Identification of borinic esters as inhibitors of bacterial cell growth and bacterial methyltransferases, CcrM and MenH

Article abstract of DOI:10.1021/jm050676a

As bacteria continue to develop resistance toward current antibiotics, we find ourselves in a continual battle to identify new antibacterial agents and targets. We report herein a class of boron-containing compounds termed borinic esters that have broad spectrum antibacterial activity with minimum inhibitory concentrations (MIC) in the low microgram/mL range. These compounds were identified by screening for inhibitors against Caulobacter crescentus CcrM, an essential DNA methyltransferase from Gram negative α-proteobacteria. In addition, we demonstrate that borinic esters inhibit menaquinone methyltransferase in Gram positive bacteria using a new biochemical assay for MenH from Bacillus subtilis. Our data demonstrate the potential for further development of borinic esters as antibacterial agents as well as leads to explore more specific inhibitors against two essential bacterial enzymes.

Full text of DOI:10.1021/jm050676a

7468
J. Med. Chem. 2005, 48, 7468-7476
Identification of Borinic Esters as Inhibitors of Bacterial Cell Growth and
Bacterial Methyltransferases, CcrM and MenH
Stephen J. Benkovic,*,† Stephen J. Baker,†,| M. R. K. Alley,| Youn-Hi Woo,† Yong-Kang Zhang,|
Tsutomu Akama,| Weimin Mao,| Justin Baboval,† P. T. Ravi Rajagopalan,†,§ Mark Wall,†, Lyn Sue Kahng,‡,#
Ali Tavassoli,† and Lucy Shapiro#
Department of Chemistry, The Pennsylvania State University, 414 Wartik Laboratory, University Park, Pennsylvania 16802,
Anacor Pharmaceuticals, Inc., 1060 East Meadow Circle, Palo Alto, California 94303, and Department of Developmental
Biology, Beckman Center, B300, Stanford University School of Medicine, Stanford, California 94305-5329
Received July 15, 2005
As bacteria continue to develop resistance toward current antibiotics, we find ourselves in a
continual battle to identify new antibacterial agents and targets. We report herein a class of
boron-containing compounds termed borinic esters that have broad spectrum antibacterial
activity with minimum inhibitory concentrations (MIC) in the low microgram/mL range. These
compounds were identified by screening for inhibitors against Caulobacter crescentus CcrM,
an essential DNA methyltransferase from Gram negative R-proteobacteria. In addition, we
demonstrate that borinic esters inhibit menaquinone methyltransferase in Gram positive
bacteria using a new biochemical assay for MenH from Bacillus subtilis. Our data demonstrate
the potential for further development of borinic esters as antibacterial agents as well as leads
to explore more specific inhibitors against two essential bacterial enzymes.
Introduction
The need to discover novel small molecules that kill
bacterial cells or prevent their growth, without affecting
the human host has been an ongoing challenge that has
reached critical dimensions as increasing numbers of
pathogens develop antibiotic resistance. Most existing
antibiotics have been derived from natural products and
are thus already associated with naturally occurring
resistance genes in the antibiotic producing microbe. As
resistance has spread, antibiotic potency and effective-
ness has been maintained by continual chemical modi-
fication.
Figure 1. A. Reaction mechanism of the methyl transfer from
We have used a mechanism-based approach to de-
AdoMet to adenine catalyzed by CcrM. B. The multisubstrate
velop inhibitors against the Caulobacter crescentus cell
adduct inhibitor designed to test the extrahelical adenine
cycle regulated methyltransferase, CcrM, an essential
DNA methyltransferase enzyme found in most R-pro-
teobacteria, including the pathogens Brucella abortus
and Agrobacterium tumefaciens.1,2 CcrM plays a central
role in cell cycle progression3,4 and virulence.1 Deletion
of CcrM results in cell death, and overexpression leads
to aberrant cell division.1,2,4-6 CcrM catalyzes the
transfer of a methyl group from S-adenosylmethionine
(AdoMet) to the 6-N-position of adenine in GANTC
sequences in DNA (Figure 1A).7 Structural8,9 studies on
other methyltransferases indicate that a dramatic con-
formational change occurs within the ternary complex
that results in the extrahelical localization of the
binding site hypothesis and that demonstrated selective
inhibition of CcrM over HhaI.13
substrate adenine base in an enzyme active site proxi-
mal to the AdoMet cofactor. Two independent stud-
ies10,11 suggest that after this base-flipping process, the
transfer of the methyl group proceeds by a direct SN2
mechanism from the cofactor AdoMet to the 6-amino
group of the target adenine base. After transfer of the
methyl group, the methylated adenine base flips back
into the DNA helix, followed by the sequential release
of S-adenosylhomocysteine (SAH) and methylated DNA.7
Known inhibitors of CcrM include the natural product
sinefungin, which is an analogue of the cofactor AdoMet;
however, it inhibits all AdoMet requiring enzymes,
including eukaryotic cytosine methyltransferases, thus
making it toxic to mammalian cells and unsuitable as
a therapeutic agent.12
* Corresponding author. Tel: (814)865-2882. Fax: (814)865-2973.
E-mail: sjb1@psu.edu.
† The Pennsylvania State University.
| Anacor Pharmaceuticals Inc.
# Stanford University School of Medicine.
§ Current address: Celera Genomics Inc, 180 Kimball Way, South
San Francisco, CA 94080.
In the absence of a crystal structure of CcrM we
hypothesized the existence of an extrahelical adenine-
binding site located between the DNA and the AdoMet
binding sites. To test this hypothesis, we previously
reported a multisubstrate adduct inhibitor of CcrM,
Current address: Johnson & Johnson Pharmaceutical Research
& Development, 665 Stockton Drive, Exton, PA 19341.
‡ Current address: Section of Digestive and Liver Diseases, 840
South Wood Street (M/C 716), University of Illinois at Chicago,
Chicago, IL 60612.
10.1021/jm050676a CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/26/2005

Borinic Ester Inhibitors of Bacterial Cell Growth
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7469
Scheme 1. Syntheses of Diphenyl Borinic Acid
Quinoline Esters (4)a
Figure 2. Evolution of diphenylborinic acid quinoline ester
library (4). Using the multisubstrate adduct inhibitor (1) as a
lead, a combinatorial library was designed based upon an
adenine scaffold (2). The best CcrM inhibitor from the library
(3) contained a borinic ester group, which was the focus of the
final iteration to give the borinic acid quinoline esters (4).
such as compound 1 (Figure 1B), which was developed
by fusing the methionine moiety of AdoMet to the 6-N-
position of adenosine monophosphate via a methylene
spacer to correctly position the functional groups.13
These compounds were found to selectively inhibit CcrM
over the C-5 cytosine methyltransferase HhaI, suggest-
ing that CcrM contains an extrahelical adenine-binding
site.
a Conditions: (a) BCl3, THF, -78 °C to r.t; (b) EtOH, reflux.
converged to a diphenylborinic acid substituent linked
via an ethylene spacer to the 6-N-position, with various
substituents at the 7- or 9-position. Upon deconvolution
of this set we found the best of these six inhibitors
contained the 2-morpholinoethyl group at the 9-position
(3, Figure 2). In an attempt to improve the physico-
chemical properties of these borinic esters (3) yet retain
the activity of our initial leads, we studied the SAR of
the borinic ester group. Keeping the diphenyl borinic
acid group constant, we replaced the ester group,
containing the adenine nucleus, with other heteroaro-
matic groups. We found that only the 8-hydroxyquino-
line ester series (4)14 provided these improvements and
also enhanced activity against CcrM. Therefore, we
focused our efforts on this quinoline series.
Diphenylborinic acid quinoline esters (4a-f) were
synthesized as shown in Scheme 1. Arylmetal reagents
(5) were treated with 0.5 mol equiv of boron trichloride
at -78 °C in THF. This was stirred overnight at room
temperature and after workup gave the diphenylborinic
acid (6). Without further purification, the diphenyl-
borinic acid (6) was treated with 8-hydroxy-
quinoline derivatives (7a-c) in ethanol at reflux. Upon
cooling, the diphenylborinic acid quinoline ester (4)
usually crystallized from the solution in high purity.
