Design, synthesis, and evaluation of stable and taste-free erythromycin proprodrugs

Article abstract of DOI:10.1021/jm049155y

Erythromycin A is normally formulated for children as its 2′-ethyl succinate, a taste-free prodrug. Unfortunately, the prodrug hydrolyzes at a measurable rate in the medicine bottle, leading to the vile-tasting erythromycin. We have prepared derivatives of erythromycin B as putative paediatric prodrugs, taking advantage of the much improved acid stability of erythromycin B relative to erythromycin A. Thus, erythromycin B enol ether ethyl succinate is very poorly soluble in water, and its hydrolysis is undetectable in conditions resembling the medicine bottle. In acid, however, it converts rapidly to erythromycin B 2′-ethyl succinate, and this is in turn hydrolyzed to erythromycin B in neutral and basic conditions. Derivatives of erythromycin B enol ether are therefore proposed as taste-free proprodrugs of erythromycin B.

Full text of DOI:10.1021/jm049155y

3878
J. Med. Chem. 2005, 48, 3878-3884
Design, Synthesis, and Evaluation of Stable and Taste-Free Erythromycin
Proprodrugs
Pranab K. Bhadra,† Gareth A. Morris,‡ and Jill Barber*,†
School of Pharmacy and Pharmaceutical Sciences and Department of Chemistry, University of Manchester,
Manchester, M13 9PL, U.K.
Received October 21, 2004
Erythromycin A is normally formulated for children as its 2′-ethyl succinate, a taste-free
prodrug. Unfortunately, the prodrug hydrolyzes at a measurable rate in the medicine bottle,
leading to the vile-tasting erythromycin. We have prepared derivatives of erythromycin B as
putative paediatric prodrugs, taking advantage of the much improved acid stability of
erythromycin B relative to erythromycin A. Thus, erythromycin B enol ether ethyl succinate
is very poorly soluble in water, and its hydrolysis is undetectable in conditions resembling the
medicine bottle. In acid, however, it converts rapidly to erythromycin B 2′-ethyl succinate,
and this is in turn hydrolyzed to erythromycin B in neutral and basic conditions. Derivatives
of erythromycin B enol ether are therefore proposed as taste-free proprodrugs of erythromycin
B.
Scheme 1. Decomposition Pathway of Erythromycin A
Erythromycin A (1) is one of the most successful drugs
of all time and has been widely used in the treatment
of bacterial infections for 50 years.1 However, there are
problems associated with its use. These may be sum-
marized as (i) its vile taste, (ii) its lack of activity against
most Gram-negative organisms, (iii) its metabolism by
liver enzymes, competing with other drugs such as
theophylline and common hay fever remedies, leading
to overdose of these drugs, and (iv) the gut motilide
activity of both erythromycin A and its metabolites.
These problems are all greatly exacerbated by the fact
that erythromycin needs to be administered in large
doses (typically 1-2 g per day for 7-14 days for adults).
Large oral doses are required because of the chemical
and metabolic instability of the drug. In particular,
erythromycin is unstable below pH 6.9, equilibrating
with erythromycin A enol ether (2) and degrading to
erythromycin A anhydride (3), as shown in Scheme 1.
Enteric-coated tablets and capsules of erythromycin
are used clinically so that the drug is released in the
intestine rather than in the stomach.4 Much of the use
of erythromycin is, however, in pediatrics, requiring a
liquid formulation. This has been achieved by esterifi-
cation of the drug at the 2′-position to give compounds
that act as prodrugs and are administered as oral
suspensions. Esterification using a suitable acid chloride
is highly selective and facile. Further, the resulting
esters are generally taste-free and have enhanced acid
stability relative to erythromycin itself.5 Erythromycin
is normally administered to children as a suspension
of erythromycin A 2′-ethyl succinate (4).
in Acidic Aqueous Solution2,3
selectivity to be obtained. From the moment the eryth-
romycin 2′-ethyl succinate suspension is made, base-
catalyzed hydrolysis proceeds at a measurable rate, and
within 3 weeks there is considerable contamination by
erythromycin free base. Most children find the bitter
taste very nasty, and in some it causes vomiting.6 This
leads to poor compliance with drug regimes, which in
turn is associated with increased resistance to the drug.
