Synthesis and antibacterial activity of novel 6-O-substituted erythomycin A derivatives

Article abstract of DOI:10.1016/S0960-894X(00)00106-2

A series of novel 6-O-substituted erythromycin A derivatives has been synthesized. Good in vitro antibacterial activity has been demonstrated for analogues incorporating a variety of structural features. The methodology disclosed is expected to find application in the design of future macrolide antibiotics that target the prevalent bacterial resistance problem. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00106-2

Bioorganic & Medicinal Chemistry Letters 10 (2000) 815±819
Synthesis and Antibacterial Activity of Novel 6-O-Substituted
Erythromycin A Derivatives
Richard F. Clark,* Zhenkun Ma, Sanyi Wang, George Griesgraber, Michael Tufano,
Hong Yong, Leping Li, Xiaolin Zhang, Angela M. Nilius, Daniel T. W. Chu
and Yat Sun Or
Infectious Diseases Research, Abbott Laboratories, 200 Abbott Park Road, Abbott Park, IL 60064-3537, USA
Received 22 December 1999; accepted 10 February 2000
AbstractÐA series of novel 6-O-substituted erythromycin A derivatives has been synthesized. Good in vitro antibacterial activity
has been demonstrated for analogues incorporating a variety of structural features. The methodology disclosed is expected to ®nd
application in the design of future macrolide antibiotics that target the prevalent bacterial resistance problem. # 2000 Elsevier
Science Ltd. All rights reserved.
Macrolide antibiotics are a preferred drug class for the
safe and e€ective treatment of respiratory tract infec-
tions. Erythromycin A (1), for example, has enjoyed
widespread clinical use for more than 40 years. This
14-membered macrolide, however, undergoes decom-
position within the acidic medium of the stomach. This
results in poor oral bioavailability and undesired gas-
trointestinal side e€ects.1 One approach to overcome
these liabilities is to improve the acid stability of the
macrolide through modi®cation of functional groups
involved in the degradation process. It is known that
the degradative pathway involves interactions between
the hydroxyl groups at C-6 and C-12 with the carbonyl
at C-9.2,3 A C-6 hydroxyl blocking strategy has been
successfully employed in the development of second
generation macrolides such as the 6-O-methyl derivative
of erythromycin A, clarithromycin (2).4 Clarithromycin
exhibits markedly improved acid stability relative to
erythromycin A resulting in better bioavailability and
pharmacokinetics, increased gastrointestinal tolerance,
and a superior spectrum of antibacterial activity.1
Despite intense interest, previously reported modi®ca-
tions of the sterically hindered C-6 hydroxyl position of
erythromycin A have been limited to methylation pro-
ducts.5 Attempts to install higher alkyl groups have
been accompanied by dramatic losses in selectivity for
the C-6 hydroxyl.3 We have recently developed a metho-
dology for the selective introduction of a variety of
substituents at the C-6 position. This communication
will highlight the synthesis and in vitro antibacterial
activity of a series of novel 6-O-substituted erythro-
mycin A derivatives with particular focus on the syn-
thetic versatility of the methodology as well as the
potential for application to newer macrolide classes6
which target the rapidly emerging problem of bacterial
resistance.7
6-O-Substituted macrolides, therefore, constitute
a
rational approach to compounds with enhanced biologi-
cal pro®les.
Our general 6-O-alkylation strategy was adapted from
the manufacturing process used for the production of
*Corresponding author. Tel.: +1-874-938-2509; fax: +1-847-938-
3403; e-mail: rick.f.clark@abbott.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00106-2

816
R. F. Clark et al. / Bioorg. Med. Chem. Lett. 10 (2000) 815±819
Table 1. Scope of 6-O-alkylation of 3
clarithromycin.8 By utilizing suitably protected erythro-
mycin A (3),8 a core group of electrophiles was selec-
tively attached to the C-6 hydroxyl as outlined in
Scheme 1 and Table 1. Although not detailed in this
account, less reactive electrophiles did not alkylate.
