)HH0), 3.76 (s, 2H, C(5b)H2), 4.09 (s, 1H,
C(10)H), 7.23–7.35 (m, 3H, ArH), 7.43 (s, 1H, ArH);
1H NMR (CD3OD) for the minor diastereomer, d 3.16
(d, J = 14.8 Hz, 1H, C(5a)HH0), 3.62 (d, J = 13.2 Hz,
1H, C(30)HH0), 4.10 (s, 1H, C(10)H), the remaining peaks
overlapped with nearby signals and were not detected;
the structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for the
major diastereomer, 23.3, 29.2, 30.6, 35.7, 51.3, 61.2,
67.7, 71.4, 77.3, 82.8, 88.5, 127.2, 127.7, 129.2, 130.1,
134.4, 141.7, 168.0, 171.3 ppm; 13C NMR (CD3OD) for
the minor diastereomer, 82.7, 88.9, 127.8 ppm, the
remaining peaks overlapped with nearby signals and
were not detected; MS (+ESI) 483 [M+Na]+; Mr
(+ESI) 483.0985 [M+Na]+ (calcd for C19H25ClN2NaO7S
483.0964 [M+Na]+).
Using General procedure 2, 27 (30 mg, 0.06 mmol) gave
14 as a mixture of diastereomers (ꢀ3:1); yield, 17 mg
(64%); Rf 0.35 (20% MeOH–CHCl3); FT-IR (KBr)
3397 (br), 2939, 1688, 1405, 1125 cmꢁ1 1H NMR
;
(CD3OD) for the major diastereomer, d 1.25 (s, 3H,
C(20)CH3), 2.00–2.09 (m, 3H, C(4)HH0, C(4)HH0,
C(5)H), 2.10 (d, J = 9.9 Hz, 1H, C(5a)HH0), 3.03 (d,
J = 9.9 Hz, 1H, C(5a)HH0), 3.43 (d, J = 13.7 Hz, 1H,
C(30)HH0), 3.56–3.87 (m, 2H, C(3)HH0, C(3)HH0),
3.58 (d, J = 13.7 Hz, 1H, C(30)HH0), 3.70 (s, 2H,
C(5b)H2), 3.95 (s, 1H, C(10)H), 7.23–7.29 (m, 2H,
ArH), 7.56–7.59 (m, 2H, ArH); 1H NMR (CD3OD)
for the minor diastereomer, the peaks overlapped with
nearby signals and were not detected; the structural
1
assignments were in agreement with the H–1H COSY
experiment; 13C NMR (CD3OD) for the major diaste-
reomer, 24.0, 29.3, 30.5, 35.6, 51.3, 61.0, 67.7, 71.3,
77.3, 82.9, 88.6, 123.7–123.9 (m, CF3), 125.8–126.0
(m), 129.4 (2C), 133.1 (2C), 140.9, 168.0, 171.0 ppm;
13C NMR (CD3OD) for the minor diastereomer, 29.3,
62.3, 71.5, 82.6, 88.8 ppm, the remaining peaks over-
lapped with nearby signals and were not detected; MS
(ꢁCI) 533 [Mꢁ1]ꢁ; Mr (ꢁCI) 533.157 67 [Mꢁ1]ꢁ (calcd
for C20H24F3N2O7S 533.156 93 [Mꢁ1]ꢁ).
4.35. 5a-(3-Cyanobenzylsulfanyl)-dihydrobicyclomycin (42)
Using General procedure 2, 47 (40 mg, 0.08 mmol) gave
42 as a mixture of diastereomers (ꢀ3:1): yield, 21 mg
(58%); Rf 0.43 (20% MeOH–CHCl3); FT-IR (KBr)
1
3271, 1697, 1404, 1042 cmꢁ1; H NMR (CD3OD) for
4.33. 5a-(2,5-Bis-trifluoromethylbenzylsulfanyl)-dihydro-
bicyclomycin (15)
the major diastereomer, d 1.40 (s, 3H, C(20)CH3),
1.96–2.37 (m, 3H, C(4)HH0, C(4)HH0, C(5)H), 2.24 (d,
J = 10.4 Hz, 1H, C(5a)HH0), 3.14 (d, J = 10.4 Hz, 1H,
C(5a)HH0), 3.59 (d, J = 11.5 Hz, 1H, C(30)HH0), 3.72–
3.91 (m, 1H, C(3)HH0), 3.73 (d, J = 11.5 Hz, 1H,
C(30)HH0), 3.82 (s, 2H, C(5b)H2), 3.97–4.05 (m, 1H,
C(3)HH0), 4.10 (s, 1H, C(10)H), 7.56 (t, J = 7.7 Hz,
1H, ArH), 7.67 (d, J = 7.7 Hz, 1H, ArH), 7.75 (d,
J = 7.7 Hz, 1H, ArH), 7.79 (s, 1H, ArH); 1H NMR
(CD3OD) for the minor diastereomer, d 3.36 (d,
J = 13.2 Hz, 1H, C(5a)HH0), 4.09 (s, 1H, C(10)H), the
remaining peaks overlapped with nearby signals and
were not detected; the structural assignments were in
agreement with the 1H–1H COSY experiment; 13C
NMR (CD3OD) for the major diastereomer, 23.3,
29.3, 30.4, 35.3, 51.2, 61.2, 67.6, 71.4, 77.3, 82.8, 88.5,
112.5, 118.8, 129.7, 130.8, 132.8, 134.1, 141.3, 167.6,
171.1 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 35.2, 71.5, 82.6, 88.8, 134.2, 166.7, 173.2 ppm, the
remaining peaks overlapped with nearby signals and
were not detected; MS (+ESI) 474 [M+Na]+; Mr
(+ESI) 474.1300 [M+Na]+ (calcd for C20H25N3NaO7S
474.1306 [M+Na]+).
