Quinazolinone fungal efflux pump inhibitors. Part 3: (N-methyl)piperazine variants and pharmacokinetic optimization

Article abstract of DOI:10.1016/j.bmcl.2007.02.047

Further structure-activity relationships of a novel series of fungal efflux pump inhibitors with respect to potentiation of the activity of fluconazole against strains of C. albicans and C. glabrata over-expressing ABC-type efflux pumps are systematically

Full text of DOI:10.1016/j.bmcl.2007.02.047

Bioorganic & Medicinal Chemistry Letters 17 (2007) 2802–2806
Quinazolinone fungal efflux pump inhibitors. Part 3:
(N-methyl)piperazine variants and pharmacokinetic optimization
William J. Watkins,a,* Lee Chong,a Aesop Cho,a Ramona Hilgenkamp,a
Maria Ludwikow,a Negar Garizi,a Nadeem Iqbal,b John Barnard,b Rajeshwar Singh,b
Deidre Madsen,a Karen Lolans,a Olga Lomovskaya,a Uma Oza,a
Padmapriya Kumaraswamy,a Andrea Blecken,a Shuang Bai,a
David J. Loury,a David C. Griffitha and Michael N. Dudleya
aEssential Therapeutics, Inc., 850 Maude Ave., Mountain View, CA 94043, USA
bNAEJA Pharmaceutical Inc., #2 4290-91 A St., Edmonton, Alta., Canada T6E 5V2
Received 7 February 2007; accepted 23 February 2007
Available online 27 February 2007
Abstract—Further structure–activity relationships of a novel series of fungal efflux pump inhibitors with respect to potentiation of
the activity of fluconazole against strains of C. albicans and C. glabrata over-expressing ABC-type efflux pumps are systematically
explored. Rat protein binding and pharmacokinetics of selected analogues are reported.
ꢀ 2007 Elsevier Ltd. All rights reserved.
The incidence of systemic fungal infections due to
organisms that are not susceptible to commonly used
antifungal agents such as fluconazole continues to in-
crease, and the association of this phenomenon with
overproduction of multidrug transporters of the ABC
and Major Facilitator superfamilies is now well estab-
lished.1,2 We have previously reported the discovery of
a series of inhibitors of efflux pumps in Candida spp.
exemplified by MC-5805 (1, Fig. 1).3 Structure–activity
relationships for in vitro potentiation of fluconazole
(FLU) versus Candida albicans and Candida glabrata
for 3-(N0-methyl-piperazinyl)quinazolinones of this class
have also been described.4 These studies demonstrated
that incorporation of an N-methyl-piperazine motif into
early screening hits broadened the spectrum of inhibi-
tion of ABC-type transporters to include all clinically
relevant Candida spp., and culminated in the identifica-
tion of 2 as a 3-(N-methyl-piperazinyl)quinazolinone
with optimal activity for the potentiation of fluconazole
in vitro.
O
N
N
O
N
N
Herein the results of further studies of analogues vary-
ing the N-methyl-piperazine moiety are reported, both
in terms of their in vitro activity and with regard to
pharmacokinetics in rats.
N
N
N
1
N
2
F
H
N
H
N
N
O
Cl
N
O
O
O
O
O
F
All but one of the analogues reported were synthesized
by acylation of the piperazine that is the product of
the route shown in Scheme 1.
Figure 1. Prototypical leads.
The route developed for the synthesis of earlier ana-
logues4 was adapted to allow for facile variation of
piperazine substituents by demethylation of the pipera-
zine and protection prior to functionalization of the
2-ethyl substituent on the quinazolinone. Acylations
were accomplished using conventional reagents.
Keywords: Quinazolinone; Antifungal; Efflux; EPI; Efflux pump inhi-
bitor; Synergy; Fluconazole; Candida.
*
Corresponding author. Present address: Gilead Sciences, 333 Lake-
side Drive, Foster City, CA 94404, USA. Tel: +1 650 522 5384; fax:
+1 650 522 5899; e-mail: will.watkins@gilead.com
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.02.047

