Modulating the development of E. coli biofilms with 2-aminoimidazoles

Article abstract of DOI:10.1016/j.bmcl.2010.08.075

The synthesis of a 20 member 2-aminoimidazole/triazole pilot library is reported. Each member of the library was screened for its ability to inhibit or promote biofilm development of either Escherichia coli and Acinetobacter baumannii. From this screen, E. coli-selective 2-aminoimidazoles were discovered, with the best inhibitor inhibiting biofilm development with an IC₅₀ of 13 μM. The most potent promoter of E. coli biofilm formation promoted biofilm development by 321% at 400 μM.

Full text of DOI:10.1016/j.bmcl.2010.08.075

Bioorganic & Medicinal Chemistry Letters 20 (2010) 6310–6312
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Modulating the development of E. coli biofilms with 2-aminoimidazoles
⇑
Catherine S. Reed, Robert W. Huigens III, Steven A. Rogers, Christian Melander
Department of Chemistry, North Carolina State University, Raleigh, NC 27695-8024, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of a 20 member 2-aminoimidazole/triazole pilot library is reported. Each member of the
library was screened for its ability to inhibit or promote biofilm development of either Escherichia coli
and Acinetobacter baumannii. From this screen, E. coli-selective 2-aminoimidazoles were discovered, with
Received 12 May 2010
Revised 13 August 2010
Accepted 17 August 2010
Available online 20 August 2010
the best inhibitor inhibiting biofilm development with an IC50 of 13
lM. The most potent promoter of
E. coli biofilm formation promoted biofilm development by 321% at 400
lM.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
2-Aminoimidazoles
E. coli
Biofilms
Bacterial biofilms, which are defined as a community of surface
attached bacteria protected by an extracellular matrix of biomole-
cules, account for ca. 80% of the world’s microbial mass at any gi-
ven time.1 When bacteria are embedded in a biofilm, they are
significantly more resistant to eradication by microbicides and
other environmental pressures that would lead to the elimination
of their free-floating (planktonic) brethren.2 Although biofilms are
typically associated with infectious disease3 (i.e., the NIH estimates
that 3 in 4 bacterial infections are biofilm-based) and represent a
tremendous hurdle for disease treatment4 (biofilm bacteria are up-
wards of 1000-fold more resistant to antibiotics); there are a num-
ber of bacterial biofilm populations, such as gut flora5 and
commensal dental bacteria,6 that provide significant benefits for
their host. In this regard, when designing small molecules that
modulate biofilm formation, it is desirable to have the ability to
tune activity for specific bacterial strains through structural mod-
ification and have the ability to generate compounds that actually
promote biofilm formation in a targeted bacterial population.
With the goal of identifying molecules that have selective activ-
ity against a particular bacterial species, we report the investiga-
tion of structural motifs based upon the 2-aminoimidazole (2-AI)
scaffold. Previous studies in our group have established that 2-ami-
noimidazoles can be designed to inhibit and disperse biofilms from
both gram-positive and gram-negative strains of bacteria.7 Based
upon these results, we wanted to design a new approach to rapidly
generate diversity on the 2-AI scaffold, which could then be evalu-
ated to identify small molecules that selectively inhibit biofilm for-
mation against a target bacterial population. In addition, we were
interested in determining if a subset of molecules from this new
2-AI library would promote biofilm formation. As a test case, we
attempted to identify molecules with differential activity between
two c-proteobacteria: Escherichia coli and Acinetobacter baumannii.
The synthetic approach to this library is outlined in Scheme 1.
We chose to synthetically elaborate chloroacetyl chloride into a di-
verse collection of azido amides due to their ease of preparation
and potential for structural diversity. The azido amides were then
planned to undergo a click reaction8 with alkyne 1 to generate a
diverse set of 2-AI small molecules.9 Alkyne
1
was rapidly
synthesized from 5-hexynoic acid by transformation into the
-bromoketone 2 followed by condensation with Boc-guanidine.10
a
Each azido amide that was prepared (Scheme 1) was then sub-
jected to the click reaction in the presence of 1. Final deprotection
and counterion exchange delivered 2-AIT conjugates 4a–p.
Each 2-AIT conjugate was initially screened for the ability to
modulate biofilm development of E. coli and A. baumannii at
400 lM. Biofilm development was monitored under static condi-
tions using a crystal violet (CV) reporter assay.11 Briefly, bacteria
were incubated in the absence or presence of compound in 96-well
microtiter plates. After 24 h, the media and planktonic bacteria
were removed and the wells were washed with water. Each well
was then treated with CV, which stains any biofilm bacteria that
are attached to the surface of the well. Following the removal of
CV and washing of the wells the remaining CV was then solubilized
by ethanol and quantified spectrophotometrically (A540). The re-
sults of this initial screen are summarized in Table 1.
From this initial assessment at 400 lM, we rapidly identified
several compounds that not only selectively inhibited E. coli bio-
film development (in comparison to A. baumannii), but we also
identified a number of compounds that promote E. coli biofilm
development. Interestingly, compounds 4k and 4p promoted bio-
film formation by 277% and 321%, respectively, while compound
4b inhibited biofilm development by 90.8%.
⇑
Corresponding author. Tel.: +1 919 513 2960; fax: +1 919 515 5079.
E-mail address: Christian_Melander@NCSU.edu (C. Melander).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.08.075