This paper describes the preparation and evaluation
of some taste-free and stable prodrug and proprodrug
alternatives to erythromycin A 2′-ethyl succinate that
hydrolyze selectively in the body and not in the formu-
lation.
Unfortunately, children are less enthusiastic about
erythromycin 2′-ethyl succinate than chemists. The
prodrug is activated by base-catalyzed hydrolysis in the
blood stream, but the environment of the blood and the
medicine bottle are not sufficiently different for good
* To whom correspondence should be addressed. Phone: 44-161-
275-2369. Fax: 44-161-275-2396. E-mail: jill.barber@man.ac.uk.
† School of Pharmacy and Pharmaceutical Sciences.
‡ Department of Chemistry.
10.1021/jm049155y CCC: $30.25 © 2005 American Chemical Society
Published on Web 04/29/2005

Taste-Free Erythromycin Proprodrugs
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 11 3879
Table 1. Solubility Data of the Enol Ether Ethyl Succinates in
Comparison with the Parent Ethyl Succinates in Deuterated
Phosphate Buffer (50 mM)
Scheme 2. Degradation of Erythromycin B in Acidic
Aqueous Solution3
solubility at 25 °C (mM)
apparent
compd
pH 6.0
apparent pH 7.0
4 (EAES)
1.9
0.2
5.3
0.3
0.8
0.1
1.2
9 (EAEEES)
10 (EBES)
8 (EBEEES)
Broad lines indicated aggregation.
Solubility could not be measured.
The first lead compound selected for synthesis was
erythromycin B enol ether 2′-ethyl succinate (8) because
the 2′-ethyl succinate of erythromycin A is well-tolerated
in the body. There was no direct evidence that the
proprodrug approach would be successful in the eryth-
romycin A series; indeed, we had previously observed
slow conversion of erythromycin B enol ether to eryth-
romycin B at neutral pH, under which conditions
erythromycin A enol ether (2) is stable.7 Nevertheless,
it was logical to prepare and test erythromycin A enol
ether 2′-ethyl succinate (9) in parallel.
Synthesis. Erythromycin B enol ether 2′-ethyl suc-
cinate (8) was prepared by the two methods shown in
Scheme 3. Better yields (72% vs 44%) were obtained
when erythromycin B enol ether (6, Scheme 3A), rather
than erythromycin B 2′-ethyl succinate (10), was used
as an intermediate. The same methodology was then
also applied to erythromycin A, yielding erythromycin
A enol ether 2′-ethyl succinate (9) in 75% yield from
erythromycin A.
Solubility Measurements. The solubilities of the
two enol ether ethyl succinates at apparent pH 6.0 and
apparent pH 7.0 (at 37 °C) were measured by NMR
spectroscopy and are shown in Table 1. It can be seen
that, as predicted, both enol ether derivatives are much
less soluble than their parent ethyl succinates.
Activation of Erythromycin B Enol Ether 2′-
Ethyl Succinate by Acid. The conversion of the enol
ether ethyl succinates to the corresponding erythromy-
cin 2′-ethyl succinates in acid is an essential part of the
drug design strategy. It was tested by incubating 8 and
9 separately at apparent pH 2.0 and 37 °C in deuterated
buffer and monitoring by 1H NMR spectroscopy at
Results
Design of Proprodrugs of Erythromycins A and
B. During investigations of the kinetics of the acid-
catalyzed degradation of erythromycin B (Scheme 2), we
demonstrated an equilibrium between erythromycin B
(5) and erythromycin B enol ether (6) at apparent pH
2.5. (Erythromycin B degrades slowly in acid by loss of
the cladinose sugar to give 7.3) Erythromycin B enol
ether was found to have a solubility of 9.0 mM in
aqueous solution at apparent pH 7.4 (25 °C), compared
with 19.9 mM for erythromycin B itself. These observa-
tions suggested the possibility of preparing sparingly
soluble enol ether esters of erythromycin B that would
convert to the normal 14-membered ring esters on
treatment with acid, for example, in the stomach. Such
compounds might be expected to serve as taste-free
proprodrugs of erythromycin B.