To understand the importance of the boron atom in
these compounds, we synthesized an analogue of com-
pound 4b whereby the boron was replaced with carbon
to give compound 9 (Scheme 2). We also surmised that
the size of substituents attached to the central boron
atom might influence the bioactivities of these com-
pounds. To test this hypothesis, compounds 12a-g
In this report we describe the design of a series of
borinic esters as inhibitors against CcrM. Furthermore,
these borinic esters showed both Gram positive and
Gram negative antibacterial activity with minimum
inhibitory concentrations (MICs) in the low microgram/
mL range. As CcrM is not found in Gram positive
bacteria, we investigated the potential target in these
bacteria and identified MenH, an essential menaquino-
ne methyltransferase. A newly developed biochemical
assay for MenH from Bacillus subtilis and preliminary
structure-activity relationship (SAR) results against
this enzyme demonstrate that the borinic esters are
inhibitors of B. subtilis both in vivo and in vitro.
Chemistry. The success of the multisubstrate inhibi-
tor (1, Figure 1B) against CcrM13 led us to choose
adenine as a scaffold to construct an informed library
of adenine derivatives substituted at the 6-N- and the
9-positions in order to extend the chemical diversity.
Using solution phase combinatorial chemistry, we syn-
thesized a small mixed library (ca. 1000 members,
relative to 1 000 000 members, typical of random librar-
ies) of adenine derivatives substituted at the 6-N-,9- or
6-N-,7-positions (the 6-N-,9-substitution pattern (2) is
shown in Figure 2). The choice of Ri and Rii deliberately
included substituents that would provide a variety of
side chain volumes and geometries. This library was
screened for activity against C. crescentus cell growth13
and in parallel, for direct inhibition of CcrM in vitro.
We identified six leads that were bactericidal toward
C. crescentus at a concentration of 100 µM and also
inhibited CcrM activity completely at this level. All

7470 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Scheme 2. Synthesis of Carbon Analogue (9) of 4ba
Benkovic et al.
Figure 3. Inhibition of DNA methyltransferases, CcrM, Dam,
and HhaI, by diphenylborinic acid quinoline esters (4a-f) and
carbon analogue (9). Concentration of inhibitor was 100 µM.
Assay measured the transfer of [3H-Me] from Adomet to DNA
substrate as a percent, relative to the absence of inhibitor
(100% activity).
tallize, the addition of diethyl ether or diisopropyl ether
to the ethanol would cause precipitation of the product.
a Conditions: (a) SOCl2, CH2Cl2. (b) K2CO3, DMF, 60 °C.
Scheme 3. Syntheses of Vinyl-phenyl Borinic Acid
Quinonline Esters (12)a
Biological Results and Discussion
In Vitro Activity against DNA Methyltrans-
ferases. Compounds 4a-f and 9 were screened in vitro
for activity against CcrM and two other DNA methyl-
transferases, Dam, a bacterial adenine DNA methyl-
transferase and HhaI, a bacterial cytosine methyltrans-
ferase. The results are shown in Figure 3, which shows
the remaining enzyme activity when screened at 100
µM.15 Compounds bearing a chloro group on the borinic
acid moiety (4b, 4c, 4e, and 4f) showed potent inhibitory
activity against CcrM, whereas compounds 4a and 4d
did not show significant inhibition. Furthermore, com-
pounds 4c, 4e, and 4f showed some selectivity for the
adenine methyltransferases, CcrM and Dam, over the
cytosine methyltransferase, HhaI. The carbon analogue
(9) showed no activity against these enzymes, demon-
strating that boron is important for the enzyme inhibi-
tion.
Inhibitor Kinetics of CcrM. Steady-state kinetic
studies utilizing CcrM and a hemimethylated DNA
substrate were conducted to assess the nature and
degree of inhibition by 4e in the presence of variable
DNA concentrations and at saturating AdoMet levels.15
Standard reciprocal plots of rate versus DNA concentra-
tion showed an uncompetitive inhibition pattern. In the
presence of variable AdoMet concentrations and at
saturating DNA levels, the effect of 4e on CcrM activity
was more complex. From the standard reciprocal plots,
the observed inhibition pattern was uncompetitive at
low concentration of inhibitor and noncompetitive at
higher levels. A global fit to the data was achieved with
a model in which the inhibitor initially binds to the
binary CcrM•DNA complex followed by a second mol-
ecule of inhibitor competing with AdoMet binding at the
active site. The calculated inhibition constants Ki’s were
9 µM (uncompetitive inhibition) and 30 µM (mixed
noncompetitive inhibition).15 The inhibitors did not
appear to bind to the free enzyme.
Microbiological Activity of the Borinic Esters.
We tested the borinic esters (4 and 12) for microbiologi-
cal activity against both Gram negative and Gram
positive bacteria. The results of this study are shown
in Table 1. We found that most of the compounds
inhibited the growth of Gram negative bacteria in the
low µg/mL range except for Yersinia pseudotuberculosis.
Francisella tularensis was most sensitive to diaryl
a Conditions: (a) vinylMgBr, THF, -78 °C to r.t.; (b) EtOH,
reflux.
(Scheme 3) in which one of the two aryl groups is re-
placed with a less sterically hindered vinyl group were
designed. As shown in Scheme 3, aryl boronic acid ethyl-
ene glycol esters (10) were reacted with vinylmagnesium
bromide at -78 °C under anhydrous conditions and al-
lowed to warm to room temperature to yield the asym-
metrical vinyl-aryl borinic acids (11) after hydrolysis
with HCl solution. As a final step, the borinic acid (11)
was treated with 8-hydroxyquinoline (7a) in ethanol at
reflux to furnish the final product (12), usually as a
crystalline solid upon cooling. If the product did not crys-

Borinic Ester Inhibitors of Bacterial Cell Growth
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7471
Table 1. Minimum Inhibitory Concentration MIC (µG/mL) of Borinic Acid Quinoline Esters (4a-c,e,f and 12a-g), and the Carbon
Analogue 9 against Gram Positive and Gram Negative Bacteria
structure
Gram positive
Gram negative
S.
Y.
S.
epider-
E.
B.
subtilis
M.
tuber-
culosis
pseudo-
H.
aureus midis faecalis
E.
tuberculosis influenzae
ATCC ATCC ATCC faecium ATCC
B.
C.
M.
F.
ATCC
29833
ATCC
4976
Riii
4a 4-F
4b 4-Cl
4c 3-Cl
4e 4-Cl
4f 4-Cl
Rvi
29213 12228 29212 CT-26 23857 anthracis H37Rv crescentus catarrhalis tularensis
H
4
2
2
32
4
32
4
0.62
0.31
0.62
12.5
4
2
2
2
4
8
8
H
2
2
2
6.25
6.25
0.03
0.01
6
64
32
H
1
2
8
32
4
2-Me
5-Cl
16
0.5
32
32
>64
64
64
16
2
2
2.6
0.25
>64
32
>64
4
4
4
5.3
8
0.62
>5
0.62
0.31
0.62
e0.125
0.01
8
>64
16
0.25
>64
4
2
2
16
2
9
carbon analogue >64
>64
4
12a 3-Cl
12b 2-Cl
12c 4-Cl
12d H
12e 3-F
12f 3-Cl, 4-F
12g 3-CN
H
H
H
H
H
H
H
2
2
2
2
2
2
1
1
1
1
1
1
1
1
4
1
1
2
1
2
2
2
16
32
>64
>64
>64
32
4
32
4
32
2
2
4
4
8
8
0.31
0.62
0.62
16
8
32
2
32
4
2
8
2
32
Table 2. Relative Increase in Sensitivity of the B. subtilis
menH Strain to Borinic Acid Quinoline Esters (4b,c and
12a-g)
borinic esters with sub µg/mL activity. As neither CcrM
nor Dam is present in Gram positive bacteria, we did
not expect this series of compounds to provide broad
spectrum antibacterial activity. Yet these compounds
had equivalent activity against these pathogens, par-
ticularly compound 4f, which showed MIC values of
below 1 µg/mL against three out of seven Gram positive
bacteria (Table 1). The two compounds that were most
selective for CcrM in vitro 4c and 4e showed different
specificities: compound 4c exhibited activity against
both Gram positive and Gram negative bacteria, but
compound 4e was active only against Gram negative
bacteria. Interestingly, only compounds 4b and 12g
showed activity against Enterococcus faecalis. Activity
against F. tularensis was greatly reduced when a vinyl
group replaced one aromatic ring. The carbon analogue
(9) was also tested but, as expected, showed no activity.
