Synthesis of 4″-O-desosaminyl clarithromycin derivatives and their anti-bacterial activities

Article abstract of DOI:10.1016/j.bmcl.2013.09.083

A series of new 4″-O-desosaminyl clarithromycin derivatives were designed and synthesized. The efficient synthesis routes of 6-deoxy-desosamine donors 8 and 11 were developed and the methodology of glycosylation of clarithromycin 4″-OH with desosamine was studied. The activities of the target compounds were tested against a series of macrolide-sensitive and macrolide-resistant pathogens. Some of them showed activities against macrolide sensitive pathogens, and compounds 19 and 22 displayed significant improvement of activities against sensitive pathogens and two strains of MRSE, which verified the importance of desosamine in the interaction of macrolide and its receptor, and offered valuable information of the SAR of macrolide 4″-OH derivatives.

Full text of DOI:10.1016/j.bmcl.2013.09.083

Bioorganic & Medicinal Chemistry Letters 23 (2013) 6274–6279
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of 400-O-desosaminyl clarithromycin derivatives and their
anti-bacterial activities
⇑
Di Zhu, Yanpeng Xu, Yi Liu, Xiaozhuo Chen, Zhehui Zhao, Pingsheng Lei
State Key Laboratory of Bioactive Substances and Functions of Natural Medicines, Beijing Key Laboratory of Active Substances Discovery and Drugability Evaluation,
Department of Medicinal Chemistry, Institute of Materia Medica, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing 100050, PR China
a r t i c l e i n f o
a b s t r a c t
A series of new 400-O-desosaminyl clarithromycin derivatives were designed and synthesized. The effi-
cient synthesis routes of 6-deoxy-desosamine donors 8 and 11 were developed and the methodology
of glycosylation of clarithromycin 400-OH with desosamine was studied. The activities of the target com-
pounds were tested against a series of macrolide-sensitive and macrolide-resistant pathogens. Some of
them showed activities against macrolide sensitive pathogens, and compounds 19 and 22 displayed sig-
nificant improvement of activities against sensitive pathogens and two strains of MRSE, which verified
the importance of desosamine in the interaction of macrolide and its receptor, and offered valuable infor-
mation of the SAR of macrolide 400-OH derivatives.
Article history:
Received 9 June 2013
Revised 9 September 2013
Accepted 25 September 2013
Available online 3 October 2013
Keywords:
Macrolide antibiotics
Clarithromycin derivatives
Desosamine
Ó 2013 Elsevier Ltd. All rights reserved.
Glycosylation
Antibacterial activity
Bacterial resistance
Macrolide antibiotics have been widely used for the treatment
of upper and lower respiratory tract infections (RTIs), renowned
for its safety and low toxicity since the discovery of erythromycin
in 1950s.1 Second generation macrolides, such as clarithromycin
and roxithromycin, improved the acid unstability and decreased
the gastric-intestine stimulation of erythromycin and as a result
were widely used clinically. However, the problem of antibacterial
resistance is becoming more and more serious due to the abuse of
antibiotics and hence great efforts have been made to develop no-
vel macrolide structures against resistant pathogens.2 The most
successful improvement of anti-resistant macrolide antibiotics is
well-known as ketolides, represented by telithromycin3 and
cethromycin.4 The structural features of ketolides include a 3-keto
group, a 11,12-cyclic or 6-carbamate and a tethered hetero-aro-
matic substituent. Ketolides, however, are not the only class of
macrolides against resistant bacterials. Among the many none
ketolides, 400-modified macrolide derivatives show a notable potent
activity against macrolide-sensitive and macrolide-resistant
pathogens. In 1989, Fernandes et al. found that compound A-
60565 had better activity than erythromycin against some consti-
tutively resistant Streptococcus pyogenes.5 During the refinement
process of A-60565, Su et al. discovered that compound CP-
544372, whose linker was as long as 6 atoms, exhibited a powerful
activity against several sensitive and resistant strains.6 Since then,
such 400-modified macrolides have been investigated by several re-
search groups. Ma et al. have designed and synthesized a series of
400-carbamate macrolide derivatives based on the structure of sec-
ond generation macrolides clarithromycin and azithromycin.7 In
their recent report, compound 1, the C-400 elongated aryl–alkyl
groups with eight atoms from the 400-oxygen atom to the terminal
benzene ring were the most effective against resistant bacteria.
Our group has also synthesized some C-400 modified clarithromycin
with significant anti-resistant activity, represented by compound 2
(Fig. 1).8
The anti-bacterial mechanism of macrolides has been illus-
trated as a reversible binding between the nucleotide A2058 in do-
main V of 23S rRNA of the 50S ribosomal subunit and the
macrolide molecule and block the protein synthesis and elonga-
tion, in which the 5-desosamine plays an important part.9 It has
been suggested that the 20-OH and 30-NMe2 form hydrogen bonds
with A2058 and A2059 and the stereo-shape of desosamine
matches with the binding pocket made of A2058, C2611 and
G2505.10 The major mechanisms of macrolide resistance are
erm-encoded methylation of 23S rRNA or mef-encoded efflux. The
expression of erm gene leads to the production of a methyltransfer-
ase which results in the methylation of A2058 and thus causing
resistance in MLSB (Macrolide–Lincosamide–Streptogramin B).11
The mechanism of anti-resistance of ketolides is through an extra
⇑
Corresponding author. Tel.: +86 10 63162006; fax: +86 10 6301775.
E-mail address: Lei@imm.ac.cn (P. Lei).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.09.083

