Influence of heterocyclic and oxime-containing farnesol analogs on quorum sensing and pathogenicity in Candida albicans

Article abstract of DOI:10.1016/j.bmc.2007.11.011

A series of synthetic molecules combining a geranyl backbone with a heterocyclic or oxime head group are quorum-sensing molecules that block the yeast to mycelium transition in the dimorphic fungus Candida albicans. A number of the analogs have an IC₅₀ ≤ 10 μM, a level of potency essentially identical to the natural quorum sensing signal, the sesquiterpene farnesol. Two of the most potent analogs, neither toxic toward healthy mice, display remarkably different effects when co-administered with C. albicans. While neither offers protection from candidiasis, one analog mimics farnesol in acting as a virulence factor, whereas the other has no effect. The results offer the first example of highly potent synthetic fungal quorum-sensing molecules, and provide the first evidence for the ability to decouple quorum sensing and virulence.

Full text of DOI:10.1016/j.bmc.2007.11.011

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 1842–1848
Influence of heterocyclic and oxime-containing farnesol analogs
on quorum sensing and pathogenicity in Candida albicans
Roman Shchepin,a Dhammika H. M. L. P. Navarathna,b Raluca Dumitru,b
Shane Lippold,b Kenneth W. Nickersonb and Patrick H. Dussaulta,*
aDepartment of Chemistry, University of Nebraska-Lincoln, Lincoln, NE 68588, USA
bSchool of Biological Sciences, University of Nebraska-Lincoln, Lincoln, NE 68588, USA
Received 5 September 2007; revised 30 October 2007; accepted 1 November 2007
Available online 6 November 2007
Abstract—A series of synthetic molecules combining a geranyl backbone with a heterocyclic or oxime head group are quorum-sens-
ing molecules that block the yeast to mycelium transition in the dimorphic fungus Candida albicans. A number of the analogs have
an IC50 6 10 lM, a level of potency essentially identical to the natural quorum sensing signal, the sesquiterpene farnesol. Two of the
most potent analogs, neither toxic toward healthy mice, display remarkably different effects when co-administered with C. albicans.
While neither offers protection from candidiasis, one analog mimics farnesol in acting as a virulence factor, whereas the other has no
effect. The results offer the first example of highly potent synthetic fungal quorum-sensing molecules, and provide the first evidence
for the ability to decouple quorum sensing and virulence.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
depending in part on whether the cell densities are above
or below 106 cells per ml.3 Our research revealed that the
population dependence of C. albicans morphology re-
sults from the continuous production and detection of
the sesquiterpene farnesol (1, Table 1),3,4 which acts to
block the yeast to mycelium transition at concentrations
as low as 1–2 lM.5,6 In accordance with the precedent
established for homoserine lactones in Gram-negative
bacteria,7 the phenomenon was termed quorum sensing.
Farnesol was the first quorum-sensing molecule (QSM)
to be discovered in an eukaryotic system.3,4
Candida albicans normally resides as a harmless yeast in
the GI and urinary tracts, and to a lesser extent on the
skin, of most humans. However, for patients with weak-
ened immune systems (e.g., AIDS patients, transplant
recipients, and premature infants) C. albicans is an
opportunistic and often deadly pathogen that will in-
vade host tissues, undergo a dimorphic shift, and then
grow as a fungal mass in the kidney, heart, or brain.
Candida albicans is the fourth leading cause of hospi-
tal-acquired infection in the United States and over
95% of AIDS patients will suffer from infections by C.
albicans.1,2 There is a pressing need for new treatments
for fungal infections.
The dimorphic nature of C. albicans is essential for path-
ogenicity.8 The discovery of a QSM able to direct
growth in the yeast phase therefore suggested the possi-
bility of a ‘single morphology therapy’ which would
force this opportunistic pathogen to remain in a mono-
morphic, nonpathogenic lifestyle. However, three sepa-
rate experiments have shown that farnesol actually
increases fungal pathogenicity. First, cultures of C. albi-
cans pretreated with subinhibitory doses of fluconazole,
a protocol that increases the levels of secreted farnesol 8-
to 10-fold,9 were 4- to 5-fold more pathogenic to mice.10
Second, mice which received farnesol either in their
drinking water or via intraperitoneal injection showed
ca. 4-fold higher lethality from candidiasis than mice
receiving control IP injections.11 Third, a knockout mu-
tant of C. albicans lacking the DPP3 gene encoding an
Candida albicans is a polymorphic fungus able to grow
as yeasts, hyphae, and pseudohyphae, and a great many
chemical and environmental factors influence the rela-
tive populations of these morphologies. In liquid cul-
ture, C. albicans develops as either yeasts or hyphae,
Keywords: Farnesol; Quorum sensing; Candida albicans; Virulence;
Candidiasis; Oxime; Tetrazole; Triazole; Thiotetrazole; Thiotriazole;
Thioimidazole.