The activity of the borinic esters is comparable to
clinically used antibiotics such as erythromycin (MIC90
B. anthracis 1.5 µg/mL),16 gentamycin (MIC90 B. an-
thracis 0.38 µg/mL16 MIC90 F. tularensis 1 µg/mL),17 and
streptomycin (MIC90 F. tularensis 4 µg/mL).17 We have
a full ADME-Tox profile study (plasma exposure, tissue
distribution, serum binding, etc.) for these compounds
underway. The data will be subject for a separate
manuscript.
Mechanism of Action Study in Gram Positive
Bacteria. As a subset of the borinic esters showed
activity against Gram positive bacteria, which do not
contain CcrM, we set out to investigate their mechanism
of action. To identify putative target enzymes, we
attempted to isolate mutants of B. subtilis that were
resistant to compounds 4b and 4c. However, even
though chemical mutagenesis was used, we were unable
to obtain resistant mutants. Therefore, as the initial
concept of the research targeted a methyltransferase,
we tested the genes encoding the essential methyltrans-
ferases in B. subtilis, which are menH, trmD, trmU,
cspR, ydiO, and ydiP. These genes were identified by
the B. subtilis functional genomics project.18 Strains
were constructed that placed each of the essential
methyltransferase genes under the control of a chro-
Rv
increase in sensitivity
4b
4-Cl
none
16×
64×
10×
none
none
16×
64×
24×
4c
3-Cl
3-Cl
2-Cl
4-Cl
H
12a
12b
12c
12d
12e
12f
12g
3-F
3-Cl, 4-F
3-CN
targeted any of these methyltransferases, the minimal
inhibitory concentrations should decrease as IPTG
concentrations were reduced. The MICs were deter-
mined for compounds 4b and 4c at different concentra-
tions of IPTG that were empirically determined to give
equivalent growth rates in the absence of inhibitor for
each of the strains. The trmD, trmU, cspR, ydiO, and
ydiP strains showed no significant changes of sensitivi-
ties for 4b and 4c (data not shown). The only strain that
showed significant change of sensitivity with altered
methyltransferase gene transcription was the menH
strain. In B. subtilis, the menH gene, which encodes the
essential menaquinone methyltransferase, is upstream
of hepT (a gene required for the synthesis of the
isoprenoid tail prior to its attachment to the menaquino-
ne ring); therefore, the Pspac promoter will regulate both
menH and hepT genes. Thus, either menH or genes
upstream in the menaquinone biosynthetic pathway
might be the target for these compounds.
Additional compounds 12a-g were selected to test
their SAR against the menH strain (Table 2). The strain
was most sensitive to the 3-chlorophenyl vinyl borinic
acid quinoline ester (12a). When the chloro group was
removed, giving 12d, the sensitivity was lost entirely,
indicating an important role for this chloro group.
Moving the chloro group from the 3-position to the
2-position, giving 12b, also reduced sensitivity from 64-
fold to 10-fold. However, all sensitivity was lost again
when the chloro group was moved to the 4-position,
giving 12c. These data suggested the optimum position
for the chloro group was meta to the boron. Replacing
the chloro group with fluoro, giving 12e, or with a cyano
group, giving 12g, also resulted in reduced sensitivity
again confirming the importance of the chloro group.
mosomally integrated copy of the isopropyl â-D-thioga-
19
lactoside (IPTG) inducible promoter Pspac
.
As the
concentration of IPTG was reduced, less protein was
expressed. We reasoned that if the borinic esters

7472 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Benkovic et al.
Scheme 4. Methyl Transfer Reaction by MenH in
Menaquinone Biosynthesisa
a Conditions: MenH, AdoMet; n ) 3-9 or larger.
Finally, addition of a fluoro group para to boron and
ortho to the 3-chloro group, giving 12f, retained equal
sensitivity to that of compound 12a. Since the menH
strain may affect transcription of both menH and hepT,
either gene products could be the target for these borinic
esters. To determine if these compounds acted on MenH
directly, we set up a biochemical assay for menaquinone
methyltransferase.
In Vitro Activity against MenH. MenH is a meth-
yltransferase that catalyzes the final step in the bio-
synthesis of menaquinone (MK). It transfers a methyl
group from AdoMet onto the aromatic ring of the
substrate demethylmenaquinone (DMMK) (Scheme 4).
MK is involved in respiratory and photosynthetic elec-
tron-transport employing membrane-bound protein com-
plexes.20 In B. subtilis, MK is essential for successful
endospore formation and germination21 and is involved
in regulation of cytochrome formation.22,23 Most Gram
positive bacteria contain menaquinone as the sole
quinone whereas most aerobically grown Gram negative
bacteria have ubiquinone (Q) as the sole quinone. Those
Gram negative bacteria, such as Klebsiella aerogenes
and E. coli, that can be grown anaerobically, contain
menaquinone.
Figure 4. HPLC traces of lipid extracts from (A) E. coli JM109
without B. subtilis menH insert (B) E. coli JM109 with B.
subtilis MenH overexpressed (C) E. coli JM109 with B. subtilis
MenH overexpressed and inhibitor (SAH) showing the levels
of Q8 (ubiquinone with eight prenyl units) and MK8 (menaquino-
ne with eight prenyl units) as a result of ubiquinone methyl-
transferase activity or menaquinone methyltransferase activ-
ity.
MenH is an attractive target for a novel class of
antibiotics because it has no direct or functional homo-
logues in mammalian cells, unlike the ubiquinone
methyltransferases. Most in vitro assays of menaquino-
ne methyltransferase reported to date use whole cell
extracts22,24 with a few exceptions.25-27
Figure 5. Inhibition of MenH by natural inhibitors SAH and
sinefungin, borinic esters 4b,c and 12a, and carbon analogue
9. Concentration of inhibitor was 100 µM. Assay measured the
transfer of [3H]-Me from AdoMet to menaquinone as a percent,
relative to the absence of inhibitor (100% activity).
Following the method of Ogura et al.22 we used whole
cell lysates of E. coli JM109 containing the overex-
pressed B. subtilis MenH protein (described in Experi-
mental Methods) and 1,4-dihydroxy-2-naphthoic acid as
the substrate. After the reaction, MK was extracted
from the cell lysate using a mixture of pentane and
2-propanol. The organic extract from each reaction was
collected, the products were separated by HPLC and
identified by MS. Using 1,4-dihydroxy-2-naphthoic acid
as the substrate, wild-type E. coli (containing no B.
subtilis menH insert) predominantly yielded Q8 (ubi-
quonone with eight prenyl subunits). In the E coli strain
containing the B. subtilis menH insert, both Q8 and MK8
(MK with eight prenyl subunits) were produced (Figure
4B). Production of MK8 instead of MK7, which is found
in B. subtilis, suggests that the prenyl units were added
according to the E. coli system, as MK8 is the predomi-
nate quinone in anaerobically grown E. coli. When the
inhibitor S-adenosylhomocysteine (SAH) was added, the
amount of MK8 was dramatically reduced (Figure 4C).
This demonstrated that the MenH was expressed and
active in E. coli whole cell lysates. Thus, this strain
would allow us to monitor the inhibition of the MenH
menaquinone methyltransferase.
we used [3H-Me]-AdoMet to monitor the transfer of the
methyl group to menaquinone by liquid scintillation
counting of the organic extraction. To determine the
effect of activity due to the co-extracted E. coli ubiquino-
ne, E. coli lysates without the menH insert were used
as a control. The methyltransferase inhibitors sinefun-
gin and SAH were also tested in this assay revealing
that SAH had greater inhibitory activity than sinefun-
gin (Figure 5). Although sinefungin is usually found to
be a better methyltransferase inhibitor than SAH, this
was found not the case for methyltransferases from
Streptomyces spp.28
Compounds 4b, 4c, 9, and 12a were tested for in vitro
inhibition of MenH (Figure 5). Although compound 4b
showed good inhibition activity against the CcrM DNA
methyltransferase in vitro, it showed no significant
inhibition of the B. subtilis MenH methyltransferase.
Compound 12a showed more than 50% inhibition, which
was comparable to the activity of sinefungin. Not
surprisingly, the carbon analogue (9) was inactive in
this in vitro assay, which we attribute to the unique
structural, conformational, and physical organic char-
acter that only the boron atom can bring to this class of
compounds.
