Synthesis and biological evaluation of novel hygromycin A antibacterial agents

Article abstract of DOI:10.1016/j.bmcl.2010.08.139

Novel hygromycin A derivatives bearing a variety of functionalized aminocyclitol moieties have been synthesized in an effort to increase the antibacterial activity and drug-like properties of this class of agents. A systematic study of the effect of alkylation and removal of the hydroxyls of the aminocyclitol directed us to a series of alkylated aminocyclitol derivatives with improved Gram-positive activity.

Full text of DOI:10.1016/j.bmcl.2010.08.139

Bioorganic & Medicinal Chemistry Letters 20 (2010) 6730–6734
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and biological evaluation of novel hygromycin A antibacterial agents
⇑
Michael S. Visser, Kevin D. Freeman-Cook , Steven J. Brickner , Katherine E. Brighty, Phuong T. Le,
Sarah K. Wade, Rhonda Monahan, Gary J. Martinelli, Kyle T. Blair, Dianna E. Moore
Pfizer Worldwide Research and Development, Eastern Point Road, Groton, CT 06340, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Novel hygromycin A derivatives bearing a variety of functionalized aminocyclitol moieties have been
synthesized in an effort to increase the antibacterial activity and drug-like properties of this class of
agents. A systematic study of the effect of alkylation and removal of the hydroxyls of the aminocyclitol
directed us to a series of alkylated aminocyclitol derivatives with improved Gram-positive activity.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 20 August 2010
Accepted 30 August 2010
Available online 21 September 2010
Keywords:
Hygromycin
Gram-positive
Bacteria
MRSA
Staphylococcus aureus
Aminocyclitol
The emergence of multidrug resistance among Gram-positive
pathogens represents a significant challenge for medical profes-
sionals.1 Of particular concern are methicillin-resistant Staphylo-
coccus aureus (MRSA)2,3 and vancomycin-resistant enterococci
(VRE), due to a combination of increasing prevalence and recalci-
trance to therapy.4 One important strategy to address these resis-
tance issues is the development of new classes of antibiotic
drugs with activity against resistant Gram-positive pathogens.
Hygromycin A (1, Table 1), isolated from Streptomyces hygro-
scopicus and first reported in 1953,5 has been demonstrated to be
a bacterial protein synthesis inhibitor6 with weak activity against
Gram-positive organisms.7 A program to develop hygromycin A
derivatives for Gram-positive pathogens led to the discovery that
the furanose portion of the natural product could be truncated to
simple lipophilic groups such as the allyl group in 2.8,9 Critical in
this work was the discovery that the phenolic hydroxyl group
could be replaced by either one or two fluorine atoms as phenyl
substituents.10,11 This resulted in the discovery of compounds such
as 3. Compound 3 showed antibacterial activity, with MIC values in
bioavailability was due primarily to poor permeability and possible
active efflux (CaCO-2 values of AB = 0.9 ꢁ 10ꢂ6 cm/s, and BA =
5.9 ꢁ 10ꢂ6 cm/s).
Here we describe initial derivatizations of the aminocyclitol
moiety,12,13 with the goal of producing more potent antibacterial
agents and improving permeability to produce analogs with en-
hanced bioavailability. A systematic study of alkylation and hydro-
xyl removal from the aminocyclitol moiety was undertaken to
determine which position of the aminocyclitol held the most
promise for further derivatization efforts.
The alkylated and deoxygenated aminocyclitol derivatives were
prepared as shown in Scheme 1A. Synthesis of methylated analogs
began with protection of aminocyclitol 4, which was derived from
degradation of hygromycin A.14 Selective protection of the amine
and 60-hydroxyl as the phenyloxazoline provided a common inter-
mediate for diversification of the remaining hydroxyls of the
aminocyclitol based on selective protecting group manipula-
tion.12,13 The 50-hydroxyl was selectively silylated with TBSCl to af-
ford 5. The free 20-hydroxyl was subsequently alkylated with MeI;
removal of all protecting groups with HF–pyridine, followed by
transfer hydrogenation,15 afforded 6. Alternatively, intermediate
5 was deoxygenated by treatment with Ph3P and DEAD, and similar
deprotection afforded compound 7. Synthesis of the 50-methoxy
hygromycin analog began with forcing bis-silyl protection of the
oxazoline aminocyclitol with TBSCl followed by selective deprotec-
tion of the 50-position, to afford mono-silyl diol 8. Subsequent
alkylation of the 50-position with NaH and MeI in THF followed
by deprotection produced methoxy aminocyclitol 9. Compound 8
was also used as an intermediate to remove the 50-hydroxyl. In this
the range of 0.2–4 lg/ml against a panel of Gram-positive bacteria.
However, 3 behaved sub-optimally in oral pharmacokinetic studies
in rats, exhibiting low bioavailability (ꢀ20%). Because this was a
low clearance, soluble compound, we expected that the low
⇑
Corresponding author. Address: Pfizer Worldwide Research and Development,
La Jolla, 10777 Science Center Drive, San Diego, CA 92121, United States. Tel.: +1
858 526 4844; fax: +1 860 686 5036.
E-mail address: kevin.freeman-cook@pfizer.com (K.D. Freeman-Cook).
Present address: SJ Brickner Consulting, LLC Ledyard, CT 06339, United States.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.08.139