*
Corresponding author. Tel.: +1 402 472 6951; fax: +1 402 472
9402; e-mail: pdussault1@unl.edu
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.11.011

R. Shchepin et al. / Bioorg. Med. Chem. 16 (2008) 1842–1848
1843
Table 1. QSM activity of oximes
2.1. Oxime leads for new QSMs
Analog
IC50 (lM)a
Toxicityb
None
While developing a series of fluorescent farnesol ana-
logs,13 we discovered a pentaene oxime 2 with QSM
activity (IC50 = 10 lM) superior to any previously inves-
tigated analog (Table 1; see the experimental section for
a description of the QSM assay). Based on this discov-
ery, we prepared the oxime of farnesaldehyde (3) and
found it was also a potent QSM (IC50 = 5 lM, Table
1). The corresponding O-methyl ether, farnesaldehyde
methoxyoxime 4, was inactive (IC50 > 100 lM), consis-
tent with the results of earlier studies pointing to a
requirement for an acidic head group.6 However, oxime
3 was found to exhibit significant toxicity (IP) in healthy
mice and this class of molecules was not pursued further.
OH
1–2
1 (farnesol)
OH
N
10
NDc
2
2
X
N
3: 5
3: Highd
4: ND
4: P100
3: X = OH
4: X = OMe
a Inhibition of germ tube formation.
2.2. Tetrazole QSMs
b Healthy mice observed over 14 d following injection (IP) with 1 ml of
20 mM solution (0.5% Tween 80 v/v in sterile, nonpyrogenic saline).
c Not determined.
The high quorum-sensing activity of oximes 2 and 3 led
us to explore replacement of the primary alcohol of far-
nesol with other acidic functional groups (Fig. 1). Our
earlier work had found farnesoic acid to be a weakly ac-
tive QSM (IC50 = 36 lM).6 5-Substituted-1H-tetrazoles
are weak acids (pK ꢀ 5) often used as lipophilic isosteres
for carboxylic acids,14,15 and therefore we investigated
several analogs incorporating a tetrazole in place of
C1–C3 of farnesol.
d Mice died within 4–5 days.
isoprenoid pyrophosphatase for conversion of farnesyl
pyrophosphate to farnesol, and thus producing only
15% as much farnesol as the wild-type, was 4- to 5-fold
less pathogenic. Reconstitution of DPP3 restored both
farnesol production and pathogenicity.11 Taken to-
gether, these observations show that farnesol can act
as a virulence factor for C. albicans. We have subse-
quently discovered a potential basis for this behavior,
in that farnesol interferes with the normal progression
of cytokine induction in mice infected with C. albicans.12
Analogs 5, 6, and 7 were synthesized via cycloaddition
of the appropriate nitrile with Et3NHN3, generated
in situ from NaN3 (Scheme 1).16
Analog 5, which incorporates a tetrazole in place of the
C1–C3 region of farnesol, was a potent QSM with
IC50 = 7 lM (Table 2). However, in agreement with con-
clusions from our earlier research,6 chain length was
clearly important. Tetrazole 6, although possessing only
one fewer backbone carbon than 5, was considerably
less active, while commercially available 5-methyl-1-tet-
razole (not shown) was inactive as a QSM at concentra-
tions up to 3 mM. Nonyl tetrazole 7 and commercially
We are interested in the relationships between quorum
sensing and virulence and, in particular, whether it is
possible to decouple the two. Specifically, we were inter-
ested in the possibility of designing analogs that retain
farnesol’s ability to block hyphal development but lack
farnesol’s activity as a virulence factor. In order to ex-
plore this problem, we required potent QSMs structur-
ally distinct from farnesol. We now report the
discovery of a series of highly potent synthetic quo-
rum-sensing molecules, as well as preliminary investiga-
tions of their activity in a mouse model of systemic
candidiasis.
Me
Me
alcohol
(farnesol)
OH
OH
4
4
N
N
Me O
acid
NH
N
N
OH
"4"
4
tetrazole
oxime
2. Results
Figure 1. Design concept for analogs.
When we initiated these studies, there were no synthetic
molecules possessing a significant fraction of farnesol’s
biological activity as a quorum-sensing molecule. Previ-
ously, we had probed farnesol’s mode of action with
over 40 natural and synthetic farnesol analogs.6 Modi-
fied structural features included the head group, chain
length, presence or absence of the three double bonds,
substitution of a backbone carbon by S, O, N, and Se
heteroatoms, presence or absence of a 3-methyl branch,
and bulkiness of the hydrophobic tail. Although, more
than half the compounds showed measurable quorum-
sensing activity (inhibition of hyphal development), the
best displayed only 7% of the activity of E,E-farnesol.6
N
N
N
NaN3,
Et3NHCl
toluene, Δ
CN
N
H
5
H
N
" "
" "
CN
N
6
N
N
N
N
N
7
Me(CH2)8CN
nonyl
N
H
Scheme 1. Synthesis of Tetrazoles.

1844
R. Shchepin et al. / Bioorg. Med. Chem. 16 (2008) 1842–1848
Table 2. Quorum sensing and toxicity of tetrazoles
Tetrazole
IC50 (lM)a
Toxicityb
5
6
7
7
Non-toxic, virulence factor
ND
Non-toxic
>50
3
N
N
NH
Ph
10
ND
N
8
a,bSee Table 1.
available styrenyl tetrazole 8 were also highly active;
however, 7 was only marginally soluble in water. The
strong quorum-sensing activity observed for analogs 5
and 8 implies a functional equivalency between styrene
and 4,8-dimethyl-3,7-nonadiene sidechains, suggesting
that 3-methyl-5-phenyl-2,4-pentadienol or similar struc-
tures might represent productive future targets as syn-
thetic QSMs.17 Along these lines, it is interesting to
note a report indicating that dodecanol, an analog of
the 12-carbon homoserine lactone secreted as a QSM
by Pseudomonas aeruginosa, is a weak QSM for C.
albicans.18
Scheme 2. Additional heterocyclic analogs.