D. Zhu et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6274–6279
6275
Figure 1. The chemical structure of important compounds mentioned above.
interaction between the 11,12-side chain and A752 in domain II of
the 23S rRNA. As for the 400-modified macrolides, Takashima et al.
proposed that the side chain might extend to the binding pocket
of choloromycetin.10
explore for novel macrolide antibiotics, we designed and synthe-
sized a series of novel clarithromycin derivatives with 400-O-desos-
amine substituents.
In this research, macrolide precursors 3 and 4 were conjugated
with 6-deoxy-desosamine derivatives 8 and 11, giving compound
12–15. Then, two target productions were yielded from each com-
pound through a reasonable de-protecting process. We focused on
the efficient synthesis of desosamine derivatives 8 and 11, as well
as a facile glycosylation at 400-OH group with desosamine exhibit-
ing weak alkaline, which differed from regular glycosylation with
neutral donors.
Macrolide precursor 3 was synthesized from commercially
available clarithromycin by selective acetylation of 20-OH with ace-
tic anhydride and triethylamine in a yield of 95%. Then we used
diphosgene to form the 11,12-carbonate without affecting 400-OH
to arrive at macrolide precursor 4 in a yield of 85% (Scheme 1).
400-OH modified macrolide derivatives, with ethers, esters and
carbamates as linkers, have been frequently investigated in recent
years. Using carbohydrate as a linker, however, is rarely seen. Our
group once reported a series of clarithromycin derivatives with
400-O-saccharide substituents, mainly some protected and unpro-
tected glucopyranosides.12 Without 11,12-modification, such
derivatives showed poor activity against either susceptible or
resistant strains. As far as we know, desosamine is involved in a
broad variety of biological phenomena.13 The neighboring hydro-
xyl and dimethylamino group is a classical structure to form
hydrogen bond with nucleotide receptors.14 To further delve into
the structure–activity relationship of 400-modified macrolides and