Alkylation of compound 3 with allylic, propargylic,
and benzylic bromides in the presence of potassium
hydroxide resulted in incomplete conversion of the C-6
hydroxy starting material. We surmised that an incom-
patability between the hydroxide base and the reactive
alkyl halides was responsible for the low yields. A sur-
vey of alternative bases applied to the allylation of 3
revealed potassium tert-butoxide to be optimal. Overall
yields for 4b were improved by more than 20% when
potassium hydroxide was replaced by the hindered alk-
oxide. A more systematic study of 6-O-alkylation with
potassium tert-butoxide is in progress. Following the
alkylation step, removal of the trimethylsilyl and cyclo-
hexylketal protecting groups under acidic conditions
provided the 9-oxime-6-O-alkylerythromycin A pro-
ducts. Subsequent deoximation using either sodium
bisul®te or sodium nitrite9 a€orded the desired 6-O-
alkylerythromycin A targets (4b±g).
Entry
R
Yielda
33%b
Entry
R
Yielda
21%
4b
4e
4c
4d
51%
19%
4f
10%
24%
4g
aPuri®ed yield over three steps.
bKOBut employed as the base in Scheme 1, step a.
suppressed by the initial addition of a stoichiometric
amount of acid to the reaction mixture.
Aldehyde 4k and epoxide 4m were further modi®ed to
generate analogues encompassing a broader range of
physicochemical properties. A series of oximes (4o±q)
and hydrazones (4r,s) was prepared via the condensa-
tion of hydroxylamines and hydrazines, respectively,
with 4k. In each instance, the products were isolated as
mixtures of E:Z isomers about the carbon±nitrogen
double bond. The parent oxime (4o) was transformed to
the corresponding nitrile (4n) by carbodiimide-mediated
dehydration in the presence of cuprous chloride. Alco-
hol and amine products were also readily derived from
4k. Reductive amination of the parent aldehyde pro-
vided a series of amine targets such as compounds 4t±v.
Alcohol 4l was obtained by the reduction of 4k in the
presence of L-Selectride at low temperature. Treatment
of epoxide 4m with selected amine nucleophiles at
elevated temperatures yielded b-substituted alcohol
derivatives represented by entries 4w and 4x. Sulfur,
oxygen and carbon nucleophiles have also been success-
fully employed and will be discussed in a future report.
Allyl compound 4b was envisioned as a versatile syn-
thetic intermediate that could be further modi®ed in
order to explore structure±activity relationships (SAR)
of the series. Speci®c targets pursued were governed by
attempts to de®ne the scope of our chemistry approach
and by a desire to assess the biological tolerance for
structural variation at the C-6 position. The strategy is
described in Scheme 2. Heck arylation10 of the allylic
ole®n provided a series of conjugated derivatives exem-
pli®ed by entries 4h and 4i. Reduction of the terminal
double bond by catalytic hydrogenation yielded the 6-
O-propyl analogue (4a). Selective oxidation of the allyl
side chain employing Wacker conditions11 delivered
methyl ketone 4j in modest yield. Ozonolysis of 4b to
the corresponding aldehyde was accompanied by com-
peting oxidation of the dimethylamino nitrogen at the 30
position of the desosamine sugar. However, subsequent
selective reduction of the N-oxide with triphenylpho-
sphine a€orded the desired aldehyde (4k) in good yield.
Similarly, epoxidation of 4b with m-chloroperbenzoic
acid produced the N-oxide of the desired compound
requiring treatment with sodium bisul®te to liberate the
target epoxide (4m) as a mixture of diastereomers
(ꢀ1:1). These N-oxidation products could be eciently
These analogues were screened for in vitro antibacterial
activity against representative erythromycin-susceptible
and erythromycin-resistant staphylococci, streptococci
and pneumococci (Table 2). Minimum inhibitory con-
centrations (MICs) were determined using standard
Scheme 1. (a) R-Br, KOH, DMSO-THF, 0 ꢁC; (b) HOAc, Ch3CN-H2O; (c) (i) NaNO2 (5 equiv), HCl (5 equiv), EtOH-H2O, 70 ꢁC, 2 h, or (ii)
NaHSO3 (4 equiv), HCO2H (2.5 equiv), EtOH-H2O, 80 ꢁC, 2 h.