Table 3. Quorum sensing and toxicity in 9–13
Azolec
IC50 (lM)a
Toxicityb
ND
N
N
9
>50
NH
RS
RS
N
N NH
N
10
1
None in P14 d;
not a virulence
factor
2.3. Other heterocyclic QSMs
The potency of the tetrazoles as quorum-sensing mole-
cules led us to investigate additional examples of five-
membered heterocyclic analogs possessing the same
approximate shape, crosssection, lipophilicity, and
molecular length (Scheme 2). Our previous studies had
found the replacement of a methylene with a sulfur atom
(thioether) to be a well-tolerated structural modifica-
tion,6 and we therefore mainly focused on sulfur-linked
heterocycles, which are readily available via nucleophilic
displacement (Scheme 2).19 A similar approach has been
employed for synthesis of a thiotriazole from farnesyl
chloride.20 The substrates tested grafted a series of het-
erocyclic head groups onto a geranyl (3,7-dimethyl-2,6-
octadienyl) tail: thio-1,2,3,4-tetrazole (9), thio-1,2,4-tria-
zole (10), thio-1,2,3-triazole (11), and thioimidazole (12).
We also investigated a carbon-linked 1,2,3-triazole (13)
prepared through a dipolar cycloaddition (Eq. 1).21
NH
11
12
13
3
Slight
N
RS
RS
N
N
>50
1–2
ND
N
H
NH
ND
R
N
N
a,bSee Table 1; R = geranyl.
c
of the most potent synthetic QSMs (5, 10, and 11) were
tested next for their potential toxicity to mice following
IP injection of 1 ml of a 20 mM solution (ca. 4.4 mg of
analog in 0.5% v/v Tween 80 in sterile, nonpyrogenic sal-
ine). Attempts to use 0.1% Triton X-100 in place of
Tween 80 were precluded by toxicity. In an initial exper-
iment with 3 mice per group, no toxicity was observed
for at least 14 days following injection. In a follow-up
experiment with 15 mice per group, also observed for
at least 14 days, injection of 11 caused a slight peritonitis
and mild hepatomegaly in some mice. Therefore, ana-
logs 5 and 10 were chosen for further testing with a stan-
dard mouse model of candidiasis,10,11 employing tail
vein injection of C. albicans cells and IP injection of
the analogs.
TMSN3
CuI (cat.)
N
N
ð1Þ
N
H
13
Although thiotetrazole 9 and thioimidazole 12 were only
minimally active (IC50 > 50 lM), triazoles 10, 11, and 13
were potent QSMs (Table 3). It is notable that two of
the analogs (10 and 13) exhibit QSM activity equal to
E,E-farnesol, with three other analogs (5, 7, and 11) dis-
playing activities nearly as great. It is important to note
that none of the compounds listed in Tables 1 and 2 had
any inhibitory effects on the rate of cell growth, even at
the highest concentrations tested (P100 lm).
Forty-five mice were inoculated with 5 · 106 cells of C.
albicans A72 through tail vein injection; an additional
45 mice were inoculated with 2 · 106 cells. Each batch
of mice was divided into 3 groups of 15 mice each.
The chosen doses provide ca. 50% mortality after 3
days with strain C. albicans A72;10 virtually identical
results were obtained with each dose. Group 1 was gi-
ven a control injection (IP, 1 ml of 0.5% Tween 80 in
saline) while groups 2 and 3 were injected with ana-
2.4. Effect of heterocyclic QSMs on candidiasis
In selecting analogs for mouse assays, we eliminated the
nonyl tetrazole 7 due to limited water solubility. Three

R. Shchepin et al. / Bioorg. Med. Chem. 16 (2008) 1842–1848
1845
hances pathogenicity) and 10 (a potent quorum-sensing
molecule which does not enhance pathogenicity) may
provide a potent tool in understanding how farnesol
functions as a virulence factor in interactions with cellu-
lar macrophages and CD4 T cells to alter adaptive
immunity.12 The analogs also provide a means of estab-
lishing differences and similarities between farnesol’s ef-
fect on virulence and its ability to block both hyphal
development and biofilm formation.3,29 Along these
lines, it is interesting to note that an analog of triazole
10 bearing five additional carbons has been reported
to be a potent activator of protein kinase C.20 Of course,
the differences in the pathogenicity of C. albicans ob-
served in the presence of farnesol, 5, and 10 could also
be due to differing pharmacokinetics or their differing
influence on mouse cytokine production.
Figure 2. Influence of 5 and 10 on systemic candidiasis. Mouse
mortality in the presence and absence of tetrazole 5 and triazole 10
following inoculation with C. albicans A72 (5 · 106 cells) and IP
injection of 1 ml of: d (saline); m (20 mM thio-1,2,4-triazole 10 in 0.5%
Tween 80); j (20 mM tetrazole 5 in 0.5% Tween 80).