To screen the inhibitory activity of our borinic esters
in E. coli JM109 with menH insert as described above,

Borinic Ester Inhibitors of Bacterial Cell Growth
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7473
(30 µL, 0.5 M solution in DMF). To each row, one different
alkyl halide (15 µL, 1 M solution in DMF) was added and a
lid was placed on top of the microtiter plate. The plate was
heated at 45 °C for 24 h to give a regioisomeric mixture of 7-
and 9-substituted chloropurines. After this time, one to three
different alkylamines (5 µL, 1 M solution in DMF) were added
to each column and a lid was placed on top of the plate. This
plate was heated at 90 °C for 24 h to furnish a mixed library
of 6-N-,9- and 6-N-,7-substituted adenine products. The plate
was cooled to room temperature, and the solution in each well
was removed from the solid potassium salts. The final con-
centration of each reaction was adjusted to approximately 17
mM in DMF, assuming quantitative yield, and these solutions
were tested for activity without further purification. We used
HPLC and LC-MS to analyze a cross-section of this library.
We determined that the first reactions proceeded with a 7-:
9-substitution ratio of around 2:3, and the overall reaction
proceeded with yields varying between 50 and 95%.
General Method for the Synthesis of Diarylborinic
Acids (4). Trichloroborane or tribromoborane (1 mol equiv)
was added to either tetrahydrofuran (0.2 M) or diethyl ether
(0.2 M) under argon and cooled to -78 °C. The arylmagnesium
bromide or aryllithium (2 mol equiv), in tetrahydrofuran,
diethyl ether, cyclohexane, toluene, or mixtures of these
solvents, was added dropwise to the cold reaction. The reaction
was allowed to warm to room temperature and stirred over-
night. The solvents were removed in vacuo, and the residue
was dissolved in diethyl ether. The reaction was stirred rapidly
and hydrolyzed by the slow addition of 1 N hydrochloric acid.
Stirring was discontinued, the layers were separated, and the
organic layer was washed with saturated aqueous NaCl. The
organic layer was dried (MgSO4) and filtered, and the solvent
was removed in vacuo to give the crude product as an oil. This
oil was dissolved in ethanol to an estimated concentration of
1 M and divided into portions to be used in the subsequent
precipitation with the various complexing agents.
Thus, MenH appears to be a target of the borinic ester
12a, which showed the greatest sensitivity against the
B. subtilis menH strain (Table 2). However, this does
not exclude the possibility that 12a may also target the
hepT gene product or another enzyme in the pathway.
Indeed, the differences between the MIC data (Table 1)
and in vitro sensitivity data (Table 2) and our inability
to obtain resistant mutant suggest that MenH may not
be the sole target of these borinic esters.
Conclusion
In this report, we showed the evolution of borinic acid
quinoline esters designed to target the essential enzyme,
CcrM, a DNA methyltransferase from Gram negative
R-proteobacteria, to its application as a broad spectrum
antibiotic inhibiting Gram positive bacteria containing
the essential enzyme menaquinone methyltransferase,
MenH. We showed these boron-containing compounds
have broad-spectrum activity against several pathogenic
strains of both Gram positive and Gram negative
bacteria and that the boron is essential for its antibac-
terial activity. In an attempt to identify a target enzyme
in Gram positive bacteria, we cloned, overexpressed, and
developed an assay for B. subtilis MenH. These borinic
esters could be used as leads for the development of
inhibitors against bacteria cell growth or more specif-
ically against the essential methyltransferase enzymes,
such as CcrM and MenH.
Experimental Methods
1H NMR spectra were recorded on a Bruker Avance 400 (400
MHz), Bruker AMX360 (360 MHz), and Varian Oxford 300
(300 MHz). MALDI mass spectra were obtained using a
Perspective Biosystems Voyager-DE STR, FAB mass spectra
were obtained using a Kratos Analytical MS-50 TC. APCI mass
spectra were recorded on a Perspective Biosystems Mariner.
Microanalyses were recorded by Atlantic Microlab Inc., P.O.
Box 2288, Norcross, GA 30091. Analytical thin-layer chroma-
tography (TLC) were performed with Whatman silica gel
aluminum backed plates of thickness 250 µm and fluorescent
at 254 nm, and by using the solvent systems indicated. Flash
column chromatography was performed with Selecto Scientific
silica gel, 32-64 µm particle size. Melting points were obtained
using a Mel-Temp II melting point apparatus with a Fluke
K1 K/J type thermocouple digital thermometer and are uncor-
rected. Purity by HPLC was determined using a betabasic-18
4.6 mm × 15 cm column from Keystone Scientific Inc. with a
linear gradient of 0 to 40% acetonitrile in 10 mM triethylam-
monium acetate over 20 min. Chemicals were purchased from
Acros Organics and Aldrich Chemical Co. and were used
without further purification. Unmethylated lambda phage
DNA and litmus 29 isolated from the E. coli strain DH5R were
purchased from New England Biolabs (Beverly, MA). Reverse
phase HPLC (Waters Corporation, pump 600, PDA detector
996) was performed with a C18 column (Nova-pak, 3.9 × 150
mm) for product identification. B. subtilis strain 168 for cloning
menH was acquired from American Type Culture Collection.
Pfu DNA polymerase was purchased from Promega (Madison,
WI) and restriction enzymes were purchased from New
England Biolabs and used as recommended by manufacturers.
E. coli JM109 strain was acquired free of charge from New
England Biolabs. Incorporation of [3H-Me]-AdoMet was moni-
tored by Beckman LS 6800 for liquid scintillation counting.
General Procedure for the 6-N-,9-Substituted Adenine
Combinatorial Library. A combinatorial library was pre-
pared in a two-step, one-pot procedure by solution phase using
a 96-well glass microtiter plate. Three plates were prepared
using the following methodology. Potassium carbonate (7.5 mg)
was dispensed into each well of a 96-well microtiter plate. DMF
(140 µL) was added to each well followed by 6-chloropurine
Bis(4-fluorophenyl)borinic Acid 8-Hydroxyquinoline
Ester (4a). Bis(4-fluorophenyl)borinic acid was prepared using
the method above and was treated with 8-hydroxyquinoline
(0.5 M solution in ethanol). A yellow precipitate formed
immediately and was collected by filtration and washed with
ethanol; mp 166-167 °C; 1H NMR (360 MHz, CDCl3,): δ 8.53
(d, J ) 5.1 Hz, 1H), 8.45 (d, J ) 8.2 Hz, 1H), 7.70 (dd, J ) 8.2,
7.7 Hz, 1H), 7.66 (dd, J ) 8.2, 5.1 Hz, 1H), 7.39 (dd, J ) 8.8,
6.7 Hz, 4H), 7.30 (d, J ) 8.2 Hz, 1H), 7.20 (d, J ) 7.7 Hz, 1H),
6.97 (t, J ) 8.8 Hz, 4H); MS (+ve APCI) m/z 345 ([M + H]+,
10B), 346 ([M + H]+, 11B), 368 ([M + Na]+, 11B); Anal. (C21H14-
NOBF2) C, H, N.
Bis(4-chlorophenyl)borinic Acid 8-Hydroxyquinoline
Ester (4b). Bis(4-chlorophenylborinic acid was prepared using
the method above and was treated with 8-hydroxyquinoline
(0.5 M solution in ethanol). A yellow precipitate formed
immediately and was collected by filtration and washed with
ethanol; mp 192-194 °C; 1H NMR (360 MHz, CDCl3,): δ 8.49
(dd, J ) 4.6, 1.0 Hz, 1H), 8.43 (dd, J ) 8.2, 1.0 Hz, 1H), 7.70
(dd, J ) 8.5, 7.7 Hz, 1H), 7.66 (dd, J ) 8.2, 4.6 Hz, 1H), 7.31
(d, J ) 8.2 Hz, 4H), 7.26 (d, J ) 8.5 Hz, 1H), 7.21 (d, J ) 8.2
Hz, 4H), 7.17 (d, J ) 7.7 Hz, 1H); MS (+ve APCI) m/z 377 ([M
+ H]+, 10B, 35Cl, 35Cl), 378 ([M + H]+, 11B, 35Cl, 35Cl), 379 ([M
+ H]+, 10B, 35Cl, 37Cl), 380 ([M + H]+, 11B, 35Cl, 37Cl), 381 ([M
+ H]+, 10B, 37Cl, 37Cl), 382 ([M + H]+, 11B, 37Cl, 37Cl); Anal.
(C21H14NOBCl2) C, H, N.
