DNA binding ligands with in vivo efficacy in murine models of bacterial infection: Optimization of internal aromatic amino acids

Article abstract of DOI:10.1016/j.bmcl.2004.02.047

DNA binding ligands with potent antimicrobial activity against Gram-positive bacteria were further optimized by variation of the internal aromatic amino acids. This modification led to compounds with improved in vivo efficacy in lethal murine models of peritonitis (methicillin-resistant S. aureus, MRSA) and lung infection (S. pneumoniae).

Full text of DOI:10.1016/j.bmcl.2004.02.047

Bioorganic & Medicinal Chemistry Letters 14 (2004) 2067–2072
DNA binding ligands with in vivo efficacy in murine models of
bacterial infection: optimization of internal aromatic amino acids
€
Roland W. Burli, Jacob A. Kaizerman, Jian-Xin Duan, Peter Jones, Kirk W. Johnson,
Mari Iwamoto, Kiet Truong, Wenhao Hu, Timothy Stanton, Alfred Chen, Sofia Touami,
*
MatthewGross, Vernon Jiang, Yigong Ge and Heinz E. Moser
Genesoft Pharmaceuticals, 7300 Shoreline Court, South San Francisco, CA 94080, USA
Received 13 December 2003; revised 10 February 2004; accepted 12 February 2004
Abstract—DNA binding ligands with potent antimicrobial activity against Gram-positive bacteria were further optimized by
variation of the internal aromatic amino acids. This modification led to compounds with improved in vivo efficacy in lethal murine
models of peritonitis (methicillin-resistant S. aureus, MRSA) and lung infection (S. pneumoniae).
ꢀ 2004 Elsevier Ltd. All rights reserved.
We have identified a novel class of antimicrobial com-
pounds with high activity against drug resistant, Gram-
positive bacteria. These molecules have been shown to
exercise bactericidal activity by interacting with func-
tionally important A/T-rich sites of bacterial DNA and
thereby inhibiting DNA replication and RNA tran-
scription.1 Previous lead optimization to improve anti-
bacterial spectrum, potency, tolerability, and in vivo
efficacy was focused on C-terminal amines and N-ter-
minal hetero-aromatic groups2;3 and resulted in well-
tolerated compounds with broad and potent activity
against Gram-positive bacteria. One of these prototypic
antibiotics, 1, was used to demonstrate the in vivo proof
of concept in a murine lethal peritonitis model using
methicillin-sensitive S. aureus (MSSA).3 However, the
same compound failed in this model using MRSA. The
increasing threat by multi-drug-resistant bacteria,
MRSA in particular,4 indicates the importance of this
milestone. This article focuses on the optimization of the
internal, aromatic amino acids using 1 as the starting
point with the goal to achieve in vivo efficacy in murine
models of MRSA infection while retaining antibacterial
spectrum and potency and in vivo tolerability. We chose
1, not the most potent compound, for our initial inves-
tigations based on the large available dataset. In addi-
tion, we anticipated that the influence of subtle
structural modifications on the biological parameters
would be easily detected with a moderately potent lead
structure.
Previous optimization studies have identified 4-(2-amino-
ethyl)morpholine and 2-carboxy-3-chlorothiophene as
valuable end caps for this class of DNA-binding anti-
biotics.3 In order to better understand the importance of
internal N-methylpyrrole-2-carboxamide units (Py), we
synthesized a small library of 1 analogs, in which a
single Py unit is replaced by various aromatic amino
acids in all three positions (Table 1, Scheme 1). In the
context of larger DNA minor-groove binding ligands,
referred to as polyamides, substitution of a Py by des-
methyl pyrrole (Ds) has led to compounds with similar
binding affinity, but reduced sequence specificity.5
Hence, demethylation of a Py unit was expected to only
minimally affect DNA interaction, but alter other
physicochemical parameters such as lipophilicity and
molecular weight. Each replacement (2–4) resulted in
loss of potency, as evidenced most dramatically for Ds
in position 3 (4). The novel building block N-isoprop-
ylpyrrole6 (iPrPy) was investigated in order to study the
effect of increased steric bulk on solubility and/or
aggregation; we anticipated a positive influence on these
parameters due to steric interference on the stacking
properties of these molecules (5–7). However, the
aqueous solubility remained low. Nevertheless, 6 and 7
retained good antibacterial potency and 7 was further
evaluated in vivo. The compound passed the acute tol-
erability screen but failed to protect infected mice in the
Keywords: Antibiotics; Lead optimization; MRSA; DNA ligands.
