Synthesis of 9-oxime-11,12-carbamate ketolides through a novel N-deamination reaction of 11,12-hydrazonocarbamate ketolide

Article abstract of DOI:10.1016/S0968-0896(03)00147-0

A series of 9-oxime-11,12-carbamate ketolides was synthesized for the first time through a key 11,12-hydrazonocarbamate intermediate that was first oximated and further deaminated to give the corresponding carbamate. The N-N bond cleavage was achieved thr

Full text of DOI:10.1016/S0968-0896(03)00147-0

Bioorganic & Medicinal Chemistry 11 (2003) 2389–2394
Synthesis of 9-Oxime-11,12-carbamate Ketolides Through a Novel
N-Deamination Reaction of 11,12-Hydrazonocarbamate Ketolide
Alexis Denis,a,* Jean-Marie Pejac,a Francois Bretina and Alain Bonnefoyb
aMedicinal Chemistry Department, Aventis Pharma 102 route de Noisy, Romainville, Cedex F-93235, France
bAnti-infective Diseases Group, Aventis Pharma 102 route de Noisy, Romainville Cedex, F-93235, France
Received 9 October 2002; accepted 28 February 2003
Abstract—A series of 9-oxime-11,12-carbamate ketolides was synthesized for the first time through a key 11,12-hydrazonocarbamate
intermediate that was first oximated and further deaminated to give the corresponding carbamate. The N–N bond cleavage was
achieved through an original new reaction using glycoaldehyde dimer as deaminating reagent. The new compounds synthesized were
shown to display improved antibacterial activities against Streptococcus pneumoniae and S. pyogenes resistant to erythromycin.
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
greatly improved (20–0.15 mg/mL) and even more, that
of telithromycin (5–0.02 mg/mL).
Macrolides, including erythromycin, are an old and
well-known family of oral antibiotics. Their spectrum of
activity covers most relevant bacterial species respon-
sible for upper and lower respiratory tract infections.1
However, the extensive clinical application of macrolide
antibiotics has resulted in an increasing emergence of
macrolide MLSB resistance in respiratory pathogens
like S. pneumoniae.2 The ketolides,3 such as teli-
thromycin4 or ABT-733,5 are a new class of semisyn-
thetic erythromycin derivatives recently generated to
address the problem of erythromycin resistance. They
are very active against erythromycin-resistant and peni-
cillin resistant S. pneumoniae, and the most advanced
compound of this new class, telithromycin has demon-
strated clinical efficacy in human respiratory tract
infections.6 During our initial optimization of the keto-
lide series,7 it has been observed that the antibacterial
activity of the 11,12carbamate I could be optimized
through the introduction of alkyl-aryl or hetero-aryl
side chains to give II and finally telithromycin (Fig. 1).
On the other hand, a second series of (E)-9-oxime deri-
vatives has been described7 among which, the (R)-3-
piperidinyl-oxime III displayed interesting activities
Whereas the carbamate I is weakly active or inactive
(Table 1) against S. pneumoniae and S. pyogenes resis-
tant to erythromycin (40–2.5 mg/mL), the activity of II is
*Corresponding author at current address: Pfizer, Global Research
and Development, 3–9 rue de la Loge, Fresnes, BP-100, F-94265,
France. Tel.: +33-1-4096-7883; fax: +33-1-4096-7687; e-mail: alexis.
denis@pfizer.com
Figure 1. Structure of ketolide series.