It is also intriguing to consider the basis for the failure of
farnesol, tetrazole 5, or thiotriazole 10 to exert any protec-
tive role on the course of systemic candidiasis. One possi-
bility is that the conditions of the experiment were not well
suited to test the potential of quorum-sensing molecules
as fungistatic therapeutics. Mycelia, the invasive form of
the fungi capable of penetrating host tissues to enter the
bloodstream, are disfavored by the addition of exogenous
farnesol or synthetic QSMs. However, the standard
experimental protocol calls for injection of C. albicans di-
rectly into the bloodstream (tail vein), potentially mini-
mizing the influence of any regimen that disfavors the
mycelial morphology. Three other clinical models of can-
didiasis, oral,30 vaginal,31 and gastrointestinal,32 may
provide a more relevant test of the potential of a ‘single
morphology’ therapy for control of fungal infections.
logs 5 and 10, respectively. Neither analog protected
the mice from candidiasis (Fig. 2). In the case of the
thiotriazole 10, mouse lethality closely paralleled that
observed for the control Tween 80 injections. In con-
trast, coadministration of tetrazole 5 enhanced the
pathogenicity of C. albicans A72 ca. 4- to 5-fold, al-
most exactly duplicating the effects of farnesol.10,11
The effect is sufficiently pronounced that the entire
group of 15 mice treated with 5 died within 60 h of
infection. It is noteworthy that 5 and 10 display
opposing effects in vitro (QSM) and in vivo (pathoge-
nicity). Thiotriazole 10 is ca. 7-fold more potent than
5 as a QSM but, unlike the tetrazole, it does not act
as a farnesol-mimic in promoting virulence. Thus, the
two analogs uncouple the in vitro and in vivo effects
of farnesol.
4. Conclusion
We have discovered a family of synthetic quorum-sens-
ing molecules for C. albicans possessing in vitro potency
comparable to E,E-farnesol, the natural signal. Investi-
gations of several of the analogs in a mouse model re-
veals no protection against systemic candidiasis, but
do demonstrate the ability to decouple quorum sensing
and virulence, suggesting this family of farnesol analogs
may prove extremely useful in dissecting the basis for
farnesol’s many biological activities.
3. Discussion
Farnesol is a biologically active, lipid-soluble molecule.
We have shown that it blocks the yeast to mycelium
transition in C. albicans,3 acts as a virulence factor in
a mouse model of systemic candidiasis,10,11 probably
by interfering with the normal progression of cytokine
expression in their immune response,12 and triggers
apoptosis in the fungus Aspergillus nidulans.22 Others
have shown that farnesol alters circadian rhythms in
Neurospora crassa and targets HMG CoA reductase
for proteolysis in both Saccharomyces cerevisiae and
Chinese hamster ovary cells.23–25 Farnesol has also been
shown to block calcium channels, stimulating cell differ-
entiation, and triggering apoptosis in other mammalian
cell lines.26 Farnesol-induced apoptosis has also been re-
ported in tobacco cells.27 In addition, farnesol has re-
cently been found to be a potent and selective
inhibitor of one class of monoamine oxidases
(MAO-B).28 The family of heterocyclic farnesol analogs
described above now provides a tool to discriminate be-
tween yeast-hyphal quorum sensing and pathogenicity.
In particular, the availability of molecules such as
tetrazole 5 (a potent quorum-sensing molecule which en-
5. Experimental
5.1. General procedures
Tetrahydrofuran (THF) was distilled from sodium/ben-
zophenone. Other solvents were used as received. All
reactions were performed under N2 in oven-dried or
flame-dried glassware. All new compounds were charac-
1
terized by H, 13C, IR, and HRMS. NMR spectra were
recorded at 400 MHz (1H) or 100 MHz (13C) in CDCl3
unless otherwise noted. IR spectra were collected on
NaCl plates. Melting points are uncorrected. Mass spec-
tra were acquired at the Nebraska Center for Mass
Spectrometry (Lincoln, NE). TLC analyses were con-
ducted on Analtech 0.25 mm GHLF plates, with ana-

1846
R. Shchepin et al. / Bioorg. Med. Chem. 16 (2008) 1842–1848
lytes visualized by UV, by dipping in 1% KMnO4 (spe-
cific for alkenes), or by charring after exposure to an
aqueous solution of ceric sulfate and ammonium molyb-
date (general indicator). (E)-5-(2-phenylethyenyl)-1H-
tetrazole (8) was purchased from Alfa Aesar (‘listed as
5-styryl-1H-1,2,3,4-tetrazole’) and used without purifi-
cation. Abbreviations: EA, ethyl acetate; hex; hexane;
HOAc, acetic acid; NBA, 3-nitrobenzylalcohol.
5.3.2. (E)-5,9-Dimethyldeca-4,8-dienenitrile. This com-
pound was prepared by a variant of a known proce-
dure.34 Into a ꢁ78 ꢁC solution of acetonitrile (0.52 ml,
10 mmol) in dry THF (10 ml) was added dropwise n-
BuLi (5 mmol, 2.0 ml, nominally 2.5 M in hexane), fol-
lowed by geranyl chloride (0.37 ml, 2.0 mmol). The reac-
tion was allowed to warm to rt, and, after 30 min,
quenched with water. The mixture was extracted with
10% EA/hex (150 ml) and the organic layer was dried
over Na2SO4. The crude residue obtained upon concen-
tration was purified by flash chromatography (5% EA/
Hex) to afford 0.28 g (79%) of the nitrile as a colorless
oil: Rf = 0.2 (5% EA/Hex).