* Corresponding author. Tel.: +1-650-837-1820; fax: +1-650-827-0476;
e-mail: hmoser@genesoft.com
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.02.047

2068
R. W. B€urli et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2067–2072
Table 1. Antimicrobial activity of single aromatic substitutions in analogs of 1
Cl
H
O
N
H
H
N
N
H
N
S
N
BB1
O
BB3
BB2
O
O
O
Entry
BB1
Py
BB2
BB3
Py
MRSAa 27660
2–4
VSEFa 29212
0.5–1
PISPa 49619
0.13–1
1
Py
2
3
4
Ds
Py
Py
Py
Ds
Py
Py
Py
Ds
8
2
2
8
16
2–8
32
0.5–2
8
5
6
7
iPrPy
Py
Py
Py
Py
iPrPy
4
1
1
4
1
1
2
iPrPy
Py
0.5
0.25
Py
8
Im
Py
Py
Py
Im
Py
Py
Py
Im
2–4
0.13–1
>32
>32
0.03–0.13
16
>32
9
10
>32b
>32
11
12
13
Pz
Py
Py
Py
Pz
Py
Py
Py
Pz
8
8
0.25
1–32
>32
32
>32
>32
>32
14
15
16
mBz
Py
Py
Py
Py
mBz
>32
>32
16
>32
>32
16
>32
>32
8
mBz
Py
Py
17
18
19
pBz
Py
Py
Py
Py
pBz
32
8
16
2
32
1
pBz
Py
Py
>32
>32
>32
a MIC values against ATCC strains are given in lg/mL and were measured according to standard guidelines.25 MRSA, methicillin-resistant S. aureus
VREF, vancomycin-resistant E. faecalis VSEF, vancomycin-susceptible E. faecalis PISP, penicillin-intermediate S. pneumoniae. Ranges of values
are given for multiple determinations.
b Determined with MSSA (ATCC 29213).
H
N
H
N
H
N
H
N
H
N
N
O
N
O
O
O
O
O
O
N
N
N
N
N
H
iPrPy
Py
Ds
Im
Pz
H
H
N
N
H
N
H
N
H
N
O
O
N
iOxa
S
2Th
O
O
β
mBz
pBz
Ala
Scheme 1. Structures and abbreviations of internal aromatic amino acids. The compounds were prepared either in solution3 or on solid support23
using a thiol resin.24 The building blocks were incorporated as N-Boc protected derivatives. Py,23 Im,23 Ds,5 Pz,11 and 2Th14 have been previously
described; mBz,14 pBz,14 and bAla are commercially available and iOxa was prepared as described.16 All compounds were characterized by 1H NMR
and mass spectrometry and showed purity of at least 90% by analytical reversed phase HPLC. The isolated yields after HPLC purification ranged
from 10–30%.3
peritonitis model (MSSA, 50 mg/kg). Similar results
were obtained with the N-isobutyl pyrrole derivative, an
analog of the previously described isoamylpyrrole7 (data
not shown). N-Methyl-imidazole-2-carboxamide (Im)
has been studied extensively in polyamides; when paired
with Py in a side-by-side mode, it specifically recognizes
a G/C base pair.8 In a one to one binding mode, how-
ever, Im has been shown to interact with A/T-rich
sequences albeit lacking the specificity of Py.9;10 Inter-
estingly, the in vitro potency of Im-containing com-
pounds is highly dependent on the position: as
compared to 1, an Im in positions 2 or 3 led to complete
loss of activity (9, 10), while the analogous modification
in position 1 improved potency (8). Unfortunately, 8

R. W. B€urli et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2067–2072
2069
was not tolerated well in the acute murine screen as
demonstrated by 11% weight loss of animals after three
days (single dose at 50 mg/kg). A similar effect was
observed for the pyrazole derivatives (Pz; 11–13).11 In
this case, even the most potent derivative 11 revealed a
decrease of antibacterial activity as compared to the
parent compound 1.
