Discovery of a new boron-containing antifungal agent, 5-fluoro-1,3-dihydro- 1-hydroxy-2,1-benzoxaborole (AN2690), for the potential treatment of onychomycosis

Article abstract of DOI:10.1021/jm0603724

A structure-activity relationship investigation for a more efficacious therapy to treat onychomycosis, a fungal infection of the toe and fingernails, led to the discovery of a boron-containing small molecule, 5-fluoro-1,3-dihydro- 1-hydroxy-2,1-benzoxaborole (AN2690), which is currently in clinical trials for onychomycosis topical treatment.

Full text of DOI:10.1021/jm0603724

© Copyright 2006 by the American Chemical Society
Volume 49, Number 15
July 27, 2006
Letters
Discovery of a New Boron-Containing Antifungal
Agent, 5-Fluoro-1,3-dihydro-1-hydroxy-2,1-
benzoxaborole (AN2690), for the Potential
Treatment of Onychomycosis
Figure 1. Boron-containing small-molecule library.
Stephen J. Baker,* Yong-Kang Zhang, Tsutomu Akama,
Agnes Lau, Huchen Zhou, Vincent Hernandez, Weimin Mao,
M. R. K. Alley, Virginia Sanders, and Jacob J. Plattner
dermatophytes showed these compounds had broad spectrum
activity against all these fungal pathogens including the major
dermatophytes that cause onychomycosis, T. rubrum and T.
mentagrophytes. Herein, we report the focused medicinal
chemistry effort to develop a first-in-class boron-containing
antifungal agent for the topical treatment of onychomycosis.
The dihydrobenzoxaborole class (2) emerged as a result of a
directed effort to expand the chemical diversity of the borinic
acid quinoline esters (1), all of which contained a boron-
nitrogen dative bond.16
Dihydrobenzoxaboroles bearing aryl, heteroaryl, or vinyl
substituents at the 1-position (9-15) were synthesized starting
from 2-bromo-5-fluorobenzaldehyde (3) or 2-bromobenzyl
alcohol (4a) as shown in Scheme 1. The hydroxy groups of 4a
and 4b were protected as the methoxymethyl ether to give 5a
and 5b, respectively. Compounds 5 were treated with butyl-
lithium at -78 °C, and the anion formed was trapped by a
boronic acid ethylene glycol ester (7), prepared from the
corresponding boronic acid and ethylene glycol, to give the
borinic acid (8). For 11 and 15, intermediate 5b was converted
into the glycol ester 6, which was reacted with Grignard reagents
to give 8. Finally, the protecting group was removed under acidic
conditions and the free alcohol spontaneously cyclized to give
the target compounds (9-15).
Anacor Pharmaceuticals Inc., 1060 East Meadow Circle,
Palo Alto, California 94303
ReceiVed March 29, 2006
Abstract: A structure-activity relationship investigation for a more
efficacious therapy to treat onychomycosis, a fungal infection of the
toe and fingernails, led to the discovery of a boron-containing small
molecule, 5-fluoro-1,3-dihydro-1-hydroxy-2,1-benzoxaborole (AN2690),
which is currently in clinical trials for onychomycosis topical treatment.
One of the most common fungal infections remains one of
the most difficult to treat.1,2 Onychomycosis is a fungal infection
of the toenails and fingernails and has an incidence of 14% of
the population in the U.S.3 In diabetic and elderly patients the
incidents ranges from 33% to 50%.3,4
Onychomycosis is caused mainly by dermatophytes, a class
of fungus that dwells on skin, hair, and nails and is the cause
of other cutaneous fungal infections such as athlete’s foot.5 The
major dermatophytes involved are Trichophyton rubrum and
Trichophyton mentagrophytes, accounting for approximately
60-90% of all cases.1,6,7 There are few treatments available,
and these include oral terbinafine, oral itraconazole, topical
ciclopirox, and topical amorolfine. However, topical treatments
have poor clinical efficacy and oral treatments have a high level
of recurrence and concerns of systemic safety.8-15
Previously, we reported the development of borinic acid
quinoline ester compounds as novel antibacterials (1, Figure
1).16 Further structural modification of these boron-containing
compounds led to the 1,3-dihydro-2,1-benzoxaborole class as
illustrated by 2. Focused screening of our library revealed these
dihydrobenzoxaboroles had good antifungal activity against C.