6276
D. Zhu et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6274–6279
Scheme 1. Synthesis of macrolide precursor 3 and 4. Reagents and conditions: (a) acetic anhydride, CH2Cl2, Et3N, rt, 12 h, 95%; (b) trichloro-methyl chloroformate, Py, CH2Cl2,
rt, 8 h, 85%.
Table 1
The modified desosamine glycosyl donors were synthesized
from commercially available methyl -glucopyranoside. The
synthesis of key compound 5 was based on the previously reported
route by Chen et al.15 Then the exhaustive acetolysis of methyl
pyranoside 5 was proceeded with a mixture of acetic anhydride,
acetic acid and sulfuric acid in proportion of 20:1:0.2, to yield 6
Methodology study of glycosylation between macrolide precursor and desosamine
donor
a-D
Conditions
Cat.
Cat. (eq.)
Donor.(eq.)
T (°C)
Yield (%)
1
2
3
4
5
TMSOTf
TESOTf
TESOTf
TESOTf
TESOTf
2
2
2
2
4
2
2
2
4
2
ꢀ20
ꢀ10
ꢀ20
ꢀ20
ꢀ20
5
24
45
47
30
for 68%, as a mixture of anomers. The product in
could be isolated with silica gel chromatography. The 1H signal
of -configuration could be observed at d 6.29 ppm, (d, J = 2.7 Hz,
a-configuration
a
1H). The acetyl of 6 was removed selectivity by hydrazine acetate
in DMF to yield compound 7 in a yield of 82%. Finally we treated
compound
7
with trichloroacetonitrile and 1,8-diazabicyclo
ize the basicity. TESOTf was chosen as a suitable catalyst, TMSOTf,
though more active, displayed stronger acidity than the 3-cladinose
could bear. The two tricholoroacetimidate donors 8 and 11 were ta-
ken respectively into glycosylation with macrolide precursor 3 and 4
according to the methodology studied. The catalyst, diluted with
CH2Cl2, was slowly added into the solution of donors at ꢀ20 °C,
and then the acceptors were added to avoid the de-cladinose side
reaction. The reaction was stirred at room temperature, to produce
compound 12–15 in a yield of 35–48%. For 2-OBz of the donors dis-
played effect of neighboring group participation in stereochemistry
of glycosylation,17 only b-configuration was obtained.
[5,4,0] undec-7-ene (DBU) to give the trichloroacetimidate 8 in a
yield of 79%, as one of our desosamine donors.
With the help of Yang et al. work,16 we developed a manipula-
tion to get compound 9 efficiently. In this case, compound 5 was
stirred with acetic anhydride, acetic acid and sulfuric acid in pro-
portion of 2:2:1 at room temperature to produce compound 9 in
a yield of 70%. Following the similar synthesis route of donor 8,
compound 11 was obtained in a yield of 83%, as another donor
(Scheme 2).
Precursor 4 and donor 11 were treated with different reagents
and conditions to study the methodology of glycosylation. The cat-
alysts were TMSOTf or TESOTf. The result is shown in Table 1. The
amount of catalyst used in the reaction was generally larger than
normal glycosylation, as the dimethylamino group existed on both
precursor and donor, thus consuming more Lewis acid to neutral-
The structures of compounds 12–15 were confirmed by 1H
NMR, 13C NMR and HRMS. For example, in compound 13 the signal
of H-1000 could be observed at d 4.63 ppm (d, J = 7.2 Hz, 1H), and in
13
the C NMR spectrum the signal of C-1000 carbon at d 109.8 ppm,
proving b-configuration.
Scheme 2. Synthesis of desosamine donors. Reagents and conditions: (a) to 6, Ac2O/AcOH/H2SO4 = 20:1:0.2, 50 °C, 8 h, 68%. Little amount of 9 was observed; (b) to 9, Ac2O/
AcOH/H2SO4 = 2:2:1, rt, 48 h, 70%; (c) H2NNH2ꢁHAc, DMF, 50 °C, 12 h, 75% for 7 and 79% for 10; (d) trichloroacetonitrile, DBU, CH2Cl2, 0 °C, 12 h, 79% for 8 and 83% for 11.

D. Zhu et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6274–6279
6277
In the following deprotection process, 12–15 were stirred in
MeOH at 65 °C for a 2–5 days to yield compound 16–23 in a yield
of 22–36%, including both all-deprotecting and part-deprotecting
products with a 2000-OBz. The difficulty to remove benzoyl group
on the 400-O-sugar was also reported in our previous work
(Scheme 3).12
Compound 24 was yielded as a side-product in the deprotecting
procedure of compound 15, under the presence of little amount of
K2CO3. We found that the 11,12-cabonate could not endure the ba-
sicity of K2CO3 thus forming the decarboxylation product before
the 2000-OBz could be removed at all. The signal of 11-H was able
to be identified at d 4.48 ppm (s, 1H), and 11-C at d 117.9 ppm
(Scheme 4).
The prepared clarithromycin 400-O-desosamine derivatives
16–24 were tested against a panel of representative respiratory
tract pathogens together with clarithromycin and telithromycin
as references. A variety of macrolide sensitive and resistant strains
were included to evaluate the activity of these novel 400-O-desos-
amine derivatives. Among them, ATCC29213, 12–32 are
methicillin-sensitive Staphylococcus aurous (MSSA), 12–31 is
Scheme 4. Reagent and conditions: (a) K2CO3 in little amount, MeOH, 65 °C, 2 days,
20%.
methicillin-reisistant Staphylococcus aurous (MRSA). ATCC12228,
11–20 and 12–5 are methicillin-sensitive Staphylococcus epidermi-
dis (MSSE), while 12–4, 12–11 are methicillin-resistant
N
N
AcO
O
HO
O
O
O
HO
HO
O
O
HO
HO
O
O
b
8/11, a
3
O
O
O
O
O
O
O
OR1
O
OR2
OCH3
OCH3
12, 13
16-19
HO
N
O
N
O
AcO
O
O
O
O
O
O
O
O
O
O
b
8/11, a
O
O
4
O
O
O
O
O
O
O
OR1
OCH3
O
OR2
OCH3
14, 15
20-23
R1
Yield (%)
37
R2
Yield (%)
34
12
14
13
15
16
N
N
BzO
OBn
BzO
HO
OBn
OBn
OH
20
17
21
18
22
19
23
29
31
28
22
26
36
34
O
O
N
35
48
47
O
N
N
O
BzO
HO
BzO
OAc
O
N
OH
O
Scheme 3. Glycosylation and deprotection. Reagents and conditions: (a) donors, TESOTf, 4 Å molecular sieves, CH2Cl2, ꢀ20 °C, 24 h. 35–48%; (b) MeOH, 65 °C, 2–5 days,
22–36%.