0968-0896/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00147-0

2390
A. Denis et al. / Bioorg. Med. Chem. 11 (2003) 2389–2394
Table 1. In vitro activity of 9-oxime modified ketolides
MIC 24 h (mg/mL)
S. a. EryS
S. a. EryRi
S. a. EryRc
S. pyo EryS
S. pyo EryRc
S. p. EryS
S. p. EryRi
S. p. EryRc
CLA
TEL
I
0.3
0.04
0.6
0.04
0.3
0.3
0.6
0.3
>40
0.08
0.6
>40
>40
>40
>40
>40
>40
>40
>40
0.08
ꢀ0.025
0.04
>40
>40
0.04
ꢀ0.02
0.022
ꢀ0.020.3
0.08
>40
ꢀ0.02
>40
ꢀ0.04
.5
5
II
0.08
0.3
0.3
0.3
0
ꢀ0.022
0.15
III
VIII
XI
0.3
0.3
>40
5
0.3
0.15
0.3
2.5
1.2
1.2
0.6
0.02
0.08
0.020.6
2.5
0.0.215
X
0.3
0.6
MIC, minimum inhibitory concentration; CLA, clarithromycin; TEL, telithromycin; EryS, susceptible; EryRc, constitutive MLS resistance; EryRi,
inducible MLS resistance. S. a., Staphylococcus aureus; S. pyo., Streptococcus pyogenes; S. p., Streptococcus pneumoniae.
against the inducibly and constitutively resistant S.
pneumoniae (2.5–0.3 mg/mL). However, III is inactive
against the constitutively resistant S. pyogenes (Table 1).
hydrazonocarbamate function. The minimized structure
of IV, built from the crystallized hydrazonocarbamate
HMR 3004,7 revealed that the overall macrocyclic
backbone was very rigid and almost identical to crys-
tallized ketolide. Accordingly, the nitrogen of the
hydrazonocarbamate and the oxygen of the carbonyl at
position 9 were shown to be, as in the parent structure,
close enough (2.03 A) to give an hydrogen bond (Fig.
2). Even if that intramolecular bond is weak (the very
long reaction time could reflect this), it could be suffi-
cient to promote the activation of the carbonyl and
allow the reaction to occur.
Whereas the synthesis and the activity of 9-oxime-
hydrazonocarbamate ketolides has already been repor-
ted,8 a combination of such chemical features has never
been reported in the 11,12carbamate series. This is why,
in continuation of our efforts to further improve the
ketolide series, we wondered whether the addition of the
(R)-3-piperidinyl-oxime feature into the cyclic carba-
mate skeleton would generate new compounds with
improved activity against resistant strains.
Having now in hand the expected 9-oxime-hydrazono-
carbamate ketolide V, we decided to use that compound
to access the corresponding carbamate through a de-
amination reaction followed by introduction of the cor-
responding side chain by nucleophilic alkylation of the
free cyclic carbamate.
Chemistry
The first attempts to introduce an oxime in position 9 of
the 11,12carbamate I were all unsuccessful. Whatever
the conditions: acidic, basic, temperature, the oximation
reaction did not occur. Therefore, we decided to come
back to the corrresponding hydrazonocarbamate IV as
starting material and to react it with an hydroxylamine
according to the published conditions.8 In contrast to I,
IV gave easily the E9 oxime V in 77% yield when reac-
ted with the CBz protected (R)-3-piperidine-O-hydroxyl-
amine in EtOH, HCl (pH 4) at reflux for 6 days
(Scheme 1).
The deamination was carried out by treating V with the
dimeric glycolaldehyde in acetic acid. This very unusual
reaction was unexpectedly discovered in our labora-
We hypothesized that the dramatic change of reactivity
of the hydrazonocarbamate series compared to the car-
bamate, could be explained by an intramolecular acti-
vation of the carbonyl group by the NH2 of the
Figure 2. Minimized conformation of IV and proposed activation of 9
carbonyl through intramolecular hydrogen bonding.
Scheme 1. Oximation reaction.

A. Denis et al. / Bioorg. Med. Chem. 11 (2003) 2389–2394
2391
tories during the synthesis of a series of hydrazono-
carbamate ketolides by reductive amination of different
aldehydes. We observed that when IV was reacted with
the dimeric glycoaldehyde and NaBH3CN to obtain VI,
20% of the deaminated compound I was identified as a
side product (Scheme 2).
The desired compound was thus obtained with 42%
yield only, 25% of starting hydrazonocarbamate being
recovered with degradations products. The mechanism
of that reaction is unknown, however, the fact that
amongst several other aldehydes, dimeric glycoaldehyde
was the only one to give that side reaction and that the
reaction is incomplete may suggest that the reaction
involves two molecules of macrolides for one glyco-
aldehyde dimer (Scheme 3).