5.2. Assay of quorum sensing
Assays for quorum sensing,3 toxicity to mice,10,11 and
mortality have been described;10,11 a brief overview fol-
lows: undifferentiated cells are inoculated to a final den-
sity of 1–2 · 107 cells/ml into a solution of N-acetyl
glucosamine (2.5 mM, buffered to pH 7) and incubated
for 4 h while shaking (200 rpm, 37 ꢁC), after which the
relative percentage of cells with germ tube formation
(GTF) is determined. Values from triplicate readings
are always within 10% of the mean value. The relative
percentage of cells displaying GTF, 95–98% for cells
grown on unsupplemented media, is reduced consider-
ably in the presence of farnesol or related QSMs.5
5.3.3. (E)-5-(4,8-Dimethylnona-3,7-dienyl)-1H-tetrazole
(5). Into a suspension of NaN3 (0.455 g, 7.00 mmol) in
toluene (5 ml) was added Et3NHCl (Fluka, 1.10 g,
8.00 mmol) followed by a solution of (E)-5,9-dimethyl-
deca-4,8-dienenitrile (see above, 2.00 mmol, 0.354 g) in
2 ml of toluene. The reaction mixture was held at reflux
for 24 h and then allowed to cool to rt. The mixture was
acidified to pH ꢀ 1 with 10% aq HCl and extracted with
20% EA/Hex (3· 100 ml). The combined organic layers
were concentrated under reduced pressure and the resi-
due purified by flash chromatography (Hex/EA/AcOH
10:2:1) to furnish 0.41 g (93%) of tetrazole 5 as a color-
5.3. Preparation of Substrates
1
The preparation of oximes 2 (anti, 2E,4E,6E,8E)-3,7,11-
trimethyldodeca-2,4,6,8,10-pentaenal oxime)13 and 3
less oil: Rf = 0.3 (5:1:1, Hex/AcOH/EA); H d 13.55 (br
s, 1H), 5.18 (t, 1H, J = 6.8), 5.04 (t, 1H, J = 6.5), 3.14 (t,
2H, J = 7.6), 2.56 (q, 2H, J = 7.2), 2.03 (m, 2H), 1.97 (m,
2H), 1.66 (s, 3H), 1.66 (s, 3H), 1.58 (s, 3H), 1.52 (s, 3H);
13C d 156.5, 138.4, 132.0, 124.0, 121.3, 39.5, 26.4, 26.1,
25.6, 23.7, 17.7, 15.9; IR 2917, 2625, 1558, 1441, 1376,
1253; HR-FABMS (3-NBA) calcd for C12H21N4
(MH+): 221.1766; found: 221.1771 (2.1 ppm).
(anti,
2E,6E)-3,7,11-trimethyldodeca-2,6,10-trienal
oxime)33 has been described.
5.3.1. anti- and syn-(2E,6E)-3,7,11-Trimethyldodeca-
2,6,10-trienal,O-methyl oxime [anti-4 and syn-4]. Into a
pyridine solution (5 ml) of methoxylamine hydrochlo-
ride (0.109 g, 1.3 mmol) was added (2E,6E)-3,7,11-trim-
ethyldodeca-2,6,10-trienal
(0.115 g,
0.52 mmol;
5.3.4. (E)-4,8-Dimethylnona-3,7-dienenitrile35. Into
a
available through oxidation of commercially available
farnesol with MnO2 or chromic acid). After 4 h the reac-
tion mixture was diluted with water (50 ml) and ex-
tracted with 30% Et2O/Hex (100 ml). The organic layer
was dried over Na2SO4 and concentrated under reduced
pressure to afford a mixture which was separated by
flash chromatography (2% Et2O/Hex) to afford a mix-
ture of anti-4 (0.080 g, 62%) and syn-4 (0.045 g, 38%).
solution of NaCN (0.196 g, 4.00 mmol) in ethanol
(10 ml) was added geranyl chloride (0.37 ml, 2.0 mmol).
The reaction mixture was stirred for 2 h, diluted with
water, and with 5% EA/Hex (twice). The separated or-
ganic layer was dried over Na2SO4 and concentrated
in vacuo. The residue was purified by column chroma-
tography (5% EA/Hex) to furnish 0.313 g (96%) of the
known nitrile: Rf = 0.3 (5% EA/Hex).
anti-4. Rf = 0.3 (5% EA/Hex); 1H d 8.01 (d, 1H,
J = 10.4), 5.92 (d, 1H, J = 9.6), 5.09 (m, 2H), 3.86 (s,
3H), 2.15 (m, 4H), 2.05 (m, 2h), 1.97 (m, 2H), 1.82 (d,
3H, J = 1.2), 1.68 (s, 3H), 1.60 (s, 6H); 13C d 147.7,
147.5, 135.8, 131.3, 124.2, 123.2, 117.8, 61.5, 40.0,
39.6, 26.7, 26.0, 25.7, 17.7, 17.2, 16.0; IR 2965, 2932,
2855, 1443, 1051; MS (HR FAB, 3-NBA) calcd for
C16H28NO (MH+): 250.2171; found 250.2160 (4.4 ppm).
5.3.5. (E)-5-(3,7-Dimethylocta-2,6-dienyl)-1H-tetrazole
(6). This compound was prepared from (E)-5-(3,7-dim-
ethylocta-2,6-dienenitrile (0.31 g, 1.9 mmol) by a similar
approach as for 5. The product (0.37 g, 95%) was a col-
orless crystal (mp = 52–54 ꢁC, uncorrected) used with-
out further purification: Rf = 0.3 (5:1:1, Hex/AcOH/
1
EA); H d 12.8 (br s, 1H), 5.45 (dt, 1H, J = 7.2, 1.2),
5.05 (m, 1H), 3.82 (d, 2H, J = 7.6), 2.05 (m, 4H), 1.69
(s, 3H), 1.68 (s, 3H), 1.57(s, 3H); 13C d 156.0, 141.5,
132.2, 123.9, 115.7, 39.3, 26.2, 25.6, 22.5, 17.6, 16.2;
IR 3417, 2971, 2916, 2732, 2253; HR-FABMS (3-
NBA) calcd for C11H19N4 (MH+): 207.1610; found:
207.1612 (1.3 ppm).