Bis(3-chlorophenyl)borinic Acid 8-Hydroxyquinoline
Ester (4c). Bis(3-chlorophenyl)borinic acid was formed using
the method above and was treated with 8-hydroxyquinoline
(1.0 M in ethanol). The solvent was removed in vacuo, and
the residue was purified by flash chromatography on silica gel
using CH2Cl2/hexane (1:1) to elute. The solvent was evaporated
and the product crystallized upon the addition of ethanol. The
solid was collected by filtration and washed with cold ethanol
to yield the title product as a yellow solid; mp 144-145 °C; 1H
NMR (360 MHz, CD3OD): δ 8.83 (d, J ) 5.0 Hz, 1H), 8.63 (d,
J ) 8.2 Hz, 1H), 7.78 (dd, J ) 8.6, 5.0 Hz, 1H), 7.67 (t, J )
8.2, 1H), 7.36 (d, J ) 8.2, 1H), 7.26-7.09 (m, 9H); MS (+ve

7474 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Benkovic et al.
ESI) m/z 377 ([M + H]+, 10B, 35Cl, 35Cl), 378 ([M + H]+, 11B,
35Cl, 35Cl), 379 ([M + H]+, 10B, 35Cl, 37Cl), 380 ([M + H]+, 11B,
35Cl, 37Cl), 381 ([M + H]+, 10B, 37Cl, 37Cl), 382 ([M + H]+, 11B,
37Cl, 37Cl); Anal. (C21H14NOBCl2) C, H, N.
was rotary evaporated. The residue was dissolved in ethyl
acetate (150 mL), washed with brine (2 × 100 mL), dried, and
evaporated. The crude solid (2.98 g) of (3-chlorophenyl)-
vinylborinic acid was mixed with 8-hydroxyquinoline (2.0 g,
13.8 mmol) in ethanol (75 mL). The mixture was stirred for
15 min and ethanol was removed by rotary evaporation. The
residue was purified by flash column chromatography over
silica gel eluted with a mixed solvent of hexane and ethyl
acetate (4:1, v/v) giving the title compound as yellow crystals
Bis(4-methoxyphenyl)borinic Acid 8-Hydroxyquino-
line Ester (4d). Bis(4-methoxyphenyl)borinic acid was formed
using the method above and was treated with 8-hydroxyquino-
line (0.5 M in ethanol) causing the title product to precipitate
from the solution. The solid was collected by filtration and
washed with cold ethanol to yield the product as a yellow solid;
1
(1.15 g, 25%): mp 145-147 °C; H NMR (300 MHz, DMSO-
1
d6): δ 5.21 (dd, 1H), 5.44 (dd, 1H), 6.40 (dd, 1H), 7.12-7.26
(m, 3H), 7.30-7.42 (m, 3H), 7.68 (t, 1H), 7.88 (dd, 1H), 8.74
(dd, 1H), 8.80 (d, 1H); MS (ESI+) m/z 294 (M + H)+; Anal.
(C17H13BClNO) C, H, N.
(2-Chlorophenyl) Vinyl Borinic Acid 8-Hydroxyquino-
line Ester (12b). This was prepared from 2-chlorophenylbo-
ronic acid using a method similar to that described for the
synthesis of 12a to give the product in 49% yield as yellow
crystals: mp 95-97 °C; 1H NMR (300 MHz, DMSO-d6): δ 5.12
(dd, 1H), 5.40 (dd, 1H), 6.54 (dd, 1H), 7.08 (d, 1H), 7.16-7.20
(m, 3H), 7.38 (d, 1H), 7.50-7.55 (m, 1H), 7.65 (t, 1H), 7.84
(dd, 1H), 8.75 (d, 1H), 8.80 (d, 1H); MS (ESI+) m/z 294 (M +
H)+; Anal. (C17H13BClNO) C, H, N.
mp 222-224 °C; H NMR (400 MHz, CDCl3): δ 8.53 (d, J )
5.0 Hz, 1H), 8.38 (d, J ) 8.5 Hz, 1H), 7.67 (t, J ) 8.0 Hz, 1H),
7.61 (dd, J ) 8.3, 5.0 Hz, 1H), 7.38 (d, J ) 8.5 Hz, 4H), 7.24
(d, J ) 8.2 Hz, 1H), 7.17 (d, J ) 7.7 Hz, 1H), 6.85 (d, J ) 8.6
Hz, 4H), 3.78 (s, 6H); MS (+ve MALDI, CHCA) m/z 369 ([M +
H]+, 10B), 370 ([M + H]+, 11B); Anal. (C23H20BNO3) C, H, N.
Bis(4-chlorophenyl)borinic Acid 8-Hydroxyquinaldine
Ester (4e). Bis(4-chlorophenyl)borinic acid was formed using
the method above and was treated with 8-hydroxyquinaldine
(0.5 M solution in ethanol). The product was collected by
filtration and washed with ethanol; mp 155-156 °C; 1H NMR
(360 MHz, CDCl3): δ 8.21 (d, J ) 8.6 Hz, 1H), 7.46 (t, J ) 8.4
Hz, 1H), 7.27 (d, J ) 8.2 Hz, 1H), 7.18-7.11 (m, 9H), 6.97 (d,
J ) 7.8 Hz, 1H) 2.38 (s, 3H); MS (+ve, APCI) m/z 392 (M+,
11B, 35Cl, 35Cl), 394 (M+, 11B, 35Cl, 37Cl), 396 (M+, 11B, 37Cl, 37Cl);
Anal. (C22H16NBOCl2) C, H, N.
(4-Chlorophenyl) Vinyl Borinic Acid 8-Hydroxyquino-
line Ester (12c). This was prepared from 4-chlorophenylbo-
ronic acid using a method similar to that described for the
synthesis of 12a giving the product in 14% yield as yellow
Bis(4-chlorophenyl)borinic Acid 5-Chloro-8-hydroxy-
quinoline Ester (4f). Bis(4-chlorophenyl)borinic acid was
formed using the method above and was treated with 5-chloro-
8-hydroxyquinoline (0.5 M solution in ethanol). The product
was collected by filtration and washed with ethanol; mp 154-
1
crystals: mp 119-121 °C; H NMR (300 MHz, DMSO-d6): δ
5.20 (dd, 1H), 5.44 (dd, 1H), 6.42 (dd, 1H), 7.13 (d, 1H), 7.22-
7.28 (m, 2H), 7.36-7.44 (m, 3H), 7.68 (t, 1H), 7.88 (dd, 1H),
8.75 (d, 1H), 8.92 (d, 1H); MS (ESI+) m/z 294 (M + H)+; Anal.
(C17H13BClNO‚0.25H2O) C, H, N.
1
156 °C; H NMR (360 MHz, CDCl3): δ 8.56 (d, J ) Hz, 1H),
Phenyl Vinyl Borinic Acid 8-Hydroxyquinoline Ester
(12d). This was prepared from phenylboronic acid using a
method similar to that described for the synthesis of 12a giving
the product in 21% yield as yellow crystals: 1H NMR (300
MHz, DMSO-d6): δ 5.20 (dd, 1H), 5.44 (dd, 1H), 6.42 (dd, 1H),
7.0-7.5 (m, 7H), 7.68 (t, 1H), 7.86 (dd, 1H), 8.73 (d, 1H), 8.92
(d, 1H); MS (ESI+) m/z 260 (M + H)+; Anal. (C17H14BNO) C,
H, N.
8.44 (d, J ) Hz, 1H), 7.64 (m, 1H), 7.57 (d, J ) Hz, 1H), 7.14
(m, 9 H), 6.98 (d, 1H); MS (+ve, ESI) m/z 412 ([M + H]+, 10B,
35Cl, 35Cl, 35Cl), 413 ([M + H]+, 11B, 35Cl, 35Cl, 35Cl); Anal.
(C21H13NBOCl2) C, H, N.
4,4′-Dichlorobenzhydryl Chloride (8b). To a solution of
4,4-dichlorobenzhydrol (2.00 g, 7.94 mmol) in dichloromethane
(20 mL) was added thionyl chloride (2.0 mL) dropwise at 0
°C, and the mixture was stirred at room temperature for
overnight. The solvent was removed under reduced pressure,
and the residue was extracted with ethyl acetate. The organic
layer was washed with water and dried on anhydrous MgSO4.
The solvent was removed under reduced pressure to afford the
desired product (2.24 g, quant), which was used for the next
step without further purification.