The deamination of a hydrazine with glycoaldehyde has
never been reported. Literature methods rather descri-
bed the use of Zn,10 Pb(OAc)2,11 Raney Ni12 or nitrous
deamination13 to carry out the N–N bond cleavage.
Whereas we do not have experimental data support-
ing this hypothesis, one can propose that the dimeric
intermediate species could undergo a complex rearrange-
ment through intramolecular activation of one nitro-
gen allowing the formation of the carbamate and an
hypothetical hydroxyacetonitrile intermediate which
may decompose into the starting hydrazonocarbamate
and degradation products. However, this hypothesis
We then decided to react V with glycoaldehyde dimer
without a hydride to avoid the formation of the 9-oxime
analogue of VI (Scheme 4).
Scheme 2. Unexpected deamination reaction during reductive amina-
tion of glycoaldehyde by IV.
Scheme 3. Proposed mechanism for deamination of hydrazonocarba-
mate ketolides by dimeric glycoaldehyde.
Scheme 4. Synthesis of 9-oxime-carbamate and hydrazonocarbamate ketolides. Reagents and conditions: (a) 1.1 equiv glycoaldehyde dimer,
CH3CO2H/CH2Cl2 1/1 vol, rt, 12h, 42%; (b) Pd(OH) 2, cyclohexene, dioxane, 80 ꢁC, 24 h, 42–93%; (c) Ac2O, CH2Cl2, rt, 93%; (d) (Boc)2O 4 equiv,
DMAP 0.4 equiv, triethylamine 3.5 equiv, THF, 24 h, rt, 50%; (e) NaH, cinnamyl bromide, DME, 0 ꢁC, 65%; (f) 1-MeOH, 4 h, reflux. 2-CH2Cl2,
TFA. 3-H2, Pd/C, dioxane, 4 h, rt, 43% overall yield; (g) 3-phenyl-propionaldehyde, NaBH3CN, AcOH 3 equiv, MeOH, 24 h, rt; (h) H2, Pd/C,
MeOH, 24 h, rt, 36%.

2392
A. Denis et al. / Bioorg. Med. Chem. 11 (2003) 2389–2394
would require further mechanistic studies to be fully
assessed.
the gain of activity obtained with X against the con-
stitutively resistant S. pyogenes (0.6 mg/mL compared to
5 mg/mL) clearly confirms the interest of this series to
further improve the ketolide family against gram posi-
tive pathogens. The potential of these oximes remains to
be evaluated in vivo.
After further deprotection with Pd(OH)2/cyclohexene
the unsubstituted oxime VIII was obtained (Scheme 4).
The introduction of the phenyl-butyl side chain was
carried out in six steps from V. After deamination, the
20OH was protected with Ac2O, and the Cbz group
removed under the same condition than VIII. Next, a
treatment with an excess of Boc2O, triethylamine and
DMAP in THF allowed the simultaneous protection of
the free piperidine nitrogen and of the nucleophilic car-
bon at position 2which was now converted into the
corresponding O-Boc enol ester VII.
Conclusion
An original chemical strategy and deamination reaction
has allowed us to synthesize for the first time a 9-oxime-
11,12carbamate ketolide. The potential of this new
modification in the ketolide series has been confirmed
by the very potent activities obtained with compound X
against the gram-positive respiratory pathogens either
constitutively or inducibly resistant to erythromycin.
This new series needs to be further optimized and eval-
uated in vivo.
We had already observed within the ketolide series,
that an O-alkylation14 or acylation can easily occur at
position 3 with various electrophiles to convert the
reactive beta-keto-ester function into the correspond-
ing enol ether or ester. The remaining nitrogen of the
cyclic carbamate function was then alkylated with
cinnamyl bromide using NaH in DME in 65% yield
to give IX.