weight, and solubility.3 The parent compound contain-
ing three internal Py units (20) demonstrated improved
in vitro potency as compared to the 3-chlorothiophene
derivative 1, but failed to protect mice in the MSSA
peritonitis model at 20 mg/kg. The Ds substitution in
position 2 (22) yielded a derivative with equal potency
against MRSA and improved in vivo efficacy in the
murine peritonitis model (MSSA: 60% protection at
10 mg/kg, MRSA: 100% at 50 mg/kg, no protection at
20 mg/kg).15 The compound was safe in mice (single dose
at 50 mg/kg), but indicated poorer tolerability in the
multiple dose evaluation (four daily doses of 30 mg/kg)
with loss of body weight (5%), elevated BUN (blood
urea nitrogen), AST (aspartate amino transferase), ALT
(alanine amino transferase), and triglycerides (all
roughly 2-fold increased) and 2–3 fold decreased glucose
levels.
Little investigation has been reported for the replace-
ment of internal five-membered ring systems by their
higher homologues in Py-containing polyamides.12–14
Six-membered aromatic amino acids have a different
geometry and consequently are expected to affect
DNA interaction by influencing the curvature of the
crescent-shaped ligands. However, derivatization and
lead optimization of six-membered rings would be
straightforward and was considered potentially benefi-
cial. Replacement of Py by meta-aminobenzoic acid
(mBz; 14–16) resulted in almost complete loss of activity,
irrespective of the position of the newly introduced ring.
The isomeric series with para-aminobenzoic acid (pBz)
revealed an unexpected result; while substitution in
positions 1 and 3 (17, 19) yielded compounds with lim-
ited or no potency, replacement in position 2 (18)
retained most activity (2–4-fold less than 1). In sum-
mary, five-membered ring analogs of pyrrole appear to
be better tolerated in positions 1 and to a lesser extent 2,
whereas para-substituted six-membered rings are best
placed in position 2.
In contrast to the findings with the 3-chlorothiophene
derivatives, Im-substitution in position 1 (23) did not
improve antibacterial potency in the isoquinoline series.
This indicates that optimized groups in various positions
are not simply additive, but have to be evaluated in the
context of the entire molecule. Replacement of Py by Pz
supported this statement: in contrast to the 3-chloro-
thiophene series, substitution in position 2 (25) resulted
in a compound with good in vitro activity. Unfortu-
nately, 25 did not protect mice in the peritonitis model
(MSSA, 20 mg/kg) and displayed mild lethargy in ani-
mals when tested for acute tolerability (50 mg/kg). Based
on antibacterial potency, the isooxazole16 (iOxa) in po-
sition 1 (26) looked promising. However, the poor sol-
ubility of this compound did not allowformulation for
an in vivo experiment. The last heterocycle evaluated
was 4-amino-2-carboxy-thiophene (2Th).14 When intro-
duced in position 1 (27), it resulted in moderately good
Next, we intended to identify the optimal five-membered
hetero-aromatic groups in positions 1 and 2 (Table 2).
Rather than repeating part of the work done with 3-
chlorothiophene, we used the isoquinoline N-terminus.