R. F. Clark et al. / Bioorg. Med. Chem. Lett. 10 (2000) 815±819
817
Scheme 2. (a) Ar-Br or Ar-I, Pd(OAc)2, P(o-tolyl)3, Et3N, CH3CN, 80 ꢁC, 73±84%; (b) H2 (1 atm), 10% Pd-C, CH3OH, 58%; (c) PdCl2CuCl2, O2,
H2O, DMF, 36%; (d) (i) O3, CH2Cl2, 78 ꢁC then Me2S, 0 ꢁC, (ii) PPh3, THF, 60 ꢁC, 61% (two steps); (e) m-CPBA, Ch2Cl2 then NaHSO3, 54%; (f)
HCl H2N-OR, NMM, CH3OH, 70±94%; (g) NH2NHR, MgSO4, THF, 59±66%; (h) (i) R-NH2, molecular sieves, CH2Cl2, (ii) NaBH3CN or H2 (1
atm), 10% Pd-C, AcOH, CH3OH, 43±69%; (i) LiB[CH(CH3)C2H5]3H, THF, 78 ꢁC, 37%; (j) 1,3-diisopropylcarbodiimide, CuCl, THF, 50%; (k)
R-NH2, CHCl3 or DMF, 60±90 ꢁC, 37±50%.
agar dilution methods.12 Resistant isolates possessing
the inducible or constitutive type of MLS (macrolide±
streptagramin±lincosamide) resistance,13 as well as those
mediated by macrolide e‚ux mechanisms,13b,14 were
included in the assay. Haemophilus in¯uenzae, another
of the prevalent respiratory tract pathogens, was also
examined.
phenyloxime 4q, phenethylamine 4u, phenylpropyl-
amine 4v, and phenethylaminoalcohol 4x were the
optimal compounds in this study, each possessing either
enhanced or retained activity across the tested panel. In
addition, 4u and 4v were the only agents examined
which demonstrated improved activity against H. in¯u-
enzae DILL and S. aureus 6538P, respectively. Simi-
larly, 4q and 4x showed unique activity enhancements
against susceptible streptococci when compared to other
entries.
In general, 6-O-substituted side chains larger than
methyl were well tolerated when appropriately functio-
nalized. Although the 6-O-propyl entry (4a) was less
active than the corresponding 6-O-methyl parent (2),
variations in the nature and length of the side chain in
4a rendered derivatives with overall antibacterial activ-
ities comparable to the reference macrolides (1, 2). In
particular, the introduction of multiple bonds (4c),
heteroatoms (4l, 4n), or conjugated aromatic systems
(4g, 4i) maintained activities over much of the bacterial
spectrum. Furthermore, those analogues containing
aromatic substituents tethered to the macrolide by a
hydrocarbon linker showed enhanced activity against
both e‚ux and constitutively resistant S. pyogenes and
S. pneumoniae. Among this group, the ¯uorobenzyl (4e)
and naphthylallyl (4h) entries were particularly e€ective,
each exhibiting greater than 16-fold improved activity
against both S. pyogenes 930 and S. pneumoniae 5737.
Those derivatives possessing both an aromatic group
and one or more heteroatoms in the side chain spacer
also exhibited very good antibacterial activity. In fact,
In summary, the synthesis of a novel series of 6-O-sub-
stituted erythromycin A derivatives has been described.
The synthetic approach has been demonstrated to be
suciently ¯exible to provide access to a variety of C-6
side chain structures that were heretofore inaccessible.