M. S. Visser et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6730–6734
6731
Table 1
In vitro antibacterial activity of hygromycin A, 1 [MIC (lg/ml)] and other early analogs in a panel of Gram-positive pathogens
Compound Structure
S. aureus
S. aureus
S. pneumoniae
S. pneumoniae
S. pyogenes
S. pyogenes
1095
1146
1046
1095
203
1079
O
OH
OH
HO
O
O
O
N
H
O
1
50
25
12.5
12.5
6.25
6.25
O
HO
OH
HO
O
OH
HO
O
O
O
N
H
2
3
50
4
50
4
1.56
0.5
1.56
0.4
1.56
0.8
HO
OH
O
OH
F
O
O
N
H
2
<0.2
F
O
F
HO
OH
OMe
H2N
HO
A
c, d, e
O
O
OH
OH
a, b
N
O
O
6
Ph
O
H2N
HO
O
O
f, d, e
OTBS
OH
2
5
OH
7
H2N
HO
O
O
6
5
OH
OH
H2N
HO
c, d, e
O
O
4
OTBS
N
O
O
OMe
a, b, d
Ph
9
O
OH
OH
H2N
HO
O
O
g, h, d, e
8
10
O
O
B
OH
O
F
F
OH
i
HN
HO
+ 9
F
O
F
F
O
F
O
OMe
11
12
Scheme 1. (A and B) Reagents and conditions: (a) PhCN, K2CO3, glycerol; (b) TBSCl, imidazole, CH2Cl2; (c) NaH, MeI, THF; (d) HF–pyridine, THF; (e) Pd black, HCO2NH4, AcOH;
(f) Ph3P, DEAD; (g) NaH, CS2, CH3I, THF; (h) Bu3SnH, toluene; (i) EDCI, DMAP, CH2Cl2.
sequence, formation of the methyl xanthate with NaH, CS2 and
MeI, followed by reduction of the xanthate with tributyltin hydride
and removal of the silyl and oxazoline protecting groups, gave the
50-deoxy aminocyclitol 10.16 Final analogs 12, 13, 14, and 16 were
prepared by coupling the appropriate aminocyclitol intermediate
(such as 9) with carboxylic acid 11 using EDCI (Scheme 1B).17 In or-
der to more efficiently explore potential substituents at these posi-
tions, a new synthetic route (Scheme 2) was developed. This
approach took advantage of a regioisomeric alkylation preference
for the 20- and 60-position isomers (with a consistent, major prefer-
ence for 20-alkylation), relative to the rather unreactive 50-position.
Purification by reverse phase HPLC consistently separated the de-
sired analogs for testing, but generally in only low to moderate
yield.