This group proved very useful in earlier studies with
respect to antibacterial potency, tolerability, molecular
Table 2. Antimicrobial activity of various five-membered heteroaromatic groups in positions 1 and 2
O
H
N
H
H
N
N
H
N
N
N
BB1
O
BB2
O
O
N
O
Entry
BB1
Py
BB2
Py
MRSAa 27660
0.5
VSEFa 29212
0.25
PISPa 49619
0.06
20
21
22
Ds
Py
Py
Ds
1–2
0.25–1
0.5
0.5–2
0.5–1
1–2
23
Im
Py
1
1
0.25
24
25
Pz
Py
Py
Pz
1
0.5
8
1
0.5
0.25
26
iOxa
Py
0.06
0.5
0.13
27
28
2Th
Py
Py
2Th
1–2b
8
0.25–0.5
>32
0.13–0.25
0.5
29
bAla
Py
32
16
8
See legend to Table 1 for details.

2070
R. W. B€urli et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2067–2072
potency, even though the S. aureus activity did not meet
our requirements. Among multiple non-aromatic sub-
stituents, b-alanine (bAla) has previously been proven a
valuable module for DNA-binding polyamides9 and
indeed performed best compared to a variety of non-
aromatic replacements (data not shown). However, 29
still revealed a dramatic loss of antibacterial potency
indicating that the use of non-aromatic internal building
blocks is not a promising avenue for advanced lead
optimization.
excellent antibacterial potency. Attempts to better
understand this unique combination of the N-terminal
isoquinoline and Bz at position 2 revealed that pyr-
idines, pyrimidines, or pyrazines (34–36) were not equal
substitutes for the isoquinoline. The position of the ring
nitrogen and connection point appeared to be crucial for
activity (37, 38). Even the dioxolane analog, a derivati-
zation that previously has led to compounds with im-
proved properties in the context of benzothiophenes,3
did not enhance potency (39).
p
p
Finally, we studied compounds containing Bz at posi-
Even though compound 33 appears to kill bacteria by
interacting with A/T-rich DNA target sites (Kd ¼
7.5 nM15;22) as demonstrated with earlier prototype
molecules,1–3 its unique properties cannot be explained
in a rational way. We observed a general correlation
with exceptions between DNA-interaction and antibac-
terial activity, however, this correlation has expectedly
tion 2 and various N-terminal caps that have led to
highly potent molecules in the Py3 series (Table 3).3
Interestingly, neither 3-chloro-benzothiophene (30), its
des-chloro-dioxolane derivative (31),17 nor 2,4-difluoro-
benzene (32) yielded compounds with good activity
against MRSA, whereas the isoquinoline 33 showed
Table 3. Antimicrobial activity of pBz containing compounds (position 2) with various end-caps
H
H
N
N-Cap
N
H
H
N
O
N
N
O
N
O
O
N
O
Entry
N-Cap
MRSAa 27660
>32
VSEFa 29212
16
PISPa 49619
0.25
Cl
30
S
O
O
31
32
32
32
>32
8
0.5
0.5
S
F
F
33
34
35
36
37
0.06–0.13
>32
0.25–0.5
>32
32
0.03–0.06
N
0.5
0.5
0.13
2
N
>32
N
N
N
0.5
2
N
N
>32
>32
38
39
1
2
0.25c
0.5
N
O
0.06
0.5
N
O
Linezolid
Vancomycin
2
1
2
1
1
0.25
See legend to Table 1 for details.
c Determined with S. pneumoniae ATCC 51422.