Using General procedure 2, 28 (45 mg, 0.07 mmol) gave
15 as a mixture of diastereomers (ꢀ3:1); yield, 28 mg
(68%); Rf 0.43 (20% MeOH–CHCl3); FT-IR (KBr)
3389 (br), 2939, 1689, 1380, 1043 cmꢁ1 1H NMR
;
(CD3OD) for the major diastereomer, d 1.24 (s, 3H,
C(20)CH3), 1.87–2.21 (m, 3H, C(4)HH0, C(4)HH0,
C(5)H), 2.10 (d, J = 11.8 Hz, 1H, C(5a)HH0), 2.97 (d,
J = 11.8 Hz, 1H, C(5a)HH0), 3.42 (d, J = 11.5 Hz, 1H,
C(30)HH0), 3.58 (d, J = 11.5 Hz, 1H, C(30)HH0), 3.68–
3.90 (m, 2H, C(3)HH0, C(3)HH0), 3.78 (s, 2H,
C(5b)H2), 3.95 (s, 1H, C(10)H), 7.75 (s, 1H, ArH), 7.89
1
(s, 2H, ArH); H NMR (CD3OD) for the minor diaste-
reomer, d 1.21 (s, 3H, C(20)CH3), 3.47 (d, J = 11.5 Hz,
1H, C(30)HH0), 3.94 (s, 1H, C(10)H), the remaining
peaks overlapped with nearby signals and were not
detected; the structural assignments were in agreement
with the 1H–1H COSY experiment; 13C NMR (CD3OD)
for the major diastereomer, 23.4, 29.3, 30.3, 34.7, 51.0,
61.1, 67.6, 71.3, 77.3, 82.7, 88.1, 120.8–120.9 (m, 2C,
CF3), 125.5, 129.8–129.9 (m), 131.2–132.6 (m, 2C),
143.0, 168.0, 171.2 ppm; 13C NMR (CD3OD) for the
minor diastereomer, 29.0, 29.8, 62.4, 71.4, 82.6, 88.8,
166.7, 173.4 ppm, the remaining peaks overlapped with
nearby signals and were not detected; MS (ꢁCI) 561
[Mꢁ1]ꢁ; Mr (ꢁCI) 561.113 77 [Mꢁ1]ꢁ (calcd for
C21H23F6N2O7S 561.113 02 [Mꢁ1]ꢁ).
4.36. 5a-(3-Methoxybenzylsulfanyl)-dihydrobicyclomycin
(43)
Using General procedure 2, 48 (35 mg, 0.07 mmol) gave
43 as a mixture of diastereomers (ꢀ3:1): yield, 16 mg

Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
17
(51%); Rf 0.45 (20% MeOH–CHCl3); FT-IR (KBr) 3411,
C(20)CH3), 1.91–2.35 (m, 3H, C(4)HH0, C(4)HH0,
C(5)H), 2.20 (d, J = 10.4 Hz, 1H, C(5a)HH0), 3.17 (d,
J = 10.4 Hz, 1H, C(5a)HH0), 3.51 (d, J = 11.5 Hz,
1H, C(30)HH0), 3.65–3.79 (m, 1H, C(3)HH0), 3.69 (d,
J = 11.5 Hz, 1H, C(30)HH0), 3.76 (s, 2H, C(5b)H2),
3.83–3.95 (m, 1H, C(3)HH0), 4.04 (s, 1H, C(10)H),
7.31 (t, J = 7.7 Hz, 1H, ArH), 7.42 (d, J = 7.7 Hz,
1H, ArH), 7.84 (d, J = 7.7 Hz, 1H, ArH), 7.94 (s,
1
2926, 1689, 1407, 1042 cmꢁ1; H NMR (CD3OD) for
the major diastereomer, d 1.32 (s, 3H, C(20)CH3),
1.95–2.09 (m, 2H, C(4)HH0, C(4)HH0), 2.12–2.19 (m,
1H, C(5)H), 2.17 (d, J = 10.9 Hz, 1H, C(5a)HH0), 3.14
(d, J = 10.9 Hz, 1H, C(5a)HH0), 3.50 (d, J = 11.0 Hz,
1H, C(30)HH0), 3.64 (d, J = 11.0 Hz, 1H, C(30)HH0),
3.62–3.70 (m, 2H, C(3)HH0, C(3)HH0), 3.66 (s, 2H,
C(5b)H2), 3.83 (s, 3H, 30ArOCH3), 4.02 (s, 1H,
C(10)H), 6.78 (d, J = 8.2 Hz, 1H, ArH), 6.89 (s, 1H,
ArH), 6.92 (d, J = 8.2 Hz, 1H, ArH), 7.19 (t,
J = 8.2 Hz, 1H, ArH); 1H NMR (CD3OD) for the minor
diastereomer, d 3.60 (d, J = 11.4 Hz, 1H, C(30)HH0), the
remaining peaks overlapped with nearby signals and
were not detected; the structural assignments were in
agreement with the 1H–1H COSY experiment; 13C
NMR (CD3OD) for the major diastereomer, 23.3,
29.5, 30.5, 36.4, 51.3, 54.8, 61.1, 67.7, 71.4, 77.3, 82.9,
88.5, 112.9, 114.6, 121.5, 129.6, 140.8, 160.4, 168.0,
171.2 ppm; 13C NMR (CD3OD) for the minor diastereo-
mer, 172.6 ppm, the remaining peaks overlapped with
nearby signals and were not detected; MS (+ESI) 479
[M+Na]+; Mr (+ESI) 479.1465 [M+Na]+ (calcd for
C20H28N2NaO8S 479.1459 [M+Na]+).
1
1H, ArH); H NMR (CD3OD) for the minor diaste-
reomer, d 4.05 (s, 1H, C(10)H), the remaining peaks
overlapped with nearby signals and were not detected;
the structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for
the major diastereomer, 24.2, 29.9, 30.5, 37.3, 52.2,
62.3, 68.4, 72.3, 78.2, 83.8, 89.3, 117.2, 119.5, 128.9,
129.2, 132.7, 139.8, 163.2, 172.2 ppm, the remaining
peaks were not detected; 13C NMR (CD3OD) for the
minor diastereomer, 78.1, 83.7 ppm, the remaining
peaks overlapped with nearby signals and were not
detected; MS (+ESI) 493 [M+Na]+; Mr (+ESI)
493.1200 [M+Na]+ (calcd for C20H26N2NaO9S
493.1251).
