Construction and screening of a 2-aminoimidazole library identifies a small molecule capable of inhibiting and dispersing bacterial biofilms across order, class, and phylum

Article abstract of DOI:10.1002/anie.200800862

(Chemical Equation Presented) A team of three: 2-Aminoimidazole, triazole, and tether units together resulted in conjugates (see picture, n=4-6 for the most active compounds) that are capable of inhibiting and dispersing bacteria biofilms without inducing bacterial death. Such biofilms have been implicated in a plethora of medical problems, including infection of implanted medical devices and the mortality of cystic fibrosis patients.

Full text of DOI:10.1002/anie.200800862

Angewandte
Chemie
DOI: 10.1002/anie.200800862
Click Chemistry
Construction and Screening of a 2-Aminoimidazole Library Identifies a
Small Molecule Capable of Inhibiting and Dispersing Bacterial
Biofilms across Order,Class,and Phylum**
Steven A. Rogers and Christian Melander*
Bacterial biofilms are defined as a surface-attached commun-
ity of bacteria that are surrounded by a protective extra-
cellular matrix.[1] Within the biofilm state, bacteria display
differential gene expression and are at least 1000-fold more
resistant to antibiotic treatment.[2] Biofilms account for more
than 80% of all bacterial infections; they drive persistent
infection of indwelling medical devices, and are responsible
for the mortality and morbidity of almost all cystic fibrosis
(CF) patients.[3–6]
Given the biomedical prominence of biofilms, there have
been significant efforts to discover small molecules that
modulate biofilm development.[1] We have shown that simple
derivatives of the marine natural product bromoageliferin will
both inhibit and disperse bacterial biofilms (Scheme 1).[7–11]
Recently, we demonstrated that dihydrosventrin (DHS)
inhibits and disperses Pseudomonas aeruginosa (multiple
strains), Acinetobacter baumannii, and Bordetella bronchi-
septica biofilms,[7] making it the first small molecule reported
to inhibit and disperse biofilms across bacterial order and
class through a nonbactericidal mechanism.
We have begun to investigate whether modifications to
the core DHS structure will lead to derivatives with enhanced
anti-biofilm activities.[8] One of the first structural variations
we have studied is replacement of the pyrrole subunit with a
triazole subunit (Scheme 2). Herein we detail the develop-
Scheme 2. Design of a 2-AIT.
ment of the synthetic protocols necessary to access 2-amino-
imidazole/triazole conjugates (2-AITs), the application of
these methods to the synthesis of a focused 2-AIT library, and
the discovery of small molecules that inhibit and disperse
bacterial biofilms across order, class, and phylum.
Scheme 1. Bromoageliferin, structural inspiration for the synthesis of
analogues. TAGE, DHS, and RA-11 are analogues that inhibit and
disperse bacterial biofilms.
Given that there is a paucity of reactions that have been
reported to be compatible with 2-aminoimidazoles, we
deemed the CuI-catalyzed [3+2] alkyne/azide cycloaddition
(click reaction)[12–14] as a promising method to generate 2-
AITs given the broad substrate range displayed by this
reaction. To test the applicability of the reaction, we
synthesized the alkynyl-substituted 2-aminoimidazole (2-AI,
1) and tested its ability to participate in a CuI-catalyzed
[3+2] cycloaddition with benzyl azide.
[*] S. A. Rogers, Prof. C. Melander
Department of Chemistry
North Carolina State University
Raleigh, NC 27695-8204 (USA)
Fax: (+1)919-515-5079
E-mail: christian_melander@ncsu.edu
[**] Financial support from NCSU and Agile Sciences, Inc. is gratefully
acknowledged. Mass spectra were obtained at the Mass Spec-
trometry Laboratory for Biotechnology at NCSU. We thank Dr.
Reza A. Ghiladi (NCSU) for spectroscopic assistance.
Amino acid 2[15] was subjected to a small-scale Akabori
reduction,[16] which, followed by condensation with cyan-
Supporting information for this article (compound synthesis,
compound characterization, initial library screening, bacterial
growth curves, bacterial colony counts, and dose–response curves)
is available on the WWW under http://www.angewandte.org or from
the author.
amide[17] delivered the target alkyne
1 in 88% yield
(Scheme 3). With 1 in hand, we explored various conditions
to elicit the Cu-catalyzed [3+2] cycloaddition between 1 and
benzyl azide (Table 1). In THF with CuI only starting material
Angew. Chem. Int. Ed. 2008, 47, 5229 –5231
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5229