Scheme 3. (A) Synthesis of Erythromycin Enol Ether 2′-Ethyl Succinates 8 and 9 via Erythromycin Enol Ethers and
(B) Synthesis of Erythromycin B Enol Ether 2′-ethyl Succinate 8 via Erythromycin B 2′-Ethyl Succinate

3880 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 11
Bhadra et al.
Figure 1. Reactions of erythromycin B enol ether 2′-ethyl succinate, 8, in deuterated Britton-Robinson buffer at apparent pH
2.0 and 37 °C. 8 is activated within 10-15 min to give erythromycin 2′-ethyl succinate (10), which is then degraded to 5-O-
desoaminylerythronolide B ethyl succinate (11). Signals A, B, and C represent H-13 in 8, 10, and 11, respectively, and D represents
H3-19.
regular intervals. A typical time course for the treat-
ment of 8 with acid is shown in Figure 1 and demon-
strates that conversion of 8 to erythromycin B 2′-ethyl
succinate (10) is very rapid and essentially complete
within 10-15 min. Unfortunately, a second reaction,
resulting in the degradation of 10, can also be detected.
In this experiment, 10 showed a half-life of 66 min. This
degradation is likely to reduce the bioavailability of the
drug significantly (a typical residence time in the
stomach is 30-120 min). Since the erythromycins are
nontoxic, this is acceptable, though not ideal.
The degradation product was identified as 5-O-des-
osaminyl erythronolide B 2′-ethyl succinate (11) by
diffusion-ordered spectroscopy (DOSY) NMR8-10 and
electrospray ionization mass spectrometry. In the DOSY
NMR spectrum (Figure 2), NMR signals are separated
according to chemical shift in one dimension and ac-
cording to their self-diffusion coefficients in the other.
It shows that the signals for the cladinose ring exhibit
much faster diffusion than those due to the macrolide
and the desosamine sugar, demonstrating that cladinose
had been hydrolyzed from the macrolide. Electrospray
ionization mass spectrometry showed signals at m/z 831,
851 (70%), and 693 (100%) corresponding to 8, 10, and
11, respectively (each containing two or three deuter-
ated hydroxyl groups because exchange of OD to OH is
incomplete in the mass spectrometer), in agreement
with the results obtained from the DOSY experiment.
The effect of acid on erythromycin B enol ether 2′-ethyl
succinate (8) is therefore analogous to the effect of acid
on erythromycin B enol ether (6), as is illustrated in
Scheme 2.
Treatment of Erythromycin A Enol Ether 2′-
Ethyl Succinate with Acid. The NMR time course at
apparent pH 2 was repeated using erythromycin A enol
ether 2′-ethyl succinate (9). Both the initial conversion
and the subsequent loss of cladinose were slower. As
can be seen in Figure 3, 9 had a half-life of 41 min under
these conditions. Loss of cladinose was not seen at
apparent pH 2 but was evident at apparent pH 1; the
half-time for loss of cladinose at this pH was 35 min.
Unfortunately, however, Figure 3 shows that the
initial hydrolysis product was not erythromycin A 2′-
ethyl succinate (4). The electrospray ionization mass
spectrum gave an M + H+ peak at m/z 848, suggesting
that 8 is converted to anhydroerythromycin A 2′-ethyl
succinate (12) under acidic conditions (see Scheme 1).
(12 had incorporated on average four deuteriums from
the deuterated solvent, and these were not washed out
in the mass spectrometer.) When 4 was incubated at
1
37 °C and pH 2, the H NMR spectrum showed that it
too was converted to 12. Thus, although esterification
at the 2′ position is reported to slow anhydride forma-
tion, it remains fast compared with the conversion of 8
to 4. This result means that the proprodrug approach
is unlikely to be successful for the delivery of erythro-
mycin A.
Degradation of Erythromycin B Enol Ether 2′-
Ethyl Succinate in Acidic H2O. It can be inferred
from the preceding data that under acidic conditions the
cladinose sugar is lost from erythromycin B 2′-ethyl

Taste-Free Erythromycin Proprodrugs
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 11 3881
Figure 2. DOSY NMR spectrum at 500 MHz of erythromycin B enol ether 2′-ethyl succinate at 25 °C after 2 h at 37 °C. Signals
due to the cladinose ring clearly show faster diffusion than those due to the macrolide and desosamine rings.