4.39. 5a-(4-Carboxybenzylsulfanyl)-dihydrobicyclomycin
(55)
4.37. 5a-(3-Nitrobenzylsulfanyl)-dihydrobicyclomycin (44)
Using General procedure 1(A), 16 (50 mg, 0.15 mmol)
and thioacetic acid S-(4-carboxybenzyl) ester (59)
(62 mg, 0.44 mmol) gave 57 as a crude mixture (Rf
0.43 (30% MeOH–CHCl3)). The reaction was not puri-
fied but immediately deprotected using General proce-
dure 2, to give 55 as a mixture of diastereomers
(ꢀ10:1): yield, 11 mg (24%); Rf 0.45 (40% MeOH–
Using General procedure 2, 49 (20 mg, 0.04 mmol) gave
44 as a mixture of diastereomers (ꢀ3:1): yield, 9 mg
(48%); Rf 0.41 (20% MeOH–CHCl3); FT-IR (KBr)
1
3271, 1698, 1401, 1045 cmꢁ1; H NMR (CD3OD) for
the major diastereomer, d 1.32 (s, 3H, C(20)CH3),
2.04–2.33 (m, 3H, C(4)HH0, C(4)HH0, C(5)H), 2.19 (d,
J = 10.4 Hz, 1H, C(5a)HH0), 3.08 (d, J = 10.4 Hz, 1H,
C(5a)HH0), 3.50 (d, J = 11.5 Hz, 1H, C(30)HH0), 3.65
(d, J = 11.5 Hz, 1H, C(30)HH0), 3.72–3.77 (m, 1H,
C(3)HH0), 3.83 (s, 2H, C(5b)H2), 3.87–3.97 (m, 1H,
C(3)HH0), 4.02 (s, 1H, C(10)H), 7.55 (t, J = 8.2 Hz,
1H, ArH), 7.74–7.79 (m, 1H, ArH), 8.09–8.13 (m, 1H,
1
CHCl3); FT-IR (KBr) 3261, 1698, 1424, 1041 cmꢁ1; H
NMR (CD3OD) for the major diastereomer, d 1.36 (s,
3H, C(20)CH3), 2.02–2.38 (m, 4H, C(4)HH0, C(4)HH0,
C(5)H, C(5a)HH0), 3.14–3.18 (m, 1H, C(5a)HH0),
3.75–3.84 (m, 4H, C(30)HH0, C(30)HH0, C(3)HH0,
C(3)HH0), 3.76 (s, 2H, C(5b)H2), 4.04 (s, 1H, C(10)H),
7.38 (s, 2H, ArH), 7.97 (s, 2H, ArH); 1H NMR
(CD3OD) for the minor diastereomer, d 1.20 (s, 3H,
C(20)CH3), the remaining peaks overlapped with nearby
signals and were not detected; the structural assignments
1
ArH), 8.23–8.27 (s, 1H, ArH); H NMR (CD3OD) for
the minor diastereomer, d 3.57 (d, J = 11.4 Hz, 1H,
C(30)HH0), 4.01 (s, 1H, C(10)H), the remaining peaks
overlapped with nearby signals and were not detected;
the structural assignments were in agreement with the
1H–1H COSY experiment; 13C NMR (CD3OD) for the
major diastereomer, 23.4, 29.3, 30.5, 35.3, 51.3, 61.2,
67.7, 71.4, 77.3, 82.8, 88.6, 122.1, 124.0, 129.8, 135.6,
141.9, 148.9, 167.9, 171.2 ppm; 13C NMR (CD3OD)
for the minor diastereomer, 29.8, 35.5, 62.4, 71.5, 82.7,
88.8, 135.7, 141.9, 166.7, 173.2 ppm, the remaining
peaks overlapped with nearby signals and were not
detected; MS (+ESI) 494 [M+Na]+; Mr (+ESI)
494.1226 [M+Na]+ (calcd for C19H25N3NaO9S
494.1204).
1
were in agreement with the H–1H COSY experiment;
13C NMR (CD3OD) for the major diastereomer, 23.0,
28.7, 30.6, 35.8, 50.1, 71.0, 76.9, 82.5, 88.1, 128.3,
128.6, 141.9, 161.2, 170.8 ppm, the remaining peaks
were not detected; 13C NMR (CD3OD) for the minor
diastereomer, 82.4 ppm, the remaining peaks overlapped
with nearby signals and were not detected; MS (+ESI)
493 [M+Na]+; Mr (+ESI) 493.1248 [M+Na]+ (calcd for
C20H26N2NaO9S 493.1251).
4.40. Thioacetic acid S-(2-fluorobenzyl) ester (29)
4.38. 5a-(3-Carboxybenzylsulfanyl)-dihydrobicyclomycin
(54)
Using General procedure 3, 2-fluorobenzyl bromide
(500 mg, 2.60 mmol) and potassium thioacetate
(362 mg, 3.12 mmol) gave yellow oil 29: yield, 416 mg
(87%); Rf 0.55 (5% EtOAc–hexanes); FT-IR (neat)
Using General procedure 2, 56 (40 mg, 0.08 mmol)
gave 54 as a mixture of diastereomers (ꢀ10:1): yield,
21 mg (58%); Rf 0.47 (40% MeOH–CHCl3); FT-IR
1
1703, 1493, 1234, 1130 cmꢁ1; H NMR (CDCl3) d 2.47
(s, 3H, C(O)CH3), 4.29 (s, 2H, SCH2Ar), 7.13–7.23
(m, 2H, ArH), 7.32–7.38 (m, 1H, ArH), 7.47–7.54 (m,
1H, ArH); 13C NMR (CDCl3) 26.1, 29.7, 114.8 (d,
(KBr) 3271, 1697, 1404, 1042 cmꢁ1
;
1H NMR
(CD3OD) for the major diastereomer, d 1.35 (s, 3H,

18
Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
J = 20.6 Hz), 123.7 (d, J = 3.4 Hz), 124.2 (d,
J = 14.9 Hz), 128.6 (d, J = 8.0 Hz), 130.5 (d,
J = 4.6 Hz), 160.2 (d, J = 247.3 Hz), 194.3 ppm; MS
(+ESI) 207 [M+Na]+; Mr (+ESI) 207.0242 [M+Na]+
(calcd for C9H9FNaOS 207.0251 [M+Na]+).