Communications
Scheme 3. Synthesis of alkynyl-substituted 2-aminoimidazole 1. a) 5%
Na/Hg, H20, pH 1.5–2.0, 0–58C. b) Cyanamide, H20, pH 4.3, 958C.
Table 1: Results of the Cu-catalyzed [3+2] cycloaddition between 1 and
benzyl azide under various conditions.
Scale [mg] CuI
source[a]
Solvent
Base[b] T [8C] Yield [%]
20
20
20
CuI
CuI
THF
THF
DIEART
DIEA40
–
–
–
CuSO4/ EtOH/H2O (1:1)
NaAsc
CuSO4/ EtOH/H2O (1:1)
NaAsc
CuSO4/ EtOH/H2O (1:1)
NaAsc
CuSO4/ tBuOH/H2O/
RT
40
40
RT
–
20
100
100
–
–
–
86
decomposition
93
Scheme 4. Construction of the initial 2-AIT library.
NaAsc
CH2Cl2 (1:1:1)
The effect of the tether length on activity was addressed
by determining the IC50 of A. baumannii biofilm inhibition for
7 and 8 using dose–response studies. Compound 7 showed an
[a] NaAsc=sodium ascorbate. [b] DIEA=diisopropylethylamine.
was returned. We then switched to using CuSO4 and sodium
ascorbate in a 1:1 solvent mixture of EtOH/H2O. Again, no
reaction was noted. However, when the reaction mixture was
heated to 408C the desired 2-AIT 3 was obtained in 86%
yield. Unfortunately, when the reaction was scaled up, a
significant amount of decomposition occurred. Room-tem-
perature click reactions have been described for a 1:1:1
solvent mixture of tBuOH/H2O/CH2Cl2.[18] With these reac-
tion conditions we observed conversion to 3 in 93% yield.
Next we synthesized the 2-AI alkynes 4 and 5 (see
Scheme 4) with longer methylene spacers between alkyne and
2-AI units analogous to 1 and used all three of them for click
reactions with the 12 azides depicted in Scheme 4 to yield an
initial 2-AIT library. Each compound was characterized
(1H NMR, 13C NMR, HRMS) prior to screening (Supporting
Information).
Each 2-AIT was initially screened at 300 mm for its ability
to inhibit P. aeruginosa PAO1 and PA14 biofilms, A. bauman-
nii biofilms and B. bronchiseptica RB50 biofilms using a
crystal-violet reporter assay.[19] From this screen, 2-AIT 6 was
identified as the most promising; it was most active against
A. baumannii, with an observed IC50 of 12 mm. This is an
order-of-magnitude increase in activity when compared to
DHS.[7] Follow-up growth curves and colony counts indicated
that 6 had no effect on planktonic growth (Supporting
Information), indicating that the inhibition of biofilm devel-
opment was not due to microbicidal activity. Compounds 6
also dispersed preformed A. baumannii biofilms with an EC50
of 400 mm.
IC50 of 220 mm, and 8 did not inhibit A. baumannii biofilm
development in a significant fashion (< 50%) at 800 mm
(highest concentration tested). Thus, increasing tether
length appears to correlate with increasing activity. In
addition, we tested the necessity of the 2-AI subunit by
synthesizing compound 9—by alkylating commercially avail-
able 1-H-1,2,3-triazole with 5-iodopent-1-yne and subjecting
the resulting alkynyl-substituted triazole to a click reaction—
and assaying for its ability to inhibit A. baumannii biofilm
development. This compound revealed minimal activity
(<30% inhibition) at the highest concentration studied
(800 mm). 2-Aminomidazole was also screened and found to
be devoid of activity up to 800 mm (highest concentration
tested).
As indicated above, as the number of methylene units
between 2-AI and triazole unit are increased (6–8), the anti-
biofilm activity against A. baumannii also increases. There-
fore, we synthesized 2-AITs 10–12 where we systematically
5230
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 5229 –5231