W. J. Watkins et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2802–2806
Table 1. Variants of the N-methyl-piperazine moiety in 1
2803
O
O
N
N
O
N
OtBu
O
(a), (b), (c), (d)
N
N
R
N
N
N
N
N
Br
H
N
O
N
O
(e), (f)
O
F
O
Compound
R
MPC8 (lg/mL)
O
N
NH
O
N
OtBu
C. albicans
C. glabrata
N
N
(g)
N
N
N
3
4
4
>32
>32
2
4
32
N
N
H
N
H
N
N
O
N
O
O
O
O
F
O
F
N
N
N
NH
Scheme 1. Reagents and conditions: (a) ClCO2CH(Cl)Me, PhCH3,
95 ꢁC, 38%; (b) 6N HCl, 80 ꢁC; (c) BOC2O, THF, 45 ꢁC; (d) Br2,
NaOAc, AcOH, 50 ꢁC, 80% over 3 steps; (e) 2,4-dimethoxyaniline,
EtOAc, reflux; (f) 4-fluorophenylisocyanate, ClCH2CH2Cl, rt, 84%; (g)
TFA, CH2Cl2, rt, 94%.
5
>32
4
6
N
O
CHO
The piperidine derivative 5 was made by the route
reported earlier,3 using 1-aminopiperidine instead of
1-amino-4-methylpiperazine.
7
0.25
0.5
N
OEt
NHEt
NH2
N
O
New compounds were initially synthesized as isomeric
mixtures via the racemic bromide. The single isomer
32 was generated using lactic acid (instead of propionic
acid) as the ultimate starting material, by an analogous
route to that previously described.3
8
1
8
1
16
4
N
N
N
O
9
N
O
H
H
N
The activity of analogues was conveniently expressed as
the minimum potentiation concentration (MPC8), which
is the lowest concentration achieving an 8-fold reduction
in FLU MIC. An initial survey of the activity of variants
of the N-methyl-piperazine moiety is summarized in
Table 1.
10
>32
N
N
distal to the piperazine is required, and that a basic
nitrogen (particularly a primary or secondary amine)
is disadvantageous even in analogues where this
requirement is fulfilled. Thus, the most potent ana-
logues against both organisms are the amide 16 and
urea 18. Even in C. glabrata, where a basic nitrogen
is better tolerated, the activity is optimal in neutral
compounds.
Both the des-methyl analogue 4 and the piperidine 5 are
significantly less active than 3, but the activity is restored
in amide or carbamate derivatives, exemplified by 6 and
7. The urea variant 8 of the ethyl carbamate also shows
good activity, but removal of the ethyl group (as in 9)
causes a reduction in potency. These data demonstrate
the benefit of incorporating an acyl motif as an H-bond
acceptor on the piperazine 4-position for retaining activ-
ity against both C. albicans and C. glabrata. Unfortu-
nately, this tactic removes the basic entity that
provides the solubility benefit in 1 relative to earlier
leads.
Accepting this design constraint, we were attracted to
the profile of the alanine derivative 23, which exhibited
good activity against both organisms while retaining
reasonable physical characteristics, and explored closely
related variants of this compound (Table 3).
It is also notable that in analogues 3–9 the SAR in
C. glabrata parallels that in C. albicans. In the proline
derivative 10, by contrast (and unlike in earlier screening
hits,3) the activity in C. glabrata is retained selectively.
This intriguing profile, coupled with the solubility bene-
fit arising from the re-introduction of a basic entity, led
us to explore related analogues in more detail (Table 2).
Activity is only marginally dependent upon the alanine
stereochemistry, but branching at this position is re-
quired (25 vs 23/24). As the lipophilicity increases, the
activity improves (26–29). The combination of the urea
motif with the methyl carbamate again gives better po-
tency (30), with C. albicans showing greater sensitivity
to the stereochemical disposition of the hydrophilic
entity (31). Overall, we concluded that 23 represents
the preferred piperazine substituent, based upon its po-
tency, balanced spectrum, and reasonable lipophilicity.
Once again it is evident that in order to achieve good
potency against C. albicans, a strong H-bond acceptor