albicans. Further screening against yeast, filamentous fungi, and
1-Hydroxydihydrobenzoxaboroles (19b-m) were synthesized
as shown in Scheme 2. The protected o-bromobenzyl alcohol
derivative (18), prepared from 16 or 17, was converted into the
corresponding phenylboronic acid. Deprotection of the meth-
oxymethyl ether using hydrochloric acid followed by spontane-
ous cyclization gave the target compounds 19b-m (19c, Rv )
Me, was prepared as a racemate). When compounds have
functional groups sensitive to butyllithium, such as a nitrile
group, an in situ trap method was applied.18
The 7-fluoro derivative (19n) was synthesized through
directed ortho metalation of 3-fluorobenzyl alcohol (20)
(Scheme 3). 19
* To whom correspondence should be addressed. Phone: (650) 739-
0711. Fax: (650) 739-0139. E-mail: sbaker@anacor.com.
10.1021/jm0603724 CCC: $33.50 © 2006 American Chemical Society
Published on Web 06/27/2006

4448 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 15
Letters
Scheme 1a
Scheme 4a
a Conditions: (a) Ph3PCH2OMeCl, t-BuOK, DMF, 0 °C to room temp;
(b) 6 N HCl, THF, reflux; (c) NaBH4, MeOH, room temp; (d) MeOCH2Cl,
i-Pr2NEt, CH2Cl2, room temp; (e) n-BuLi, 7a, THF, -78 °C to room temp;
(f) 6 N HCl, room temp; (g) n-BuLi, (i-PrO)3B, THF, -78 °C to room
temp.
Table 1. Minimum Inhibitory Concentration (µg/mL) of
Boron-Containing Compounds Compared to Ciclopirox
T.
C.
C.
A.
T.
rubrum
mentagro- albi- neofor- fumi-
Ri
other
phytes
cans mans gatus
ciclopirox
9a
9b
10a
10b
11
12
13
14
15
19a
19b
19c
23a
23b
24
0.5
4
1
8
e1
4
4
1
1
16
8
1
32
8
8
32
0.5
4
2
8
2
2
4
4
2
32
4
1
16
8
8
0.5
4
0.5
2
0.5
1
4
1
1
4
2
0.5
16
8
8
64
0.5
8
2
4
0.5
4
4
1
1
16
1
4
2
4
1
2
4
1
1
H
Riv ) H
5-F Riv ) H
H
5-F
5-F
5-F
5-F
5-F
5-F
a Conditions: (a) NaBH4, MeOH, room temp; (b) MeOCH2Cl, i-Pr2NEt,
CH2Cl2, room temp; (c) sec-BuLi, (MeO)3B, THF, -78 °C to room temp;
(d) ethylene glycol, THF or toluene, reflux; (e) n- or t-BuLi, 7, THF, -78
°C to room temp; (f) 3-bromopyridine, i-PrMgCl, THF, 0 °C (ref 17), or
vinylmagnesium bromide, THF, -78 °C to room temp; (g) 6 N HCl, THF,
room temp.
16
2
H
Rv ) H
1
5-F Rv ) H
0.25 0.25
32
8
16
>64
5-F Rv ) Me
16
8
16
32
Scheme 2a
6-F
H
6-F
32
resulting alcohol gave the methoxymethyl ether (22a), which
was converted to the final products 23a and 24 using the same
chemistry described previously. The unsubstituted derivative
(23b) was synthesized from commercially available 2-bro-
mophenylacetaldehyde (21b).
To determine the antifungal activity of these compounds, we
screened for their minimum inhibitory concentrations (MIC)
against the major dermatophytes that cause onychomycosis, T.
rubrum and T. mentagrophytes, and against the yeasts and molds
C. albicans, C. neoformans, and A. fumigatus to test for their
broad spectrum activity. The antifungal agent ciclopirox,
currently in use for the topical treatment of onychomycosis, was
used as a reference.
a Conditions: (a) NaBH4, MeOH, room temp (when X ) H), or BH3-
THF, THF, room temp (when X ) OH); (b) MeOCH2Cl, i-Pr2NEt, CH2Cl2,
room temp; (c) MeMgBr, THF, -78 °C to room temp; (d) NBS, AIBN,
CCl4, reflux; (e) NaOAc, DMF, 70 °C; (f) NaOH, MeOH, reflux; (g) n-BuLi,
(i-PrO)3B, THF, -78 °C to room temp; (h) 6 N HCl, THF, room temp.
Our initial lead compound was the 1-phenyldihydrobenzox-
aborole (9a) (Table 1). This showed modest broad spectrum
activity with MIC values of 4-8 µg/mL. One of the first
modifications made was to install a 5-fluoro group, giving 9b.