6732
M. S. Visser et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6730–6734
Scheme 2. Regioisomeric alkylation of the 20- and 60-positions of hygromycin derivatives.
Table 2
In vitro antibacterial activity [MIC (
lg/ml)] of aminocyclitol derivatives
O
F
NH
R
F
O
F
Compound
R-group
OH
S. aureus 1095
S. aureus 1146
S. pneumoniae 1046
S. pneumoniae 1095
S. pyogenes 203
S. pyogenes 1079
O
O
3
4
4
0.5
0.39
2
<0.20
HO
HO
OH
OH
O
O
12
13
14
>100
64
100
32
25
4
25
8
100
16
50
8
O
OH
O
O
HO
HO
O
O
O
6.25
3.13
0.78
1.56
1.56
1.56
OH
O
O
O
15
8
8
1
4
4
1
HO
HO
OH
O
O
16
17
16
8
8
8
ND
2
8
8
4
1
8
2
OH
OH
O
O
O
OH
The new hygromycin A analogs were assessed for antibacterial
activity against a panel of Gram-positive strains (Table 2). It imme-
diately became clear that the 50-hydroxyl was critical to maintain
useful antibacterial activity, as the methylation of that hydroxyl
(12) or its removal (13) were both very detrimental modifications,
leading to significant loss in activity.13 In contrast, the 20-position
seemed far more tolerant of manipulation, as the methyl and ethyl
derivatives (compounds 14 and 15) retained much of the potency
of the parent analog. Deoxygenation at the 20-position (16) de-
creased activity relative to the parent hydroxyl, but not to the same
extent as the 50-regioisomer (13). Alkylation of the 60-position to
provide the ethyl ether (17) was also tolerated. Compounds 14
and 17 had profiles that argued for further exploration at the 20-
and 60-positions of the aminocyclitol.
the SAR of the 20- and 60-position alkyl analogs (Table 3). The activ-
ity of the analogs that were alkylated in the 60-position was notice-
ably worse than the corresponding 20-derivatives. While this
difference was not readily apparent with the ethyl group initially
used to probe these two positions (i.e., compounds 15 and 17 in Ta-
ble 2), a substantial divergence in activity was observed when the
alkyl group was even one carbon larger (18 and 19, Table 3). The
bulkier 60-position analogs 23 and 25 had almost no activity rela-
tive to their 20-position isomers 22 and 24. In general, a variety
of motifs were tolerated in the 20-position, including functionalized
alkyl (20, 27, 28), cycloalkylmethyl (22), and heterocycloalkyl-
methyl (26). The simple benzyl analog (24) was relatively weakly
active against all three organisms, and the functionalized benzyl
derivative (29) was one of the most potent analogs produced in
this program, representing a significant improvement relative to
compound 3. In all cases, the analogs were more potent versus
Building on the knowledge gained in the early derivatization, a
more focused set of analogs was synthesized to further elucidate

M. S. Visser et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6730–6734
6733
Table 3
In vitro antibacterial activity [MIC (
lg/ml)] of advanced, alkylated aminocyclitol derivatives
O
OR1
OH
F
O
O
N
H
F
O
F
R2O
Compound
R1
R2
S. aureus 1095
S. aureus 1146
S. pneumoniae 1046
S. pneumoniae 1095
S. pyogenes 203
S. pyogenes 1079
18
19
–H
4
8
1
2
4
1
–H
–H
>64
>64
64
64
64
32
20
21
F
–H
–H
2
1
8
0.25
8
0.25
8
0.5
8
0.125
2
F
16
22
23
4
4
0.5
64
1
1
0.5
32
–H
–H
>64
>64
>64
>64
24
–H
16
16
4
8
16
2
25
26
>64
4
64
2
32
32
1
64
1
32
–H
0.5
0.25
O
27
28
O
–H
–H
16
2
16
1
4
4
1
2
2
0.5
0.5
0.25
N
N
29
–H
1
1
<0.0625
<0.0625
0.25
<0.0625
Scheme 3. Reagents and conditions: (a) pTsOH, 2,2-dimethoxypropane, DMF; (b) DMPMOC(@NH)CCl3, PPTS, CH2Cl2; (c) 2-chloromethylbenzonitrile, NaH, THF; (d) DDQ,
CH2Cl2, H2O; (e) PPTS, MeOH.
streptococci than staphylococci, which parallels the intrinsic spec-
trum seen in the natural product and the parent aminocyclitol ana-
logs. In some cases this MIC difference between pathogens was
only one or two dilutions, but in other cases (such as 29) the differ-
ential was much greater.
Others were more permeable but still showed efflux potential
(29, CaCO-2 AB = 2.5 ꢁ 10ꢂ6 cm/s, and BA = 22 ꢁ 10ꢂ6 cm/s). Based
on overall profile, the 3-methoxypropyl-substituted analog (27)
was submitted to a rat pharmacokinetic study and demonstrated
improved bioavailability (F = 90%), although higher in vivo clear-
ance ultimately limited oral exposure for this compound.
In summary, we have reported the antibacterial activity of a no-
vel series of hygromycin A analogs. A systematic study of the effect
of removal and substitution of the hydroxyl groups of 3 led to fur-
ther exploration of the 20-position of the aminocyclitol. Com-
pounds with improved antibacterial activity as compared to 1
and 3 and improved oral bioavailability have been identified. Fur-
ther lead optimization details will be reported in due course.
To better enable functionalization of the 20-position of the
aminocyclitol, a new selective protection scheme was developed
(Scheme 3). Protection of the 60-hydroxyl and the amide nitrogen
of the aminocyclitol was accomplished by formation of an aceto-
nide with 2,2-dimethoxypropane and p-TsOH. Selective protection
of the 50-hydroxyl with a 3,5-dimethoxybenzyl group was followed
by alkylation of the free 20-hydroxyl, to afford intermediates such
as 30. Global deprotection provided alkylated hygromycin analogs,
and this sequence could be carried out on multigram scale to en-
able advanced in vivo studies.
References and notes
Beyond changes in antibacterial potency, the addition of lipo-
philicity to the aminocyclitol portion of the hygromycin analogs
was designed to increase the permeability of the compounds.
Although the effect was small in some cases, all analogs studied
did show improved CaCO-2 permeability. Some alkylated com-
pounds, such as 25, showed a balanced permeability profile
(CaCO-2 values of AB = 10 ꢁ 10ꢂ6 cm/s, and BA = 14 ꢁ 10ꢂ6 cm/s).
1. Klevens, R. M.; Morrison, M. A.; Nadle, J.; Petit, S.; Gershman, K.; Ray, S.;
Harrison, L. H.; Lynfield, R.; Dumyati, G.; Townes, J. M.; Craig, A. S.; Zell, E. R.;
Fosheim, G. E.; McDougal, L. K.; Carey, R. B.; Fridkin, S. K. J. Am. Med. Assoc.
2007, 298, 1763.
2. Levy, S. B.; Marshall, B. Nat. Med. 2004, 10, S122.
3. Tomasz, A. N. Eng. J. Med. 1994, 330, 1247.
4. Fridkin, S. K.; Edwards, J. R.; Courval, J. M.; Hill, H.; Tenover, F. C.; Lawton, R.;
Gaynes, R. P.; McGowan, J. E., Jr. Ann. Int. Med. 2001, 135, 175.