1
syn-4. Rf = 0.2 (5% EA/Hex) H d 7.27 (d, 1H, J = 10),
6.41 (dd, 1H, J = 10, 1.2), 5.09 (m, 2H), 3.90 (s, 3H),
2.17 (m, 4H), 2.06 (m, 2H), 1.98 (m, 2H), 1.87 (d, 3H,
J = 1.2), 1.69 (d, 3H, J = 0.8), 1.61 (s, 6H); 13C d
149.8, 144.8, 135.9, 131.4, 124.2, 123.2, 113.3, 61.8,
40.2, 39.7, 26.7, 26.1, 25.7, 17.7, 17.1, 16.0; IR 2964,
2932, 2855, 1639, 1441, 1056; MS (HR FAB, 3-NBA)
calcd for C16H28NO (MH+): 250.2171; found 250.2166
(1.8 ppm).
5.3.6. 5-Nonyl-1H-tetrazole (7). This compound was pre-
pared from decanonitrile (0.38 ml, 2.0 mmol) by a simi-
lar method as employed for synthesis of 5 and 6. The
crude material (0.359 g, 91%) displayed spectra identical

R. Shchepin et al. / Bioorg. Med. Chem. 16 (2008) 1842–1848
1847
to literature reports36 and was used without further
purification: Rf = 0.2 (5:1:1, Hex/AcOH/EA).
NMR and was utilized without further purification:
Rf = 0.4 (50% EA/Hex); H d 12.06 (br s, 1H), 7.17 (s,
1
1H), 5.29 (dt, 1H, J = 1.2, 7.6), 5.03 (m, 1H), 3.62 (d,
2H, J = 8), 2.05–1.92 (m, 4H), 1.65 (s, 3H), 1.56 (s, 3H),
1.46 (s, 3H); 13C d 140.5, 139.6, 131.6, 123.9, 123.8,
119.0, 39.4, 33.4, 26.2, 25.6, 17.6, 15.6; IR 3094, 2916,
2637, 1447, 1328, 1098; HR-FABMS (3-NBA) calcd for
C13H20N2S (MH+): 237.1313; found 237.1310 (1.1 ppm).
5.3.7. (E)-5-(3,7-Dimethylocta-2,6-dienylthio)-1H-tetra-
zole (9). Into a solution of 5-mercapto-1H-tetrazole
(0.20 g, 2.0 mmol) in ethyl alcohol (10 ml) were added
K2CO3 (0.55 g, 4.0 mmol) and geranyl chloride
(0.19 ml, 1.0 mmol). After 12 h, the reaction mixture
was acidified to pH 7 and extracted with 50% EA/Hex
(3· 50 ml). The combined organic extracts were dried
with Na2SO4. The residue obtained upon concentration
in vacuo was purified by flash chromatography (5:1:1
Hex/AcOH/EA) to furnish 0.213 g (89%) of mercapto-
tetrazole 9 as a colorless oil: Rf = 0.2 (5:1:1, Hex/
5.3.11. (E)-4-(4,8-Dimethylnona-3,7-dienyl)-1H-1,2,3-tri-
azole (13). Into a solution of (E)-6,10-dimethylundeca-
5,9-dien-1-yne37 (0.176 g, 1.00 mmol) in 9:1 DMF/
MeOH (10 ml) was added TMSN3 (1.3 ml, 10 mmol)
followed by CuI (0.1 g, 0.5 mmol).21 The mixture was
held at 100 ꢁC for 12 h and then allowed to cool to rt.
The reaction mixture was diluted with water and then
extracted with 20% EA/Hex (3· 70 ml). The combined
organic layers were dried with Na2SO4 and concen-
trated. The residue was purified by flash chromatogra-
phy (20% EA/Hex) to furnish 0.168 g (77%) of triazole
13: Rf = 0.2 (20% EA/Hex); 1H (CD3OD) d 7.53 (s,
1H), 5.15 (dt, 1H, J = 1.2, 7.2), 5.05 (m, 1H), 2.76 (t,
2H, J = 7.6), 2.35 (q, 2H, J = 7.2), 2.04 (m, 2H), 1.98
(m, 2H), 1.65 (s, 3H), 1.57 (s, 3H), 1.54 (s, 3H); 13C
(CD3OD, 100 MHz) d 144.2, 137.7, 132.3, 128.7,
125.4, 124.2, 40.9, 28.8, 27.8, 26.1, 24.2, 17.9, 16.3; IR
3140, 2919, 2856, 1448, 1377, 1108, 983, 838; HR-FAB-
MS (3-NBA) calcd for C13H21N3 (MH+): 220.1814;
found: 220.1811 (1.1 ppm).