This substitution led to a 2- to 8-fold increase in potency against
the strains tested. Subsequently, most of the following analogues
synthesized contained this fluoro substitution.
We set out to determine the effect of replacing the 1-phenyl
group of 9 with various substitutions. The 1-styryl-substituted
dihydrobenzoxaboroles 10a and 10b led to approximately
equivalent activity to our leads 9a and 9b, respectively, against
all fungi tested (Table 1). Again, when Ri ) 5-F (10b), potency
was improved 4- to 8-fold. Replacing the 1-phenyl group of 9b
with 1-vinyl (11) or 1-(furan-3-yl) (12) led to an approximate
Scheme 3a
a Conditions: (a) sec-BuLi, (i-PrO)3B, THF, -78 °C to room temp, then
HCl.
Six-membered benzoxaborin analogues were synthesized as
shown in Scheme 4. For 6-fluoro analogues, the benzaldehyde
(3) was subjected to a Wittig reaction and the resulting enol
ether was hydrolyzed to give the phenylacetaldehyde (21a).
Reduction of the carbonyl group, followed by protection of the

Letters
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 15 4449
Table 2. Minimum Inhibitory Concentration (µg/mL) of
1,3-Dihydro-1-phenyl-2,1-benzoxaborole Compounds (9)
Table 3. Minimum Inhibitory Concentration (µg/mL) of
1,3-Dihydro-1-hydroxy-2,1-benzoxaborole Compounds (19)
T.
C.
C.
A.
T.
rubrum
mentagro-
phytes
albi-
cans
neofor-
mans
fumi-
gatus
T.
C.
C.
A.
compd
Ri
T.
rubrum
mentagro- albi- neofor- fumi-
compd Ri
Riv
phytes
cans
mans
gatus
19b
19d
19e
19f
19g
19h
19i
19j
19k
19l
5-F
5-Cl
5-Me
5-CF3
5-NC
5-MeO
5-HOCH2
6,7-benzo
5-F-6-F
4-F
1
1
8
1
2
4
0.5
1
2
0.25
2
8
16
8
>64
>64
32
2
32
0.25
1
2
9a
9b
9c
9d
9e
9f
9g
9h
9i
H
F
H
H
H
F
F
F
F
F
H
H
4
1
16
8
8
4
1
4
2
2
4
2
8
16
4
8
2
4
4
2
4
0.5
8
4
4
0.25
0.5
1
1
0.5
8
2
4
4
2
0.5
1
1
0.5
0.5
4
2
8
8
2
2
1
2
8
8
16
8
>64
>64
32
4
64
16
32
8
3′-Cl
3′-F
4′-F
3′-Cl
3′-F
4′-F
3′-Me
4′-Me
16
64
64
4
16
32
64
2
16
>64
>64
32
2
32
8
4
4
4
16
16
16
16
32
16
0.5
0.5
19m
19n
6-F
7-F
32
32
9j
2- to 8-fold decrease in activity, while replacement with
1-(thiophen-3-yl) (13) or 1-(4-methylthiophen-3-yl) (14) led to
approximately equal activity against all fungi (Table 1). Interest-
ingly, replacement of the 1-phenyl group of 9b with 1-(pyrid-
3-yl) (15) showed selectivity toward nondermatophyte strains;
there was a 16- to 64-fold decrease in activity against the
dermatophytes T. rubrum and T. mentagrophytes but no change
in activity against C. albicans and only a 4-fold reduction in
activity against C. neoformans and A. fumigatus (Table 1).
In another modification to enhance hydrophilicity, we re-
placed the 1-phenyl group of 9a and 9b with a 1-hydroxy group
to give 19a and 19b, respectively. Compounds 19a and 19b
proved to have a more broad spectrum profile than 9a and 9b,
respectively. Both 19a and 19b showed an 8-fold increase in
activity against C. neoformans, and 19b showed an 8-fold
increase in activity against A. fumigatus. (Table 1).
Figure 2. Structure of AN2690, currently in clinical trials for
onychomycosis.