Experimental
Molecular modelling IV
All the protecting groups were then removed without
any purification starting first with a treatment with
MeOH followed by hydrolysis of the Boc groups with
TFA in dichloromethane. Finally, the double bond was
hydrogenated with H2 in the presence of palladium/
charcoal to give the desired 9-(R)-3-piperidinyl oxime
11,12carbamate X in 43% yield for the three steps. The
corresponding hydrazonocarbamate XI, was synthe-
sized according to the previous conditions,7 by reductive
amination of 3-phenyl propanaldehyde with NaBH3CN
followed by deprotection of the CBz group by catalytic
hydrogenation.
The calculations were run on a silicon graphics compu-
ter using the Accelerys software Insight II with the
CVFF forcefield. The initial starting conformation was
built from the crystallographic coordinates7 of the N-
substituted hydazonocarbamate HMR 3004.
V. A solution of IV (1 g, 1.59 mM) and (3R)-(amino-
oxy)-1-(carbobenzyloxy)piperidine (1.19 g, 4.77 mM) in
10 mL of EtOH was adjusted to pH 4 with 5 mL of a
0.5 N HCl/EtOH solution. The mixture was stirred at
reflux for 6 days. After evaporation to dryness, the
residue was taken up with ethyl acetate, washed with
aqueous ammonium hydroxide. After drying over
MgSO4 and evaporation of the solvent the residue was
purified by column chromatography over silica eluting
with 96/4/0.5 CH2Cl2/MeOH/ammonium hydroxide to
give 1 g (77%) of V as a white foam.
Results and Discussion
All the ketolides were tested in vitro by standard agar
dilution method against both erythromycin-susceptible
and erythromycin-resistant S. aureus, S. pyogenes and
S. pneumoniae including constitutive (EryRc) and indu-
cible (EryRi) phenotypes. In addition, one strain of
Haemophilus influenzae was also tested. As already
shown for all the ketolides, the compounds were inac-
tive against the erythromycin-resistant (MLSB con-
stitutive type) strains of S. aureus (MIC >40 mg/mL).
Spectral data for V. MS=860+ (MH+); 1H NMR
(400 MHz, CDCl3): d 0.88 (sl, 3H) CH3CH2, 0.97 (d,
3H) 8-CH3, 1.13 (s, 3H) 10-CH3, 1.23 (3H) 50-Me, 1.29
(d, 3H) 4-CH3, 1.36 (d, 3H) 2-CH3, 1.39–1.44 (6H) 6-
CH3 and 12-CH3, 2.27 (s, 6H) N(CH3)2, 2.45 (m, 1H)
0
H3 , 2.61 (s, 1H) H10, 2.68 (s, 3H) 6-OCH3, 3.07 (m, 1H)
0
3.86 (q, 1H) H2, 3.89 (s, 1H) H11, 4.24 (d, 1H) H5, 4.3
0
H4, 3.18 (m, 1H) H2 , 3.54 (s, 1H) H5 , 3.66 (m, 1H) H8,
The positive impact of the combination of the piper-
idinyl-oxime and the carbamate was first observed with
the non substituted carbamate VIII. The MICs against
the S. pyogenes and S. pneumoniae resistant strains were
2–3 times lower than with the starting analogues I. In
agreement with the previous observations made in the
ketolide series, the addition of a lateral side chain gave
the expected increase of activity. Compound X was very
potent against all the resistant streptococci tested what-
ever the phenotype (MICs 0.15–0.6 mg/mL). Similar to
the non-oxime series, the hydrazonocarbamate analo-
gue XI was two times less active than the carbamate X
against the resistant strains. Compared to telithromycin,
0
(d, 1H) H1 , 5.1 (m, 3H) H13 and OCH2Ph, 7.35 (m, 5H)
phenyl.