R. W. B€urli et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2067–2072
2071
limitations due to the comparison between a molecular
Control
10 mg/kg
60 mg/kg
interaction and a whole cell assay. Whereas we never
characterized compounds in this series that had anti-
bacterial activity but did not bind A/T-rich DNA, we
found a fewcompounds iwth opposite properties
(strong DNA interaction with little or no antibacterial
activity). These observed differences originate presum-
ably in varying properties of membrane transport; a
hypothesis that still needs to be validated experi-
mentally. We also assumed a beneficial influence of
isoquinoline on the physico-chemical properties of cor-
responding compounds. Even though isoquinoline
derivatives appeared generally to have slightly higher
solubility as compared to other bicyclic moieties, aque-
ous solubility of 33 at pH 7.4 was still poor (<1 ng/mL,
data not shown) and protein binding remained high at
94% within the same range as measured for other rep-
resentative compounds of this class.15
30 mg/kg
10
8
6
4
2
0
0
1
2
3
4
5
days
Figure 2. Dose dependence of 33 in the mouse pneumococcal pneu-
monia model. Murine lung infection model with S. pneumoniae (ATCC
6303, penicillin sensitive):27 The organism was grown overnight in
€
Muller–Hinton broth (BD Biosciences, Sparks, MD) containing 5%
lysed horse blood to a concentration of ꢀ5 · 106 cfu/mL. Mice (ICR)
were anesthetized with isoflourane (Baxter, Deerfield, IL) and 50 lL/
mouse of the inoculum above was instilled via the intranasal route.
Compound or vehicle was administered IV at 1 and 5 h post-inocul-
ation and survival was monitored over 5 days.
Based on its promising profile, 33 was examined in a
recently isolated strain of vancomycin-resistant S. aur-
eus (VRSA) and, not surprisingly, found to be active
(MIC ¼ 0.25 lg/mL).21
In the murine lethal peritonitis model (MRSA infection,
Fig. 1), 33 showed remarkable efficacy with an ED50 of
roughly 11 mg/kg (IV administration) even though it is
still inferior to vancomycin (ED50 varies between 1.0 and
4.4 mg/kg).15 The compound was also efficacious in a
lethal murine model of lung infection by S. pneumoniae
(Fig. 2). Compound 33 was the first one to demonstrate
in vivo efficacy in these tough murine models at dose
levels that are well tolerated.15
Control
5 mg/kg
9 mg/kg
12.5 mg/kg
25 mg/kg
Safety profiling was undertaken with 33 to assess its
potential as candidate for therapeutic development.
Acute and short-term multi-day toxicity studies in mice
were encouraging in that no adverse effects were
observed at dose levels (30 mg/kg) above efficacious
doses.15 Importantly, the extensive evaluation of the
potential for genotoxic properties was negative, indi-
cating that reversible DNA binding does not necessarily
result in undesirable genetic effects on mammalian cells.
100
80
60
40
20
0
In conclusion, optimization of internal building blocks
combined with optimized terminal groups yielded 33,
the first DNA-binding ligand with potent Gram-positive
antibacterial potency and in vivo efficacy against MRSA
at safe doses in relevant murine models of infection.
Further optimization of this promising scaffold has been
initiated and will be reported shortly.
0
1
2
3
4
5
days
Figure 1. Dose dependence of 33 in the mouse peritonitis model
(MRSA). The mouse peritionitis model26 was performed as previously
described.15 Fresh colonies of ATCC 27660 (MRSA, b-lactamaseþ)
were grown in Brain Heart Infusion broth (BHI, International Bio-
Products, Bothell, WA) at 37 ꢁC with gentle agitation and adjusted to a
concentration of 2 · 107 cfu/mL. A solution of 10% gastric mucin
porcine (ICN Biomedicals Inc., Aurora, OH) was added at an equal
volume to achieve the final inoculum of 1.0 · 107 cfu/mL in 5% mucin,
representing a 5–10· LD50 inoculum with 100% mortality rate in
vehicle treated animals by 48 h. Groups of 5 ICR female mice (ꢀ25 g
from Taconic Farms) were inoculated intraperitoneally with 0.5 mL of
the inoculum described above. Groups were treated intravenously with
vehicle, positive control (Vancomycin, Linezolid) or test articles with a
single dose 1 h post infection. Clinical observations were monitored
over 5 days with a primary endpoint of survival. For all in vivo studies
described herein, test antibacterials were formulated in 40% hydroxy-
propyl-b-cyclodextrin/10% D-mannitol/50 mM NaOAc buffer pH 4.5
0.1 to achieve concentrations necessary for in vivo profiling. Number
of animals represented in the graph above: control (vehicle alone; 27),
5 mg/kg (5), 9 mg/kg (6), 12.5 mg/kg (9), 25 mg/kg (24).