Figure 3. Reactions of erythromycin A enol ether 2′-ethyl succinate, 9, in deuterated Britton-Robinson buffer at apparent pH
2.0 and 37 °C. Signals A and B represent H3-15 in 9 and anhydroerythromycin A 2′-ethyl succinate (12), respectively. Signal C
is due to H3-19 of 9 and is highly characteristic of erythromycin enol ethers.
succinate (10) rather than from the enol ether ethyl
succinate (8). Further, loss of cladinose from erythro-
mycin B esters in acid is much slower than anhydride
formation by erythromycin A esters. This was now
tion mass spectrometry. The half-life of 10 under these
conditions was 71 ( 5 min; that of 10 generated in situ
from 8 was 73 ( 5 min. Thus, compounds 8 and 10 lose
cladinose at essentially the same rate, supporting the
proposal that 8 is rapidly converted to 10 under acidic
conditions and that this conversion is followed by slow
loss of the cladinose sugar to give 11. An apparent pH
2 in D2O-based buffer corresponds to an actual pD of
about 2.4, so a shorter half-life in H2O at pH 2 was
expected, whereas similar half-lives (71 ( 5 vs 66 ( 5
1
confirmed by H NMR spectroscopy in 95% protiated
buffer (to avoid solvent and any other isotope effects).
Compounds 8 and 10 were incubated separately in
Britton Robinson buffer at pH 2 containing 5% D2O. The
WATERGATE pulse sequence was used to obtain time
courses, which were confirmed by electrospray ioniza-

3882 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 11
Bhadra et al.
Table 3. Hydrolysis of Suspensions (125 mg in 5 mL) of
Erythromycin B 2′-Ethyl Succinate (10), Erythromycin B Enol
Ether 2′-Ethyl Succinate (8), Erythromycin A 2′-Ethyl
Succinate (4), and Erythromycin A Enol Ether 2′-Ethyl
Succinate (9) under Conditions Mimicking a Medicinal
Suspensiona
Table 2. Half-Lives for the Acid-Catalyzed Activation of
Erythromycin B Enol Ether 2′-Ethyl Succinate (8) to
Erythromycin B 2′-Ethyl Succinate (10) and Its Subsequent
Degradation to 5-O-Desosaminylerythronolide B 2′-Ethyl
Succinate (11) and the Degradation of Erythromycin A Enol
Ether 2′-Ethyl Succinate (9) to Anhydroerythromycin A 2′-Ethyl
Succinate (12) under the Same Conditions
extent of hydrolysis after 21 days at 4 °C (%)
half-life (min)
compd
pH 6.0
pH 8.0
reaction
pH 1.0
pH 2.0
pH 3.0
pH 4.0
pH 5.0
10
8
68.2
0.0
26.2
20.4
23.1
0.0
30.1
10.1
8 f 10
10 f 11
9 f 12
a
<7
10
10
66
41
175
c
315
301
c
495
b
b
b
4
9
a Reaction completed before the first spectrum could be ac-
quired. b No change in the spectrum observed after 2 h. c Forma-
tion of 11 in the spectrum was not evident.
a 4 and 10 are both hydrolyzed to the corresponding erythro-
mycin. The main product of the hydrolysis of 9 is anhydroeryth-
romycin A (3).
min) were observed. Solvent isotope effects and/or small
changes in buffer composition or temperature are prob-
ably at work here. Great care was always taken to use
the same buffer stock and to run experiments on the
same day in order to minimize systematic errors when-
ever direct comparisons of rates were being made, but
buffers based on D2O and those based on H2O are
necessarily derived from different stocks, so experiments
run in the two solvents are not directly comparable.
Effect of pH on Erythromycin B Enol Ether 2′-
Ethyl Succinate (8) and Erythromycin A Enol
Ether 2′-Ethyl Succinate (9). Erythromycin B enol
ether 2′-ethyl succinate (8) was now incubated at 37 °C
at various different pH values in deuterated buffer.