VIII-NCbz. To a solution of V (9 g, 10.41 mM) in 40
mL of CH2Cl2 and 40 mL of acetic acid was added 1.38
g (11.5 mM) of glycoaldehyde dimer (Aldrich). After
stirring for one night at room temperature, the reaction
was transfered over 350 mL of 2N sodium hydroxide,
washed with water and brine. Drying over MgSO4 and
evaporation of the solvent afforded 9.97 g of crude
product. The residue was purified by column chroma-
tography over silica eluting with 98/2/0.5 CH2Cl2/

A. Denis et al. / Bioorg. Med. Chem. 11 (2003) 2389–2394
2393
MeOH/ammonium hydroxide to give 3.67g (42%) of
VIII–NCbz as a white foam.
VII. To a solution of VIII-20OAc (1.38 g, 1.84 mM) in
50 mL of THF, were added 0.9 mL (3.5 equiv) of tri-
ethylamine and 44 mg (0.37 mM, 0.2equiv) of 4- N,N-
dimethyl-amino-pyridine. The reaction was stirred at
room temperature for 24 h and evaporated to dryness.
The crude product was purified by column chromato-
graphy over silica eluting with 9/1 ethyl acetate/triethyl-
amine to afford 0.866 g (50%) of VII.
1
Spectral data for VIII NCbz. MS=845+ (MH+); H
NMR (400 MHz, CDCl3): d 0.88 (t, 3H) CH3CH2, 0.94
(d, 3H) 8-CH3, 1.19 (d, 3H) 10-CH3, 1.2–1.7 (m, 4H)
CH2 piperidine, 1.23 (3H) 50-Me, 1.22 and 1.66 (2H) 40-
CH2, 1.28 (d, 3H) 4-CH3, 1.37 (d, 3H) 2-CH3, 1.39 and
1.62(2H) 7-CH2, 1.36-1.48 (6H) 6-CH3 and 12-CH3,
0
2.27 (s, 6H) N(CH3)2, 2.45 (m, 1H) H3 , 2.44 (m, 1H)
H10, 2.64 (s, 3H) 6-OCH3, 3.06 (m, 1H) H4, 3.18 (dd,
Spectral data for VII. MS=953.5+ (MH+).
0
1H) H2 , 3.3–3.65 and 3.98 (5H) CH2NCO and
IX. To a solution of VII (0.82g, 0.86 mM) in dry
dimethoxy-ethane under nitogen were added at 0 ꢁC
55 mg (1.37 mM) of sodium hydride 60% in oil. The
reaction was stirred 30 mn at 0 ꢁC and 236 mg (1.11
mM) of cynnamyl bromide diluted in 3 mL of dime-
thoxy-ethane were added. The reaction was then stirred at
room temperature for 3 h and stooped with the addition
of a 10% NaHPO4 solution. After extraction with ethyl
acetate, drying over MgSO4 and evaporation of the sol-
vent, the residue was purified by column chromatography
over silica eluting with 9/1 isopropylic ether/triethylamine
to give 0.607 g (65%) of IX.
0
1H) H2, 3.87 (s, 1H) H11, 4.22 (d, 1H) H5, 4.3 (d, 1H)
¼NOCH, 3.54 (m, 1H) H5 , 3.64 (m, 1H) H8, 3.85 (q,
0
5H) phenyl.
H1 , 4.83 (dl, 1H) H13, 5.12(mL, 2H) OCH2Ph, 7.35 (m,
VIII. To a solution of VIII-NCbz (0.6 g, 0.71 mM) in 20
mL of dioxane, were added 0.2g of Palladium hydroxide
and 0.605 mL of cyclohexene (7.36 mM,10 equiv). After
stirring 16 h at 80ꢁC, the reaction was filtered of and eva-
porated. The residue was purified by column chromato-
graphy over silica eluting with 97/3/0.5 CH2Cl2/MeOH/
ammonium hydroxide to give 0.21g (42%) of VIII.
Spectral data for IX. MS=1083.8 (MH+).