In a final study, we synthesized various analogues of 5-fluoro-
1,3-dihydro-1-hydroxy-2,1-benzoxaborole (19b) to determine the
structure-activity relationship of this scaffold. The results of
this study are shown in Table 3. We first substituted the 5-F
group with other groups, giving 19d-i, to determine the
optimum substituent for this position. The results showed that
the 5-F compound 19b remained the most potent with only the
5-Cl analogue 19d approaching similar activity. The 5-methyl,
5-trifluoromethyl, and 5-cyano analogues 19e, 19f, and 19g,
respectively, showed only weak activity, while 5-methoxy and
5-hydroxymethyl analogues 19h and 19i, respectively, were
inactive. Interestingly, the 6,7-benzo analogue (19j) showed
moderate activity toward the dermatophytes but was effectively
inactive against the nondermatophyte strains.
In an effort to understand the effect of the 3-substiution on
the oxaborole ring, we added a methyl group to the 3-position
to give 19c. However, this modification led to an 8- to 32-fold
decrease in activity (Table 1).
We then increased the ring size from a five-membered
oxaborole of 9a, 9b, and 19b to the corresponding six-membered
oxaborin, giving 23b, 23a, and 24, respectively. The results of
these are shown in Table 1. The 1-phenyl substituted oxaborin
23b was only approximately 2-fold less active than the
oxaborole 9a. In contrast, the 5-fluoro-1-phenyloxaborin 23a
was 4- to 16-fold less active than the corresponding oxaborole
9b, showing that in this case, the 5-fluoro group gave no
advantage on potency over the unsubstituted oxaborin 23b.
Finally, the 1-hydroxyoxaborin 24, was 32- to 256-fold less
active than the corresponding oxaborole 19b.
The results shown in Table 1 led us to conclude that a five-
membered benzoxaborole ring, with no substitution at the
3-position, was optimum for activity. Furthermore, we found
that the 1-phenyl or 1-hydroxy substituents gave the best
potency.
Next, we focused our attention on the 1-phenyldihydroben-
zoxaborole scaffold (9) to determine the effect of substitutions
on the 1-phenyl ring, and examples of these are shown in Table
2. As before, in all cases, compounds with Ri ) F were more
potent than compounds with Ri ) H. Compounds 9f-j showed
activity similar to that of the lead compound 9b; however,
because of in vitro cytotoxicity of 9f-j compared to 9a (data
not shown), these compounds were not considered for further
development.
In another modification, we found that addition of a second
fluoro group at the 6-position, giving 19k, effectively offset
the additional potency provided by the 5-fluoro substituent. In
a final modification we moved the fluoro group to other
positions around the benzo ring, giving 19l-n, and found that
the optimum position for the fluoro group remained at the
5-position.
From the results shown in Tables 2 and 3, we concluded that
the 1-phenyl- and 1-hydroxy- 5-fluoro-1,3-dihydro-2,1-benzox-
aborole compounds 9b and 19b, respectively, were the most
active against fungi and especially against the dermatophytes
T. rubrum and T. mentagrophytes, the primary fungal pathogens
causing onychomycosis. To evaluate the ability of the dihy-
drobenzoxaborole antifungals to treat onychomycosis by topical
application, we have examined these compounds in a number
of in vitro experiments including nail penetration and keratin
binding (results to be reported elsewhere). From these preclinical
studies, 5-fluoro-1,3-dihydro-1-hydroxy-2,1-benzoxaborole (19b)
(Figure 2) was identified as having a unique profile of in vitro
antidermatophyte activity, maintenance of this activity in the
presence of keratin, and exceedingly good penetration of human
nails. This compound is currently undergoing clinical trials for
the treatment of onychomycosis.

4450 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 15
Letters
Rebell, G.; Lesher, J.; Gerlach, B.; Ponce-de-leon, G. F.; Ghannoum,
A.; Warner, J.; Isham, N.; Elewski, B. A large scale North American
study of fungal isolates from nails: The frequency of onychomycosis,
fungal distribution and antifungal susceptibility patterns. J. Am. Acad.
Dermatol. 2000, 43, 641-648.
In conclusion, we present a novel class of boron-containing
compounds with broad spectrum antifungal activity. We report
the synthesis and describe the structure-activity relationship
for in vitro antifungal activity. One member of this new family
of antifungal compounds, 19b, was identified as our clinical
candidate for onychomycosis. Future reports from our labs will
describe other preclinical studies, mechanism of action, ap-
plication to systemic diseases, and clinical trial results for
onychomycosis.
(8) Roberts, D. T. Onychomycosis: current treatment and future chal-
lenges. Br. J. Dermatol. 1999, 141, (Suppl. 56), 1-4.
(9) Gupta, A. K.; Ryder, J. E.; Baran, R. The use of topical therapies to
treat onychomycosis. Dermatol. Clin. 2003, 21, 481-489.