30.7, 33.1, 103.2 (t, J = 25.2 Hz, C(4)), 112.7 (dd,
J = 15.9, 8.0 Hz, C(2), C(6)), 142.2 (t, J = 9.1 Hz,
C(1)), 163.4 (dd, J = 248.7, 12.9 Hz, C(3), C(5)),
194.9 ppm; MS (+CI) 220 [M+NH4]+; Mr (+CI)
220.0607 [M+NH4]+ (calcd for C9H12F2NOS 202.0608
[M+NH4]+).
4.41. Thioacetic acid S-(3-fluorobenzyl) ester (30)
4.45. Thioacetic acid S-(2,3,6-trifluorobenzyl) ester (35)
Using General procedure 3, 3-fluorobenzyl bromide
(500 mg, 2.6 mmol) and potassium thioacetate
(362 mg, 3.12 mmol) gave yellow oil 30: yield, 397 mg
(83%); Rf 0.45 (5% EtOAc–hexanes); FT-IR (neat)
Using General procedure 3, 2,3,6-trifluorobenzyl bro-
mide (500 mg, 2.20 mmol) and potassium thioacetate
(301 mg, 2.60 mmol) gave yellow oil 35: yield, 339 mg
(70%); Rf 0.40 (10% EtOAc–hexanes); FT-IR (neat)
1
1692, 1483, 1234, 1130 cmꢁ1; H NMR (CDCl3) d 2.48
1
(s, 3H, C(O)CH3), 4.23 (s, 2H, SCH2Ar), 7.03–7.21
(m, 3H, ArH), 7.34–7.42 (m, 1H, ArH); 13C NMR
(CDCl3) 29.7, 32.3, 113.6 (d, J = 20.1 Hz), 115.1 (d,
J = 21.9 Hz), 123.9 (d, J = 3.4 Hz), 129.5 (d,
J = 8.0 Hz), 139.7 (d, J = 6.9 Hz), 162.2 (d,
J = 246.2 Hz), 194.1 ppm; MS (+CI) 202 [M+NH4]+;
Mr (+CI) 202.0711 [M+NH4]+ (calcd for C9H13FNOS
202.0702 [M+NH4]+).
1682, 1433, 1344, 1125 cmꢁ1; H NMR (CDCl3) d 2.43
(s, 3H, C(O)CH3), 4.28 (s, 2H, SCH2Ar), 6.86–6.94
(m, 1H, ArH), 7.08–7.20 (m, 1H, ArH); 13C NMR
(CDCl3) 20.6, 30.1, 109.9 (ddd, J = 24.0, 6.9, 4.6 Hz),
114.9 (dd, J = 21.0, 16.4 Hz), 115.3 (dd, J = 19.4,
11.3 Hz), 146.4 (ddd, J = 245.1, 12.6, 3.4 Hz), 148.1
(ddd, J = 251.9, 13.7, 8.0 Hz), 155.8 (ddd, J = 246.2,
5.7, 2.3 Hz), 193.9 ppm; MS (+ESI) 243 [M+Na]+; Mr
(+ESI) 243.0081 [M+Na]+ (calcd for C9H7F3NaOS
243.0062 [M+Na]+).
4.42. Thioacetic acid S-(2,4-difluorobenzyl) ester (32)
Using General procedure 3, 2,4-difluorobenzyl bromide
(207 mg, 1 mmol) and potassium thioacetate (137 mg,
1.2 mmol) gave yellow oil 32: yield, 147 mg (73%); Rf
0.40 (5% EtOAc–hexanes); FT-IR (neat) 1696, 1508,
4.46. Thioacetic acid S-(3,4,5-trifluorobenzyl) ester (36)
Using General procedure 3, 3,4,5-trifluorobenzyl bro-
mide (500 mg, 2.20 mmol) and potassium thioacetate
(301 mg, 2.60 mmol) gave yellow solid 36: yield,
315 mg (65%); mp 132 ꢁC; Rf 0.40 (10% EtOAc–hex-
1130 cmꢁ1
;
1H NMR (CDCl3)
d
2.36 (s, 3H,
C(O)CH3), 4.07 (s, 2H, SCH2Ar), 6.62–6.72 (m, 2H,
ArH), 7.01–7.11 (m, 1H, ArH); 13C NMR (CDCl3)
26.6, 30.6, 104.2 (t, J = 25.2 Hz), 111.7 (dd, J = 21.2,
4.0 Hz), 121.4 (dd, J = 15.2, 3.7 Hz), 132.2 (dd,
J = 9.7, 5.2 Hz), 161.2 (dd, J = 250.2, 12.0 Hz), 162.7
(dd, J = 248.7, 11.5 Hz), 195.2 ppm; MS (+CI) 220
[M+NH4]+; Mr (+CI) 220.0608 [M+NH4]+ (calcd for
C9H12F2NOS 220.0608 [M+NH4]+).
1
anes); FT-IR (KBr) 1696, 1443, 1324, 1118 cmꢁ1; H
NMR (CDCl3) d 2.49 (s, 3H, C(O)CH3), 4.14 (s, 2H,
SCH2Ar), 7.03–7.08 (m, 2H, ArH); 13C NMR (CDCl3)
29.7, 31.8, 112.4 (dd, J = 14.9, 6.9 Hz, 2C), 133.9 (dd,
J = 12.6, 8.0 Hz), 138.5 (dt, J = 251.9, 14.9 Hz), 150.5
(ddd, J = 249.6, 10.3, 3.9 Hz), 193.9 ppm; MS (+CI)
238 [M+NH4]+; Mr (+CI) 238.0503 [M+NH4]+ (calcd
for C9H11F3NOS 243.0513 [M+NH4]+).