1
Spectral data for VIII. MS=711+ (MH+). H NMR
(400 MHz, CDCl3): d 0.88 (t, 3H) CH3CH2, 0.97 (d, 3H)
8-CH3, 1.2(d, 3H) 10-CH3, 1.25 (3H) 50-Me, 1.27 (d,
3H) 4-CH3, 1.22 and 1.66 (2H) 40-CH2, 1.37 (d, 3H) 2-
CH3, 1.41 and 1.63 (2H) 7-CH2, 1.39–1.43 (6H) 6-CH3
and 12-CH3, 1.47–1.58–1.66 (m, 4H) ¼NOCHCH2CH2,
X. Step A—A solution of IX (0.6 g, 0.554 mM) in 20
mL of methanol was stirred at room temperature for 4 h
and evaporated to dryness to obtain 0.48 g of crude
product. Step B—0.48 g of product of step A were stir-
red in 5 mL of dichloromethane and 5 mL of tri-
fluoroacetic acid for 2h at room temperature. The
reaction was taken up with ethyl acetate, washed with
sodium hydroxide, dried over MgSO4 and evaporated
to dryness to afford 0.465 g of crude product. Step C—
A solution of 0.465 g of product of step B and 0.2g of
palladium on charcoal in 15 mL of dioxane was stirred
under 1.5 atm of hydrogen for 4 h. After filtration and
and evaporation to dryness, the residue was purified by
column chromatography over silica eluting with 96/4/
0.5 CHCl3/MeOH/ammonium hydroxide to give 0.2g
(43%) of X.
0
2.24 (s, 6H) N(CH3)2, 2.43 (m, 1H) H3 , 2.44 (m, 1H)
H10, 2.65 (sl, 3H) 6-OCH3, 2.65–2.83–3.11 (4H)
¼NOCHCH2NHCH2, 3.05 (m, 1H) H4, 3.18 (dd, 1H)
0
0
H2 , 3.90 (m, 1H) ¼NOCH, 3.55 (m, 1H) H5 , 3.68 (m,
1H) H8, 3.83 (q, 1H0) H2, 3.87 (s, 1H) H11, 4.22 (d, 1H)
H5, 4.3 (d, 1H) H1 , 5.14 (dd, 1H) H13, 5.81 m, 1H)
NHCO. Anal. calcd (%) for C36H62N4O10: C 60.82, H
8.79, N 7.88. Found: C 60.7, H 8.9, N 7.9.
VIII-NCbz-20OAc. To a solution of VIII-NCBz (3.66 g,
4.33 mM) in 60 mL of CH2Cl2 was added 0.87 mL (11.5
mM) of acetic anhydride. After stirring 3 h at room
temperature, the reaction was washed with ammonium
hydroxide and dried over MgSO4 to afford after eva-
poration 3.59 g (93%) of VIII-NCbz-20OAc. Spectral
data for VIII-NCbz-20OAc: MS=887+ (MH+); 1H
NMR (400 MHz, CDCl3): d 0.88 (t, 3H) CH3CH2, 0.95
(d, 3H) 8-CH3, 1.13–1.19–1.24–1.37 (12H) 10-CH3, 50-
Me, 4-CH3 and 2-CH3, 1.33–1.49 (6H) 6-CH3 and 12-
CH3, 2.04 (s, 3H) OCOCH3, 2.25 (s, 6H) N(CH3)2, 2 .43
Spectral data for X. MS=843+ (MH+); 1H NMR
(400 MHz, CDCl3): d 0.85 (t, 3H) CH3CH2, 0.97 (d,
3H) 8-CH3, 1.05 (d, 3H) 10-CH3, 1.26 (d, 3H) 50-Me,
1.28 (d, 3H) 4-CH3, 1.35 (d, 3H) 2-CH3;1.40–1.46 (s,
6H) 6-CH3 and 12-CH3, 1.23 and 1.68 (m, 2H)
40-CH2, 1.35 and 1.65 (m, 2H) 7-CH2, 1.35–1.64 (m,
4H) ¼NOCHCH2CH2, 1.63 and 1.82(m, 4H)
NCH2CH2CH2CH2phenyl, 2.26 (s, 6H) N(CH3)2, 2.45
0
0
(q, 1H) H10, 2.62 (sl, 3H) 6-OCH3, 2.6–2.9 (m, 1H) H3 ,
0
(m, 1H) H3 , 2.62 (m, 1H) H10, 2.66 (s, 3H) 6-OCH3,
3.01 (m, 1H) H4, 3.35–3.7 (m, 6H) H5 , CH2NCO and H8,
2.67–2.78 (m, 2H) CH2Ph, 2.64 and 2.76 (m, 4H)
¼NOCHCH2NHCH2, 3.12(m, 1H) H4, 3.18 (dd, 1H)
3.98 (1H) ¼NOCH, 3.81 (q, 1H)0H2, 4.13 (d, 1H) H5, 4.38
OCH2Ph, 5.76–6.05 (m, 1H) NHCO, 7.35 (m, 5H) phenyl.