6734
M. S. Visser et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6730–6734
5. Pittenger, R. C.; Wolfe, R. N.; Hoehn, M. M.; Marks, P. N.; Daily, W. A.; McGuire,
J. M. Antibiot. Chemother. (Washington, DC) 1953, 3, 1268.
6. Guerrero, M. d. C.; Modolell, J. Eur. J. Biochem. 1980, 107, 409.
7. Mann, R. L.; Gale, R. M.; Van Abeele, F. R. Antibiot. Chemother. (Washington, DC)
1953, 3, 1279.
12. Hecker, S. J.; Lilley, S. C.; Minich, M. L.; Werner, K. M. Bioorg. Med. Chem. Lett.
1992, 2, 1015.
13. Hecker, S. J.; Lilley, S. C.; Werner, K. M. Bioorg. Med. Chem. Lett. 1992, 2, 1043.
14. Kakinuma, K.; Sakagami, Y. Agric. Biol. Chem. 1978, 42, 279.
15. Bieg, T.; Szeja, W. Synthesis 1985, 76.
8. Jaynes, B. H.; Cooper, C. B.; Hecker, S. J.; Blair, K. T.; Elliott, N. C.; Lilley, S. C.;
Minich, M. L.; Schicho, D. L.; Werner, K. M. Bioorg. Med. Chem. Lett. 1993, 3,
1531.
9. Jaynes, B. H.; Elliott, N. C.; Schicho, D. L. J. Antibiot. 1992, 45, 1705.
10. Hecker, S. J.; Cooper, C. B.; Blair, K. T.; Lilley, S. C.; Minich, M. L.; Werner, K. M.
Bioorg. Med. Chem. Lett. 1993, 3, 289.
11. Stone, G. G.; Girard, D.; Finegan, S.; Duignan, J.; Schafer, J.; Maloney, M.;
Zaniewski, R. P.; Brickner, S. J.; Wade, S. K.; Le, P. T.; Huband, M. D. Antimicrob.
Agents Chemother. 2008, 52, 2663.
16. Barton, D. H. R.; Jang, D. O.; Jaszberenyi, J. C. Tetrahedron Lett. 1990, 31, 3991.
17. Compound 11 was prepared by the following synthetic sequence: base-
promoted addition of 3-fluoropropanol to 2,4,5-trifluorobenzaldehyde
afforded 2,5-difluoro-4-(3-fluoropropoxy)benzaldehyde. This intermediate
was then used in a Horner–Wadsworth–Emmons olefination reaction with
ethyl 2-(diethoxyphosphoryl)propanoate. The resulting ethyl ester was
saponified with LiOH to provide (E)-3-[2,5-difluoro-4-(3-fluoropropoxy)
phenyl]-2-methylacrylic acid (11).