4.43. Thioacetic acid S-(3,4-difluorobenzyl) ester (33)
4.47. Thioacetic acid S-(2,3,5,6-tetrafluorobenzyl) ester (37)
Using General procedure 3, 3,4-difluorobenzyl bromide
(207 mg, 1 mmol) and potassium thioacetate (137 mg,
1.2 mmol) gave yellow oil 33: yield, 163 mg (81%); Rf
0.41 (5% EtOAc–hexanes); FT-IR (neat) 1704, 1347,
1118 cmꢁ1; 1H NMR (CDCl3) d 2.36 (s, 3H, C(O)CH3),
4.05 (s, 2H, SCH2Ar), 7.00–7.01 (m, 1H, ArH), 7.06–
7.08 (m, 1H, ArH), 7.11–7.15 (m, 1H, ArH); 13C NMR
(CDCl3) 30.7, 32.8, 117.7 (d, J = 17.2 Hz), 118.2 (d,
J = 17.2 Hz), 125.3 (dd, J = 6.3, 3.4 Hz), 135.4 (dd,
J = 5.7, 4.0 Hz), 150.3 (dd, J = 247.9, 12.6 Hz), 150.6
(dd, J = 248.4, 12.6 Hz), 195.1 ppm; MS (+CI) 220
[M+NH4]+; Mr (+CI) 220.0606 [M+NH4]+ (calcd for
C9H12F2NOS 220.0608 [M+NH4]+).
Using General procedure 3, 2,3,5,6-tetrafluorobenzyl
bromide (500 mg, 1.90 mmol) and potassium thioacetate
(259 mg, 2.60 mmol) gave yellow oil 37: yield, 272 mg
(60%); Rf 0.50 (10% EtOAc–hexanes); FT-IR (neat)
1
1703, 1463, 1324, 1115 cmꢁ1; H NMR (CDCl3) d 2.38
(s, 3H, C(O)CH3), 4.08 (s, 2H, SCH2Ar), 7.00–7.09
(m, 1H, ArH); 13C NMR (CDCl3) 25.3, 29.7, 111.5
(dt, J = 20.6, 3.5 Hz), 121.1–121.4 (m), 137.7–138.5
(m), 143.6–148.0 (m), 193.8 ppm; no signal was detected
by either CI or ESI MS.
4.48. Thioacetic acid S-(2,3,4,5,6-pentafluorobenzyl) ester
(38)
4.44. Thioacetic acid S-(3,5-difluorobenzyl) ester (34)
Using General procedure 3, 2,3,4,5,6-pentafluorobenzyl
bromide (206 mg, 1.00 mmol) and potassium thioacetate
(136 mg, 1.20 mmol) gave yellow oil 38: yield, 153 mg
(60%); Rf 0.50 (10% EtOAc–hexanes); FT-IR (neat)
Using General procedure 3, 3,5-difluorobenzyl bromide
(207 mg, 1 mmol) and potassium thioacetate (137 mg,
1.2 mmol) gave yellow oil 34: yield, 146 mg (72%); Rf
0.40 (5% EtOAc–hexanes); FT-IR (neat) 1697, 1456,
1
1698, 1463, 1344, 1123 cmꢁ1; H NMR (CDCl3) d 2.39
1323, 1153 cmꢁ1
;
C(O)CH3), 4.01 (s, 2H, SCH2Ar), 6.65–6.72 (m, 1H,
1H NMR (CDCl3) d 2.37 (s, 3H,
(s, 3H, C(O)CH3), 4.22 (s, 2H, SCH2Ar); 13C NMR
(CDCl3) 19.5, 29.4, 110.9 (ddd, J = 17.2, 17.2, 3.4 Hz),
135.1–138.6 (m), 138.7–142.1 (m, 2C), 142.8–146.5 (m,
ArH), 6.81–6.84 (m, 2H, ArH); 13C NMR (CDCl3)

Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
19
2C), 192.7 ppm; MS (+CI) 274 [M+NH4]+; Mr (+CI)
274.0319 [M+NH4]+ (calcd for C9H9F4NOS 274.0325
[M+NH4]+).
(374 mg, 2.4 mmol) gave brown oil 52: yield, 317 mg
(81%); Rf 0.65 (5% EtOAc–hexanes); FT-IR (neat)
1
1689, 1450, 1265, 1110 cmꢁ1; H NMR (CDCl3) d 2.24
(s, 3H, C(O)CH3), 3.70 (s, 3H, OCH3), 4.00 (s, 2H,
SCH2Ar), 6.66–6.70 (m, 1H, ArH), 6.74–6.78 (m, 2H,
ArH), 7.08–7.13 (m, 1H, ArH); 13C NMR (CDCl3)
29.7, 32.9, 54.6, 112.3, 113.8, 120.5, 129.1, 138.6,
159.2, 194.4 ppm; MS (+ESI) 219 [M+Na]+; Mr
(+ESI) 219.0460 [M+Na]+ (calcd for C10H12NaO2S
219.0451 [M+Na]+).
4.49. Thioacetic acid S-(4-trifluoromethylbenzyl) ester (39)
Using General procedure 3, 4-trifluoromethylbenzyl
bromide (239 mg, 1.00 mmol) and potassium thioacetate
(136 mg, 1.20 mmol) gave brown oil 39: yield, 166 mg
(71%); Rf 0.30 (5% EtOAc–hexanes); FT-IR (neat)
1
1701, 1463, 1324, 1118 cmꢁ1; H NMR (CDCl3) d 2.36
(s, 3H, C(O)CH3), 4.14 (s, 2H, SCH2Ar), 7.49–7.54
(m, 4 H); 13C NMR (CDCl3) 29.1, 32.0, 123.2–123.4
(m, CF3), 124.7–124.9 (m), 128.4 (2C), 129.5-130.7 (m,
2C), 131.6, 193.5 ppm; MS (+CI) 252 [M+NH4]+; Mr
(+CI) 252.0655 [M+NH4]+ (calcd for C10H13F3NOS
252.0607 [M+NH4]+).