1
AcOH/EA); H d 13.0 (br s, 1H), 5.37 (t, 1H, J = 7.6),
5.03 (t, 1H, J = 6.4), 3.97 (d, 2H, J = 8), 2.03 (m, 4H),
1.68 (s, 3H), 1.66 (s, 3H), 1.56 (s, 3H); 13C d 155.1,
143.4, 132.1, 123.5, 116.8, 39.4, 31.4, 26.2, 26.6, 17.7,
16.2; IR 3055, 2917, 2727, 2560, 1504, 1447, 1038;
HR-FABMS (3-NBA) calcd for C11H19N4 (MH+):
239.1330; found 239.1336 (2.2 ppm).
5.3.8. (E)-3-(3,7-Dimethylocta-2,6-dienylthio)-1H-1,2,4-
triazole (10). Into a solution of 1H-1,2,4-triazole-3-thiol
(0.152 g, 1.50 mmol) in ethyl alcohol (10 ml) were added
K2CO3 (0.276 g, 2.00 mmol) and geranyl chloride
(0.19 ml, 1.0 mmol). After 12 h, the reaction mixture
was worked up as for 9 and the residue purified by flash
chromatography (40% EA/Hex) to furnish 0.235 g (99%)
1
of thiotriazole 10: Rf = 0.3 (50% EA/Hex); H d 13.70
(br s, 1H), 8.18 (s, 1H), 5.33 (dt, 1H, J = 1.2, 2), 5.02
(m, 1H), 3.80 (d, 2H, J = 8), 2.00 (m, 4H), 1.64 (s,
3H), 1.56 (s, 3H); 13C d 156.4, 147.7, 141.5, 131.8,
131.8, 123.6, 118.1, 39.4, 31.2, 26.2, 25.6, 17.6, 16.0;
IR 3111, 2924, 1449, 1276, 910, 734; MS (HR-FABMS,
3-NBA) calcd for C12H20N3S, (MH+): 238.1378; found:
238.1383 (2.3 ppm).
Acknowledgments
This work was supported by the University of Nebraska
Tobacco Settlement Biomedical Research Enhancement
Fund, the John C. and Nettie V. David Memorial Trust
Fund, and the Farnesol and Candida albicans Research
Fund, University of Nebraska Foundation. NMR spec-
tra were acquired, in part, on spectrometers purchased
with NSF support (MRI 0079750 and CHE 0091975).
A portion of this research was conducted in facilities
remodeled with support from NIH (RR016544-01).
We thank Jessica Dussault for assistance with graphics.
5.3.9. (E)-4-(3,7-Dimethylocta-2,6-dienylthio)-1H-1,2,3-
triazole (11). Into a 0 ꢁC solution of the sodium salt of
4-mercapto-1H-1,2,3-triazole (0.262 g, 2.2 mmol) in eth-
anol (10 ml) was added geranyl chloride (0.19 ml,
1.0 mmol). The reaction mixture was stirred for 2 h
and then allowed to warm to RT over the course of
1 h. The reaction mixture was diluted with water and
the mixture extracted with 20% EA/Hex (3· 50 ml).
The combined organic extracts were dried with Na2SO4
and concentrated, with the residue purified by flash
chromatography (20% EA/Hex) to furnish 0.218 g
References and notes
1. Gudlaugsson, O.; Gillespie, S.; Lee, K.; Vande Berg, J.;
Hu, J.; Messer, S.; Herwaldt, L.; Pfaller, M.; Diekema, D.
Clin. Infect. Dis. 2003, 37, 1172–1177.
1
(92%) of thiotriazole 11. Rf = 0.2 (20% EA/Hex); H d
14.23 (br s, 1H), 7.72 (s, 1H), 5.28 (d of t, 1H, J = 1.2,
8), 5.02 (m, 1H), 3.54 (d, 2H, J = 7.6), 1.97 (m, 4H),
1.64 (d, 3H, J = 0.8), 1.55 (s, 3H), 1.47 (s, 3H); 13C d
140.7, 139.3, 133.5, 131.6, 123.7, 118.8, 39.3, 33.0,
26.2, 25.5, 17.6, 15.7; IR 3130, 2916, 1661, 1448, 1228,
1109, 839; HR-FABMS (3-NBA) calcd for C12H20N3S,
MH+: 238.1378; found: 238.1369 (3.8 ppm).
2. Macphail, G. L.; Taylor, G. D.; Buchanan-Chell, M.;
Ross, C.; Wilson, S.; Kureishi, A. Mycoses 2002, 45, 141–
145.
3. Hornby, J. M.; Jensen, E. C.; Lisec, A. D.; Tasto, J. J.;
Jahnke, B.; Shoemaker, R.; Dussault, P.; Nickerson, K.
W. Appl. Environ. Microbiol. 2001, 67, 2982–2992.
4. Nickerson, K. W.; Atkin, A. L.; Hornby, J. M. Appl.
Environ. Microbiol. 2006, 72, 3805–3813.
5. Mosel, D. D.; Dumitru, R.; Hornby, J. M.; Atkin, A. L.;
Nickerson, K. W. Appl. Environ. Microbiol. 2005, 71,
4938–4940.
6. Shchepin, R.; Hornby, J. M.; Burger, E.; Niessen, T.;
Dussault, P.; Nickerson, K. W. Chem. Biol. 2003, 10, 743–
750.