0
0
H2 , 3.62–3.80 (m, 2H) CH2NCO, 3.97 (m, 1H)
(d, 1H) H1 , 4.74 (dd, 1H) H2 , 4.9–5.2(2H) H13 and
0
1H) H2, 3.76 (s, 1H) H11, 4.24 (d, 1H) H5, 4.29 (d, 1H)
H1 , 5.02(dd, 1H) H13, 7.16–7.22 (m, 5H) phenyl. Anal.
¼NOCH, 3.54 (m, 1H) H5 , 3.73 (m, 1H) H8, 3.87 (q,
VIII-20OAc. To a solution of VIII-NCbz-20OAc (1.5g,
1.69 mM) in 60 mL of dioxane, were added 1.2g of
Palladium hydroxide and 2.7 mL of cyclohexen. After
stirring 24 h at 80 ꢁC, the reaction was filtered of and
evaporated to give 1.18 g (93%) of crude product.
Spectral data for VIII-20OAc: MS=753+ (MH+).
0
calcd (%) for C46H74N4O10: C 65.53, H 8.85, N 6.65.
Found: C 65.45, H 8.8, N 6.55.
XI. Step A—A solution of V ( 0.311 mg, 0.367 mM) and
0.099 g (0.734 mM) of 3-phenylpropanaldehyde in 10

2394
A. Denis et al. / Bioorg. Med. Chem. 11 (2003) 2389–2394
mL of methanol and 66 mL of acetic acid was stirred at
romm temperature for 24 h. Sodium cyanoborohydride
(0.069 g, 1.1 mM) was added and the reaction was stir-
red for 3 days at room temperature. The reaction was
taken up with ethyl acetate, washed with 1 N sodium
hydroxide, water and brine and evaporated to dryness
to yield 0.4 g of crude product.
References and Notes
1. Bryskier, A. J.; Butzler, J.-P.; Neu, H. C.; Tulkens, P. M.,
Eds.; Macrolides: Chemistry, Pharmacology and Clinical Uses.
Arnette Blackwell: Paris, 1993.
2. (a) Amsden, G. W. J. Antimicrob. Chemother. 1999, 44, 1.
(b) Perez-Trallero, E. J. Antimicrob. Chemother. 2000, 45, 401.
3. (a) For recent reviews, see: Schonfeld, W.; Kirst, H. A.,
Eds., Macrolide Antibiotics. Birhauser: Basel, 2002, p 97. (b)
Kaneko, T.; McArthur, H.; Sutcliffe, J. Exp. Opin. Ther.
Patents 2000, 10, 403.
4. Denis, A.; Agouridas, C.; Auger, J.-M.; Benedetti, Y.;
Bonnefoy, A.; Bretin, F.; Chantot, J.-F.; Dussarat, A.; Fro-
`
mentin, C.; Gouin D’Ambrieres, S.; Lachaud, S.; Laurin, P.;
Le Martret, O.; Loyau, V.; Tessot, N.; Pejac, J.-M.; Perron, S.
Biorg. Med. Chem, Lett. 1999, 9, 3075.
Step B—A solution of 0.4 g (0.409 mM) of product of
step A and 0.2g of palladium on charcoal in 10 mL of
methanol was stirred under 1.5 atm of hydrogen for 24 h.
After filtration and and evaporation to dryness, the residue
was purified by column chromatography over silica eluting
with 92/8/0.5 CH2Cl2/MeOH/ammonium hydroxide to
give 0.127 g (36% overall yield for steps A and B) of XI.