Angewandte
Chemie
extended the methylene linker to investigate if additional
methylene unit would deliver a 2-AIT with even greater
biological activity than 6 (Table 2). Compound 10 showed an
IC50 value of 2.8 mm against A. baumannii as well as IC50
chemistry we have assembled a focused library of 2-AITs and,
with an initial hit from this library as lead, derived compounds
that are able to inhibit and disperse bacterial biofilms across
order, class, and phylum. Mechanistic studies are currently
underway to determine how 2-AITs 10–12 inhibit and
disperse biofilms. Furthermore, given the promising anti-
biofilm activity displayed by these and other 2-AI deriva-
tives,[7–11] we are continuing to develop methodology to access
further functionalized libraries based upon the 2-AI core
motif. These studies will be disclosed in due course.
Table 2: Effect of chain length on the activity of 2-AITs against several
biofilms.[a]
n
A. baumannii
PAO1
PA14
RB50
S. aureus
Received: February 21, 2008
Published online: June 4, 2008
4 (10)
5 (11)
6 (12)
2.8
0.98
6.8
15
5.6
2.7
4.0
0.53
22
23
9.5
70
7.0
0.81
4.6
Keywords: biofilms · combinatorial chemistry · dispersion ·
.
inhibition · marine natural products
[a] IC50 values are given in mm.
[1] D. J. Musk, Jr., P. J. Hergenrother, Curr. Med. Chem. 2006, 13,
2163 – 2177.
values of 15, 4.0, and 23 mm against PAO1, PA14, and RB50,
respectively. We also tested its ability to inhibit Staphylococ-
cus aureus biofilm development and found an IC50 value of
7.0 mm. Colony counts and growth curves of each bacterial
strain grown in the presence of 10 revealed that its activity
was not due to bactericidal activity, which, to the best of our
knowledge, is the first example of a nonbactericidal small
molecule that will inhibit biofilm development across order,
class, and phylum.[1] Increasing the methylene spacer to 5
carbon atoms (11) again led to an increase in activity (see
Table 2). Further addition of a methylene group (12) did not
lead to an increase in activity (see Table 2). Follow-up colony
count and growth curve analysis revealed that inhibition of
biofilm development for both 11 and 12 was not due to
microbicidal activity.
[2] T. B. Rasmussen, M. Givskov, Int. J. Med. Microbiol. 2006, 296,
149 – 161.
[3] J. W. Costerton, P. S. Stewart, E. P. Greenberg, Science 1999, 284,
1318– 1322.
[4] T. F. Mah, G. A. OꢀToole, Trends Microbiol. 2001, 9, 34 – 39.
[5] R. C. Boucher, Eur. Respir. J. 2004, 23, 146 – 158.
[6] S. S. Yoon, R. F. Hennigan, G. M. Hilliard, U. A. Ochsner, K.
Parvatiyar, M. C. Kamani, H. L. Allen, T. R. DeKievit, P. R.
Gardner, U. Schwab, J. J. Rowe, B. H. Iglewski, T. R. McDer-
mott, R. P. Mason, D. J. Wozniak, R. E. W. Hancock, M. R.
Parsek, T. L. Noah, R. C. Boucher, D. J. Hassett, Dev. Cell 2002,
3, 593 – 603.
[7] J. J. Richards, R. W. Huigens, T. E. Ballard, A. Basso, J.
Cavanagh, C. Melander, Chem. Commun. 2008, 9(8), 1267 –
1279.
[8] J. J. Richards, T. E. Ballard, C. Melander, Org. Biomol. Chem.
2008, 6(8), 1356 – 1363
[9] J. J. Richards, T. E. Ballard, R. W. Huigens, C. Melander,
ChemBioChem 2008, 14, 1698– 1700.
[10] R. W. Huigens, L. Ma, C. Gambino, A. Basso, P. D. R. Moeller, J.
Cavanagh, D. J. Wozniak, C. Melander, Mol. Biosystems 2008,
4(6), 614 – 621.
[11] R. W. Huigens, G. Parise, J. J. Richards, T. E. Ballard, W. Zeng,
R. Deora, C. Melander, J. Am. Chem. Soc. 2007, 129, 6966 – 6967.
[12] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
2056 – 2075; Angew. Chem. Int. Ed. 2001, 40, 2004 – 2021.
[13] V. O. Rodionov, V. V. Fokin, M. G. Finn, Angew. Chem. 2005,
117, 2250 – 2255; Angew. Chem. Int. Ed. 2005, 44, 2210 – 2215.
[14] V. D. Bock, H. Hiemstra, J. H. van Maarseveen, Eur. J. Org.
Chem. 2006, 51 – 68.
[15] S. Kotha, S. Halder, E. Brahmachary, Tetrahedron 2002, 58,
9203 – 9208.
[16] S. Akabori, Chem. Ber. 1933, 66, 151 – 158.
[17] Y. Xu Yz, K. Yakushijin, D. A. Horne, J. Org. Chem. 1997, 62,
456 – 464.
Finally, we have tested for the ability of a single admin-
istration of 2-AITs 10–12 to disperse preformed bacterial
biofilms. The summary of these experiments is outlined in
Table 3. As can be seen, each compound was able to disperse
the preformed biofilm, regardless of bacterial order, class, or
phylum.
Table 3: Efficiency of compounds 10–12 in dispersing preformed
biofilms.[a]
Compound
A. baumannii
PAO1
PA14
RB50
S. aureus
10
11
12
210
120
36
81
11
51
35
22
48
59
55
75
16
2.6
37
[a] EC50 values are given in mm.
[18] B. Y. Lee, S. R. Park, H. B. Jeon, K. S. Kim, Tetrahedron Lett.
2006, 47, 5105 – 5109.
[19] G. A. OꢀToole, R. Kolter, Mol. Microbiol. 1998, 30, 295 – 304.
In conclusion, we have developed a synthetic approach to
2-aminoimidazole/triazole conjugates that is underpinned by
CuI-catalyzed [3+2] alkyne–azide cycloaddition. Using this
Angew. Chem. Int. Ed. 2008, 47, 5229 –5231
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5231