C. S. Reed et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6310–6312
6311
Scheme 1. Synthesis of 2-AIT library. Reagents and conditions: (a) (COCl)2, CH2Cl2, cat. DMF; (b) CH2N2, Et2O; (c) HBr; (d) Boc-guanidine, DMF; (e) chloroacetyl chloride, Et3N,
CH2Cl2; (f) NaN3, DMF; (g) sodium ascorbate, CuSO4Á5H2O, CH2Cl2, EtOH, H2O; (h) TFA, CH2Cl2; then MeOH/HCl.
Table 1
Initial
%
biofilm inhibition for compounds 5a–5u at 400
l
M
in comparison to
untreated controls
a
a
Compound
E. coli biofilm
A. baumannii biofilm
4a
4b
4c
4d
4e
4f
4g
4h
4i
4j
4k
4l
4m
4n
4o
4p
À30.3 7.50
90.8 0.52
57.8 2.42
À127 18.7
17.6 4.73
À47.8 8.49
À169 22.1
2.6 5.6
À183 61.2
À142 19.9
À278 31.05
À41.8 30.7
À176 22.7
À169 22.1
À19.9 6.89
À321 34.6
31.4 6.5
28.3 6.8
36.4 6.1
13.6 3.5
21.2 7.5
15.9 8.0
17.7 3.3
6.2 8.9
30.3 10.8
23.0 3.2
19.5 3.3
20.8 12.3
23.2 3.1
13.9 3.5
17.5 7.8
8.5 3.7
Scheme 2. Reagents and conditions: (a) ethylene diamine, NaH; (b) CrO3, H2SO4;
(c) (COCl)2, CH2Cl2, cat. DMF; (d) CH2N2, Et2O; (e) HBr; (f) Boc-guanidine, DMF; (g)
sodium ascorbate, CuSO4Á5H2O, CH2Cl2, EtOH, H2O; (h) TFA, CH2Cl2, MeOH/HCl.
a
Values are means of three experiments.
Each of the new 2-AIT derivatives were screened at 400 lM for
their ability to modulate E. coli biofilm formation. Results from
these experiments are summarized in Table 2. In terms of biofilm
inhibition, 5b displayed a significant reduction in anti-biofilm
activity (58.9%) in comparison to 4b (90.8%), while the biofilm pro-
motion activity of 5j was essentially identical to that of 4j (143% vs
142%). Compound 5l, however, demonstrated a significant increase
in activity in comparison to 4l (174% vs 41.8%). Therefore, it ap-
pears that a 5-methylene unit spacer is not optimal for all 2-AIT
conjugates and must be tuned based on the application (inhibition
or promotion) and target bacterium.
Given this observation, we elected to probe whether the activity
of 2-AIT conjugate was governed by overall molecular length. We
have previously observed this phenomenon with our reverse
amide series of 2-aminoimidazole conjugates.13 To answer this
question, compound 6 was synthesized and assayed for its ability
With preliminary information, we performed a brief structure–
activity relationship (SAR) study in attempt to understand the
structural components necessary to elicit control of biofilm devel-
opment on this 2-AIT scaffold. First, we wanted to study the impact
that the number of methylene units between the 2-AI head and the
triazole had upon biofilm activity. A previous study from our group
has demonstrated that anti-biofilm activity is dependent upon the
linker distance between the 2-AI head and triazole subunit.9 Spe-
cifically, we demonstrated that activity increased as a function of
linker length from 1 to 5 methylene units, at which point any fur-
ther increase in linker length led to decreased activity. This library
employed a 3-methylene unit spacer and, given the observation
that 5 methylene units was optimal in our previous library, we
were curious as to the impact that altering the linker to 5 methy-
lene units had upon anti-biofilm activity of our lead compounds
from this library.
to inhibit E. coli biofilm developmental 400 lM (Figure 1). At this
concentration, compound 6 inhibited the formation of E. coli
The synthetic approach to these second generation analogs is
outlined in Scheme 2. 3-Octyn-1-ol was first isomerized12 to 7-oc-
tyn-1-ol followed by Jones oxidation resulting in 7-octynoic acid.
Installation of the Boc-protected 2-aminoimidazole was identical
to that of alkyne 1 to deliver 5. We then subjected 5 to the click
reaction with azido amides 2b, 2j, and 2l generating 2-AITs 5b,
5j, and 5l, respectively. Compound 2b was selected due to their
interesting biofilm inhibiting activity from the initial study (4b),
while 2j and 2l were chosen as two representative tail groups that
promoted biofilm formation (4j and 4l).
Table 2
SAR biofilm inhibition results for compounds 5b, 5j, and 5l
Compound
E. coli biofilm inhibition at 400 l
M (%)a
5b
5j
5l
58.9 8.8
À142.9 52.5
À174.1 59.3
a
Values are means of three experiments.