8-[Bis(4-chlorophenyl)methoxy]quinoline (9). To a so-
lution of compound 8b (1.04 g, 3.83 mmol) and 8-hydroxy-
quinoline (500 mg, 3.44 mmol) in DMF (5 mL) was added K2-
CO3 (529 mg, 3.83 mmol), and the mixture was stirred 60 °C
for overnight. The solvent was removed under reduced pres-
sure, and the residue was extracted with ethyl acetate. The
organic layer was washed with water (three times) and dried
on anhydrous MgSO4. The solvent was removed under reduced
pressure, and the residue was purified by silica gel column
chromatography (5:1 hexane/ethyl acetate) to give the desired
product (800 mg, 61%) as a white powder; mp 106-107 °C;
1H NMR (300 MHz, DMSO-d6,): δ 6.89 (s, 1H), 7.17 (d, J )
6.7 Hz, 1H), 7.37 (d, J ) 8.2 Hz, 1H), 7.42 (d, J ) 8.5 Hz, 4H),
7.47 (t, J ) 7.3 Hz, 1H), 7.56 (dd, J ) 8.2, 4.1 Hz, 1H), 7.59 (d,
J ) 8.5 Hz, 4H), 8.30 (dd, J ) 8.2, 1.8 Hz, 1H), 8.94 (dd, J )
4.1, 1.8 Hz, 1H); MS (+ve APCI) m/z 380 ([M + H]+, 35Cl, 35Cl),
382 ([M + H]+, 35Cl, 37Cl), 384 ([M + H]+, 37Cl, 37Cl). Anal.
(C22H15NOCl2) C, H, N.
(3-Chlorophenyl) Vinyl Borinic Acid 8-Hydroxyquino-
line Ester (12a). To a THF solution (50 mL) of 3-chlorophe-
nylboronic acid ethylene glycol ester (10) (2.89 g, 15.8 mmol),
which was prepared by refluxing the mixture of 3-chlorophe-
nylboronic acid and ethylene glycol (1:1 molar ratio) in toluene
under nitrogen for 3 h, was added a vinylmagnesium bromide
THF solution (1.0 M, 17.5 mL, 17.5 mmol) at -78 °C under
nitrogen. The reaction mixture was stirred for 2 h from - 78
°C to room temperature and for 1 h with a water bath.
Hydrochloric acid (6 M, 6 mL) was added, and the mixture
(3-Fluorophenyl) Vinyl Borinic Acid 8-Hydroxyquino-
line Ester (12e). This was prepared from 3-fluorophenylbo-
ronic acid using a method similar to that described for the
synthesis of 12a giving the product in 40% yield as yellow
1
crystals: mp 111-114 °C; H NMR (300 MHz, DMSO-d6): δ
5.21 (dd, 1H), 5.44 (dd, 1H), 6.42 (dd, 1H), 6.94 (tt, 1H), 7.10-
7.28 (m, 4H), 7.40 (d, 1H), 7.67 (t, 1H), 7.87 (dd, 1H), 8.74 (d,
1H), 8.97 (d, 1H); MS (ESI+) m/z 278 (M + H)+; Anal. (C17H13-
BFNO) C, H, N.
(3-Chloro-4-fluorophenyl) Vinyl Borinic Acid 8-Hy-
droxyquinoline Ester (12f). This was prepared from 3-chloro-
4-fluorophenylboronic acid using a method similar to that
described for the synthesis of 12a giving the product in 48%
1
yield as yellow crystals: mp 86-88 °C; H NMR (300 MHz,
DMSO-d6): δ 5.20 (dd, 1H), 5.46 (dd, 1H), 6.40 (dd, 1H), 7.14
(d, 1H), 7.20 (t, 1H), 7.33-7.49 (m, 3H), 7.68 (t, 1H), 7.88 (dd,
1H), 8.74 (d, 1H), 8.98 (d, 1H); MS (ESI+) m/z 312 (M + H)+;
Anal. (C17H12BClFNO) C, H, N.
(3-Cyanophenyl) Vinyl Borinic Acid 8-Hydroxyquino-
line Ester (12g). This was prepared from 3-cyanophenylbo-
ronic acid using a method similar to that described for the
synthesis of 12a giving the product in 53% yield as yellow
1
crystals: mp 115-117 °C; H NMR (300 MHz, DMSO-d6): δ
5.23 (dd, 1H), 5.47 (dd, 1H), 6.42 (dd, 1H), 7.17 (d, 1H), 7.38-
7.45 (m, 2H), 7.60-7.80 (m, 4H), 7.89 (dd, 1H), 8.76 (d, 1H),
9.60 (d, 1H); MS (ESI+) m/z 285 (M + H)+; Anal. (C18H13BN2O)
C, H, N.
CcrM Inhibition Assay. Methyltransferase activity was
measured by monitoring the incorporation of [3H]-Me from [3H]-
AdoMet into DNA. A stock solution containing 250 nM CcrM,
5 µM N645/50 mer, 150 mM potassium acetate, 5 mM
2-mercaptoethanol in pH 7.5 HEPES buffer was prepared.
Aliquots were placed in eppendorf tubes and inhibitors were
added from concentrated stock solutions (16.7 mM in DMF or

Borinic Ester Inhibitors of Bacterial Cell Growth
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23 7475
microdilution antibacterial susceptibility tests, with the ad-
dition of 40 µg/mL spectinomycin and either 1 mM IPTG or
30 µM IPTG. The sensitivity data was determined by dividing
the MIC for the menH strain in MHB media containing 40
µg/mL spectinomycin and 1 mM IPTG by the MIC in MHB
media containing 40 µg/mL spectinomycin and 30 µM IPTG.
30 µM IPTG was determined empirically as the concentration
that allowed cells to grow where their A600 end points were
not significantly different from cells grown in 1 mM IPTG.
Furthermore the control compounds gentamycin and 8-hy-
droxyquinoline showed no change in MIC.
Cloning B. subtilis menH. Boiled cells of B. subtilis strain
168 were used as a template for PCR. The PCR conditions were
30 cycles of denaturation at 94 °C for 30 s, annealing at 55 °C
for 30 s, and extension at 72 °C for 2 min, followed by a final
extension at 72 °C for 15 min using Pfu DNA polymerase. The
primers used were menH forward (5′-GCATGAATTCCATAT-
GCAGGACTCAAAAGAACAGCGCG-3′) and menH reverse
(5′-GCATGTGGATTCTTTCCATCCGATATGCGTGGCAGCG-
3′). The PCR product was cut with NdeI and BamHI restriction
enzymes and cloned into pMAL-c2x (New England Biolabs)
vector before transformed into E. coli JM109.
Expression and Isolation of MenH. E. coli JM109 cells
with menH were grown in LB media at 37 °C to OD600 ) 0.6-
0.8 and induced with IPTG (final concentration 1 mM). The
cells were collected by centrifugation after 3 h, resuspended
in a solution of 20 mM Tris-HCl, pH 7.5 with 1 mM EDTA
and 10 mM BME, disrupted with a sonicator, and stored at
-80 °C until they were used in an assay. The level of MenH
overexpression was monitored by 12% SDS-PAGE electro-
phoresis.
MenH Assay. The inhibitor candidates (final concentration
of 100 µM) were added to 500 µL of solution containing 200
µL of disrupted E. coli cells from previous procedure, 0.1 M
Tris-HCl buffer pH 8.0, 10 mM MgCl2, 10 mM DTT, 1 µM FPP,
1 µM 1,4-dihydroxy-2-naphthoic acid (in EtOH:ether 2:1, v/v),
1 µM IPP, and 20 µM unlabeled AdoMet mixed with [3H]-
labeled AdoMet (3.01 × 103 dpm/reaction). The samples were
incubated at 37 °C for 1 h with occasional rocking. The
reactions were quenched by adding 500 µL of 0.1 M acetic acid
in methanol. The expected menaquinone product was extracted
with hexane/2-propanol (3:2, v/v, 3 × 3 mL), and the level of
[3H]-Me incorporation was monitored by a liquid scintillation
counter. The assays were performed in triplicate.22
Chart 1
DMSO) to reach the appropriate final concentrations (500 or
100 µM) in a total of 15 µL. Reactions were initiated by the
addition of [3H]AdoMet at a final concentration of 50 µM. After
40 min at 30 °C, 5 µL aliquots were removed from the reaction
and spotted onto DE81 anion exchange filter circles. The filters
were allowed to dry and then washed with 3 × 200 mL of 0.3
M ammonium formate to remove unreacted [3H]AdoMet fol-
lowed by 2 × 200 mL 95% ethanol wash and finally a 200 mL
ether wash. The filters were allowed to air-dry and counted
by standard liquid scintillation techniques. Control reactions
in the absence of inhibitors were used to determine the extent
of inhibition. High throughput screening was carried out
similarly in Tris-HCl buffer (50 µM, pH 7.5) with plasmid DNA
as substrate Litmus 29 (New England Biolabs): 250 µM DNA,
3 µM sites) 100 µM inhibitor candidate and enzyme, potassium
acetate, and 2-mercaptoethanol concentrations as described
above. Assays were initiated with [14CH3]AdoMet (50 µM, 34
Ci/mol) in a volume of 10 µL in a PCR plate and incubated for
30 min at 30 °C. Four microliter aliquots were then spotted
on DE81 paper with a multichannel pipet and washed and
dried as described above. Data were collected with a Molecular
Dynamics model 425S phosphorimager and analyzed with the
spotfinder utility in ImageQuant 3.3.