5. Or, Y. S.; Clark, R. F.; Wang, S.; Chu, D. T.; Nilius, A. M.;
Flamm, R. K.; Mitten, M.; Ewing, P.; Alder, J.; Ma, Z. J.
Med. Chem. 2000, 43, 1045.
6. Barman Balfour, J. A.; Figgitt, D. P. Drug 2001, 61, 815.
7. Agouridas, C.; Denis, A.; Auger, J.-M.; Benedetti, Y.; Bon-
nefoy, A.; Bretin, F.; Chantot, J.-F.; Dussarat, A.; Fromentin,
1
Spectral data for XI. MS=844.7+ (MH+); H NMR
(400 MHz, CDCl3): d 0.85 (t, 3H) CH3CH2, 1.00 (d, 3H)
8-CH3, 1.12(d, 3H) 10-CH3, 1.26 (d, 3H) 50-Me, 1.30 (d,
3H) 4-CH3, 1.36 (d, 3H) 2-CH3, 1.41-1.47 (s, 6H) 6-CH3
and 12-CH3, 1.22 and 1.68 (m, 2H) 40-CH2, 1.38 and 1.7
(m, 2H) 7-CH2, 1.5–2(m, 4H) ¼NOCHCH2CH2, 1.83
(m, 2H) NCH2CH2CH2phenyl, 2.27 (s, 6H) N(CH3)2, 2 .45
C.; Gouin D’Ambrieres, S.; Lachaud, S.; Laurin, P.; Le Mar-
`
tret, O.; Loyau, V.; Tessot, N. J. Med. Chem. 1998, 41, 4080.
8. (a) Agouridas, C.; Denis, A.; Chantot, J.-F.; Fromentin, C.;
Pejac, J.-M. WO Patent 9838199, 1998. (b) Kaneko, T.;
McMillen, W. T. WO Patent 200044761, 2000.
9. Stereochemistry of oxime confirmed by NMR according to
`
ref Gasc, J. C.; Gouin d’Ambrieres, S.; Lutz, A.; Chantot, J. F.
New ether oxime of erythromycin A and: a structure activity
relationship study. J. Antibiot. 1991, 44, 313.
10. Evans, D. A.; Johnson, D. S. Organic Lett. 1999, 1, 595.
11. Gmeiner, P.; Bollinger, B.; Mierau, J.; Hofner, G. Arch.
Pharm. 1995, 328, 609.
0
(m, 1H) H3 , 2.67 (m, 1H) H10, 2.67 (s, 3H) 6-OCH3, 2 .67–
2.87–3.04 (m, 8H) CH2NHNCO, ¼NOCHCH2NHCH2
0
and CH2Ph, 3.10 (m, 1H) H4, 3.18 (dd, 1H) H2 , 3.55 (m,
0
1H) H5 , 3.73 (m, 1H) H8, 3.86 (m, 1H) =NOCH, 3.87 (s,
1H) H11, 3.88 (q, 1H) H2, 4.27 (d, 1H) H5, 4.29 (d, 1H)
H1 , 5.07 (dd, 1H) H13, 7.15–7.25 (m, 5H) phenyl. Anal.
calcd (%) for C45H73N5O10: C 64.03, H 8.72, N 8.03.
Foundnbsp;: C 64.49, H 9.55, N 7.49.
0
12. Atkinson, R. S.; Judkins, B. D. J. Chem. Soc., Perkin
Trans. 1981, 509.
13. Milcent, R.; Guevrekian-Soghomoniantz, M.; Barbier, G.
J. Heterocycl. Chem. 1986, 23, 1845.
Acknowledgements
The authors thank C. Lang and F. Maquin for the
analytical experiments and Pr. Peter Goeckjian for
helpful discussion.
14. Denis, A.; Bretin, F.; Fromentin, C.; Bonnet, A.; Piltan,
G.; Agouridas, C.; Bonnefoy, A. Bioorg. Med. Chem. Lett.
2000, 2019.