6312
C. S. Reed et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6310–6312
(highest concentration tested). The biological data for all three of
these compounds are summarized in Table 3.
In summary, we have developed an approach to the introduc-
tion of diversity on the 2-aminoimidazole/triazole scaffold through
the use of chloroacetyl chloride as a key building block for struc-
tural diversification. Furthermore, we have shown that 2-amino-
imidazole-derived anti-biofilm agents can be designed to be
selective for a target bacterial population, and that both inhibitors
and promoters of biofilm formation can be identified. Based upon
these studies, compounds 4b, 6, 4k, and 4p represent promising
small molecule probes to study the role of E. coli biofilms in
E. coli pathogenesis.
Figure 1. Structure/activity comparison of 4b and 6.
Table 3
Summary of biological data for 4b, 4k, 4p, and 6
Acknowledgments
a
a
a
Compound
E. coli IC50
(lM)
E. coli EC50
(l
M)
HD50 (lM)
4b
6
4k
4p
36.9 1.0
13.0 1.0
—
—
120.4 12.7
NA
—
487 87.0
675 21.6
>800
Financial support for this work is gratefully acknowledged from
the Jimmy V. Cancer Therapeutics Training Program (predoctoral
fellowship to R.W.H. and S.A.R.) and the University of North Caro-
lina Competiveness Research Fund.
—
>800
a
Values are means of three experiments.
Supplementary data
biofilms by 95%. To compare the activity of compounds 4b and 6,
we then performed a dose-response study to determine the IC50
Supplementary data associated with this article, including rep-
resentative examples of experimental procedures, characterization
data, inhibition and dispersion assay protocols, and bacterial
growth curves can be found, in the online version, at doi:10.1016/
j.bmcl.2010.08.075.
value for each compound. For 2-AIT 4b, an IC50 value of 36.9
lM
was observed, while for 6 we observed an IC50 of 13.0 M (ca.
l
three-fold increase in activity). Therefore, precise placement of
the triazole and amide within the structure of the 2-AIT conjugate
can significantly impact activity.
Next, we analyzed the ability of 4b and 6 to disperse preformed
E. coli biofilms. Compounds that have the ability to disperse pre-
formed biofilms are significant because they have the potential
to remediate established biofilms.14 Compound 4b was able to dis-
References and notes
1. Costerton, J. W.; Lewandowski, Z.; Caldwell, D. E.; Korber, D. R.; Lappinscott, H.
M. Annu. Rev. Microbiol. 1995, 49, 711.
2. Davies, D. Nat. Rev. Drug Discov. 2003, 2, 114.
3. Davies, D. G.; Parsek, M. R.; Pearson, J. P.; Iglewski, B. H.; Costerton, J. W.;
Greenberg, E. P. Science 1998, 280, 295.
perse pre-established E. coli biofilms with an EC50 of 120 lM,
where we define EC50 as the concentration of compound necessary
to elicit 50% biofilm dispersion. Compound 6 showed no dispersal
activity against E. coli, paralleling our previous results with reverse
amide 2-AI anti-biofilm agents.13 Specifically, biofilm dispersion is
dependent on tail composition.
Finally, red blood cell hemolysis of 4b, 6, 4k, and 4p was per-
formed.15 Compounds 4b and 6 were chosen because these com-
pounds had the greatest inhibition activity against E. coli. On the
other hand, compounds 4k and 4p exhibited the greatest biofilm
promotion of E. coli. The HD50 (the dose that lyses 50% of the red
4. Rasmussen, T. B.; Givskov, M. Int. J. Med. Microbiol. 2006, 296, 149.
5. Sansonetti, P. J. Curr. Opin. Gastroenterol. 2008, 24, 435.
6. Soysa, N. S.; Samaranayake, L. P.; Ellepola, A. N. B. Oral Dis. 2008, 14, 138.
7. Richards, J. J.; Melander, C. Anti-Infect. Agents Med. Chem. 2009, 8, 295.
8. Finn, M. G.; Kolb, H. C.; Fokin, V. V.; Sharpless, K. B. Prog. Chem. 2008, 20, 1.
9. Rogers, S. A.; Melander, C. Angew. Chem., Int. Ed. 2008, 47, 5229.
10. Birman, V. B.; Jiang, X. T. Org. Lett. 2004, 6, 2369.
11. O’Toole, G. A.; Kolter, R. Mol. Microbiol. 1998, 30, 295.
12. Brown, C. A.; Yamashita, A. J. Am. Chem. Soc. 1975, 97, 891.
13. Ballard, T. E.; Richards, J. J.; Wolfe, A. L.; Melander, C. Chem. Eur. J. 2008, 14,
10745.
14. Hett, E. C.; Hung, D. T. Chem. Biol. 2009, 16, 1216.
15. Liu, Z. G.; Brady, A.; Young, A.; Rasimick, B.; Chen, K.; Zhou, C. H.; Kallenbach, N.
R. Antimicrob. Agents Chemother. 2007, 51, 597.
blood cells) for 4b and 6 was found to be 391
lM and 675 lM,
respectively, while compounds 4k and 4p had HD50’s >800
lM