4.54. Thioacetic acid S-(3-carboxybenzyl) ester (58)
Using General procedure 3, 3-chloromethylbenzoic acid
(340 mg, 2.0 mmol) and potassium thioacetate (274 mg,
2.4 mmol) gave white solid 58: yield, 257 mg (58%); mp
143–145 ꢁC; Rf 0.45 (1:20:50 AcOH–CHCl3-THF); FT-
IR (KBr) 1689, 1413, 1299, 1128 cmꢁ1 1H NMR
;
4.50. Thioacetic acid S-(3,5-bis-trifluoromethylbenzyl)
ester (40)
(CD3OD) d 2.45 (s, 3H, C(O)CH3), 4.27 (s, 2H,
SCH2Ar), 7.48 (t, J = 7.5 Hz, 1H), 7.60 (d, J = 7.5 Hz,
1 H), 8.01 (d, J = 7.5 Hz, 1H), 8.07 (s, 1 H); 13C NMR
(CD3OD) 29.9, 32.7, 127.8, 128.9, 129.5, 130.9, 143.7,
168.7, 195.3 ppm; MS (+ESI) 233 [M+Na]+; Mr
(+ESI) 233.0252 [M+Na]+ (calcd for C10H10NaO3S
233.0243).
Using General procedure 3, 3,5-bis-trifluoromethyl ben-
zyl bromide (307 mg, 1.00 mmol) and potassium thioac-
etate (136 mg, 1.20 mmol) gave brown oil 40: yield,
220 mg (73%); Rf 0.27 (5% EtOAc–hexanes); FT-IR
1
(neat) 1704, 1452, 1334, 1123 cmꢁ1; H NMR (CDCl3)
d 2.51 (s, 3H, C(O)CH3), 4.33 (s, 2H, SCH2Ar), 7.90
(s, 3H, ArH); 13C NMR (CDCl3) 29.6, 32.0, 120.6–
120.9 (m, 2C, CF3), 124.5, 128.5–128.6 (m), 130.7–
132.0 (m, 2C), 140.2, 193.7 ppm; MS (+CI) 320
[M+NH4]+; Mr (+CI) 320.0545 [M+NH4]+ (calcd for
C11H12F6NOS 320.0544 [M+NH4]+).
4.55. Thioacetic acid S-(4-carboxybenzyl) ester (59)
Using General procedure 3, 4-chloromethylbenzoic acid
(85 mg, 0.50 mmol) and potassium thioacetate (68 mg,
0.6 mmol) gave white solid 59: yield, 132 mg (50%);
mp 155–158 ꢁC; Rf 0.43 (1:20:50 AcOH–CHCl3-THF);
1
FT-IR (KBr) 1691, 1421, 1287, 1121 cmꢁ1; H NMR
4.51. Thioacetic acid S-(3-chlorobenzyl) ester (50)
(CD3OD) d 2.34 (s, 3H, C(O)CH3), 4.16 (s, 2H,
SCH2Ar), 7.37 (s, 2 H), 7.90 (s, 2 H); 13C NMR (CDCl3)
d 30.2, 33.8, 129.9, 131.1, 131.2, 144.8, 169.9, 196.3 ppm;
MS (+ESI) 233 [M+Na]+; Mr (+ESI) 233.0251 [M+Na]+
(calcd for C10H10NaO3S 233.0243).
Using General procedure 3, 3-chlorobenzyl bromide
(205 mg, 1.0 mmol) and potassium thioacetate
(137 mg, 1.2 mmol) gave brown oil 50: yield, 176 mg
(88%); Rf 0.55 (5% EtOAc–hexanes); FT-IR (neat)
1
1691, 1425, 1234, 1129 cmꢁ1; H NMR (CDCl3) d 2.49
4.56. Inhibitory properties of bicyclomycin and bicyclo-
mycin derivatives in the poly(C)-dependent ATPase assay32
(s, 3H, C(O)CH3), 4.21 (s, 2H, SCH2Ar), 7.34–7.42
(m, 4H, ArH); 13C NMR (CDCl3) 29.8, 32.3, 126.5,
126.9, 128.4, 129.3, 133.8, 139.3, 194.2 ppm; MS (+CI)
218 [M+NH4]+; Mr (+CI) 218.0422 [M+NH4]+ (calcd
for C9H13ClNOS 218.0406 [M+NH4]+).
The ability of wild-type rho to hydrolyze [c-32P]ATP
was assayed in 100 lL reactions containing ATPase
buffer (40 mM Tris–HCl, pH 7.9, 50 mM KCl, and
12 mM MgCl2), 250 lM ATP, 0.5 lCi [c-32P]ATP,
300 nM poly(C), and 200 nM rho based on monomer.
Reactions were preincubated at 32 ꢁC for 90 s prior to
the addition of ATP. Aliquots (1 lL) were removed at
various times (15, 30, 45, 60, and 75 s) during the
reaction and spotted onto PEI-TLC plates. [c-32P]ATP
and 32Pi were separated by chromatography on PEI-
TLC plates using 0.75 M KH2PO4, pH 3.5, as the mo-
bile phase. The developed PEI-TLC plates were used
to expose PhosphorImager Plates (Fuji and Molecular
Dynamics) (3 h), scanned on a Storm 860 PC Phos-
phorImager, and analyzed using Molecular Dynam-
icsÕs ImageQuant 5.0. The initial rates of reactions
were determined by plotting the amount of ATP
hydrolyzed versus time. Relative percent activities
were calculated from the initial velocities. Each assay
was performed in duplicate and the results were
averaged.
4.52. Thioacetic acid S-(3-cyanobenzyl) ester (51)
Using General procedure 3, 3-cyanobenzyl bromide
(392 mg, 2.0 mmol) and potassium thioacetate
(374 mg, 2.4 mmol) gave brown oil 51: yield, 143 mg
(75%); Rf 0.53 (5% EtOAc–hexanes); FT-IR (neat)
1
1697, 1426, 1234, 1131 cmꢁ1; H NMR (CDCl3) d 2.47
(s, 3H, C(O)CH3), 4.21 (s, 2H, SCH2Ar), 7.51–7.71
(m, 4H, ArH); 13C NMR (CDCl3) 29.7, 31.9, 112.1,
118.0, 128.3, 130.3, 131.7, 132.7, 139.0, 193.8 ppm; MS
(+ESI) 214 [M+Na]+; Mr (+ESI) 214.0290 [M+Na]+
(calcd for C10H9NNaOS 214.0297 [M+Na]+).