Acknowledgements
The authors would like to thank Steve Ley and Heinz
Gschwend for helpful discussions during this project.
We are grateful for the excellent technical support from
Bernadette Batiste, Hsiu Chen, Kiran Dhillon, Stacey
Difuntorum, Tod Flak, Rebecca Foss, Martin Garcia,
Lin Lin, Geoffrey Nosrati, Quinn Pack, and Zhijun Ye.
This work was supported in part by the Defense
Advanced Research Projects Agency (DARPA Grant
Number N65236-99-1-5427).

2072
R. W. B€urli et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2067–2072
p-TsOH; reflux, 24 h; 94%), formylation (1.07 equiv BuLi,
References and notes
Et2O; 0 ꢁ C, 30 min; then, )78 ꢁ C, 1.3 equiv DMF; then
rt, 2 h; 69%)19 and deprotection (dioxane/H20/85% H3PO4
3:3:2; rt, 12 h; 84%)20 to give benzo[1,3]dioxole-5,6-
dicarbaldehyde. Treatment of the dialdehyde with
1 equiv glycine methylester hydrochloride (dioxane/
DBU 5:1; reflux, 2 h; 10%) followed by hydrolysis
(MeOH/2 M aq NaOH 1:5; 70 ꢁC, 12 h; 43%) gave the
title compound.
€
1. Ge, Y.; Difuntorum, S.; Touami, S.; Critchley, I.; Burli,
R.; Jiang, V.; Drazan, K.; Moser, H. Antimicrob. Agents
Chemother. 2002, 46, 3168.
€
2. Burli, R. W.; Ge, Y.; White, S.; Baird, E. E.; Touami, S.
M.; Taylor, M.; Kaizerman, J. A.; Moser, H. E. Bioorg.
Med. Chem. Lett. 2002, 12, 2591.
3. Kaizerman, J. A.; Gross, M. I.; Ge, Y.; White, S.; Hu, W.;
Duan, J.-X.; Baird, E. E.; Johnson, K. W.; Tanaka, R. D.;
18. Harrowven, D. C.; Newman, N. A.; Knight, C. A.
Tetrahedron Lett. 1998, 39, 6757.
€
Moser, H. E.; Burli, R. W. J. Med. Chem. 2003, 46, 3914.
19. Boger, D. L.; Wolkenberg, S. E. J. Org. Chem. 2000, 65,
9120.
4. Eady, E. A.; Cove, J. H. Curr. Opin. Infect. Dis. 2003, 16,
103.
20. Dallacker, F.; Schleuter, H.-J.; Schneider, P. Z. Natur-
forsch. B Anorg. Chem. Org. Chem. 1986, 41, 1273.
21. Bozdogan, B.; Esel, D.; Whitener, C.; Browne, F. A.;
Appelbaum, P. C. J. Antimicrob. Chemother. 2003, 52,
864.
5. Bremer, R. E.; Szewczyk, J. W.; Baird, E. E.; Dervan, P.
B. Bioorg. Med. Chem. 2000, 8, 1947.
6. iPrPy was incorporated as the N-Boc protected amino
acid. This compound was prepared starting from ethyl
4-nitropyrrole-2-carboxylate5 by N-alkylation with
2-bromopropane (2.1 equiv, 1 equiv KI, 2 equiv K2CO3
in DMF; 75 ꢁC) followed by hydrogenation (H2, 5% Pd–C,
THF; rt), protection (Boc2O, DMF; rt), and hydrolysis
(4 equiv KOH in EtOH/H2O; 60 ꢁC, 72 h) in 62% overall
yield.
22. Compound 33 was shown to interact with A/T-rich DNA:
DNase
I
footprinting2 on the r70 target sequence
50-ACAATTAA-30 revealed a Kd (app.) of ꢀ7.5 nM.