5.3.10. (E)-2-(3,7-Dimethylocta-2,6-dienylthio)-1H-imid-
azole (12). This compound was prepared from 2-merca-
ptoimidazole (0.12 g, 1.2 mmol) and geranyl chloride
(0.19 ml, 1.0 mmol) in 96% yield (0.227 g) by a similar
procedure as for 9. The crude product was pure by

1848
R. Shchepin et al. / Bioorg. Med. Chem. 16 (2008) 1842–1848
7. Fuqua, W. C.; Winans, S. C.; Greenberg, E. P. J.
Bacteriol. 1994, 176, 269–275.
24. Shearer, A. G.; Hampton, R. Y. EMBO J. 2005, 24, 149–
159.
8. Saville, S. P.; Lazzell, A. L.; Monteagudo, C.; Lopez-
Ribot, J. L. Eukaryot. Cell 2003, 2, 1053–1060.
9. Hornby, J. M.; Nickerson, K. W. Antimicrob. Agents
Chemother. 2004, 48, 2305–2307.
25. Meigs, T. E.; Simoni, R. D. Arch. Biochem. Biophys. 1997,
345, 1–9.
26. Edwards, P. A.; Ericsson, J. Ann. Rev. Biochem. 1999, 68,
157–185.
10. Navarathna, D. H.; Hornby, J. M.; Hoerrmann, N.;
Parkhurst, A. M.; Duhamel, G. E.; Nickerson, K. W. J.
Antimicrob. Chemother. 2005, 56, 1156–1159.
11. Navarathna, D. H.; Hornby, J. M.; Krishnan, N.;
Parkhurst, A.; Duhamel, G. E.; Nickerson, K. W. Infect.
Immun. 2007, 75, 1609–1618.
27. Hemmerlin, A.; Bach, T. J. Plant Physiol. 2000, 123, 1257–
1268.
´
28. (a) Binda, C.; Hubalek, F.; Li, M.; Castagnoli, N.;
Edmondson, D. E.; Mattevi, A. Neurology 2006, 67(Suppl.
2), S5–S7; (b) Khalil, A. A.; Davies, B.; Castagnoli, N.
´
Bioorg. Med. Chem. 2006, 14, 3392–3398; (c) Hubalek, F.;
12. Navarathna, D. H.; Nickerson, K. W.; Duhamel, G. E.;
Jerrels, T. R.; Petro, T. M. Infect. Immun. 2007, 75, 4006–
4011.
13. Shchepin, R.; Dumitru, R.; Nickerson, K. W.; Lund, M.;
Dussault, P. H. Chem. Biol. 2005, 12, 639–641.
14. Herr, R. J. Bioorg. Med. Chem. 2002, 10, 3379–3393.
15. Thornber, C. W. Chem. Soc. Rev. 1979, 8, 563–580.
Binda, C.; Khalil, A.; Li, M.; Mattevi, A.; Castagnoli, N.;
Edmondson, D. E. J. Biol. Chem. 2005, 280, 15761–15766.
29. Ramage, G.; Saville, S. P.; Wickes, B. L.; Lopez-Ribot, J.
L. Appl. Environ. Microbiol. 2002, 68, 5459–5463.
30. Elahi, S.; Pang, G.; Clancy, R.; Ashman, R. B. Infect.
Immun. 2000, 68, 5771–5777.
31. Cheng, G.; Woznaik, K.; Wallig, M. A.; Fidel, P. L., Jr.;
Trupin, S. R.; Hoyer, L. L. Infect. Immun. 2005, 73, 1656–
1663.
16. Furmeier, S.; Metzger, J. O. Eur. J. Org. Chem. 2003, 885–
¨
893.
17. We thank a reviewer for this suggestion.
18. Hogan, D. A.; Vik, A.; Kolter, R. Mol. Microbiol. 2004,
54, 1212–1223.
19. Kume, M.; Kubota, T.; Kimura, Y.; Nakashimizu, H.;
Motokawa, K. J. Antibiot. 1993, 46, 316–330.
20. Gilbert, B. A.; Lim, Y.-H.; Ding, J.; Badwey, J. A.;
Rando, R. R. Biochemistry 1995, 34, 3916–3920.
21. Jin, T.; Kamijo, S.; Yamamoto, Y. Eur. J. Org. Chem.
2004, 3789–3791.
22. Semighini, C. P.; Hornby, J. M.; Dumitru, R.; Nickerson,
K. W.; Harris, S. D. Mol. Microbiol. 2006, 59, 753–764.
23. Granshaw, T.; Tsukamoto, M.; Brody, S. J. Biol. Rhythms
2003, 18, 287–296.
32. Samonis, G.; Anaissie, E. J.; Rosenbaum, B.; Bodey, G. P.
Infect. Immun. 1990, 58, 1514–1517.
33. Zheng, G. C.; Kakisawa, H. Bull. Chem. Soc. Jpn. 1989,
62, 602–604.
34. Shin, D. S. J. Kor. Chem. Soc. 1992, 36, 579–583.
35. (a) Mori, K.; Funaki, Y. Tetrahedron 1985, 41, 2369–2377;
(b) Takahashi, T.; Tomida, S.; Doi, T. Synlett 1999, 644–
646.
36. Kiuchi, F.; Miyashita, N.; Tsuda, Y.; Kondo, K.;
Yoshimura, H. Chem. Pharm. Bull. 1987, 35, 2880–2886.
37. Lipshutz, B. H.; Bulow, G.; Fernandez-Lazaro, F.; Kim,
S. K.; Lowe, R.; Mollard, P.; Stevens, K. L. J. Am. Chem.
Soc. 1999, 121, 11664–11673.