6278
D. Zhu et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6274–6279
Table 2
In vitro antibacterial activity (MICs,
lg/mL) of compounds 16–24
Pathogens
Clar.
Teli.
16
17
18
19
20
21
22
23
24
S. aureus
ATCC29213 MSSA
12–32 MSSA
12–31 MRSA
61
4
>32
62
62
>64
>32
>32
>32
>32
>32
>32
16
>32
>32
4
16
>32
>32
>32
>32
16
16
>32
2
2
>32
16
32
>32
>32
>32
>32
S. epidermidis
ATCC12228 MSSE
11–20 MSSE
12–5 MSSE
12–4 MRSE
12–11 MRSE
61
8
61
32
4
62
62
62
62
62
>32
>32
>32
>32
>32
>32
>32
>32
>32
>32
>32
32
32
16
16
16
8
16
4
>32
>32
>32
>32
>32
>32
16
16
>32
16
2
4
2
8
32
16
16
32
16
32
>32
32
>32
32
2
61
S. pyogenes
ATCC19625 ESSPy
11–2 ERSPy
61
4
62
2
>32
>32
32
>32
8
>32
2
>32
>32
>32
2
>32
1
32
2
16
1
>32
S. pneumoniae
12–1 ESSP
12–4 ESSP
12–8 ESSP
12–10 ESSP
12–14 ESSP
12–5 ERSP
12–6 ERSP
1
2
1
1
1
>32
>32
62
62
62
62
62
62
62
>32
32
2
16
32
32
32
32
32
32
32
32
32
32
>32
32
4
16
16
32
>32
2
16
2
2
4
>32
>32
16
>32
>32
>32
>32
2
32
2
1
4
1
16
1
2
2
2
4
1
1
2
16
16
1
>32
16
>32
>32
>32
>32
16
32
32
32
32
32
Staphylococcus epidermidis (MRSE). ATCC19625 is erythromycin-
sensitive Streptococcus pyogenes (ESSPy), and 11–2 is erythromy-
cin-resistant Streptococcus pyogenes (ERSPy). 12–1, 12–4, 12–8,
12–10 and 12–14 are erythromycin-sensitive Streptococcus pneu-
moniae (ESSP), while 12–5 and 12–6 are erythromycin-sensitive
Streptococcus pneumoniae (ERSP). All the strains chosen in this test
were supplied by Institute of Medicinal Biotechnology, Academy of
Medical Sciences & Peking Union Medical College. The in vitro
antibacterial activities were reported as minimum inhibitory con-
centrations (MICs), which was determined by the broth micro-
dilution method as recommended by the NCCLS.18 The result is
illustrated in Table 2.
the conformational fixation effect of 11,12-carbonate could be ben-
eficial. Finally, the anti-resistant activity of compound 19 and 22
implied that the 4000-OH of desosamine substituent played a vital
role in forming an extra hydrogen bonds.
In summary, with efficient synthesis of macrolide precursors
and desosamine donors, proper method of glycosylation, and rea-
sonable strategy of deprotection, a series of novel clarithromycin
400-O-desosaminyl derivatives were obtained. Some of them
showed activity against macrolide sensitive and resistant patho-
gens. Compounds 19 and 22 displayed significant improvement
of activity against macrolide sensitive pathogens and resistant S.
pneumoniae (MRSE), which offered us valuable information of the
SAR of macrolide 400-OH derivatives.
Compared with clarithromycin and telithromycin, compound
17 and 20 showed no activity against any susceptible or resistant
strains. Compound 16 preserved activity against a strain of S. pneu-
moniae (ESSP) 12–8. Compound 18 displayed better activity than
clarithromycin against a S. epidermidis resistant strain 12–4. Com-
pound 19 kept activity against several susceptible strains, like
MSSE 11–20, S. pyogenes (ESSPy) ATCC19625, S. pneumoniae (ESSP)
12–1, 12–8 and 12–10. Moreover, its activity against resistant S.
pneumoniae (MRSE) 12–4 and 12–11 was more powerful than clar-
ithromycin, or even equal to telithromycin. Compounds 21 and 23
showed activity against sensitive S. pyogenes strains (ESSPy)
ATCC19525, S. pneumoniae (ESSP) 12–1, 12–8 and 12–10. The same
with 24, which displayed activity against S. pyogenes (ESSPy)
ATCC19525 and S. pneumoniae (ESSP) 12–1. Among the nine sam-
ples compound 22 exhibited the most significant improvement of
activity against both susceptible and resistant strains. Its activity
was preserved or enhanced against almost all the sensitive strains,
and demonstrated a notable potent activity against resistant S.
pneumoniae (MRSE) 12–4 and 12–11.
From the data mentioned above it could be seen that seven of
nine compounds showed activity, against some sensitive strains,
and compound 19 and 22 displayed potent activity against resis-
tant strain MRSE. It might be concluded that desosamines were
better substituents than glucopyranosides,12 which might be for
their stronger ability to form hydrogen bonds. The compounds
having free 4000-OH (compound 18, 19, 22, 23, 24) possessed higher
anti-bacterial activity than those with 4000-OBn (16, 17, 20, 21),
probably 4000-OBn were too bulky to fit the binding pocket, and
Acknowledgments
This work was supported by the National S&T Major Special
Project on Major New Drug Innovation (Item Number:
2008ZX09401-004 and 2009ZX09301-003) and National Natural
Science Foundation of China (81072524). And we are grateful to
Professor Xuefu You for the antibacterial activity test.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.
09.083.
References and notes
1. Galm, U.; Shen, B. Chem. Biol. 2007, 10, 1098.
2. Walsh, C.; Wright, G. Chem. Rev. 2005, 2, 391.
3. Velvadapu, V.; Glassford, I.; Lee, M.; Paul, T.; DeBrosse, C.; Klepacki, D.; Small,
M. C.; MacKerell, A. D.; Andrade, R. B. Med. Chem. Lett. 2012, 3, 211.
4. Plata, D. J.; Leanna, M. R.; Rasmussen, M.; McLaughlin, M. A.; Condon, S. L.;
Kerdesky, F. A. J.; King, S. A.; Peterson, M. J.; Stoner, E. J.; Wittenberger, S. J.
Tetrahedron 2004, 45, 10171.
5. Omura, S. Macrolide Antibiotics; Academic Press, 2002. Chapter 6.
6. Ju, Y.; Xian, R.; Zhang, L.; Ma, R.; Cao, J.; Ma, S. Bioorg. Med. Chem. Lett. 2010, 20,
3212.
7. Ma, S.; Jiao, B.; Ju, Y.; Zheng, M.; Ma, R.; Liu, L.; Zhang, L.; Shen, X.; Ma, C.; Meng,
Y.; Wang, H.; Qi, Y.; Ma, X.; Cui, W. Eur. J. Med. Chem. 2011, 46, 556.
8. Xu, P.; Liu, L.; Jin, Z.; Wang, G.; Liu, J.; Li, Y.; Lei, P. Bioorg. Med. Chem. Lett. 2008,
18, 5507.