Furthermore, SAR studies con®rm that the 6-O posi-
tion displays a high tolerance for structural variation
with regard to both size and nature of substituents.
Several analogues have been identi®ed that feature
in vitro antibacterial activities comparable to those
for clarithromycin across a spectrum of representative
pathogens. Those side chains containing aromatic
groups produced the most potent antibacterial e€ects
including marked activity increases against resistant
organisms. Ultimately, the ability to modify this key
position in the macrolide a€ords new opportunities to
exploit the SAR of other macrolide classes currently
showing promise in overcoming drug resistance.15

818
Table 2. In vitro antibacterial activity of 6-O-substituted erythromycin A derivatives 4a
Minimum inhibitory concentration (mg/mL)
R. F. Clark et al. / Bioorg. Med. Chem. Lett. 10 (2000) 815±819
Entry
R
S. aureus
S. pyogenes
S. pneumoniae
H. in¯uenzae
6538P
A5177
EES61
PIU 2548
930
ATCC6303
5649
5737
DILL
1
Erythromycin A
Clarithromycin
0.2
0.2
3.1
6.2
6.2
25
0.03
0.03
0.05
32
32
64
>128
>128
>128
0.06
0.06
Ð
16
16
Ð
>128
>128
Ð
4
8
2
4a
32
4b
4c
0.78
0.2
12.5
12.5
Ð
64
Ð
>128
>128
Ð
Ð
16
Ð
32
16
0.03
0.06
>128
4d
4e
1.56
3.1
12.5
12.5
0.06
1
16
8
>128
8
Ð
1
16
8
64
8
Ð
64
4g
4h
0.39
0.78
6.2
0.03
0.06
16
8
32
8
0.06
0.25
4
32
8
4
4
12.5
16
4i
4j
0.2
3.1
3.1
Ð
0.06
Ð
4
64
0.06
Ð
4
128
Ð
8
128
>128
Ð
Ð
4k
4l
0.78
0.39
25
6.2
25
Ð
0.06
0.125
4
128
64
>128
>64
0.5
64
32
Ð
16
>128
>64
Ð
8
8
0.125
Ð
4m
4n
25
>128
32
>128
>128
>128
8
0.2
0.2
0.78
0.2
0.03
0.06
>128
4o
4p
4q
Ð
25
0.015
0.125
0.008
64
64
32
Ð
0.06
32
32
16
>128
>128
128
16
16
4
>128
128
0.125
0.008
3.1
4r
4s
0.78
0.78
25
25
0.06
128
128
>128
>128
0.125
0.25
64
64
>128
>128
8
8
0.125
4t
0.78
0.2
12.5
12.5
3.1
0.25
0.015
0.03
64
8
>128
64
0.125
0.03
0.03
32
4
>128
64
8
2
4
4u
4v
0.05
32
>128
16
>128
4w
4x
0.39
0.2
6.2
6.2
0.125
0.03
Ð
16
Ð
0.125
0.008
64
32
>128
32
8
4
>64
aS. aureus 6538P, S. pyogenes EES61 and S. pneumoniae ATCC6303: erythromycin susceptible; S. pyogenes 2548 and S. pneumoniae 5649: macrolide
e‚ux resistance; H. in¯uenzae DILL: ampicillin-resistant; S. aureus A5177: inducible MLS resistance; S. pyogenes 930 and S. pneumoniae 5979:
constitutive MLS resistance.

R. F. Clark et al. / Bioorg. Med. Chem. Lett. 10 (2000) 815±819
819
References and Notes
Plattner, J. J. J. Med. Chem. 1996, 39, 3853. (d) Doern, G. V.;
Jones, R. N.; Pfaller, M. A.; Kugler, K. Antimicrob. Agents
Chemother. 1999, 43, 385.
8. Morimoto, S.; Adachi, T.; Matsunaga, T.; Kashimura, M.;
Asaka, T.; Watanabe, Y.; Sota, K.; Sekiuchi, K. US Patent
4,990,602, 1991.