For the in-depth steady-state kinetic inhibition study the
DNA concentration was varied from 0.5 to 5 times KM of 1 µM,
4e levels ranged 0 to 75 µM, and AdoMet was at 50 µM. From
reciprocal plots of the rates vs DNA levels. Ki ≈ 11.2 ( 0.2
µM for 4e.
The AdoMet concentration was varied from 1 to 10 times
KM of 6 µM, 4e levels ranged from 0 to 75 µM and DNA was
at 5 µM. From reciprocal plots Ki ≈ 9.7 ( 2.3 µM (uncompeti-
tive) and Ki ≈ 30.2 ( 6.7 µM (noncompetitive) for 4e.
DNA Substrate [N645/50 mer] (Chart 1). Synthesis of
DNA was acheived on an Expedite BioSystems DNA synthe-
sizer and purified as previously described.29
Determination of Minimum Inhibitory Concentration
(MIC) against Caulobacter cresentus and Other Bacte-
ria. Liquid cultures of C. crescentus were grown to mid-log
phase in PYE media30 at 30 °C and diluted for use in broth
macrodilution assays with the different compounds.31 The
standard inoculum used was 105 cfu/mL, and visible growth
was assayed after 18 h. In all other bacteria the MICs were
determined according to the NCCLS guidelines.
MenH Product Identification. The extract from each
reaction above was concentrated to dryness and dissolved into
100 µL of acetonitrile (AN). The samples were injected into
HPLC reverse phase column (C18) and eluted with a gradient
(70% AN/H2O to 100% AN) over 30 min at 1 mL/min flow rate.
The eluants were monitored by following absorbance at 220
and 270 nm followed by MS analysis: Q8 (APCI(-), calculated
for C49H74O4 726.56 observed M+ [726.42], M + 2 [728.47]);MK8
(APCI(-), calculated for C51H72O2 716.5532, observed M+
[716.4231]); demethyl-MK, (DMMK8, APCI(-), calculated for
C50H70O2 702.5376, observed M+ [702.3421]).
Construction of the menH Strain. The spectinomycin
resistance gene was amplified from pPE3032 with the primers
GAGGAATTCTAAAAAATTTAGAAGCCAATGAAATCT GAG-
GAATTCTAAAATCTGATTACCAATTAGAATGAA and the re-
sulting PCR amplification product was digested with Eco RI
and cloned into the Mun I digested plasmid pMUTIN4.33 The
spectinomycin derivative pMUTIN4 was designated pMUTINS1.
The ribosomal binding site, start codon, and 331 base pairs of
menH coding sequence were amplified from the B. subtilis
genome using the following primers TCAAAGCTTGGAA-
GAAGGGTAAACCATTATG and CATGAATTCAAGGAAGCTC-
CATCGCATTTCC and cloned into the Hind III and Eco RI
sites in pMUTINS1. The plasmid bearing this insert was
confirmed by DNA sequence and designated pMENS1. The Eco
RI primer was designed so it would introduce an in-frame stop
codon into the menH coding region once integrated into the
B. subtilis genome and therefore placing the only functional
copy of menH under Pspac control. The plasmid pMENS1 was
introduced into B. subtilis,34 and transformants were selected
on low salt LB agar containing 150 µg/mL spectinomycin and
1 mM IPTG. Transformants were tested for IPTG growth
dependence, and the pMENS1 insertion was checked using the
following primers AAAGGGGGATGTGCTGCAAGGCGATT
and TTTGGGGACAAGGGTGATATTTTTGC.
Acknowledgment. This work was supported by
Defense Advanced Research Project Agency Research
Grant MDA972-97-1-0008 and by NIH Grant
R01GM51426 to L.S. We acknowledge Dr. Anders
Sjo¨stedt who obtained the data for the inhibition studies
with F. tularensis and Dr. Theresa M. Koehler who
obtained the data for B. anthracis. We would like to
thank Dr. Kirk R. Maples and Dr. Ving J. Lee for helpful
discussions and support. We also acknowledge Maureen
Kully, Lydia Kohut, and Jehangir Khan at NEAJA
Pharmaceuticals, Inc., for obtaining MIC data on other
strains of pathogenic bacteria.
Determination of Minimum Inhibitory Concentration
(MIC) of IPTG Conditional B. subtilis. MIC assays were
performed as described by the NCCLS protocol M7-A5 for
Supporting Information Available: Complete elemental
analyses for compounds 4a-f, 9, and 12a-g. This material is
available free of charge via the Internet at http://pubs.acs.org.

7476 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 23
Benkovic et al.
wood: C. R.; Hecker, M.; Hosoya, D.; Hullo, M. F.; Kakeshita,
H.; Karamata, D.; Kasahara, Y.; Kawamura, F.; Koga, K.; Koski,
P.; Kuwana, R.; Imamura, D.; Ishimaru, M.; Ishikawa, S.; Ishio,
I.; Le Coq, D.; Masson, A.; Maue¨l, C.; Meima, R.; Mellado, P.;
Moir, A.; Moriya, S.; Nagakawa, E.; Nanamiya, H.; Nakai, S.;
Nygaard, P.; Ogura, M.; Ohanan, T.; O’Reilly, M.; O’Rourke, M.;
Pragai, Z.; Pooley, H. M.; Rapoport, G.; Rawlins, J. P.; Rivas, L.
A.; Rivolta, C.; Sadaie, A.; Sadaie, Y.; Sarvas, M.; Sato, T.; Saxild,
H. H.; Scanlan, E.; Schumann, W.; Seegers, J. F. M. L.;
Sekiguchi, J.; Sekowska, A.; Se´ror, S. J.; Simon, M., Stragier,
P.; Studer, R.; Takamatsu, H.; Tanaka, T.; Takeuchi, M.;
Thomaides, H. B.; Vagner, V.; Dijl, J. M. van; Watabe, K.; Wipat,
A.; Yamamoto, H.; Yamamoto, M.; Yamamoto, Y.; Yamane, K.;
Yata, K.; Yoshida, K.; Yoshikawa, H.; Zuber, U.; Ogasawara, N.
Essential Bacillus subtilis genes. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 4678-4683.
References
(1) Robertson, G. T.; Reisenauer, A.; Wright, R.; Jensen, R. B.;
Jensen, A.; Shapiro, L.; Martin Roop, R., II. The Brucella abortus
CcrM DNA Methyltransferase Is Essential for Viability, and Its
Overexpression Attenuates Intracellular Replication in Murine
Macrophages. J. Bacteriol. 2000, 182, 3482-3489.
(2) Kahng, L. S.; Shapiro, L. The CcrM DNA Methyltransferase of
Agrobacterium tumefaciens Is Essential, and Its Activity Is Cell
Cycle Regulated. J. Bacteriol. 2001, 183, 3065-3075.
(3) Reisenauer, A.; Kahng, L. S.; McCollum, S.; Shapiro, L. Bacterial
DNA Methylation: a Cell Cycle Regulator. J. Bacteriol. 1999,
181, 5135-5139.
(4) Zweiger, G.; Marczynski, G.; Shapiro, L. A Caulobacter DNA
Methyltransferase that Functions only in the Predivisional Cell.
J. Mol. Biol. 1994, 235, 472-485.
(5) Stephens, C.; Reisenauer, A.; Wright, R.; Shapiro, L. A cell cycle-
regulated bacterial DNA methyltransferase is essential for
viability. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 1210-1214.
(6) Wright, R.; Stephens, C.; Shapiro, L. The CcrM DNA methyl-
transferase is widespread in the alpha subdivision of proteo-
bacteria, and its essential functions are conserved in Rhizobium
meliloti and Caulobacter crescentus. J. Bacteriol. 1997, 279,
5869-5877.