4.53. Thioacetic acid S-(3-methoxybenzyl) ester (52)
Using General procedure 3, 3-methoxybenzyl bromide
(402 mg, 2.0 mmol) and potassium thioacetate

20
Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
16. Hudlicky, M., Pavlath, A. E., Chemistry of Organic
Fluorine Compounds II. A Critical Review, ACS Mono-
graph 187; American Chemical Society: Washington, DC,
1995.
17. Grever, M. R.; Kopecky, K. J.; Coltman, C. A.; Files, J.
C.; Greenberg, B. R.; Hutton, J. J.; Talley, R.; Von Hoff,
D. D.; Balcerzak, S. P. Nouv. Revu. Franc. dÕHemat. 1988,
30, 457.
4.57. Inhibitory properties of bicyclomycin and bicyclo-
mycin derivatives in the antimicrobial assay33
Aliquots (200 lL) from suspensions of overnight LB
broth cultures (E. coli W3350) were diluted into LB
broth (2 mL). The suspension was poured onto 15 mL
volume LB agar plates. The solution was gently rocked
to distribute the cells evenly over the plates surface, and
any excess cell solution was removed by pipet. The
plates were incubated at 37 ꢁC (30 min) and an antibiot-
ic-assay disk (0.25 in. diameter) containing 20 lL to the
test compound (1, 2, 4, 8, 16, and 32 mg/mL in 50–100%
DMSO) was placed on the agar surface. The plates were
incubated at 37 ꢁC (18 h). Data plots of the zone of
inhibited bacterial growth (cm3) versus log(1000 · C),
where C is the concentration of the test compound
(mg/mL), yielded linear slopes to provide the minimal
inhibitory concentrations (MIC) for 1. The dihydrobicy-
clomycin derivatives showed no zones of growth inhibi-
tion. Each assay was performed in duplicate and the
results were averaged.
18. Montgomery, J. A. Cancer Res. 1982, 42, 3911.
19. Tsuruo, T.; Sato, S.; Yusa, K. Jpn. J. Cancer Res. 1989,
80, 686.
20. Zhou, W.; Gumina, G.; Chong, Y.; Wang, J.; Schinazi, R.
F.; Chu, C. K. J. Med. Chem. 2004, 47, 3399.
21. Van Aerschot, A.; Herdewijn, P.; Balzarini, J.; Pauwels,
R.; De Clercq, E. J. Med. Chem. 1989, 32, 1743.
22. Herdewijn, P. A. M.; Van Aerschot, A.; Balzarini, J.; De
Clercq, E.; Everaert, D. H.; De Winter, H. L.; Peeters, O.
M.; Blaton, N. M.; De Ranter, C. J. Med. Chem. Res.
1991, 1, 9.
23. Chu, D. T. W. In Organofluorine Compounds in Medicinal
Chemistry and Biomedical Applications; Filler, R., Kobay-
ashi, Y., Yagupolskii, L. M., Eds.; Elsevier: Amsterdam,
1993, pp 165–197.
24. Floyd, D. M.; Kimball, S. D.; Krapcho, J.; Das, J.; Turk,
C. F.; Moquin, R. V.; Lago, M. W.; Duff, K. J.; Lee, V.
G.; White, R. E.; Ridgewell, R. E.; Moreland, S.; Brittain,
R. J.; Normandin, D. E.; Hedberg, S. A.; Cucinotta, G. G.
J. Med. Chem. 1992, 35, 756.
25. Smart, B. E. J. Fluorine Chem. 2001, 109, 3.
26. Zhang, Z.; Kohn, H. J. Am. Chem. Soc. 1994, 116, 9815.
27. Park, H. G.; Zhang, X.; Widger, W. R.; Kohn, H. J. Org.
Chem. 1996, 61, 7750.
Acknowledgments
We thank Dr. Y. Itoh and the Fujisawa Pharmaceutical
Co., Ltd., Japan, for the gift of bicyclomycin, and Dr. T.
Platt (University of Rochester) for the overproducing
strain of rho. This work was supported by the National
Institutes of Health Grant GM37934 and the Robert A.
Welch Foundation Grant E1381 (W.R.W.).
´
28. Santillan Jr, A.; Park, H. G.; Zhang, X.; Lee, O.-S.;
Widger, W. R.; Kohn, H. J. Org. Chem. 1996, 61, 7756.
´
29. Santillan Jr, A.; Zhang, X.; Hardesty, J.; Widger, W. R.;
Kohn, H. J. Med. Chem. 1998, 41, 1185.
30. Park, H. G.; Zhang, Z.; Zhang, X.; Widger, W. R.; Kohn,
H. J. Org. Chem. 1996, 61, 7764.
References and notes
´
31. Vincent, F.; Srinivansan, J.; Santillan Jr, A.; Widger, W.
1. Miyoshi, T.; Miyairi, N.; Aoki, H.; Kohsaka, M.; Sakai,
H.; Imanaka, H. J. Antibiot. 1972, 25, 569.
2. Miyamura, S.; Ogasawara, N.; Otsuka, H.; Niwayama, S.;
Tanaka, H.; Take, T.; Uchiyama, T.; Ochiai, H.; Abe, K.
J. Antibiot. 1972, 25, 610.
R.; Kohn, H. J. Org. Chem. 2001, 66, 2251.
32. Sharp, J. A.; Galloway, J. L.; Platt, T. J. Biol. Chem. 1983,
258, 3482.
33. Ericsson, H. M.; Sherris, J. C. Acta Pathol. Microbiolol.
Scand. Section B: Microbiol. Immun. 1971, 217, 1.