1. (a) Chu, D. T. W. Exp. Opin. Invest. Drugs 1995, 4, 65. (b)
Lartey, P. A.; Perun, T. In Studies in Natural Products Chem-
istry; Rahman, A., Ed.; Elsevier Science: Amsterdam, 1993;
Vol. 13, pp 155±185.
2. Kurath, P.; Jones, P.; Egan, R.; Perun, T. Experienta 1971,
27, 362.
3. Itoh, Z.; Nakaya, M.; Suzuki, T.; Hiral, H.; Wakabayashi,
K. Am. J. Physiol. 1984, 10, G688.
4. (a) Morimoto, S.; Takahashi, Y.; Watanabe, Y.; Omura, S.
J. Antibiot. 1984, 37, 187. (b) Watanabe, Y.; Morimoto, S.;
Adachi, T.; Kashimura, M.; Asaka, T. J. Antibiot. 1993, 46,
647.
5. Morimoto, S.; Takahashi, Y.; Adachi, T.; Nagate, T.;
Watanabe, Y.; Omura, S. J. Antibiot. 1990, 43, 286.
6. (a) Fernandes, P. B.; Baker, W. R.; Freiberg, L. A.; Hardy,
D. J.; McDonald, E. D. Antimicrob. Agents Chemother. 1989,
33, 78. (b) Agouridas, C.; Denis, A.; Auger, J.-M.; Benedetti,
Y.; Bonnefoy, A.; Bretin, F.; Chantot, J.-F.; Dussarat, A.;
Fromentin, C.; Gouin D'Ambrieres, S.; Lachaud, S.; Laurin,
P.; Le Martret, O.; Loyau, V.; Tessot, N. J. Med. Chem. 1998,
41, 4080. (c) Asaka, T.; Kashimura, M.; Misawa, Y.; Ono, T.;
Suzuki, K.; Yoshida, T.; Akashi, T.; Yokoo, C.; Nagate, T.;
Morimoto, S. Abstracts of the 35th Interscience Conference
on Antimicrobial Agents and Chemotherapy, San Francisco,
CA, 17±20 September 1995, Abstract No. F177.
9. In general, sodium nitrite was found to be the preferred
reagent giving higher yields of the deoximated products.
10. Heck, R. F. In Comprehensive Organic Synthesis; Trost, B.
M., Ed.; Pergamon: New York, 1991; Vol. 4, Chapter 4.3.
11. Tsuji, J.; Shimizu, I.; Yamamoto, K. Tettrahedron Lett.
1976, 2975.
12. Methods for Dilution Antimicrobial Susceptibility Tests for
Bacteria that Grow Aerobically; National Committee for Clin-
ical Laboratory Standards: Villanova, PA, 1985: M7-A, Vol.
5, No. 22.
13. (a) Leclerc, R.; Courvalin, P. In Macrolides: Chemistry,
Pharmacology and Clinical Uses; Bryskier, A. J.; Butzler, J.-P.;
Neu, H. C.; Tulkens, P. M., Eds.; Arnette Blackwell: Paris,
1993; pp 125±141. (b) Weisblum, B. Drug Resistance Updates
1998, 1, 29.
14. Sutcli€e, J.; Tait-Kamradt, A.; Wondrack, L. Antimicrob.
Agents Chemother. 1996, 40, 1817.
15. An early example of the successful application of this
methodology has recently appeared: Ma, Z.; Clark, R. F.;
Wang, S.; Nilius, A.; Flamm, R. K.; Or, Y. Abstracts of the
39th Interscience Conference on Antimicrobial Agents and
Chemotherapy, San Francisco, CA, 26±29 September 1999,
Abstract No. 2133.
7. (a) Neu, H. C. Science 1992, 257, 1064. (b) Appelbaum, P.
C. Clin. Infect. Dis., 1992, 15, 77. (c) Katz, L.; Chu, D. T. W.;