2804
W. J. Watkins et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2802–2806
Table 2. Variants of 10
Table 3. Non-basic variants of 10
R
R
O
N
O
N
N
N
N
N
N
N
H
N
H
N
N
O
N
O
O
O
O
O
F
F
Compound
R
MPC8 (lg/mL)
Compound
R
MPC8 (lg/mL)
C. albicans
C. glabrata
C. albicans C. glabrata
O
O
O
O
O
O
H
H
H
H
H
10
11
12
13
>32
>32
4
23
24
25
26
27
2
4
0.25
0.5
2
N
NHCO2Me
NHCO2Me
H
N
2
O
O
0.5
0.125
2
>32
O
N
NHCO2Me
32
0.5
0.5
2
N
CO2Me
O
O
NHCO2Me
NHCO2Me
1
O
H
H
14
1
0.25
N
H
N
28
29
32
8
2
15
16
17
18
32
2
HO
O
O
0.25
0.125
0.25
<0.03
NHCO2Me
2
O
O
HO
O
8
N
30
31
0.25
0.5
N
N
NHCO2Me
NHCO2Me
0.0625
4
O
N
4
0.25
O
19
20
1
4
N
O
O
H
N
>32
The protein binding of the analogues measured was in a
reasonable range, but the sensitivity to metabolic degra-
dation varied widely. The more lipophilic (and more ac-
tive) compounds (18, 30) tended to be less stable upon
microsomal incubation. Notably, the rate of degrada-
tion was sensitive to the stereochemistry of the side
chain (23 vs 24; 30 vs 31).
21
22
23
>32
>32
2
0.5
4
N
H
H
O
O
NH2
The superior activity and pharmacological properties of
the (L)-alanine carbamate in 23 were combined with the
structural features of other regions of the lead that had
previously been shown to be optimal,4 culminating in
the synthesis of 32. The in vitro and in vivo characteris-
tics of this compound are compared with those of the
original lead 1 in Table 5.
0.25
NHCO2Me
In parallel with measurement of fluconazole potentia-
tion in vitro, pharmacological assays expected to be pre-
dictive of pharmacokinetic profiles were also performed
and are summarized in Table 4.
The increased potency and improved pharmacokinetic
profile of 32 over 1 translated into better efficacy as a

W. J. Watkins et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2802–2806
2805
Table 4. Rat serum protein binding and hepatocyte stability data for
selected analogues
such as 1 have been outlined and optimized. By carefully
modulating the hydrophilicity of the piperazinyl substi-
tuent, it proved possible to identify potent analogues
with a balanced spectrum of activity in clinically rele-
vant Candida spp., combined with reasonable protein
binding and metabolic stability. These studies culmi-
nated in the synthesis of 32, an excellent tool for studies
into the feasibility of potentiation of the activity of
fluconazole in animal models of fungal infection.
R
O
N
N
N
N
H
N
N
O
O
O
F
Chemicals. All analogues were purified by reverse-phase
MPLC using
Compound
R
Protein
bindinga (%) stabilitya
(% remaining
Hepatocyte
a Phenomenex Synergy Hydro-RP
50 · 21.2 mm column, eluting with MeCN/water (10–
100%, ramped over 20 min) at 20 mL/min, and tested
as the TFA or methanesulfonic acid salt; no precipita-
tion was observed at the test concentrations employed
for assessment of antifungal activity. The structural
identity of each compound was confirmed by 1H
NMR and MS.
after 24 h)
3
10
18
23
24
30
31
74
18
Me
O
H
H
77
77
0
N
O
O
The purity of the single isomer 32 was assessed as >95%
Chiralpack AD
ND
82
N
by chromatography using
a
250 · 4.6 mm column, eluting with 30% 2-propanol in
hexanes.
39
7
NHCO2Me
NHCO2Me
In vitro potentiation of fluconazole. FLU MICs for
C. albicans strain YEM15 (over-expressing both
CDR1 and CDR2) and C. glabrata strain YEM19
(over-expressing both CgDR1 and CgDR2) were mea-
sured in the presence and absence of varying concentra-
tions of efflux pump inhibitor (checkerboard format).
Drugs were tested in RPMI 1640 according to CLSI ref-
erence methods.5 None of the analogues displayed
intrinsic antifungal activity (MIC > 32 lg/ml). Not all
of the analogues reported in this study were assessed
in full checkerboard format; an abbreviated method of
evaluation involved testing varying concentrations of
efflux pump inhibitor in the presence of fixed concentra-
tions of FLU equivalent to 1/8 and 1/32 MIC. This
method was validated on a set of eight analogues by
comparison with the MPC8 values derived from full
checkerboard studies; the results were identical.
O
O
82
ND
ND
4
N
N
NHCO2Me
NHCO2Me
O
23
ND, not determined.
a See Serum protein binding and hepatocyte stability.
potentiator of the activity of fluconazole in vivo. These
experiments will be the subject of a future publication.
In this and the previous studies,3,4 the absolute struc-
tural requirements for satisfactory efflux pump inhibi-
tion leading to potentiation of the antifungal activity
of fluconazole in vitro by 3-(piperazinyl)quinazolinones
Serum protein binding and hepatocyte stability. Serum
protein binding was measured by adding compounds
Table 5. In vitro and in vivo profiles of 1 and 32
O
NHCO2Me
O
N
N
O
N
N
N
N
N
N
F
H
N
H
N
Cl
N
O
N
O
O
O
O
O
F
32
1
Compound
MPC8 (lg/mL)
Rat protein binding (%)
Rat CL (L/h/kg)
Rat CLfree(L/h/kg)
C. albicans
C. glabrata
1
32
1
0.5
4
0.0625
98
87
5
7.5
225
60