(7) Berdis, A. J.; Lee, I.; Coward, J. K.; Stephens, C.; Wright, R.;
Shapiro, L.; Benkovic, S. J. A cell cycle-regulated adenine DNA
methyltransferase from Caulobacter crescentus processively
methylates GANTC sites on hemimethylated DNA. Proc. Natl.
Acad. Sci. U.S.A. 1998, 95, 2874-2879.
(8) Klimasauskas, S.; Kumar, S.; Roberts, R. J.; Cheng, X. Hhal
methyltransferase flips its target base out of the DNA helix. Cell.
1994, 76, 357-369.
(9) Pues, H.; Bleimling, N.; Holz, B.; Wo¨lcke, J.; Weinhold, E.
Functional Roles of the ConservedAromatic Amino Acid Residues
at Position 108 (Motif IV) and Position 196 (Motif VIII) in Base
Flipping and Catalysis by the N6-Adenine DNA Methyltrans-
ferase from Thermus aquaticus. Biochemistry 1999, 38, 1426-
1434.
(10) Pogolotti, A. L.; Ono, A.; Subramaniam, R.; Santi, D. V. On the
mechanism of DNA-adenine methylase. J. Biol. Chem. 1988, 263,
7461-7464.
(11) Ho, D. K.; Wu, J. C.; Santi, D. D. V.; Floss, H. G. Stereochemical
studies of the C-methylation of deoxycytidine catalyzed by Hha
I methylase and the N-methylation of deoxyadenosine catalyzed
by EcoRI methylase. Arch. Biochem. Biophys. 1991, 284, 264-
269.
(12) Ghosh, A. K.; Liu, W. Total Synthesis of (+)-Sinefungin. J. Org.
Chem. 1996, 61, 6175-6182.
(13) Wahnon, D. C.; Shier, V. K.; Benkovic, S. J. Mechanism-Based
Inhibition of an Essential Bacterial Adenine DNA Methyltrans-
ferase: Rationally Designed Antibiotics. J. Am. Chem. Soc. 2001,
123, 976.
(14) Although drawn as the acyclic isomer, such compounds can exist
in equilibrium with five-membered cyclic adducts where boron
forms an adaptive bond with the 6-amino group of adenosine.
The actual geometry of such cyclic adducts is exquisitely
sensitive to the strength of the B-N bond. For a more detailed
investigation into the tetrahedral characteristic of the B-N
bond, see: H. Ho¨pfl The tetrahedral character of the boron atom
newly definedsa useful tool to evaluate the NfB bond. J.
Organomet. Chem. 1999, 581, 129-149.
(15) Shier, V. K. Thesis. Pennsylvania State University, 2001.
(16) Turnbull, P. C. B.; Sirianni, N. M.; LeBron, C. I.; Samaan, M.
N.; Sutton, F. N.; Reyes, A. E.; Peruski, L. F., Jr. MICs of
Selected Antibiotics for Bacillus anthracis, Bacillus cereus,
Bacillus thuringiensis, and Bacillus mycoides from a Range of
Clinical and Environmental Sources as Determined by the Etest.
J. Clin. Microbiol. 2004, 42, 3626-3634.
(17) Ikaheimo, I.; Syrjala, H.; Karhukorpi, J.; Schildt, R.; Koskela,
M. In vitro antibiotic susceptibility of Francisella tularensis
isolated from humans and animals. J. Antimicrob. Chemother.
2000, 46, 287-290.
(19) Yansura, D. G.; D. J. Henner. Use of the Escherichia coli lac
repressor and operator to control gene expression in Bacillus
subtilis. Proc. Natl. Acad. Sci. U.S.A. 1984, 81, 439-443.
(20) Søballe, B.; Poole, R. K. Microbial ubiquinones: multiple roles
in respiration, gene regulation and oxidative stress management.
Microbiology 1999, 145, 1817-1830.
(21) Leatherbarrow, A. J.; Yazdi, M. A.; Curson, J. P.; Moir, A. The
gerC locus of Bacillus subtilis, required for menaquinone bio-
synthesis, is concerned only indirectly with spore germination.
Microbiology 1998, 144, 2125-2130.
(22) Koike-Takeshita, A.; Koyama, T.; Ogura, K. Identification of a
Novel Gene Cluster Participating in Menaquinone (Vitamin K2)
Biosynthesis. Cloning And Sequence Determination Of The
2-Heptaprenyl-1,4-Naphthoquinone Methyltransferase Gene Of
Bacillus Stearothermophilus. J. Biol. Chem. 1997, 272, 12380-
12383.
(23) Farrand, S. K.; Taber, H. W. Changes in Menaquinone Concen-
tration during Growth and early Sporulation in Bacilus subtilis.
J. Bacteriol. 1974, 117, 324-326.
(24) Daves, G. D., Jr.; Moore, H. W.; Schwab, D. E.; Olsen, R. K.;
Wilczynski, J. J.; Folkers, K. Synthesis of 2-Multiprenylphenols
and 2-multiprenyl-6-methoxyphenols, biosynthetic precursors of
the Ubiquinones. J. Org. Chem. 1967, 32, 1414-1417.
(25) Azerad, R.; Samuel, O. C-me´thylation des desme´thylme´naquino-
nes: II. Spe´cificite´ du syste`me enzymatique de me´thylation de
Mycobacterium phlei vis-a`-vis du substrat quinonique. Biochimie
1972, 54, 305-317.
(26) Hsu, A. Y.; Poon, W. W.; Shepherd, J. N.; Myles, D. C.; Clarke,
C. F. Complementation of coq3 mutant Yeast by Mitochondrial
Targeting of the Escherichia coli UbiG Plypeptide: Evidence
That UbiG Catalyzes both O-Methylation Steps in Ubiquinone
Biosynthesis. Biochemistry 1996, 35, 9797-9806.
(27) Barkovich, R. J.; Shtanko, A.; Shepherd, J. N.; Lee, P. T.; Myles,
D. C.; Tzagoloff, A.; Clarke, C. F. Characterization of the COQ5
Gene from Saccharomyces cerevisiae Evidence For A C-Methyl-
transferase In Ubiquinone Biosynthesis. J. Biol. Chem. 1997,
272, 9182-9188.
(28) Barbe´s, C.; Sa´nches, J.; Yebra, M. J.; Robert-Gero´, M.; Hardisson,
C. Effects of sinfungin and S-adenosylhomocysteine an DNA and
protein methyltransferases from Streptomyces and other bacte-
ria. FEMS Microbiology Lett. 1990, 69, 239-244.
(29) Capson, T. L.; Peliska, J. A.; Kaboord, B. F.; Frey, M. W.; Lively,
C.; Dahlberg, M.; Benkovic, S. J. Biochemistry 1992, 31, 10984-
10994.
(30) Ely, B. Genetics of Caulobacter crescentus. Methods Enzymol.
1991, 304, 372-384.
(31) Amsterdam, D. Susceptibility testing of antimicrobials in liquid
media. In Antibiotics in laboratory medicine, 4th ed.; Lorian, V.,
Ed.; Williams and Wilkins: Baltimore, 1996; pp 52-111.
(32) Britton, R. A.; Eichenberger, P.; Gonzalez-Pastor, J. E.; Fawcett,
P.; Monson, R.; Losick, R.; Grossman, A. D. Genome-Wide
Analysis of the Stationary-Phase Sigma Factor (Sigma-H) Regu-
lon of Bacillus subtilis. J. Bacteriol. 2002, 184, 4881-4890.
(33) Vagner, V.; Dervyn, E.; Ehrlich, S. D. A vector for systematic
gene inactivation in Bacillus subtilis. Microbiology 1998, 144,
3097-3104.
(18) Kobayashi, K.; Ehrlich, S. D.; Albertini, A.; Amati, G.; Andersen,
K. K.; Arnaud, M.; Asai, K.; Ashikaga, S.; Aymerich, S.; Bes-
sieres, P.; Boland, F.; Brignell, S. C.; Bron, S.; Bunai, K.;
Chapuis, J.; Christiansen, L. C.; Danchin, A.; De´barbouille´, M.;
Dervyn, E.; Deuerling, E.; Devine, K.; Devine, S. K.; Dreesen,
O.; Errington, J.; Fillinger, S.; Foster, S. J.; Fujita, Y.; Galizzi,
A.; Gardan, R.; Eschevins, C.; Fukushima, T.; Haga, K.; Har-
(34) Kunst, F.; Rapoport, G.; Salt stress is an environmental signal
affecting degradative enzyme synthesis in Bacillus subtilis. J.
Bacteriol. 1995, 177, 2403-2407.
JM050676A