Table 2 shows that, unfortunately, conversion to the
parent ethyl succinate esters is very slow above appar-
ent pH 2.0 and that degradation by loss of cladinose is
fast at apparent pH 1.0. The acid-catalyzed loss of
cladinose, however, proceeds slowly compared with
anhydride formation by erythromycin A esters. It is
therefore reasonable to suggest that the main limitation
placed on the use of 8 as a proprodrug is that the pH of
the stomach must be maintained below 3 for 30 min
following each dose.
romycin B enol ether 2′-ethyl succinate (8) is very stable
at both pH 6 and pH 8, suggesting that complete taste-
masking of erythromycin B could be achieved by using
this derivative. Erythromycin A enol ether 2′-ethyl
succinate hydrolyzes slowly under these conditions;
anhydroerythromycin A (3) is the major product.
Discussion
The success of erythromycin A as an antibacterial
was, for the first 30 years of its production, unremark-
able. The drug acts by inhibiting bacterial protein
synthesis, a process necessary for cellular survival, not
just cell division. It has excellent activity against a wide
range of Gram-positive organisms and is generally well-
tolerated. The continued success of this drug over the
past 20 years does, however, require explanation. Three
distinct modes of resistance11-13 are well-known in the
clinic, and the second-generation erythromycin A de-
rivatives clarithromycin (13) and azithromycin (14) are
in many respects superior to the parent drug. Despite
this, erythromycin remains hugely popular in part
because it is cheap and (usually) effective and in part
because its spectrum of action is to some extent comple-
mentary to those of clarithromycin and azithromycin.
Erythromycin A enol ether 2′-ethyl succinate (9) was
converted to anhydroerythromycin A 2′-ethyl succinate
(12) under all the conditions studied. The obligatory
intermediate, erythromycin A 2′-ethyl succinate (4), was
undetectable, showing that the hydration of the enol
ether ester was rate-determining.
Stability of Erythromycin A and B Esters and
Enol Ether Esters under Storage Conditions. To
assess the stability of the enol ether derivatives in
storage conditions, the two erythromycin esters and the
two enol ether esters 4, 10, 8, and 9 were stored
individually in a refrigerator (4 °C for 21 days) as
suspensions at a concentration of 125 mg mL-1 in buffer
pH 6.0 and pH 8.0. These conditions mimic those of the
medicine bottle. The suspensions were shaken on al-
ternate days to avoid localized concentration zones. At
the end of 21 days, the suspensions were extracted with
organic solvent and the extent of hydrolysis to eryth-
romycin base was determined by NMR spectroscopy.
The observations are summarized in Table 3.
The results suggest that because 10 (erythromycin B
2′-ethyl succinate) is more soluble than 4 (erythromycin
A 2′-ethyl succinate) at pH 6, it hydrolyzes more rapidly
to the vile-tasting erythromycin. Successful formulation
of 10 would require the use of mildly basic conditions;
the extent of hydrolysis is acceptable at pH 8. Eryth-
While the pharmacokinetic, pharmacodynamic, and
metabolic differences among these three drugs are
complex, it is fair to point to the large differences in
solubility and in acid stability as being key to the
advantages and disadvantages of each. The second-
generation compounds are both relatively acid-stable.
Clarithromycin is modified at C6 and azithromycin at
C9 so that neither is able to undergo facile acid-
catalyzed degradation to anhydro compounds analogous
to 3. In common with erythromycin B (which lacks a
C12 hydroxyl group), they are deactivated slowly in acid
by loss of the cladinose sugar.3 The relatively poor
aqueous solubilities of clarithromycin and azithromycin
are almost certainly important in their activity against

Taste-Free Erythromycin Proprodrugs
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 11 3883
solvent was removed in vacuo. The crystalline product, 8, was
recrystallized from methanol. Yield 44%, mp 132-134 °C.