2806
W. J. Watkins et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2802–2806
to fresh rat serum at three concentrations then separat-
ing bound from unbound by ultrafiltration using an
Amicon Centrifree Micropartition device. Unbound fil-
trates were assayed using LC/MS/MS.
References and notes
1. (a) Klepser, M. E. Pharmacotherapy 2001, 21, 124S; (b)
Stevens, D. A.; Holmberg, K. Curr. Opin. Anti-Infect.
Invest. Drugs 1999, 1, 306; (c) Vanden Bossche, H.;
Dromer, F.; Improvisi, I.; Lozano-Chiu, M.; Rex, J. H.;
Sanglard, D. Med. Mycol. 1998, 36, 119.
2. (a) Katiyar, S. K.; Edlind, T. D. Med. Mycol. 2001, 39, 109;
(b) Barchiesi, F.; Calabrese, D.; Sanglard, D.; di Francesco,
L. F.; Caselli, F.; Giannini, D.; Giacometti, A.; Gavaudan,
S.; Scalise, G. Antimicrob. Agents Chemother. 2000, 44,
1578; (c) Sanglard, D.; Ischer, F.; Calabrese, D.; Maj-
cherczyk, P. A.; Bille, J. Antimicrob. Agents Chemother.
1999, 43, 2753; (d) Sanglard, D.; Kuchler, K.; Ischer, F.;
Pagani, J. L.; Monod, M.; Bille, J. Antimicrob. Agents
Chemother. 1995, 39, 2378.
3. Lemoine, R. C.; Glinka, T. W.; Watkins, W. J.; Cho, A.;
Yang, J.; Iqbal, N.; Singh, R.; Madsen, D.; Lolans, K.;
Lomovskaya, O.; Oza, U.; Dudley, M. N. Bioorg. Med.
Chem. Lett. 2004, 14, 5127.
4. Watkins, W. J.; Lemoine, R. C.; Chong, L.; Cho, A.;
Renau, T. E.; Kuo, B.; Wong, V.; Ludwikow, M.; Garizi,
N.; Iqbal, N.; Barnard, J.; Jankowska, R.; Singh, R.;
Madsen, D.; Lolans, K.; Lomovskaya, O.; Oza, U.; Dudley,
M. N. Bioorg. Med. Chem. Lett. 2004, 14, 5133.
Freshly isolated rat hepatocytes attached to collagen
coated 24-well plates were purchased from Cedra
Corporation (Austin, TX) and maintained in nutrient
culture medium. Dosing solutions of each compound
were prepared at 1 and 5 lg/mL, and added to dupli-
cate cultures. At 3, 6, and 24 h after the start of
exposure, samples of the culture medium were taken
and added to methanol, vortexed, and centrifuged.
The supernatant was quantified using an LC/UV
system.
Pharmacokinetic studies in rats. Serum pharmacokinet-
ics were evaluated in male Sprague-Dawley rats (3 per
compound) with implanted femoral and jugular vein
cannulas. Compounds were administered by bolus injec-
tion via the femoral vein cannula at dose of 5 mg/kg.
Blood samples were collected for up to 8 h from the jug-
ular vein cannula and were assayed for concentration of
compound using LC/MS/MS. Total clearance (CL) val-
ues were determined using a two-compartment pharma-
cokinetic model.
5. NCCLS (now CLSI). Reference method for broth dilution
antifungal susceptibility testing of yeasts: approved stan-
dard. NCCLS document M27-A. 1997.