(10) Baker, S. J.; Hui, X.; Maibach, H. I. Progress on new therapeutics
for fungal nail infections. Annu. Rep. Med. Chem. 2005, 40, 323-
335.
(11) Baran, R.; Kaoukhov, A. Topical antifungal drugs for the treatment
of onychomycosis: an overview of current strategies for monotherapy
and combination therapy. J. Eur. Acad. Dermatol. Venereol. 2005,
19, 21-29.
Acknowledgment. We acknowledge Maureen Kully and
Jehangir Khan at NAEJA Pharmaceuticals, Inc., for obtaining
some MIC data.
(12) Gupta, A. K.; Lynch, L. E. Onychomycosis: review of recurrence
rates, poor prognostic factors and strategies to prevent disease
recurrence. Cutis 2004, 74 (Suppl. 1), 10-15.
(13) Scher, R. K.; Baran, R. Onychomycosis in clinical practice: factors
contributing to recurrence. Br. J. Dermatol. 2003, 149 (Suppl. 65),
5-9.
(14) Gupta, A. K.; Baran, R.; Summerbell, R. Onychomycosis: strategies
to improve efficacy and reduce recurrence. J. Eur. Acad. Dermatol.
Venereol. 2002, 16, 579-586.
(15) Prescribing information for Lamisil tablets. http://www.pharma.us-
.novartis.com/product/pi/pdf/Lamisil_tablets.pdf.
(16) Benkovic, S. J.; Baker, S. J.; Alley, M. R. K.; Woo, Y. H.; Zhang,
Y. K.; Akama, T.; Mao, W.; Baboval, J.; Rajagopalan, P. T. R.; Wall,
M.; Kahng, L. S.; Tavassoli, A.; Shapiro, L. Identification of borinic
esters as inhibitors of bacterial cell growth and bacterial methyl-
transferases, CcrM and MenH. J. Med. Chem. 2005, 48, 7468-7476.
(17) Trecourt, F.; Breton, G.; Bonnet, V.; Mongin, F.; Marsais, F.;
Queguiner, G. New syntheses of substituted pyridines via bromine-
magnesium exchange. Tetrahedron 2000, 56, 1349-1360.
(18) Li, W.; Nelson, D. P.; Jensen, M. S.; Hoerrner, R. S.; Cai, D.; Larsen,
R. D.; Reider, P. J. An improved protocol for the preparation of
3-pyridyl- and some arylboronic acids. J. Org. Chem. 2002, 67,
5394-5397.
Supporting Information Available: Experimental procedures
and characterization data for all compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Faergemann, J.; Baran, R. Epidemiology, clinical presentation and
diagnosis of onychomycosis. Br. J. Dermatol. 2003, 149 (Suppl. 65),
1-4.
(2) Arrese, J. E.; Pie´rard, G. E. Treatment failures and relapses in
onychomycosis: a stubborn clinical problem. Dermatology 2003, 207,
255-260.
(3) Elewski, B. E.; Charif, M. A. Prevalence of onychomycosis in patients
attending a dermatology clinic in northeastern Ohio for other
conditions. Arch. Dermatol. 1997, 133, 1172-1173.
(4) Gupta, A. K.; Jain, H. C.; Lynde, C. W.; MacDonald, P.; Cooper, E.
A.; Summerbell, R. C. Prevalence and epidemiology of onychomy-
cosis in patients visiting physicians’ offices: A multicenter Canadian
survey of 15,000 patients. J. Am. Acad. Dermatol. 2000, 43, 244-
248.
(5) Richardson, M. D.; Warnock, D. W. Fungal Infections, Diagnosis
and Management, 3rd ed.; Blackwell Publishing: Malden, MA, 2003.
(6) Summerbell, R. C.; Kane, J.; Krajden, S.; Onychomycosis, tinea pedis
and tinea manuum caused by non-dermatophytic filamentous fungi
Mycoses. 1989, 32, 609-619.
(7) Ghannoum, M. A.; Hajjeh, R. A.; Scher, R.; Konnikov, N.; Gupta,
A. K.; Summerbell, R.; Sullivan, S.; Daniel, R.; Krusinski, P.;
Fleckman, P.; Rich, P.; Odom, R.; Aly, R.; Pariser, D.; Zaiac, M.;
(19) Austin, P. W.; Kneale, C. J.; Crowley, P. J.; Clough, J. M. Oxaboroles
and Salts and Their Use as Biocides for Plastics. PCT Int. Appl.
WO 95/33754, 1995; 50 pp.
JM0603724