Synthesis of Erythromycin A Enol Ether 2′-Ethyl
Succinate (9). The synthesis procedure (first method) outlined
above was used to prepare erythromycin A enol ether 2′-ethyl
succinate (9) from erythromycin A enol ether. Yield 75%, mp
114-116 °C.
certain Gram-negative bacteria but may also be respon-
sible for the absence of 2′-esters in the clinic. 2′-Esters
of erythromycin A, especially erythromycin A 2′-ethyl
succinate, have sufficiently favorable pharmacokinetics
to be very successful in pediatric medicine.
We have pointed to the advantages of erythromycin
B as a natural product sharing the relative acid-stability
of clarithromycin and azithromycin and the therapeutic
profile of erythromycin A. We have now shown that it
may be particularly advantageous in pediatric medicine.
Erythromycin B 2′-ethyl succinate (10) compares favor-
ably with the market leader, erythromycin A 2′-ethyl
succinate (4). Still more interesting, however, is the
prospect of using erythromycin B enol ether esters (for
example 8) as taste-free proprodrugs.
Solubility Studies. A Varian Unity 500 spectrometer was
used to determine the solubilities of 4 and 8-10 at apparent
pH 6.0 and 7.0. A saturated solution of each compound in
deuterated phosphate buffer (apparent pH 6.0, 7.0) was
prepared by adding successive10 µL aliquots of a solution of
the compound in acetone-d6 (10 mg/100 µL). The solution was
filtered, and a one-dimensional 1H spectrum was acquired. The
peaks were referenced to TSP (1 mM), and the solubility of
the compound was determined by measuring the relative
1
integrals in the one-dimensional H spectrum.
Acid-Catalyzed Activation Studies. Compounds 8 and
9 were dissolved in deuterated Britton-Robinson buffer at
apparent pH 2.0 or in deuterated phosphate buffer at apparent
pH 2.01 to give 8 at 4 mM and 9 at 2 mM (concentrations
higher than 2 mM resulted in precipitation of 9 during the
experiment). The degradation experiments were each per-
Experimental Section
General Procedures. All chemicals were purchased from
Sigma Aldrich unless otherwise stated. 1H NMR spectra were
acquired using a Varian Unity 500 spectrometer operating at
500 MHz, a Varian Inova 400 spectrometer operating at 400
MHz, or a Bruker AVANCE 300 specrometer operating at 300
MHz. Electrospray-ionization mass spectra (ESI-MS) were
acquired on a Micromass Platform mass spectrometer, and the
data were analyzed using the program PLATFORM with a
Masslynx data system. Ten microliters of the sample was
injected using a Hewlett-Packard autosampler, and the ma-
chine was operated at a cone value of 30 eV at 80 °C. For
identification purposes, all samples (0.2 mg) were prepared
in acetonitrile (1 mL). For activation studies, deuterated
Britton-Robinson buffer was used unless otherwise stated.
Water was used as the solvent for running the samples.
Synthesis of Erythromycin B Enol Ether (6). Erythro-
mycin B enol ether (6) was prepared by an adaptation of a
literature procedure.14 A solution of erythromycin B (5 g, 3.58
mmol) in glacial acetic acid (25 mL) (AnalaR, BDH, U.K.) was
stirred at room temperature for 4 h. The progress of the
reaction was monitored by TLC (EtOAc-CH3OH-25% NH3,
85:10:5). The excess of glacial acetic acid was removed under
vacuum, and the mixture was neutralized using saturated
NaHCO3 (GPR grade, BDH, U.K.) solution. The product was
extracted using 3 × 100 mL dichloromethane (GPR grade,
BDH, U.K.). The organic layers were combined and washed
with saturated NaHCO3 solution and with water and finally
dried over anhydrous Na2SO4. Then the solvent was removed
in vacuo. The colorless crystalline 6 was recrystallized from
ethanol. Yield 85%, mp 128-130 °C (lit. 126-128 °C).11
Synthesis of Erythromycin B Enol Ether 2′-Ethyl
Succinate (8). First Method. 6 (1 g, 0.69 mmol) was
dissolved in acetone (75 mL, previously dried over anhydrous
Na2SO4), and anhydrous sodium bicarbonate (0.5 g) was added
with stirring. Ethyl succinyl chloride (0.8 mL, 0.91 mmol) was
added. The reaction mixture was stirred overnight at ambient
temperature, the volume of the mixture was then reduced to
2-3 mL in vacuo, and the ester was precipitated in 150 mL of
50 mM phosphate buffer (pH 6.5). The precipitate was dis-
solved in ethyl acetate (25 mL) and was washed successively
with 2 × 25 mL of water and 25 mL of brine. The organic layer
was separated and dried over anhydrous Na2SO4, and the
solvent was removed in vacuo. The colorless crystals of 8 were
recrystallized from methanol. Yield 72.3%, mp132-134 °C.
