J. Maresh et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3281–3286
3285
tion of xenognosis in A. tumefaciens.5 Therefore, HDM-
BOA does not accumulate to the point where selection
pressure could create resistance; but merely disables
the sensory system of those pathogens very close to
the host. As researchers overcome the difficulties inher-
ent in studying unstable natural products, the generality
of this apparent resistance strategy may be revealed.
Given the instability of HDMBOA and the generality of
these assumptions, HDI may in fact function as an ana-
log of decomposition intermediate 3, which has been
shown to possess in vivo inhibitory activity. In a previ-
ous report, an analog of imidoquinone 3 which lacked
the aldehyde functional group was an active vir inhibi-
tor.7 In these experiments, we have now uncoupled the
aldehyde from the imidoquinone and demonstrated that
stable aromatic aldehydes are also vir inhibitors.
The emergence of antibiotic resistance in pathogenic
bacteria has inspired research and development of
two-component system inhibitors. The results of these
efforts have been reviewed,6,25 and unfortunately all
compounds identified by several different approaches
were either too hydrophobic or detergent-like for further
development as therapeutic agents. As a natural two-
component system inhibitor, HDMBOA may hold the
key to novel strategies for development of anti-virulence
two-component system inhibitors with improved phar-
macokinetics. Accordingly, this study examined the spe-
cific features of HDMBOA required for inhibition of the
host recognition elements VirA–VirG.
The high potency of HDMBOA is likely to depend on
both functional elements, though they may act on differ-
ent steps of signal transduction. In vivo and in vitro as-
says are being developed to assess the specificity of HDI,
HDMBOA, and DIMBOA for a particular step in two-
component system signal perception and transmission,
and for specificity to other two-component systems.
While HDI is an apparently specific inhibitor of VirA/
VirG signal recognition, these results raise the possibil-
ity that the unique reversible electrophilic reactivity of
3H-isobenzofuran-1-ones may find wider application
as an initial scaffold for library construction.
Simple in vivo assays suggested that the inhibitory activ-
ity of 2 and the weaker inhibitor 5 can be removed by
media exchange, implying a reversible inhibition mecha-
nism (Fig. 3). Yet most of the possible chemical reactiv-
ities ascribed to benzoxazinones are irreversible in
nature and associated with the unstable iminoquinone
nucleus.26–28 We therefore focused inhibitor design on
the reversible aldehyde electrophile. Indeed, the alde-
hyde was found to be essential to the activity of analogs
12–16 in Table 1B, with HDI (12) being the most potent
of these analogs. Replacement of the aldehyde with a ke-
tone abolished activity (compounds 17–19). Synthesis of
additional HDI analogs revealed that structural effects
of the methoxy substituents, possibly for active-site
recognition, are apparently more critical than their
electronic effects as the 4,6-diethoxy analog 26 was
completely inactive. The 4-methoxy group is required
for activity (Table 1C). Moreover, many ketone substi-
tutions at R1 were tolerated. Therefore, derivatization
of this position will permit convenient access to labeled
compounds and screening for more potent analogs.
Acknowledgment
We are indebted to Professor Andrew Binns and his lab-
oratory at the University of Pennsylvania for resources,
insight and advice, we thank Andrew G Palmer for his
assistance with this manuscript, and acknowledge the
valuable support of NIH GM 47369 and NSF CHE
0121013.
Supplementary data
Supplementary data associated with this article can be
References and notes
1. Hansen, G.; Wright, M. Trends Plant Sci. 1999, 4, 226.
2. Birch, R. G. Annu. Rev. Plant Physiol. Plant Mol. Biol.
1997, 48, 297.
3. D’Halluin, K.; Botterman, J. In Rhizobiaceae; Spaink, H.
P., Kondorosi, A., Hooykaas, P. J. J., Eds.; Kluwer
Academic: Boston, 1998; p 339.
Despite the simplicity of the assumptions in its design,
HDI is in fact a reversible inhibitor with higher po-
tency for vir inhibition than DIMBOA and all other
synthetic vir inhibitors. The design of HDI assumed
that HDMBOA was competitive with phenol induc-
tion through reversible covalent binding to VirA via
an electrophilic aldehyde. Although a site of phenol
binding on VirA has not been precisely identified,
the structural requirements and limits have been iden-
tified. Methoxy groups proximal to the phenol
hydroxy contribute positively to phenol sensitivity.
This dependence is observed in the trend of increasing
inducer potency for 4-hydroxy-acetophenone ꢀ 4-hy-
droxy-3-methoxy-acetophenone < 4-hydroxy-3,5-dime-
thoxy-acetophenone (AS).29 By simple analogy, the
3H-isobenzofuran-1-one derivative HDI (12) may also
be described as a stable structure combining the
critical features of HDMBOA with known structural
features of phenol inducers.
4. Hansen, G.; Chilton, M. D. Curr. Top. Microbiol. 1999,
240, 21.
5. Zhang, J.; Boone, L.; Kocz, R.; Zhang, C.; Binns, A. N.;
Lynn, D. G. Chem. Biol. 2000, 7, 611.
6. Hubbard, J.; Burnham, M. K. R.; Throup, J. P. In
Histidine Kinases in Signal Transduction; Inouye, M.,
Dutta, R., Eds.; Elsevier Science: San Diego, 2003;
p 459.
7. Maresh, J.; Zhang, J.; Lynn, D. G. ACS Chem. Biol. 2006,
1, 165.
8. Maresh, J. Ph.D. thesis, University of Chicago, Chicago,
IL, 2004.
9. Sicker, D.; Schulz, M. Stud. Nat. Prod. Chem. 2002, 27,
185.
10. Hedin, P. A.; Davis, F. M.; Williams, W. P. J. Chem. Ecol.
1993, 19, 531.