D. Zhu et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6274–6279
6279
9. Katz, L.; Ashley, G. W. Chem. Rev. 2005, 2, 499.
10. Takashima, H. Curr. Top. Med. Chem. 2003, 3, 991.
15. Chen, X.; Xu, P.; Xu, Y.; Liu, Y.; Zhu, D.; Lei, P. Bioorg. Med. Chem. Lett. 2012, 22,
7402.
11. Zhanel, G. G.; Walters, M.; Noreddin, A.; Vercaigne, L. M.; Wierzbowski, A.;
Embil, J. M.; Gin, A. S.; Douthwaite, S.; Hobanl, D. J. Drugs 2002, 62, 1771.
12. Xu, P.; Chen, X. Z.; Liu, L.; Jin, Z. P.; Lei, P. S. Bioorg. Med. Chem. Lett. 2010, 20,
5527.
13. Velvadapu, V.; Andrade, R. B. Carbohydr. Res. 2008, 343, 145.
14. Achatz, S.; Domling, A. Bioorg. Med. Chem. Lett. 2006, 24, 6360.
16. Cao, Y.; Okada, Y.; Yamada, H. Carbohydr. Res. 2006, 341, 2219.
17. Chen, L.; Kong, F. Tetrahedron Lett. 2003, 44, 3691.
18. National committee for clinical laboratory standards, methods for dilution
antimicrobial susceptibility tests for bacteria that grow aerobically, 5th ed.
2000; Vol. 20. approved standard: NCCLS document M7–A5.