Second Method. A solution of erythromycin B 2′-ethyl
succinate (10, 2 g, 1.68 mmol) in glacial acetic acid (10 mL,
0.62 mmol) (AnalaR, BDH, U.K.) was allowed to stand at room
temperature for 4 h. The excess glacial acetic acid was removed
by treating the reaction mixture with saturated NaHCO3
solution followed by extraction with dichloromethane (3 × 100
mL). The organic layers were combined and washed with 100
mL of saturated NaHCO3 solution to remove any remaining
traces of acetic acid. This was followed by washing the organic
layer with 2 × 100 mL of water and 1 × 100 mL of brine. The
organic layer was dried over anhydrous Na2SO4, and the
1
formed by acquiring an array of 1D H spectra. Each experi-
ment consisted of 17 spectra of 64 increments and each
spectrum took 7 min to acquire, making a total experiment
time of nearly 2 h. All degradation studies were carried out
at 37 °C. TSP, at a concentration of 1 mM, was used as a
reference standard. The spectra were recorded with a spinning
sample, using the PRESAT pulse sequence with a preaquisi-
tion saturation delay of 2 s, a 90° pulse of width 12 µs, and a
spectral width of 6000 Hz. The spectra were processed with
reference deconvolution using the FIDDLE (free induction
decay deconvolution for lineshape enhancement) algorithm15
for line shape correction with a Gaussian time constant of 0.15
s, using the TSP signal as the reference followed by cubic
spline baseline correction. The FIDDLE algorithm compares
the experimental time-domain signal of a reference with that
predicted by theory, multiplying the raw experimental data
by the complex ratio of the two signals to produce a corrected
free induction decay (FID).
Compounds 8 and 10 were dissolved in protiated Britton-
Robinson buffer containing 10% D2O at 37 °C and pH 2 to give
a concentration of 4 mM. Time courses were measured on a
Varian Inova 400 MHz spectrometer by acquiring an array of
1D 1H spectra using a WATERGATE pulse sequence. The
spectra were recorded with a selective 180° pulse of 4.15 ms,
a 90° pulse of width 5.2 µs, a gradient pulse of width 2µs, and
a spectral width of 5000 Hz. Each experiment involved the
acquisition of an array of 24 spectra, each of 64 transients
occupying 5 min, making a total experiment time of 2 h. The
spectra were processed as above.
DOSY Spectroscopy of the Degradation Products of
Erythromycin B Enol Ether 2′-Ethyl Succinate (8).
Erythromycin B enol ether 2′-ethyl succinate (8) was dissolved
in deuterated Britton-Robinson buffer at an apparent pH 2.0
to give a concentration of 4 mM and was degraded over 2 h at
37 °C. The degraded samples were cooled to ambient temper-
ature (25 °C), and a high-resolution DOSY (diffussion-ordered
spectroscopy) spectrum was acquired using the ONESHOT
DOSY pulse sequence.9 Twelve spectra were acquired, with
gradient pulses of 2.9 ms ranging in strength from 1 to 15 G
cm-1 and a diffusion delay of 0.2 s. The FIDDLE algorithm15
was used to correct line shapes using TSP as a reference
standard. DOSY spectra were constructed after baseline
correction by taking the first echo spectrum and distributing
the intensities of the individual signals in the second dimen-
sion according to their respective diffusion coefficients, using
Gaussian line shapes with widths determined by the standard
errors obtained from the fitting process.15
Comparative Stability Study. Suspensions (5 mL at 25
mg mL-1) of each of the drugs 4 and 8-10 were prepared in
50 mM protiated phosphate buffer at pH 6.0 and 8.0, and the
samples were stored in a refrigerator at 4 °C for 21 days. The

3884 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 11
Bhadra et al.
in aqueous solution and human plasma. Int. J. Pharm. 1989,
56, 159-168.
samples were extracted with chloroform, 2 × 2.5 mL, and the
organic layers were combined and dried under a gentle stream
of nitrogen at 40 °C. 1D 1H NMR spectra in deuterated
chloroform were acquired, and the extent of hydrolysis to free
erythromycin in each sample was determined by comparisons
of integrals. These results are summarized in Table 3. All
spectra were referenced to TMS as an internal standard.