34. Kamiya, T.; Maeno, S.; Kataura, Y. Belgium Patent 847-
475, 1976.
3. Kohn, H.; Widger, W. Curr. Drug Targ. Inf. Disord. in
press.
4. Park, H. G.; Zhang, X.; Moon, H. S.; Zwiefka, A.; Cox,
K.; Gaskell, S. J.; Widger, W. R.; Kohn, H. Arch.
Biochem. Biophys. 1995, 323, 447.
5. Magyar, A.; Zhang, X.; Kohn, H.; Widger, W. R. J. Biol.
Chem. 1996, 271, 25369.
6. Weber, T. P.; Widger, W. R.; Kohn, H. Biochemistry 2002,
41, 12377.
7. Skordalakes, E.; Brogan, A. P.; Park, B. S.; Kohn, H.;
Berger, J. M. Structure 2005, 13, 99.
35. Hansch, C.; Leo, A. Exploring QSAR: Fundamentals and
Applications in Chemistry and Biology, American Chemical
Society: Washington, DC, 1995; Vol. 1 and Vol. 2, pp 19–
35.
36. McLellan, A.L. Table of Experimental Dipole Moment,
W.H. Freeman and Co.: London. 1963; Vol 1.
37. Onda, M.; Mukaida, H.; Yamaguchi, I. J. Mol. Spectrosc.
1996, 176, 146.
38. Freiser, H.; Hobbs, M. E.; Gross, P. M. J. Am. Chem. Soc.
1949, 71, 111.
39. Smyth, C. P. J. Phys. Chem. 1955, 59, 1121.
40. Zatsepina, N. N.; Kane, A. A.; Tupitsyn, I. F. Zhur. Org.
Khim. 1977, 13, 1793.
8. Zwiefka, A.; Kohn, H.; Widger, W. R. Biochemistry 1993,
32, 3564.
9. Zhu, A. Q.; von Hippel, P. H. Biochemistry 1998, 37, 11202.
10. Richardson, J. P. Biochim. Biophys. Acta 2002, 1577, 251.
11. Skordalakes, E.; Berger, J. M. Cell 2003, 114, 135.
12. Xu, Y.; Kohn, H.; Widger, W. R. J. Biol. Chem. 2002, 277,
30023.
41. Bolton, R.; Carter, G. B. M.; Sandall, J. P. B. J. Chem.
Soc. Perkin II 1979, 389.
42. Jain, S. R.; Walker, S. J. Chem. Phys. 1971, 75, 2942.
43. Granzhan, V. A.; Laktionova, S. K. Rus. J. Phys. Chem.
1973, 47, 294.
13. Xu, Y.; Johnson, J.; Kohn, H.; Widger, W. R. J. Biol.
Chem. 2003, 278, 13719.
14. Brogan, A. P.; Widger, W. R.; Kohn, H. J. Org. Chem.
2003, 68, 5575.
15. Filler, R.; Kobayashi, Y.; Yagupolskii, L. M. Organoflu-
orine Compounds in Medicinal Chemistry and Biomedicinal
Applications; Elsevier: Amsterdam, 1993.
44. Cheng, L. T.; Wilson, T.; Stevenson, S. H.; Meredith, G. R.;
Rikken, G.; Marder, S. R. J. Phys. Chem. 1991, 95, 10631.
45. Cheryukanova, G. Y.; Chmutova, G. A.; Vereshchagin,
A. N. Zhur. Fizich. Khim. 1975, 49, 234.
´
46. Jaffe, H. H. Chem. Rev. 1953, 53, 191.

Boon-Saeng Park et al. / Bioorg. Med. Chem. 13 (2006) 1–21
21
47. Marriott, S.; Reynolds, W. F.; Taft, R. W.; Topsom, R. D.
J. Org. Chem. 1984, 49, 959.
59. Parsch, J.; Engels, J. W. J. Am. Chem. Soc. 2002, 124, 5664.
60. Alkorta, I.; Rozas, I.; Elguero, J. J. Fluorine Chem. 2000,
101, 233.
61. Carosati, E.; Sciabola, S.; Cruciani, G. J. Med. Chem.
2004, 47, 5114.
62. Lai, J. S.; Kool, E. T. J. Am. Chem. Soc. 2004, 126, 3040.
63. Schmidt, K. S.; Sigel, R. K. O.; Filippov, D. V.; van der
Marel, G. A.; Lippert, B.; Reedijk, J. New J. Chem. 2000,
24, 195.
48. Inamoto, N.; Masuda, S. Chem. Lett. 1982, 1003.
49. Mullay, J. J. Am. Chem. Soc. 1985, 107, 7271.
50. Mullay, J. J. Am. Chem. Soc. 1984, 106, 5842.
51. Babij, C.; Poe¨, A. J. J. Phys. Org. Chem. 2004, 17, 162.
52. Smyth, C. P. J. Phys. Chem. 1937, 41, 209.
53. Boyd, R. J.; Edgecombe, K. E. J. Am. Chem. Soc. 1988,
110, 4182.
54. Sproul, G. J. Phys. Chem. 1994, 98, 6699.
55. Wells, P. R. Prog. Phys. Org. Chem. 1968, 6, 111.
56. Huheey, J. E. J. Phys. Chem. 1965, 69, 3284.
57. Jeffrey, G. A. An Introduction to Hydrogen Bonding;
Oxford University Press: Oxford, 1997, pp. 95–96.
58. Howard, J. A. K.; Hoy, V. J.; OÕHagan, D.; Smith, G. T.
Tetrahedron 1996, 52, 12613.
64. Brogan, A. P.; Widger, W. R.; Bensadek, D.; Riba-Garcia,
I.; Gaskell, S. J.; Kohn, H. J. Am. Chem. Soc. 2005, 127,
2741.
65. Nehrke, K. W.; Seifried, S. E.; Platt, T. Nucl. Acids Res.
1992, 20, 6107.
66. Magyar, A.; Zhang, X.; Abdi, F.; Kohn, H.; Widger, W.
R. J. Biol. Chem. 1999, 274, 7316.