23. Baird, E. E.; Dervan, P. B. J. Am. Chem. Soc. 1996, 118,
6141.
24. To avoid the incorporation of a C-terminal b-alanine as
described,23 a thiol containing resin (thiomethyl-polysty-
rene used as scavenger for alkylating agents, NovaBio-
chem #01-64-0319) was used. After washing the resin with
N-methyl pyrrolidine (NMP), the first Boc-protected
amino acid (1.3 equiv) was coupled in NMP/iPrEt2N
(4:1) for 24 h at 37 ꢁC. The resin was treated with Ac2O
(1 h, rt) and thoroughly washed (2· each with NMP,
MeOH, and CH2Cl2) prior to the next step. Deprotection
was achieved by incubating the resin twice with
CF3COOH (10 min, rt) followed by washing. The follow-
ing coupling steps were accomplished as described.23
Cleavage of the final product from the resin was achieved
by aminolysis (RNH2/NMP 2:1, 60 ꢁC, 48–96 h). This
procedure was mainly used for compounds carrying Py,
iPrPy, or Pz in position 3.
7. Dyatkina, N. B.; Roberts, C. D.; Keicher, J. D.; Dai, Y.;
Nadherny, J. P.; Zhang, W.; Schmitz, U.; Kongpachith,
A.; Fung, K.; Novikov, A. A.; Lou, L.; Velligan, M.;
Khorlin, A. A.; Chen, M. S. J. Med. Chem. 2002, 45, 805.
8. White, S.; Baird, E. E.; Dervan, P. B. Chem. Biol. 1997, 4,
569.
9. Urbach, A. R.; Dervan, P. B. Proc. Natl. Acad. Sci. U.S.A.
2001, 98, 4343.
10. Maeshima, K.; Janssen, S.; Laemmli, U. K. The EMBO J.
2001, 20, 3218.
11. Nguyen, D. H.; Szewczyk, J. W.; Baird, E. E.; Dervan, P.
B. Bioorg. Med. Chem. 2001, 9, 7.
12. Reddy, B. S. P.; Sharma, S. K.; Lown, J. W. Curr. Med.
Chem. 2001, 8, 475.
13. Neidle, S. Nat. Prod. Rep. 2001, 18, 291.
14. Boger, D. L.; Fink, B. E.; Hedrick, M. P. J. Am. Chem.
Soc. 2000, 122, 6382.
25. National Committee for Clinical Laboratory Standards
2000. Methods for dilution antimicrobial susceptibility
tests for bacteria that growaerobically; Approved Stan-
dard (M7-A5)-Fifth Edition. National Committee for
Clinical Laboratory Standards, Wayne, PA.
€
15. Gross, M.; Burli, R.; Jones, P.; Garcia, M.; Batiste, B.;
Kaizerman, J.; Moser, H.; Jiang, V.; Hoch, U.; Duan,
J.-X.; Tanaka, R.; Johnson, K. Antimicrob. Agents Che-
mother. 2003, 47, 3448.
26. Frimodt-Møller, N.; Knudsen, J. D.; Espersen, F. In
Handbook of Animal Models of Infection; Zak, O., Sande,
M. A., Eds.; Academic: London, San Diego, 1999; pp 127–
136.
27. Azoulay-Dupuis, E.; Moine, P. In Handbook of Animal
Models of Infection; Zak, O., Sande, M. A., Eds.;
Academic: London, San Diego, 1999; pp 481–493.
16. Adolphe, P. S.; Hublot, B.; Lepage, F. FR Patent
2,750,425, 1998; (CAN 128:192643 AN 1998:195243).
17. The synthesis of [1,3]dioxolo[4,5-g]isoquinoline-7-carbox-
ylic acid was performed as follows: Commercially avail-
able benzo[1,3]dioxole-5-carbaldehyde was brominated in
AcOH (2.5 equiv Br2 in AcOH; rt, 24 h, then 50 ꢁC, 2 h;
85%)18 followed by acetal formation (HC(OEt)3, 5 mol %