(6) Washington, J. A.; Wilson, W. R. Erythromycin: a microbial and
clinical perspective after 30 years of clinical use (first of two
parts). Mayo Clin. Proc. 1985, 60, 189-203.
(7) Tyson, P.; Barber J. Unpublished data.
(8) Morris, K. F.; Johnson, C. S. Diffusion-ordered-two-dimensional
nuclear magnetic resonance spectroscopy. J. Am. Chem. Soc.
1992, 114, 3139-3141.
Supporting Information Available: 1H NMR and ESI-
MS data, TLC Rf values, and results from elemental analysis
of erythromycin B enol ether, erythromycin B enol ether 2′-
ethyl succinate, and erythromycin A enol ether 2′-ethyl suc-
cinate. This material is available free of charge via the Internet
at http://pubs.acs.org.
(9) Barjat, H.; Morris, G. A.; Smart, S.; Swanson, A. G.; Williams,
S. C. R. High-resolution-diffusion-ordered 2D spectroscopy (HR
DOSY)sa new tool for analysis of complex mixtures. J. Magn.
Reson., Ser. B. 1995, 108, 170-172.
(10) Pelta, M. D.; Morris, G. A.; Stchedroff, M. D.; Hammond, S. J.
A
one-shot sequence for high resolution diffusion ordered
spectroscopy. Magn. Reson. Chem. 2002, 40, S147-S152.
(11) Taubman, S. B.; Young, F. E.; Corcoran, J. W. Antibiotic
glycosides IV: Studies on the mechanism of erythromycin
resistance in Bacillus subtilis. Biochemistry 1963, 50, 955-962.
(12) Liljas, A.; Al-Karadaghi, S. Structural aspects of protein syn-
thesis. Nat. Struct. Biol. 1997, 4, 767-771.
(13) Goldman, R. C.; Capobianco, J. O. Role of an energy-dependent
efflux pump in plasmid pNE24 mediated resistance to 14- and
15-membered macrolides in Staphylococcus epiderdimis. Anti-
microb. Agents Chemother. 1990, 34, 1973-1980.
(14) Kurath, P.; Jones, P. H.; Egan, R. S.; Perun, T. J. Acid
degradation of erythromycin A and B. Experientia 1971, 27, 362.
(15) Morris, G. A.; Barjat, H.; Horne, T. J., Reference deconvolution
methods. Prog. Nucl. Magn. Reson. Spectrosc. 1997, 31, 197-257.
References
(1) McGuire, J. M.; Bunch, R. L.; Anderson, R. C.; Boaz, H. E.
Ilotycin, a new antibiotic. Antibiot. Chemother. 1952, 6, 281-
283.
(2) Cachet, T. H.; Van den Mooter, G.; Hauchecorne, R.; Vinckier,
C.; Hoogmartens, J. Decomposition kinetics of erythromycin A
in acidic aqueous solutions. Int. J. Pharm. 1989, 55, 59-65.
(3) Mordi, M. N.; Pelta, M. D.; Boote, V.; Morris, G. A.; Barber, J.
Acid-catalysed degradation of clarithromycin and erythromycin
B: a comparative study using NMR spectroscopy. J. Med. Chem.
2000, 43, 467-474.
(4) Fraser, D. G. Selection of an oral erythromycin product. Am. J.
Hosp. Pharm. 1980, 37, 1199-1205.
(5) Steffansen, B.; Bundgaard, H. Erythromycin prodrugs: kinetics
of hydrolysis of erythromycin and various erythromycin 2′-esters
JM049155Y